Download to read offline
![Preface
15
These installation instructions assume
You have a CD-ROM drive on your computer and the Official Red Hat Linux or OpenNA Linux
CD-ROM. Installations were tested on the Official Red Hat Linux version 7.3 and OpenNA Linux.
You should familiarize yourself with the hardware on which the operating system will be installed.
After examining the hardware, the rest of this document guides you, step-by-step, through the
installation process.
Obtaining the example configuration files
In a true server environment and especially when Graphical User Interface is not installed, we will
often use text files, scripts, shell, etc. Throughout this book we will see shell commands, script
files, configuration files and many other actions to execute on the terminal of the server. You can
enter them manually or use the compressed archive file that I made which contains all
configuration examples and paste them directly to your terminal. This seems to be useful in many
cases to save time.
The example configuration files in this book are available electronically via HTTP from this URL:
http://www.openna.com/products/books/securing-optimizing-linux/3rdedition/index.htm
• In either case, extract the files into your Linux server from the archive by typing:
[root@deep /]# cd /var/tmp
[root@deep tmp]# tar xzpf floppy-3.0.tgz
If you cannot get the examples from the Internet, please contact the author at this email address:
gmourani@openna.com
Problem with Securing & Optimizing Linux
When you encounter a problem in "Securing & Optimizing Linux" we want to hear about it. Your
reports are an important part in making the book more reliable, because even with the utmost
care we cannot guarantee that every part of the book will work on every platform under every
circumstance.
We cannot promise to fix every error right away. If the problem is obvious, critical, or affects a lot
of users, chances are that someone will look into it. It could also happen that we tell you to
update to a newer version to see if the problem persists there. Or we might decide that the
problem cannot be fixed until some major rewriting has been done. If you need help immediately,
consider obtaining a commercial support contract or try our Q&A archive from the mailing list for
an answer.
Below are some important links:
OpenNA web site: http://www.openna.com/
Mailing list: http://www.openna.com/support/mailing/
Support: http://www.openna.com/support/
RPM Download: http://www.openna.com/download/
Acknowledgments
I would like to thank all the OpenNA staff for their hard works and patience. A special gratitude
and many thanks to Colin Henry who made tremendous efforts to make this book grammatically
and orthographically sound in a professional manner. Adrian Pascalau for its time and help in the
open source community and all Linux users around the world who have participated by providing
good comments, ideas, recommendations and suggestions.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-15-320.jpg)
![Introduction 0
CHAPTER 1
23
Libraries
Programs can be linked not only to object files (*.o) but also to libraries that are collections of
object files. There are two forms of linking to libraries: static, where the code goes in the
executable file, and dynamic, where the code is collected when the program starts to run.
Patches
It was common for executable files to be given corrections without recompiling them. Now this
practice has died out; in modern days, people change a small portion of the source code, putting
a change into a file called a “patch”. Where different versions of a program are required, small
changes to code can be released this way, saving the trouble of having two large distributions.
Errors in Compilation and Linking
Errors in compilation and linking are often due to typos, omissions, or misuse of the language.
You have to check that the right “includes file” is used for the functions you are calling.
Unreferenced symbols are the sign of an incomplete link step. Also check if the necessary
development libraries (GLIBC) or tools (GCC, DEV86, MAKE, etc) are installed on your system.
Debugging
Debugging is a large topic. It usually helps to have statements in the code that inform you of what
is happening. To avoid drowning in output you might sometimes get them to print out only the first
3 passes in a loop. Checking that variables have passed correctly between modules often helps.
Get familiar with your debugging tools.
Build & install software on your system
You will see in this book that we use many different compile commands to build and install
programs on the server. These commands are UNIX compatible and are used on all variants of
*NIX machines to compile and install software.
The procedures to compile and install software tarballs on your server are as follows:
1. First of all, you must download the tarball from your trusted software archive site. Usually
from the main site of the software you hope to install.
2. After downloading the tarball, change to the /var/tmp directory (note that other paths
are possible, at personal discretion) and untar the archive by typing the commands (as
root) as in the following example:
[root@deep /]# tar xzpf foo.tar.gz
The above command will extract all files from the example foo.tar.gz compressed archive and
will create a new directory with the name of the software from the path where you executed the
command.
The “x” option tells tar to extract all files from the archive.
The “z” option tells tar that the archive is compressed with gzip utility.
The “p” option maintains the original permissions the files had when the archive was created.
The “f” option tells tar that the very next argument is the file name.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-23-320.jpg)
![32
Since this, is a relatively rare occurrence, you will save time if you try to use the standard diskette
images first, and then review the Errata only if you experience any problems completing the
installation. Below, we will show you two methods to create the installation Boot Disk, the first
method is to use an existing Microsoft Windows computer and the second using an existing Linux
computer.
Making a Diskette under MS-DOS:
Before you make the boot disk, insert the Official Red Hat Linux CD-ROM Disk 1 in your
computer that runs the Windows operating system. When the program asks for the filename,
enter boot.img for the boot disk. To make the floppies under MS-DOS, you need to use these
commands (assuming your CD-ROM is drive D: and contain the Official Red Hat Linux CD-ROM).
• Open the Command Prompt under Windows: Start | Programs | Command Prompt
C:> d:
D:> cd dosutils
D:dosutils> rawrite
Enter disk image source file name: ..imagesboot.img
Enter target diskette drive: a:
Please insert a formatted diskette into drive A: and press -ENTER- :
D:dosutils>exit
The rawrite.exe program asks for the filename of the disk image: Enter boot.img and insert
a blank floppy into drive A. It will then ask for a disk to write to: Enter a:, and when complete,
label the disk “Red Hat boot disk”, for example.
Making a Diskette under a Linux-Like OS:
To make a diskette under Linux or any other variant of Linux-Like operating system, you must
have permission to write to the device representing the floppy drive (known as /dev/fd0H1440
under Linux).
This permission is granted when you log in to the system as the super-user “root”. Once you
have logged as “root”, insert a blank formatted diskette into the diskette drive of your computer
without issuing a mount command on it. Now it’s time to mount the Red Hat Linux CD-ROM on
Linux and change to the directory containing the desired image file to create the boot disk.
• Insert a blank formatted diskette into the diskette drive
Insert the Red Hat Linux CD Part 1 into the CD-ROM drive
[root@deep /]# mount /dev/cdrom /mnt/cdrom
[root@deep /]# cd /mnt/cdrom/images/
[root@deep images]# dd if=boot.img of=/dev/fd0H1440 bs=1440k
1+0 records in
1+0 records out
[root@deep images]# cd /
[root@deep /]# umount /mnt/cdrom
Don’t forget to label the diskette “Red Hat boot disk”, for example.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-32-320.jpg)
![Installation Issues 0
CHAPTER 2
39
Disk Partition (Manual Partitioning)
Now that we know exactly what partitions we need to create for our new Linux server, it is time to
choose the partitioning software we will use to make these partitions. With Red Hat Linux two
programs exist to assist you with this step:
• Manually partition with Disk druid
• Manually partition with fdisk [experts only]
Disk Druid is new software used by default in Red Hat Linux to partition your disk drive, this
program is easy to use, and allows you to use a graphical interface to create your partitions
tables.
fdisk was the first partitioning program available on Linux. It is more powerful then Disk
Druid and allows you to create your partition table in exactly the way you want it (if you want to
put your swap partition near the beginning of your drive, then you will need to use fdisk).
Unfortunately, it is also a little more complicated than Disk Druid and many Linux users prefer
to use Disk Druid for this reason.
Personally, I prefer to create the partitions with the fdisk program and I recommend you use
and be familiar with it, because if, in the future you want to add or change some file systems you
will need to use fdisk.
Partitioning with Disk Druid
This section applies only if you chose to use Disk Druid to partition your system. Disk Druid
is a program that partitions your hard drive for you. Choose “New” to add a new partition, “Edit”
to edit a partition, “Delete” to delete a partition and “Reset” to reset the partitions to the original
state. When you add a new partition, a new window appears on your screen and gives you
parameters to choose.
Mount Point: for where you want to mount your new partition in the filesystem.
Filesystem Type: Ext3 for Linux filesystem and Swap for Linux Swap Partition
Size (MB): for the size of your new partition in megabytes.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-39-320.jpg)
![Installation Issues 0
CHAPTER 2
53
Boot Disk Creation
From this section of the installation, we have the possibility to create a boot disk for our newly
installed operating system. If you do not want to create a boot disk, you should check the “Skip
boot disk creation” checkbox before you click Next. Creating a boot disk must be made if you
decide to not install GRUB or LILO on the MBR (the Master Boot Record) or if you are not installing
GRUB or LILO at all.
How to use RPM Commands
This section contains an overview of using RPM for installing, uninstalling, upgrading, querying,
listing, and checking RPM packages on your Linux system. You must be familiar with these RPM
commands now because we’ll use them often in this book and especially later in this chapter for
software that must be uninstalled after installation of the server.
Install a RPM package:
Note that RPM packages have a file of names like foo-1.0-2.i386.rpm, which include the
package name (foo), version (1.0), release (2), and architecture (i386).
• To install a RPM package, use the command:
[root@deep /]# rpm -ivh foo-1.0-2.i386.rpm
foo ##################################################
Uninstall a RPM package:
Notice that we used the package name “foo”, not the name of the original package file “foo-
1.0-2.i386.rpm”.
• To uninstall a RPM package, use the command:
[root@deep /]# rpm -e foo](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-53-320.jpg)
![54
Upgrade a RPM package:
With this command, RPM automatically uninstalls the old version of foo package and installs the
new one. Always use “rpm -Uvh” command to install packages, since it works fine even when
there are no previous versions of the package installed. This is the recommended method of
installing package on the system.
• To upgrade a RPM package, use the command:
[root@deep /]# rpm -Uvh foo-1.0-2.i386.rpm
foo ##################################################
Force the installation of a RPM package:
With this command, RPM will force the installation of the specified package even if some conflict
or other kind of problem exists. This command should be used with care and only if you know
what you do. In most case, RPM can correctly guest problem and refuse to install. To bypass RPM
warning, you can use the RPM command below.
• To force the installation of a RPM package, use the command:
[root@deep /]# rpm -Uvh --force foo-1.0-2.i386.rpm
foo ##################################################
Avoid RPM package dependency:
With this command, RPM will not take care of package dependency and will install the RPM
software on your system. Package dependency is an important concept in the RPM world.
Dependency is when some other packages depend of the RPM package you are trying to install.
By default, RPM check if all other RPM packages required for the RPM you try to install are present
before installing the RPM. If some required packages are not present, RPM will inform you. This is
made to avoid problem and be sure that the software you want to install will perfectly work. In
some special case, we don’t need to take care of dependency and can use the option below to
inform it to skip the dependency check when installing the software.
• To avoid RPM package dependency, use the command:
[root@deep /]# rpm -Uvh --nodeps foo-1.0-2.i386.rpm
foo ##################################################
Query a RPM package:
This command will print the package name, version, and release number of installed package
foo. Use this command to verify that a package is or is not installed on your system.
• To query a RPM package, use the command:
[root@deep /]# rpm -q foo
foo-2.3-8](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-54-320.jpg)
![Installation Issues 0
CHAPTER 2
55
Display RPM package information:
This command displays package information; includes name, version, and description of the
installed program. Use this command to get information about the installed package.
• To display RPM package information, use the command:
[root@deep /]# rpm -qi foo
Name : foo Relocations: none
Version : 2.3 Vendor: OpenNA.com, Inc.
Release : 8 Build Date: Thu 24 Aug 2000 11:16:53 AM EDT
Install date: Mon 12 Feb 2001 01:17:24 AM EST Build Host: openna.com
Group : Applications/Archiving Source RPM: foo-2.3-8.src.rpm
Size : 271467 License: distributable
Packager : OpenNA.com, Inc. <http://www.openna.com/>
Summary : Here will appears summary of the package.
Description : Here will appears the description of the package.
Display RPM package information before installing the program:
This command displays package information; includes name, version, and description of the
program without the need to install the program first. Use this command to get information about
a package before you install it on your system.
• To display package information before installing the program, use the command:
[root@deep /]# rpm -qpi foo-2.3-8.i386.rpm
Name : foo Relocations: none
Version : 2.3 Vendor: OpenNA.com, Inc.
Release : 8 Build Date: Thu 24 Aug 2000 11:16:53 AM EDT
Install date: Mon 12 Feb 2001 01:17:24 AM EST Build Host: openna.com
Group : Applications/Archiving Source RPM: foo-2.3-8.src.rpm
Size : 271467 License: distributable
Packager : OpenNA.com, Inc. <http://www.openna.com/>
Summary : Here will appears summary of the package.
Description : Here will appears the description of the package.
List files in a installed RPM package:
This command will list all files in a installed RPM package. It works only when the package is
already installed on your system.
• To list files in a installed RPM package, use the command:
[root@deep /]# rpm -ql foo
/usr/bin/foo
/usr/bin/foo1
/usr/sbin/foo2
List files in RPM package that is not already installed:
This command will list all files in a RPM package that is not already installed on your system. It is
useful when you want to know which components are included in the package before installing it.
• To list files in RPM package that is not already installed, use the command:
[root@deep /]# rpm -qpl foo
/usr/lib/foo
/usr/bin/foo1
/usr/sbin/foo2](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-55-320.jpg)
![56
Know which files is part of which RPM package:
This command will show you from which RPM package the file comes from. It works only when the
package is already installed on your system and it is very useful when you see some files into
Linux that you do not know about it and want to get more information about its RPM provenance.
• To know which files is part of which RPM package, use the command:
[root@deep /]# rpm -qf /etc/passwd
setup-2.3.4-1
Check RPM signature package:
This command checks the PGP signature of specified package to ensure its integrity and origin.
Always use this command first before installing new RPM package on your system. GnuPG or PGP
software must be already installed on your system before you can use this command. See the
chapter related to GnuPG installation and configuration for more information.
• To check a RPM signature package, use the command:
[root@deep /]# rpm --checksig foo
Examine the md5sum of RPM package:
The RPM md5sum is useful to verify that a package has not been corrupted or tampered with. You
can use it to be sure that the download of your new RPM package was not corrupted during
network transfer.
• To examine only the md5sum of the package, use the command:
[root@deep /]# rpm --checksig --nogpg foo
Starting and stopping daemon services
The init program of Linux (also known as process control initialization) is in charge of starting
all the normal and authorized processes that need to run at boot time on your system. These may
include the APACHE daemons, NETWORK daemons, and anything else that must be running
when your machine boots.
Each of these processes has a script file under the /etc/init.d directory written to accept an
argument, which can be start, stop, restart, etc. As you can imagine, those script are made
to simplify the administration of the server and the way we can start or stop services under Linux.
Of course, we can use the native way to start all required services under our server, but it is much
simple to have some kind of script files that should provide us some easy method to automate
and control the procedures. This is why init program and all initialization script files available
under the /etc/init.d directory exist.
Below are some examples showing you how to execute those scripts by hand.
For example:
• To start the httpd web server daemon manually under Linux, you’ll type:
[root@deep /]# /etc/init.d/httpd start
Starting httpd: [OK]
• To stop the httpd web server daemon manually under Linux, you’ll type:
[root@deep /]# /etc/init.d/httpd stop
Shutting down http: [OK]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-56-320.jpg)
![Installation Issues 0
CHAPTER 2
57
• To restart the httpd web server daemon manually under Linux, you’ll type:
[root@deep /]# /etc/init.d/httpd restart
Shutting down http: [OK]
Starting httpd: [OK]
Check inside your /etc/init.d directory for services available and use the commands start |
stop | restart to work around. You will see along this book that we often use initialization
script file to administration and control the way we start, restart, stop, etc services under Linux.
Software that must be uninstalled after installation of the server
Red Hat Linux installs other programs on your system by default and doesn’t give you the choice
to uninstall them during the install setup. For this reason, you must uninstall the following
software after the complete installation of our Linux server.
Below is the list of programs and a short description of their purpose. We must uninstall them for
increased security and to make more space on our server. For more information and an
explanation of their capabilities and uses, please see your Red Hat manual or query the package
by making an “rpm -qi foo” command before uninstalling them.
The anacron package:
The anacron package is similar to the cron package but differ in the way that it does not
assume that the system is running continuously and it is a good command scheduler for system
which doesn’t run 24 hours a day. In server environment, the system should absolutely run 24/24
hours; therefore we simply don’t need to have this kind of package installed on a server.
• To remove the anacron package from your system, use the following commands:
[root@deep /]# /etc/init.d/anacron stop
[root@deep /]# rpm -e anacron
[root@deep /]# rm -rf /var/spool/anacron/
The apmd package:
The apmd package or Advanced Power Management Daemon utilities is used on notebook
computer. It can watch your notebook's battery and warn all users when the battery is low. As you
can imagine, there is no need to have it installed on a server machine.
• To remove the apmd package from your system, use the following commands:
[root@deep /]# /etc/init.d/apmd stop
[root@deep /]# rpm -e apmd
The at package:
The at package is a utility that will do time-oriented job control by scheduling a command to run
later. Unfortunately, it has had a rich history of problems and we can achieve the same
functionality with the more secure vixie-cron package. For this reason I recommend you to
uninstall it.
• To remove the at package from your system, use the following commands:
[root@deep /]# /etc/init.d/atd stop
[root@deep /]# rpm -e at](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-57-320.jpg)
![58
The gpm package:
The gpm package provides mouse support to text-based Linux applications. It’s the software that
allows you to cut-and-paste operations using the mouse on your terminal. If most of your entire
administration of the server is made via remote connection, you can remove this package to save
some processes and memories. We can continue to use cut-and-paste operation via remote
connection to the server without problem. The gpm package is only useful if you stay at the
console terminal of your server to make administration tasks.
• To remove the gpm package from your system, use the following commands:
[root@deep /]# /etc/init.d/gpm stop
[root@deep /]# rpm -e gpm
The dhcpcd package:
The dhcpcd package contains the protocol, which allows systems to get their own network
configuration information from a DHCP server. If you are going to use DHCP on your network, it is
recommended to install the DHCP client included in the pump package, which provides a faster
and simpler DHCP client. For more information about DHCP, see its related chapter in this book.
• To remove the dhcpcd package from your system, use the following command:
[root@deep /]# rpm -e dhcpcd
The eject package:
The eject package contains an eject program that allows the user to eject removable media
(typically CD-ROMs, floppy disks, Iomega Jaz or Zip disks) using software. This is an unneeded
program to have installed in a server environment. You should keep it installed only if you’re
intended to run a tape backup on your system.
• To remove the eject package from your system, use the following command:
[root@deep /]# rpm -e eject
The hotplug package:
The hotplug package is a helper application for loading modules for USB devices. On a server
environment, USB devices are not used at all and are required only on Linux workstation.
• To remove the hotplug package from your system, use the following command:
[root@deep /]# rpm -e hotplug
The lokkit package:
The lokkit package is a Firewall configuration application for an average end user and it is not
designed to configure arbitrary firewalls since it is solely designed to handle typical dialup user
and cable modem set-ups. It is not the answer to a complex firewall configuration, and it is not the
equal of an expert firewall designer. Therefore remove it from your server and read the chapter
related to GIPTables in this book for more information about secure firewall configuration.
• To remove the lokkit package from your system, use the following command:
[root@deep /]# rpm -e lokkit](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-58-320.jpg)
![Installation Issues 0
CHAPTER 2
59
The ipchains package:
The ipchains package is the old tool used with Linux kernel 2.2 for managing Linux kernel
packet filtering capabilities and to set up firewalling on the network. A new and more powerful tool
called “IPTables” exists and this is the one that we will use later in the book to set up our
firewall on Linux.
• To remove the ipchains package from your system, use the following command:
[root@deep /]# rpm -e ipchains
The ksymoops package:
The ksymoops package is a small tool used to report kernel oops and error message decoder.
This package is useful for developers that work on the Linux kernel and want to debug or for
users that want to report bugs with the kernel. The same result can be achieved with the dmesg
command of Linux. Therefore, you can remove this package from your secure server.
• To remove the ksymoops package from your system, use the following command:
[root@deep /]# rpm -e ksymoops
The kudzu package:
The kudzu package is a hardware-probing tool that runs at system boot time to determine what
hardware has been added or removed from the system. Again, in server environment, we don’t
upgrade, add or remove hardware every time. Therefore, we can safety remove this small
package from our server.
• To remove the kudzu package from your system, use the following command:
[root@deep /]# rpm -e kudzu
The mailcap package:
Metamail is a program that uses the mailcap file to determine how it should display non-text or
multimedia material. We don’t need to have it installed at all. Remove it.
• To remove the mailcap package from your system, use the following command:
[root@deep /]# rpm -e mailcap
The pciutils package:
The pciutils package contains various utilities for inspecting and setting devices connected to
the PCI bus. Keep it installed if you want, but I recommend removing it form the server.
• To remove the pciutils package from your system, use the following command:
[root@deep /]# rpm -e pciutils
The raidtools package:
The raidtools package includes the tools you need to set up and maintain a software RAID
device on a Linux system. You should keep this package only if you have configured your server
to run with RAID support. Otherwise, remove it.
• To remove the raidtools package from your system, use the following command:
[root@deep /]# rpm -e raidtools](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-59-320.jpg)
![60
The redhat-logos package:
The redhat-logos package contains files of the Red Hat "Shadow Man" logo and the RPM logo.
• To remove the redhat-logos package from your system, use the following command:
[root@deep /]# rpm -e redhat-logos
The redhat-release package:
The redhat-release package contains the Red Hat Linux release files. Please note that if you
remove this package, the boot process of the system will generate error messages because it will
look for a file called “redhat-release” which will not be available anymore. To solve this
problem, we recreate the required file under the appropriated directory and add as content of this
file the word “Red Hat Linux”. Of course you can change it for whatever you want.
• To remove the redhat-release package from your system, use the command:
[root@deep /]# rpm -e redhat-release
[root@deep /]# echo Red Hat Linux > /etc/redhat-release
The setserial package:
The setserial package is a basic system utility for displaying or setting serial port information.
• To remove the setserial package from your system, use the command:
[root@deep /]# rpm -e setserial
The hdparm package:
The program hdparm is needed by IDE hard disks but not SCSI hard disks. If you have an IDE
disk on your system you must keep this program (hdparm), but if you don’t have an IDE hard
disk you can remove it safely from your system. hdparm is a small Linux tool used to optimize
your IDE hard drive. SCSI hard drives don’t need to be optimized since they are capable to run at
their full speed (80 Mps to 160 Mps) without modification.
• To remove the hdparm package from your system, use the following command:
[root@deep /]# rpm -e hdparm
The mkinitrd package:
The program mkinitrd is needed by SCSI or RAID hard disk but not IDE hard disks. If you
have a SCSI or RAID disk on your system you must keep this program (mkinitrd), but if you
don’t have a SCSI or RAID hard disk you can safely remove it from your system.
• To remove the mkinitrd package from your system, use the following command:
[root@deep /]# rpm -e --nodeps mkinitrd
The xconfig packages:
Use the programs kbdconfig, mouseconfig, timeconfig, netconfig, authconfig,
ntsysv, and setuptool in order to set your keyboard language and type, your mouse type,
your default time zone, your Ethernet devices, your NIS and shadow passwords, your numerous
symbolic links in /etc/rc.d directory, and text mode menu utility which allow you to access all
of these features.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-60-320.jpg)
![Installation Issues 0
CHAPTER 2
61
After those configurations have been set during the installation stage of your Linux server it’s rare
that you would need to change them again. So, you can uninstall them, and if in the future you
need to change your keyboard, mouse, default time, etc again via test mode menu, all you have
to do is to install the program with the RPM from your original CD-ROM.
• To remove all the above programs from your system, use the following command:
[root@deep /]# rpm -e kbdconfig mouseconfig timeconfig netconfig
authconfig ntsysv setuptool
The newt package:
The newt package provides a development library for text mode user interfaces. It’s mainly used
by all the above config packages. Since all the config packages are removed from the server, we
can safety remove newt from our system. If you have decided to keep the above config
packages (kbdconfig, mouseconfig, etc), then you should keep newt, otherwise you should
remove it.
• To remove the newt package from your system, use the following command:
[root@deep /]# rpm -e newt
The lilo package:
The lilo package provides a boot loader for Linux. Remember that during our Linux installation,
we have chosen to go with GRUB instead of LILO as our boot loader for Linux. Therefore, you can
safety remove this package from your system since it’s really not used now.
• To remove the lilo package from your system, use the following command:
[root@deep /]# rpm -e lilo
[root@deep /]# rm -f /etc/lilo.conf.anaconda
The reiserfs-utils package:
The reiserfs-utils package contains a number of utilities for creating, checking, modifying,
and correcting any inconsistencies in ReiserFS file-systems. ReiserFS is another file-system
like Ext3 for Linux. In our configuration of Linux we use Ext3 as our file-system therefore we
don’t need to keep this package installed. Keep this utility package only if you intended to run
ReiserFS instead of Ext3 file-system on your Linux server.
• To remove the reiserfs-utils package from your system, use the command:
[root@deep /]# rpm -e reiserfs-utils
The quota package:
The program quota is a system administration tools for monitoring and limiting user/group disk
usage, per file system. This program must be installed only on servers where the need for
monitoring and restricting amount of disk space in user’s directories is require.
• To remove the quota package from your system, use the following command:
[root@deep /]# rpm -e quota](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-61-320.jpg)
![62
The indexhtml package:
The indexhtml package contains the HTML page and graphics for a welcome page shown by
your browser under graphical interface installation. These HTML pages are information about Red
Hat software. You really don’t need this package under server installation and especially when
GUI is not available. Therefore, you can safety remove this package from your system.
• To remove the indexhtml package from your system, use the following command:
[root@deep /]# rpm -e indexhtml
The usbutils package:
The usbutils package provides Linux USB utilities for inspecting devices connected to a USB
bus on your system. In server environment, we really don’t use any USB devices and can safety
remove this package from our server installation. USB will usually be used only in Linux
workstation installation where you want to plug printer, camera and any other media of this type.
• To remove the usbutils package from your system, use the following command:
[root@deep /]# rpm -e usbutils
The hwdata package:
The hwdata package contains various hardware identification and configuration data mainly
used with USB and XFree86. Remember that XFree86 is related to graphical interface and since
we don’t use any GUI into our Linux server, we can remove this package from our system.
• To remove the hwdata package from your system, use the following command:
[root@deep /]# rpm -e hwdata
The parted package:
The parted package contains various utilities to create, destroy, resize, move and copy hard
disk partitions. It is useful when you need to play with your hard disk structure. In most case, we
partition and set hard disk organization at the installation of the operating system and don’t need
to play or change something once everything are installed and running properly. It’s rare to have
to use this package utility on a production server and this is the reason why I recommend you to
uninstall it. If you prefer to keep it installed for future possible usage, you are free to do it.
• To remove the parted package from your system, use the following command:
[root@deep /]# rpm -e parted
The hesiod package:
The hesiod package is another one that we can uninstall from our Linux server configuration
setup. It’s a system which uses existing DNS functionality to provide access to databases of
information that change infrequently. In most cases, we don’t need it and you should keep it
install only if you think that you will need it for some special configuration of your server.
• To remove the hesiod package from your system, use the following command:
[root@deep /]# rpm -e hesiod](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-62-320.jpg)
![Installation Issues 0
CHAPTER 2
63
The mt-st package:
The mt-st package provides tools for controlling and managing tape drives on your system. You
should keep it installed only if you have and want to run a tape backup media on your system.
• To remove the mt-st package from your system, use the following command:
[root@deep /]# rpm -e mt-st
The man-pages package:
The man-pages package provides a large collection of additional manual pages (man pages)
from the Linux Documentation Project (LDP). By default many manual pages are installed with
the operating system and the man-pages package provides additional documentation for those
who want to read them on the system. In server environment, I really don’t see the need to have
additional manual pages installed since these manual pages can be read online from the Internet
or even on another computer running as a development or workstation.
• To remove the man-pages package from your system, use the following command:
[root@deep /]# rpm -e man-pages
The sendmail package:
Even if you don’t want to run your system as a full mail server, mailer software is always needed
for potential messages sent to the root user by different software services installed on your
machine.
Sendmail is a Mail Transport Agent (MTA) program that sends mail from one machine to another
and it’s the default mail server program installed on Red Hat Linux. Unfortunately, this software
has a long history of security problem and for this reason I highly recommend you to not use it on
your Linux server. You must uninstall Sendmail and see the part in this book that is related to
Mail Transfer Agent configuration and installation for some good alternative like Exim or Qmail.
• To remove the sendmail package from your system, use the following commands:
[root@deep /]# /etc/init.d/sendmail stop
[root@deep /]# rpm -e sendmail
The procmail package:
Procmail is a mail-processing program, which can be used by Sendmail for all local mail
delivery and filtering. This program is required only if you decide to install and use Sendmail on
your server and only if Sendmail is installed. Since we’ve decided to not go with Sendmail as
our MTA for security reason, we can uninstall procmail from our Linux server.
• To remove the procmail package from your system, use the following command:
[root@deep /]# rpm -e procmail
The openldap package:
The OpenLDAP software is a set of protocols for accessing directory services like phone book
style information and other kinds of information over the Internet. This useful program is not
suitable for everyone and depends of what you want to do with your system. If you want to give it
a try, see later in this book under the chapter related to databases for more information.
• To remove the OpenLDAP package from your system, use the following command:
[root@deep /]# rpm -e --nodeps openldap](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-63-320.jpg)
![64
The cyrus-sasl packages:
The Cyrus SASL implementation is the Simple Authentication and Security Layer, a method for
adding authentication support to connection-based protocols. It is used in conjunction with
Cyrus, which is an electronic messaging program like Sendmail. Since Cyrus SASL is made to
be used with Sendmail that we have removed previously for security reason, we can safety
remove it.
• To remove the Cyrus SASL package from your system, use the following command:
[root@deep /]# rpm -e --nodeps cyrus-sasl cyrus-sasl-md5 cyrus-sasl-plain
The openssl package:
OpenSSL is an SSL encryption mechanism which ensures and provides safe and secure
transactions of data over networks. This piece of software is one of the most important tools for a
Linux server and it is highly recommended that it is installed. Unfortunately, the one that comes
with Red Hat Linux is not up to date and not optimized for our specific server. For this reason, we
will uninstall it now and see later in this book, under the chapters related to security software, how
to install, secure, optimize and use it.
• To remove the OpenSSL package from your system, use the following command:
[root@deep /]# rpm -e --nodeps openssl
[root@deep /]# rm -rf /usr/share/ssl/
The ash package:
The ash package is a smaller version of the bourne shell (sh). Since we already use sh, we can
uninstall this package from our system. If you use this program in your regular administration
task, then keep it installed on your server. In most case, we can remove it.
• To remove the ash package from your system, use the following command:
[root@deep /]# rpm -e ash
The tcsh package:
The tcsh package is an enhanced version of csh, another C shell. We already have bash as our
default shell program on Linux, and I don’t find any reason to keep another variant installed if we
don’t have any program or services that need it to run.
Most services under Linux can easily run with our default bash shell program and if you don’t
have any program that required tcsh to run, then I recommend you to uninstall it. If in the future,
you see that you need to have tcsh installed on your server for some specific program to run,
then all you have to do is to install it from your CD-ROM. In most cases, there is no program that
needs tsch to run, therefore you can remove it.
• To remove the tcsh package from your system, use the following command:
[root@deep /]# rpm -e tcsh](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-64-320.jpg)
![Installation Issues 0
CHAPTER 2
65
The specspo package:
The specspo package contains the portable object catalogues used to internationalize Red Hat
packages. I don’t think that this kind of package is really required on a production server.
• To remove the specspo package from your system, use the following command:
[root@deep /]# rpm -e specspo
The krb5-libs package:
The krb5-libs package contains the shared libraries needed by Kerberos 5. Because we’re
not using Kerberos, we’ll need to uninstall this package. Kerberos is not secure as you can
think and can be cracked easily with some good knowledge of this program. Anyway it is yours to
decide if you really need it.
• To remove the krb5-libs package from your system, use the following command:
[root@deep /]# rpm -e krb5-libs
[root@deep /]# rm -rf /usr/kerberos/
The MAKEDEV package:
The MAKEDEV package contains the MAKEDEV program, which makes it easier to create and
maintain the files in the /dev directory. Program provided by this package is used for creating
device files in the /dev directory of your server. In general, we use it under development server
when we build new package for our Linux system. On production servers, it’s rare to use it.
Therefore we can remove it from our system without any problem.
• To remove the MAKEDEV package from your system, use the following command:
[root@deep /]# rpm -e MAKEDEV
Remove unnecessary documentation files
By default the majority of each RPM’s packages installed under Linux come with documentation
files related to the software. This documentation contains original files from the program tar
archive like README, FAQ, BUG, INSTALL, NEWS, PROJECTS and more.
Many of them can be easily retrieved from the website where the program has been downloaded
and it makes no sense for them to be kept on your system. I know that hard drives costs have
come down considerably recently, but why keep this kind of documentation on a secure server if
it unlikely they will not be read more than once. Anyway, have a look inside those files and decide
for yourself if you want to remove them or not.
• To remove all documentation files from your system, use the following commands:
[root@deep /]# cd /usr/share/doc/
[root@deep doc]# rm -rf *](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-65-320.jpg)
![66
Remove unnecessary/empty files and directories
There are some files and directories we can remove manually from the file system of Linux to
make a clean install. These files and directories are not needed but still exist after our secure
installation of Linux and can be removed safely. Some are bugs from the Red Hat installation
script and others are created by default even if you don’t use them.
• To remove all unnecessary files and directories from your system, use the commands:
[root@deep /]# rm -f /etc/exports
[root@deep /]# rm -f /etc/printcap
[root@deep /]# rm -f /etc/ldap.conf
[root@deep /]# rm -f /etc/krb.conf
[root@deep /]# rm -f /etc/yp.conf
[root@deep /]# rm -f /etc/hosts.allow
[root@deep /]# rm -f /etc/hosts.deny
[root@deep /]# rm -f /etc/csh.login
[root@deep /]# rm -f /etc/csh.cshrc
[root@deep /]# rm -f /etc/fstab.REVOKE
[root@deep /]# rm -f /etc/pam_smb.conf
[root@deep /]# rm -rf /etc/xinetd.d/
[root@deep /]# rm -rf /etc/opt/
[root@deep /]# rm -rf /etc/X11/
[root@deep /]# rm -rf opt/
[root@deep /]# rm -rf /var/opt/
[root@deep /]# rm -rf /var/nis/
[root@deep /]# rm -rf /var/yp/
[root@deep /]# rm -rf /var/lib/games/
[root@deep /]# rm -rf /var/spool/lpd/
[root@deep /]# rm -rf /usr/lib/python1.5/
[root@deep /]# rm -rf /usr/lib/games/
[root@deep /]# rm -rf /usr/X11R6/
[root@deep /]# rm -rf /usr/etc/
[root@deep /]# rm -rf /usr/games/
[root@deep /]# rm -rf /usr/local/
[root@deep /]# rm -rf /usr/dict/
[root@deep /]# rm -f /usr/bin/X11
[root@deep /]# rm -f /usr/lib/X11
NOTE: If in the future you want to install a program which needs some of the files/directories we
have removed, then the program will automatically recreate the missing files or directories. Good!
Software that must be installed after installation of the server
There are certain programs required to be able to compile programs on your server, for this
reason you must install the following RPM packages. This part of the installation is very important
and requires that you install all the packages described below.
These are on your Red Hat Part 1 and Part 2 CD-ROMs under RedHat/RPMS directory and
represent the necessary base software needed by Linux to compile and install programs. Please
note that if you don’t want to compile software in your server or if you only use RPM’s packages to
update programs or if you use a dedicated server to develop, compile or create your own RPM’s
packages which will be installed later along your network on the servers, then you DON’T need to
install the packages described here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-66-320.jpg)
![Installation Issues 0
CHAPTER 2
67
Step 1
First, we mount the CD-ROM drive and move to the RPMS subdirectory of the CD-ROM.
• To mount the CD-ROM drive and move to RPM directory, use the following commands:
[root@deep /]# mount /dev/cdrom /mnt/cdrom/
had: ATAPI 32X CD-ROM drive, 128kB Cache
mount: block device dev/cdrom is write-protected, mounting read-only
[root@deep /]# cd /mnt/cdrom/RedHat/RPMS/
These are the packages that we need to be able to compile and install programs on the Linux
system. Remember, this is the minimum number of packages that permits you to compile most of
the tarballs available for Linux. Other compiler packages exist on the Linux CD-ROM, so verify
with the README file that came with the tarballs program you want to install if you receive error
messages during compilation of the specific software.
The compiler packages:
Compiler packages contain programs and languages used to build software on the system.
Remember to uninstall the entire following compiler packages after successful installation of all
software required on your Linux server.
binutils-2.11.93.0.2-11.i386.rpm
bison-1.35-1.i386.rpm
byacc-1.9-19.i386.rpm
cdecl-2.5-22.i386.rpm
cpp-2.96-110.i386.rpm
cproto-4.6-9.i386.rpm
ctags-5.2.2-2.i386.rpm
dev86-0.15.5-1.i386.rpm
flex-2.5.4a-23.i386.rpm
gcc-2.96-110.i386.rpm
gcc-c++-2.96-110.i386.rpm
glibc-kernheaders-2.4-7.14.i386.rpm
m4-1.4.1-7.i386.rpm
make-3.79.1-8.i386.rpm
patch-2.5.4-12.i386.rpm
perl-5.6.1-34.99.6.i386.rpm
The development packages:
Development packages contain header and other files required during compilation of software. In
general, development packages are needed when we want to add some specific functionality to
the program that we want to compile. For example if I want to add PAM support to IMAP, I’ll need
pam-devel, which contains the required header files for IMAP to compile successfully.
As for compiler packages, all development packages must be uninstalled after successful
compilation of all the software that you need on your Linux server. Remember to uninstall them
since they are not needed for proper functionality of the server, but just to compile the programs.
aspell-devel-0.33.7.1-9.i386.rpm
db3-devel-3.3.11-6.i386.rpm
freetype-devel-2.0.9-2.i386.rpm
gdbm-devel-1.8.0-14.i386.rpm
gd-devel-1.8.4-4.i386.rpm
glibc-devel-2.2.5-34.i386.rpm
libjpeg-devel-6b-19.i386.rpm
libpng-devel-1.0.12-2.i386.rpm
libstdc++-devel-2.96-110.i386.rpm
ncurses-devel-5.2-26.i386.rpm
pam-devel-0.75-32.i386.rpm
pspell-devel-0.12.2-8.i386.rpm
zlib-devel-1.1.3-25.7.i386.rpm](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-67-320.jpg)
![68
Dependencies packages:
Dependencies packages are other RPM packages needed by the RPM packages that we want to
install. This happens because some RPM’s are directly linked with others and depend on each
one to function properly. The following packages are required by the above RPM packages and
we will install them to satisfy dependencies. After proper compilation and installation of all needed
software on the Linux server, we can uninstall them safety (if not needed by special program that
we will install).
aspell-0.33.7.1-9.i386.rpm
freetype-2.0.9-2.i386.rpm
gd-1.8.4-4.i386.rpm
libjpeg-6b-19.i386.rpm
libpng-1.0.12-2.i386.rpm
libtool-libs-1.4.2-7.i386.rpm
pspell-0.12.2-8.i386.rpm
Step 2
It is better to install the software described above together if you don’t want to receive
dependencies error messages during the install. Some of the RPMs reside on CD-ROM Part 1
and other on CD-ROM Part2 of Red Hat. For easy installation, I recommend you to copy all of the
required packages (compilers and development) to your hard drive and install them from there.
• These procedures can be accomplished with the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rpm -Uvh *.rpm
Preparing... ########################################### [100%]
1:binutils ########################################### [ 2%]
2:bison ########################################### [ 5%]
3:byacc ########################################### [ 8%]
4:cdecl ########################################### [ 11%]
5:cpp ########################################### [ 13%]
6:cproto ########################################### [ 16%]
7:ctags ########################################### [ 19%]
8:db3-devel ########################################### [ 22%]
9:dev86 ########################################### [ 25%]
10:flex ########################################### [ 27%]
11:freetype ########################################### [ 30%]
12:freetype-devel ########################################### [ 33%]
13:gdbm-devel ########################################### [ 36%]
14:glibc-kernheaders ########################################### [ 38%]
15:glibc-devel ########################################### [ 41%]
16:gcc ########################################### [ 44%]
17:libjpeg ########################################### [ 47%]
18:libjpeg-devel ########################################### [ 50%]
19:libpng ########################################### [ 52%]
20:gd ########################################### [ 55%]
21:gd-devel ########################################### [ 58%]
22:libstdc++-devel ########################################### [ 61%]
23:gcc-c++ ########################################### [ 63%]
24:libtool-libs ########################################### [ 66%]
25:m4 ########################################### [ 69%]
26:make ########################################### [ 72%]
27:pam-devel ########################################### [ 75%]
28:patch ########################################### [ 77%]
29:perl ########################################### [ 80%]
30:ncurses-devel ########################################### [ 83%]
31:pspell ########################################### [ 86%]
32:aspell ########################################### [ 88%]
33:pspell-devel ########################################### [ 91%]
34:aspell-devel ########################################### [ 94%]
35:zlib-devel ########################################### [ 97%]
36:libpng-devel ########################################### [100%]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-68-320.jpg)
![General Security 0
CHAPTER 3
75
Linux General Security
Abstract
A secure Linux server depends on how the administrator makes it. Once we have eliminated the
potential security risk by removing unneeded services, we can start to secure our existing
services and software on our server. Within a few hours of installing and configuring your system,
you can prevent many attacks before they occur. In this chapter we will discuss some of the more
general, basic techniques used to secure your system. The following is a list of features that can
be used to help prevent attacks from external and internal sources.
BIOS
It is recommended to disallow booting from floppy drives and set passwords on BIOS features.
You can check your BIOS manual or look at it thoroughly the next time you boot up your system
to find out how to do this. Disabling the ability to boot from floppy drives and being able to set a
password to access the BIOS features will improve the security of your system.
This will block unauthorized people from trying to boot your Linux system with a special boot disk
and will protect you from people trying to change BIOS features like allowing boot from floppy
drive or booting the server without prompt password. It is important to note that there is a
possibility to bypass this security measure if someone has a physical access to your server since
they can open the computer and unplug the BIOS battery. This will reset all features to their initial
values.
Unplug your server from the network
It is not wise to apply security changes in your newly installed Linux server if you are online. So it
is preferable to deactivate all network interfaces in the system before applying security changes.
• To stop specific network devices manually on your system, use the command:
[root@deep /]# ifdown eth0
• To start specific network devices manually on your system, use the command:
[root@deep /]# ifup eth0
• To shut down all network interfaces, use the following command:
[root@deep /]# /etc/init.d/network stop
Shutting down interface eth0 [OK]
Disabling Ipv4 packet forwarding [OK]
• To start all network interfaces, use the following command:
[root@deep /]# /etc/init.d/network start
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-75-320.jpg)
![General Security 0
CHAPTER 3
79
The single-user login mode of Linux
This part applies for those who use LILO as their boot loader. Linux has a special command
(linux single) also known as ‘single-user mode’, which can be entered at the boot prompt
during startup of the system. The single-user mode is generally used for system maintenance.
You can boot Linux in single-user mode by typing at the LILO boot prompt the command:
LILO: linux single
This will place the system in Run level 1 where you'll be logged in as the super-user 'root', and
where you won't even have to type in a password!
Step 1
Requiring no password to boot into root under single-user mode is a bad idea! You can fix this by
editing the inittab file (vi /etc/inittab) and change the following line:
id:3:initdefault:
To read:
id:3:initdefault:
~~:S:wait:/sbin/sulogin
The addition of the above line will require entering the root password before continuing to boot
into single-user mode by making init (8) run the program sulogin (8) before dropping
the machine into a root shell for maintenance.
Step 2
Now, we have to restart the process control initialization of the server for the changes to take
effect.
• This can be done with the following command:
[root@deep /]# /sbin/init q
Disabling Ctrl-Alt-Delete keyboard shutdown command
Commenting out the line (with a “#”) listed below in your /etc/inittab file will disable the
possibility of using the Control-Alt-Delete command to shutdown your computer. This is
pretty important if you don't have the best physical security to the machine.
Step 1
• To do this, edit the inittab file (vi /etc/inittab) and change/comment the line:
ca::ctrlaltdel:/sbin/shutdown -t3 -r now
To read:
#ca::ctrlaltdel:/sbin/shutdown -t3 -r now](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-79-320.jpg)
![80
Step 2
Now, we have to restart the process control initialization of the server for the changes to take
effect.
• This can be done with the following command:
[root@deep /]# /sbin/init q
Limiting the default number of started ttys on the server
Default installed Linux system comes with six virtual consoles in standard run levels. This means
that six mingetty processes will always by available at every time on the server. These virtual
consoles allow you to login with six different virtual consoles on the system.
Step 1
On secure server, we can limit the number to two virtual consoles and save some resources
which may be used for other work by the server when required.
• To do this, edit the inittab file (vi /etc/inittab) and remove/comment the lines:
1:2345:respawn:/sbin/mingetty tty1
2:2345:respawn:/sbin/mingetty tty2
3:2345:respawn:/sbin/mingetty tty3
4:2345:respawn:/sbin/mingetty tty4
5:2345:respawn:/sbin/mingetty tty5
6:2345:respawn:/sbin/mingetty tty6
To read:
1:2345:respawn:/sbin/mingetty tty1
2:2345:respawn:/sbin/mingetty tty2
#3:2345:respawn:/sbin/mingetty tty3
#4:2345:respawn:/sbin/mingetty tty4
#5:2345:respawn:/sbin/mingetty tty5
#6:2345:respawn:/sbin/mingetty tty6
Step 2
Now, we have to restart the process control initialization of the server for the changes to take
effect.
• This can be done with the following command:
[root@deep /]# /sbin/init q
The LILO and /etc/lilo.conf file
This part applies for those who use LILO as their boot loader. LILO is a boot loader that can be
used to manage the boot process, boot Linux kernel images from floppy disks, hard disks or can
even act as a "boot manager" for other operating systems.
LILO is very important in the Linux system and for this reason, we must protect it the best we
can. The most important configuration file of LILO is the lilo.conf file. It is with this file that we
can configure and improve the security of our LILO program and Linux system. Following are
three important options that will improve the security of our valuable LILO program.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-80-320.jpg)
![General Security 0
CHAPTER 3
81
• Adding: timeout=00
This option controls how long (in seconds) LILO waits for user input before booting to the default
selection. One of the requirements of C2 security is that this interval be set to 0 unless the system
dual boots something else.
• Adding: restricted
This option asks for a password only, if parameters are specified on the command line (e.g.
linux single). The option “restricted” can only be used together with the “password”
option. Make sure you use this one on each additional image you may have.
• Adding: password=<password>
This option asks the user for a password when trying to load the image. Actually the effect of
using the password parameter in /etc/lilo.conf will protect the Linux image from booting.
This means, it doesn't matter if you load Linux in single mode or if you just do a normal boot. It
will always ask you for the password.
Now this can have a very bad effect, namely you are not able to reboot Linux remotely any more
since it won't come up until you type in the root password at the console. It is for this reason that
adding “restricted” with “password” is very important since the option "restricted" relaxes
the password protection and a password is required only if parameters are specified at the LILO
prompt, (e.g. single).
Passwords are always case-sensitive, also make sure the /etc/lilo.conf file is no longer
world readable, or any user will be able to read the password. Here is an example of our
protected LILO with the lilo.conf file.
Step 1
• Edit the lilo.conf file (vi /etc/lilo.conf) and add or change the three options
above as show:
boot=/dev/sda
map=/boot/map
install=/boot/boot.b
prompt remove this line if you don’t want to pass options at the LILO prompt.
timeout=00 change this line to 00 to disable the LILO prompt.
linear
message=/boot/message remove this line if you don’t want the welcome screen.
default=linux
restricted add this line to relaxes the password protection.
password=<password> add this line and put your password.
image=/boot/vmlinuz-2.4.2-2
label=linux
initrd=/boot/initrd-2.4.2-2.img
read-only
root=/dev/sda6
Step 2
Because the configuration file /etc/lilo.conf now contains unencrypted passwords, it should
only be readable for the super-user “root”.
• To make the /etc/lilo.conf file readable only by the super-user “root”, use the
following command:
[root@deep /]# chmod 600 /etc/lilo.conf (will be no longer world readable).](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-81-320.jpg)
![82
Step 3
Now we must update our configuration file /etc/lilo.conf for the change to take effect.
• To update the /etc/lilo.conf file, use the following command:
[root@deep /]# /sbin/lilo -v
LILO version 21.4-4, copyright © 1992-1998 Wernerr Almesberger
‘lba32’ extentions copyright © 1999,2000 John Coffman
Reading boot sector from /dev/sda
had : ATAPI 32X CD-ROM drive, 128kB Cache
Merging with /boot/boot.b
Mapping message file /boot/message
Boot image : /boot/vmlinuz-2.2.16-22
Mapping RAM disk /boot/initrd-2.2.16-22.img
Added linux *
/boot/boot.0800 exists – no backup copy made.
Writing boot sector.
Step 4
One more security measure you can take to secure the lilo.conf file is to set it immutable,
using the chattr command.
• To set the file immutable simply, use the following command:
[root@deep /]# chattr +i /etc/lilo.conf
And this will prevent any changes (accidental or otherwise) to the lilo.conf file. If you wish to
modify the lilo.conf file you will need to unset the immutable flag:
• To unset the immutable flag, use the following command:
[root@deep /]# chattr -i /etc/lilo.conf
WARNING: When you use the password option, then LILO will always ask you for the password,
regardless if you pass options at the LILO prompt (e.g. single) or not EXCEPT when you set
the "restricted" option in /etc/lilo.conf.
The option "restricted" relaxes the password protection and a password is required only if
parameters are specified at the LILO prompt, (e.g. single).
If you didn't had this option set "restricted", Linux will always ask you for the password and
you will not be able to remotely reboot your system, therefore don’t forget to add the option
"restricted” with the option "password" into the /etc/lilo.conf file.
The GRUB and /boot/grub/grub.conf file
This part applies for those who use GRUB as their boot loader. GRUB is another boot loader like
LILO but with many more useful feature and power. One of its main advantages compared to
LILO is the fact that it provides a small shell interface to manage the operating system. Also, it
doesn’t need to be updated each time you recompile a new kernel on your server.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-82-320.jpg)
![General Security 0
CHAPTER 3
83
GRUB is very important since it is the first software program that runs when the computer starts
and we have to secure it as much as possible to avoid any possible problem. In its default
installation it’s already well protected and below I explain how its configuration file is made. In
regard to LILO, GRUB is really easy to use and configure. Below is a default GRUB configuration
file and security I recommend you to apply. The text in bold are the parts of the configuration file
that must be customized and adjusted to satisfy our needs.
• Edit the grub.conf file (vi /boot/grub/grub.conf) and set your needs. Below is
what we recommend you:
default=0
timeout=0
splashimage=(hd0,0)/grub/splash.xpm.gz
password --md5 $1$oKr0ÝmFo$tPYwkkvQbtqo1erwHj5wb/
title Red Hat Linux (2.4.18-3)
root (hd0,0)
kernel /vmlinuz-2.4.18-3 ro root=/dev/sda5
initrd /initrd-2.4.18-3.img
This tells the grub.conf file to set itself up for this particular configuration with:
default=0
The option “default” is used to define the default entry of the configuration file. The number “0”
mean that the following parameters are the default entry for the configuration of GRUB. In a server
configuration where Linux is the only operating system to boot, the default entry of “0” will be the
only one to use and we don’t need to define any additional entry.
timeout=0
This option “timeout” is used to define the timeout, in sec seconds to wait, before automatically
booting the default entry. As for LILO, one of the requirements of C2 security is that this interval
be set to 0 unless the system dual boots something else. One of the disadvantages to set this
option to “0” is that you will no longer be able to have access at boot time to the shell interface of
the software but this is not really a problem since all we need from the GRUB software is to boot
our operating system.
splashimage=(hd0,0)/grub/splash.xpm.gz
This option “splashimage” is an option added by Red Hat to boot the system with a graphical
image. The value is the path of the compressed image to use when booting GRUB. It’s to you to
keep this parameter on your system or to remove it. If you want to remove it, just delete the
above line with the compressed image from your server.
password --md5 $1$bgGCL/$4yF3t0py.IjU0LU.q7YfB1
This option “password” is used to inform GRUB to ask for a password and disallows any
interactive control, until you press the key <p> and enter a correct password. The option --md5
tells GRUB that a password in MD5 format is required as a value. If it is omitted, GRUB assumes
the specified password is in clear text.
When we have installed the operating system, we have already configured GRUB with a
password protection. This password is what you see here. If you want to change it, you have to
use the “grub-md5-crypt” command to generate a new encrypt password it in MD5 format.
• This can be done with the following command:
[root@dev /]# grub-md5-crypt
Password:
$1$bgGCL/$4yF3t0py.IjU0LU.q7YfB1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-83-320.jpg)
![84
Once the above command has been issued, you have to cut and paste the encrypted password
to your configuration file.
title Red Hat Linux (2.4.18-3)
This option “title” is used to define a name to the contents of the rest of the line. It is directly
related to the default boot entry. What you enter here is what you will see during boot time. This
option is useful when we have more than one OS to start on our computer and allow us to give
the name that we want to distinguish them. You are free to enter whatever name you like.
root (hd0,0)
This option “root” is one of the most important parameter with GRUB and without it nothing will
work. It is used to define the current root device to use for booting the operating system. Its
definition and configuration is a little bit special as you can see. Here is an explanation of its
meaning. The “hd0” parameter represents using the entire disk and the “hd0,0” represents using
the partition of the disk (or the boot sector of the partition when installing GRUB). Don’t be
confused here because “hd” is valid for IDE and SCSI drives. There is no difference; you always
use “hd” even on SCSI drive.
kernel /vmlinuz-2.4.18-3 ro root=/dev/sda5
This option “kernel” is used to load the primary boot image (our kernel). The parameter to this
option is simply the path where GRUB should find the kernel image from which we want it to boot.
The additional lines are to inform it that kernel image is located on the sda5 partition on our
server and that we want to load it as read only for security reason.
initrd /initrd-2.4.18-3.img
This option “initrd” is optional and will appear into your GRUB configuration file only if you run a
SCSI computer. For IDE computer, this option is not required and should not be defined inside
the configuration file of GRUB. The parameter simply informs GRUB software where our initial ram
disk image is located on the server. GRUB reads this initial ram disk and loads it during startup.
The /etc/services file
The port numbers on which certain "standard" services are offered are defined in the RFC 1700
"Assigned Numbers". The /etc/services file enables server and client programs to convert
service names to these numbers (ports). The list is kept on each host and it is stored in the file
/etc/services. Only the "root" user is allowed to make modifications to this file. It is rare to
edit the /etc/services file since it already contains the more common service names / port
numbers. To improve security, we can set the immutable flag on this file to prevent unauthorized
deletion or modification.
• To immunize the /etc/services file, use the following command:
[root@deep /]# chattr +i /etc/services](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-84-320.jpg)
![General Security 0
CHAPTER 3
85
The /etc/securetty file
The /etc/securetty file allows you to specify which TTY and VC (virtual console) devices the
“root” user is allowed to login on. The /etc/securetty file is read by the login program
(usually /bin/login). Its format is a list of the tty and vc devices names allowed, and for all
others that are commented out or do not appear in this file, root login is disallowed.
Disable any tty and vc devices that you do not need by commenting them out (# at the
beginning of the line) or by removing them.
• Edit the securetty file (vi /etc/securetty) and comment out or remove the lines:
vc/1
#vc/2
#vc/3
#vc/4
#vc/5
#vc/6
#vc/7
#vc/8
#vc/9
#vc/10
#vc/11
tty1
#tty2
#tty3
#tty4
#tty5
#tty6
#tty7
#tty8
#tty9
#tty10
#tty11
Which means root is allowed to login on only tty1 and vc/1. This is my recommendation,
allowing “root” to log in on only one tty or vc device and use the su or sudo command to
switch to “root” if you need more devices to log in as “root”.
Special accounts
It is important to DISABLE ALL default vendor accounts that you don’t use on your system
(some accounts exist by default even if you have not installed the related services on your
server). This should be checked after each upgrade or new software installation. Linux provides
these accounts for various system activities, which you may not need if the services are not
installed on your server. If you do not need the accounts, remove them. The more accounts you
have, the easier it is to access your system.
We assume that you are using the shadow password suite on your Linux system. If you are not,
you should consider doing so, as it helps to tighten up security somewhat. This is already set if
you’ve followed our Linux installation procedure and selected, under the “Authentication
Configuration”, the option to “Enable Shadow Passwords” (see the chapter related to the
“Installation of your Linux Server” for more information).
• To delete user on your system, use the following command:
[root@deep /]# userdel username
• To delete group on your system, use the following command:
[root@deep /]# groupdel username](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-85-320.jpg)
![86
Step 1
First we will remove all default vendor accounts into the /etc/passwd file that are unnecessary
for the operation of the secure server configuration that we use in this book.
• Type the following commands to delete all default users accounts listed below.
[root@deep /]# userdel adm
[root@deep /]# userdel lp
[root@deep /]# userdel shutdown
[root@deep /]# userdel halt
[root@deep /]# userdel news
[root@deep /]# userdel mailnull
[root@deep /]# userdel operator
[root@deep /]# userdel games
[root@deep /]# userdel gopher
[root@deep /]# userdel ftp
[root@deep /]# userdel vcsa
WARNING: By default, the userdel command will not delete a user’s home directory. If you want
the home directories of accounts to be deleted too, then add the -r option to the userdel
command. Finally, the -r option must be used only when you have added a new user to the
server and want to remove them. It doesn’t need to be used for the removal of the above default
user’s accounts since they do not have a home directory.
Once the above list of users has been deleted from your Linux system, the /etc/passwd file will
look like this:
root:x:0:0:root:/root:/bin/bash
bin:x:1:1:bin:/bin:/sbin/nologin
daemon:x:2:2:daemon:/sbin:/sbin/nologin
sync:x:5:0:sync:/sbin:/bin/sync
mail:x:8:12:mail:/var/spool/mail:/sbin/nologin
uucp:x:10:14:uucp:/var/spool/mail:/sbin/nologin
nobody:x:99:99:Nobody:/:/sbin/nologin
rpm:x:37:37::/var/lib/rpm:/bin/bash
Step 2
After that we have removed all the unnecessary default vendor accounts into the /etc/passwd
file from our system, we will remove all default vendor accounts into the /etc/group file.
• Type the following commands to delete all default users groups accounts listed below.
[root@deep /]# groupdel adm
[root@deep /]# groupdel lp
[root@deep /]# groupdel news
[root@deep /]# groupdel games
[root@deep /]# groupdel dip](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-86-320.jpg)
![General Security 0
CHAPTER 3
87
Once the above list of group users has been deleted from your Linux system the /etc/group
file will look like this:
root:x:0:root
bin:x:1:root,bin,daemon
daemon:x:2:root,bin,daemon
sys:x:3:root,bin
tty:x:5:
disk:x:6:root
mem:x:8:
kmem:x:9:
wheel:x:10:root
mail:x:12:mail
uucp:x:14:uucp
man:x:15:
lock:x:54:
nobody:x:99:
users:x:100:
slocate:x:21:
floppy:x:19:
utmp:x:22:
rpm:x:37:
Step 3
Now, you can add all the necessary and allowed users into the system. Below I show you how
you should add new user into your Linux server. Adding a new user into your server mean that
you have to create the username and assign him/her a password.
• To add a new user on your system, use the following command:
[root@deep /]# useradd username
For example:
[root@deep /]# useradd admin
• To add or change password for user on your system, use the following command:
[root@deep /]# passwd username
For example:
[root@deep /]# passwd admin
The output should look something like this:
Changing password for user admin
New UNIX password:
Retype new UNIX password:
passwd: all authentication tokens updated successfully
Step 4
The immutable bit can be used to prevent accidentally deleting or overwriting a file that must be
protected. It also prevents someone from creating a symbolic link to this file, which has been the
source of attacks involving the deletion of /etc/passwd, /etc/shadow, /etc/group or
/etc/gshadow files.
• To set the immutable bit on the passwords and groups files, use the following commands:
[root@deep /]# chattr +i /etc/passwd
[root@deep /]# chattr +i /etc/shadow
[root@deep /]# chattr +i /etc/group](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-87-320.jpg)
![88
[root@deep /]# chattr +i /etc/gshadow
WARNING: In the future, if you intend to add or delete users, passwords, user groups, or group
files, you must unset the immutable bit on all those files or you will not be able to make and
update your changes. Also if you intend to install an RPM program that will automatically add a
new user to the different immunized passwd and group files, then you will receive an error
message during the install if you have not unset the immutable bit from those files.
• To unset the immutable bit on the passwords and groups files, use the commands:
[root@deep /]# chattr -i /etc/passwd
[root@deep /]# chattr -i /etc/shadow
[root@deep /]# chattr -i /etc/group
[root@deep /]# chattr -i /etc/gshadow
Control mounting a file system
You can have more control on mounting file systems like /var/lib, /home or /tmp partitions
with some nifty options like noexec, nodev, and nosuid. This can be setup in the /etc/fstab
file. The fstab file contains descriptive information about the various file system mount options;
each line addresses one file system.
Information related to security options in the fstab text file are:
defaults Allow everything (quota, read-write, and suid) on this partition.
noquota Do not set users quotas on this partition.
nosuid Do not set SUID/SGID access on this partition.
nodev Do not set character or special devices access on this partition.
noexec Do not set execution of any binaries on this partition.
quota Allow users quotas on this partition.
ro Allow read-only on this partition.
rw Allow read-write on this partition.
suid Allow SUID/SGID access on this partition.
NOTE: For more information on options that you can set in this file (fstab) see the man pages
about mount (8).
Step 1
• Edit the fstab file (vi /etc/fstab) and change it depending of your needs.
For example change:
LABEL=/home /home ext3 defaults 1 2
LABEL=/tmp /tmp ext3 defaults 1 2
LABEL=/var/lib /var/lib ext3 defaults 1 2
To read:
LABEL=/home /home ext3 defaults,nosuid 1 2
LABEL=/tmp /tmp ext3 defaults,nosuid,noexec 1 2
LABEL=/var/lib /var/lib ext3 defaults,nodev 1 2](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-88-320.jpg)
![General Security 0
CHAPTER 3
89
Meaning, <nosuid>, do not allow set-user-identifier or set-group-identifier bits to take effect,
<nodev>, do not interpret character or block special devices on this file system partition, and
<noexec>, do not allow execution of any binaries on the mounted file system.
Step 2
Once you have made the necessary adjustments to the /etc/fstab file, it is time to inform the
Linux system about the modifications.
• This can be accomplished with the following commands:
[root@deep /]# mount /var/lib -oremount
[root@deep /]# mount /home -oremount
[root@deep /]# mount /tmp -oremount
Each file system that has been modified must be remounted with the command show above. In
our example we have modified the /var/lib, /home, and /tmp file system and it is for this
reason that we remount these files systems with the above commands.
• You can verify if the modifications have been correctly applied to the Linux system with
the following command:
[root@deep /]# cat /proc/mounts
rootfs / rootfs rw 0 0
/dev/root / ext3 rw 0 0
/proc /proc proc rw 0 0
/dev/sda1 /boot ext3 rw 0 0
/dev/sda8 /chroot ext3 rw 0 0
none /dev/pts devpts rw 0 0
/dev/sda7 /home ext3 rw,nosuid 0 0
none /dev/shm tmpfs rw 0 0
/dev/sda11 /tmp ext3 rw,nosuid,noexec 0 0
/dev/sda6 /usr ext3 rw 0 0
/dev/sda9 /var ext3 rw 0 0
/dev/sda10 /var/lib ext3 rw,nodev 0 0
This command will show you all the files systems on your Linux server with parameters applied to
them.
Mounting the /usr directory of Linux as read-only
It is allowable to mount your /usr partition read-only since this is, by definition, static data. Of
course, anyone with root access can remount it as writable, but a generic attack script may not
know this. On many Linux variants this directory resides in its own partition and the default
parameter is to mount it as read-write. We can change this parameter to make it read-only for
better security. Please be sure that your /usr directory is on its own partition or the following
hack will not work for you.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-89-320.jpg)
![90
Step 1
Mounting the /usr partition as read-only eliminates possible problems that someone may try to
change or modify vital files inside it. To mount the /usr file system of Linux as read-only, follow
the simple steps below.
• Edit the fstab file (vi /etc/fstab) and change the line:
LABEL=/usr /usr ext3 defaults 1 2
To read:
LABEL=/usr /usr ext3 defaults,ro 1 2
We add the “ro” option to this line to specify to mount this partition as read-only.
Step 2
Make the Linux system aware about the modification you have made to the /etc/fstab file.
• This can be accomplished with the following command:
[root@deep /]# mount /usr -oremount
• Then test your results with the following command:
[root@deep /]# cat /proc/mounts
rootfs / rootfs rw 0 0
/dev/root / ext3 rw 0 0
/proc /proc proc rw 0 0
/dev/sda1 /boot ext3 rw 0 0
/dev/sda8 /chroot ext3 rw 0 0
none /dev/pts devpts rw 0 0
/dev/sda7 /home ext3 rw,nosuid 0 0
none /dev/shm tmpfs rw 0 0
/dev/sda11 /tmp ext3 rw,nosuid,noexec 0 0
/dev/sda6 /usr ext3 ro 0 0
/dev/sda9 /var ext3 rw 0 0
/dev/sda10 /var/lib ext3 rw,nodev 0 0
If you see something like: /dev/sda6 /usr ext3 ro 0 0, congratulations!
WARNING: If in the future you want to install some RPM package or program from source code, it is
important to reset the modification you have made to the /usr directory to its initial state (read-
write) or you will not be able to install new software because the /usr partition is set as read-
only. All you have to do if you want to put the /usr partition to its original state is to edit the
/etc/fstab file again and remove the “ro” option then remount the /usr file system with the
“mount -oremount” command again.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-90-320.jpg)
![General Security 0
CHAPTER 3
91
Tighten scripts under /etc/init.d
Fix the permissions of the script files that are responsible for starting and stopping all your normal
processes that need to run at boot time.
• To fix the permissions of those files, use the following command:
[root@deep /]# chmod 0700 /etc/init.d/*
Which means just the super-user “root” is allowed to Read, Write, and Execute scripts files on
this directory. I don’t think regular users need to know what’s inside those script files.
WARNING: If you install a new program or update a program that use the init system V script
located under /etc/init.d/ directory, don’t forget to change or verify the permission of this
script file again.
Tighten scripts under /etc/cron.daily/
As for the above hack, we can tighten the security of all script files that are responsible for
executing scheduled job on our server. Those files have a default permission mode of (0755-
rwxr-xr-x), which is too high for what they should accomplish.
• To fix the permissions of those files, use the following command:
[root@deep /]# chmod 0550 /etc/cron.daily/*
The same is true for other cron directories under the /etc directory of your system. If files exist
under the other cron directories, then use the above command to change their default permission
mode for better security.
WARNING: If you install a new program or update a program that provides and install a cron file on
your server, don’t forget to change or verify the permission of this script file again.
Bits from root-owned programs
A regular user will be able to run a program as root if it is set to SUID root. All programs and files
on your computer with the ’s’ bits appearing on its mode, have the SUID (-rwsr-xr-x) or SGID
(-r-xr-sr-x) bit enabled. Because these programs grant special privileges to the user who is
executing them, it is important to remove the 's' bits from root-owned programs that won't
absolutely require such privilege.
This can be accomplished by executing the command chmod a-s with the name(s) of the
SUID/SGID files as its arguments.
Such programs include, but aren't limited to:
Programs you never use.
Programs that you don't want any non-root users to run.
Programs you use occasionally, and don't mind having to su (1) to root to run.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-91-320.jpg)
![92
Step 1
We've placed an asterisk (*) next to each program we personally might disable and consider
being not absolutely required for the duty work of the server. Remember that your system needs
some suid root programs to work properly, so be careful.
• To find all files with the ‘s’ bits from root-owned programs, use the command:
[root@deep]# find / -type f ( -perm -04000 -o -perm -02000 ) -exec ls -l {} ;
*-rwsr-xr-x 1 root root 34296 Mar 27 20:40 /usr/bin/chage
*-rwsr-xr-x 1 root root 36100 Mar 27 20:40 /usr/bin/gpasswd
-rwxr-sr-x 1 root slocate 25020 Jun 25 2001 /usr/bin/slocate
-r-s--x--x 1 root root 15104 Mar 13 20:44 /usr/bin/passwd
*-r-xr-sr-x 1 root tty 6920 Mar 14 15:24 /usr/bin/wall
*-rws--x--x 1 root root 12072 Apr 1 18:26 /usr/bin/chfn
*-rws--x--x 1 root root 11496 Apr 1 18:26 /usr/bin/chsh
*-rws--x--x 1 root root 4764 Apr 1 18:26 /usr/bin/newgrp
*-rwxr-sr-x 1 root tty 8584 Apr 1 18:26 /usr/bin/write
-rwsr-xr-x 1 root root 21080 Apr 15 00:49 /usr/bin/crontab
*-rwsr-xr-x 1 root root 32673 Apr 18 17:40 /usr/sbin/ping6
*-rwsr-xr-x 1 root root 13994 Apr 18 17:40 /usr/sbin/traceroute6
-rwxr-sr-x 1 root utmp 6604 Jun 24 2001 /usr/sbin/utempter
-rws--x--x 1 root root 22388 Apr 15 18:15 /usr/sbin/userhelper
*-rwsr-xr-x 1 root root 17461 Apr 19 12:35 /usr/sbin/usernetctl
*-rwsr-xr-x 1 root root 35192 Apr 18 17:40 /bin/ping
*-rwsr-xr-x 1 root root 60104 Apr 1 18:26 /bin/mount
*-rwsr-xr-x 1 root root 30664 Apr 1 18:26 /bin/umount
-rwsr-xr-x 1 root root 19116 Apr 8 12:02 /bin/su
-r-sr-xr-x 1 root root 120264 Apr 9 23:24 /sbin/pwdb_chkpwd
-r-sr-xr-x 1 root root 16992 Apr 9 23:24 /sbin/unix_chkpwd
*-rwxr-sr-x 1 root root 14657 Apr 19 12:35 /sbin/netreport
Step 2
• To disable the suid bits on selected programs above, use the following commands:
[root@deep /]# chmod a-s /usr/bin/chage
[root@deep /]# chmod a-s /usr/bin/gpasswd
[root@deep /]# chmod a-s /usr/bin/wall
[root@deep /]# chmod a-s /usr/bin/chfn
[root@deep /]# chmod a-s /usr/bin/chsh
[root@deep /]# chmod a-s /usr/bin/newgrp
[root@deep /]# chmod a-s /usr/bin/write
[root@deep /]# chmod a-s /usr/sbin/ping6
[root@deep /]# chmod a-s /usr/sbin/traceroute6
[root@deep /]# chmod a-s /usr/sbin/usernetctl
[root@deep /]# chmod a-s /bin/ping
[root@deep /]# chmod a-s /bin/mount
[root@deep /]# chmod a-s /bin/umount
[root@deep /]# chmod a-s /sbin/netreport](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-92-320.jpg)
![General Security 0
CHAPTER 3
93
Don’t let internal machines tell the server what their MAC address is
To avoid the risk that a user could easily change a computers IP address and appear as
someone else to the firewall, you can force the ARP cache entries of Linux using the arp
command utility. A special option can be used with the arp utility to avoid letting INTERNAL
machines tell the server what their MAC (Media Access Control) address is and the IP address
associated with it.
Step1
ARP is a small utility, which manipulates the kernel’s ARP (Address Resolution Protocol) cache.
Through all possible options associated with this utility, the primary one is clearing an address
mapping entry and manually setting up one. In the hope to more secure our server from the
INTERNAL, we will manually set MAC address (sometimes called Hardware addresses) of all
known computers in our network statically by using static ARP entries.
• For each IP address of INTERNAL computers in your network, use the following
command to know the MAC address associate with the IP address:
[root@deep /]# ifconfig
eth0 Link encap:Ethernet HWaddr 00:50:DA:C6:D3:FF
inet addr:207.35.78.3 Bcast:207.35.78.32 Mask:255.255.255.224
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:1887318 errors:0 dropped:0 overruns:1 frame:0
TX packets:2709329 errors:0 dropped:0 overruns:0 carrier:1
collisions:18685 txqueuelen:100
Interrupt:10 Base address:0xb000
eth1 Link encap:Ethernet HWaddr 00:50:DA:C6:D3:09
inet addr:192.168.1.11 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:182937 errors:0 dropped:0 overruns:0 frame:0
TX packets:179612 errors:0 dropped:0 overruns:0 carrier:0
collisions:7434 txqueuelen:100
Interrupt:11 Base address:0xa800
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:3924 Metric:1
RX packets:7465 errors:0 dropped:0 overruns:0 frame:0
TX packets:7465 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
The MAC (Media Access Control) address will be the letters and numbers that come after
“HWaddr” (the Hardware Address). In the above example our MAC address are:
00:50:DA:C6:D3:FF for the interface eth0 and 00:50:DA:C6:D3:09 for the interface eth1.
Step 2
Once we know the MAC (Media Access Control) address associated with IP address, we can add
them manually to the ARP entries of the Linux server.
• To add manually MAC address to ARP entries, use the following commands:
[root@deep /]# arp -s 207.35.78.3 00:50:DA:C6:D3:FF
[root@deep /]# arp -s 192.168.1.11 00:50:DA:C6:D3:09](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-93-320.jpg)
![94
The “-s” option means to manually create an ARP address mapping entry for host hostname with
hardware address set to hw_addr class. You can add you ARP commands to the
/etc/rc.local file if you want to keep your configuration if the system reboots.
Step 3
• To verify if the modifications have been added to the system, use the following command:
[root@deep /]# arp
Address Hwtype Hwaddress Flags Mask Iface
207.35.78.3 ether 00:20:78:13:86:92 CM eth1
192.168.1.11 ether 00:E0:18:90:1B:56 CM eth1
WARNING: If you receive error message like: SIOCSARP: Invalid argument, it is because the MAC
(Media Access Control) address you want to add is the one of your server. You must add only MAC
address of INTERNAL computers in your private network. This hack doesn’t apply to external
node on the Internet.
You can now be reassured that someone will not change the system's IP address of an
INTERNAL system and get through. If they do change the IP address, the server simply won't
talk to them. With the new iptables tool of Linux, which replace the old ipchains utility for
packet filter administration and firewall setup, MAC addresses can be filtered and configured in the
firewall rules too.
Unusual or hidden files
It is important to look everywhere on the system for unusual or hidden files (files that start with a
period and are normally not shown by the “ls” command), as these can be used to hide tools and
information (password cracking programs, password files from other systems, etc.). A common
technique on UNIX systems is to put a hidden directory or file in a user's account with an unusual
name, something like '...' or '.. ' (dot dot space) or '..^G' (dot dot control-G). The find
program can be used to look for hidden files.
• To look for hidden files, use the following commands:
[root@deep /]# find / -name ".. " -print -xdev
[root@deep /]# find / -name ".*" -print -xdev | cat –v
/etc/skel/.bash_logout
/etc/skel/.bash_profile
/etc/skel/.bashrc
/etc/.pwd.lock
/root/.bash_logout
/root/.Xresources
/root/.bash_profile
/root/.bashrc
/root/.cshrc
/root/.tcshrc
/root/.bash_history
/usr/lib/perl5/5.6.1/i386-linux/.packlist
/home/admin/.bash_logout
/home/admin/.bash_profile
/home/admin/.bashrc
/.autofsck](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-94-320.jpg)
![General Security 0
CHAPTER 3
95
Finding Group and World Writable files and directories
Group and world writable files and directories, particularly system files (partitions), can be a
security hole if a cracker gains access to your system and modifies them. Additionally, world-
writable directories are dangerous, since they allow a cracker to add or delete files as he or she
wishes in these directories. In the normal course of operation, several files will be writable,
including some from the /dev, /var/mail directories, and all symbolic links on your system.
• To locate all group & world-writable files on your system, use the command:
[root@deep /]# find / -type f ( -perm -2 -o -perm -20 ) -exec ls -lg {} ;
-rw-rw-r-- 1 root utmp 107904 Jun 17 12:04 /var/log/wtmp
-rw-rw-r-- 1 root utmp 4608 Jun 17 12:04 /var/run/utmp
• To locate all group & world-writable directories on your system, use the command:
[root@deep /]# find / -type d ( -perm -2 -o -perm -20 ) -exec ls -ldg {} ;
drwxrwxr-x 12 root man 4096 Jun 17 06:50 /var/cache/man/X11R6
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat1
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat2
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat3
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat4
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat5
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat6
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat7
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat8
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat9
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/catn
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat1
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat2
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat3
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat4
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat5
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat6
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat7
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat8
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat9
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/catn
drwxrwxr-x 12 root man 4096 Jun 17 06:50 /var/cache/man/local
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat1
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat2
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat3
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat4
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat5
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat6
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat7
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat8
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat9
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/catn
drwxrwxr-x 3 root lock 4096 Jun 17 06:49 /var/lock
drwxrwxr-x 2 root root 4096 Apr 19 12:35 /var/run/netreport
drwxrwxr-x 2 root 12 4096 Jun 17 12:30 /var/spool/mail
drwxrwxrwt 2 root root 4096 Jun 17 11:29 /var/tmp
drwxrwxrwt 2 root root 4096 Jun 17 06:52 /tmp
WARNING: A file and directory integrity checker like “Tripwire” software can be used regularly to
scan, manage and find modified group or world writable files and directories easily. See later in
this book for more information about Tripwire.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-95-320.jpg)
![96
Unowned files
Don’t permit any unowned file on your server. Unowned files may also be an indication that an
intruder has accessed your system. If you find unowned file or directory on your system, verify its
integrity, and if all looks fine, give it an owner name. Some time you may uninstall a program and
get an unowned file or directory related to this software; in this case you can remove the file or
directory safely.
• To locate files on your system that do not have an owner, use the following command:
[root@deep /]# find / -nouser -o -nogroup
WARNING: It is important to note that files reported under /dev/ directory don’t count.
Finding .rhosts files
Finding all existing .rhosts files that could exist on your server should be a part of your regular
system administration duties, as these files should not be permitted on your system. Remember
that a cracker only needs one insecure account to potentially gain access to your entire network.
Step 1
If you are doing a new install of Linux (like we did) you should not have any .rhosts files on
your system. If the result returns nothing, then you are safe and your system contain no .rhosts
files in the /home directory at this time.
• You can locate all existing .rhosts files on your system with the following command:
[root@deep /]# find /home -name .rhosts
Step 2
You can also use a cron job to periodically check for, report the contents of, and delete
$HOME/.rhosts files. Also, users should be made aware that you regularly perform this type of
audit, as directed by your security policy.
• Create the rhosts.cron file (touch /etc/cron.daily/rhosts.cron) and add the
following lines inside the script file.
#!/bin/sh
/usr/bin/find /home -name .rhosts | (cat <<EOF
This is an automated report of possible existent ..rhosts files on
the server deep.openna.com, generated by the find utility command.
New detected ..rhosts. files under the ./home/. directory include:
EOF
cat
) | /bin/mail -s "Content of .rhosts file audit report" root
• Now make this script executable, verify the owner, and change the group to “root”.
[root@deep /]# chmod 550 /etc/cron.daily/rhosts.cron
[root@deep /]# chown 0.0 /etc/cron.daily/rhosts.cron
Each day mail will be sent to “root” with a subject:” Content of .rhosts file audit report” containing
potential new .rhosts files.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-96-320.jpg)
![General Security 0
CHAPTER 3
97
Physical hard copies of all-important logs
One of the most important security considerations is the integrity of the different log files under
the /var/log directory on your server. If despite each of the security functions put in place on
our server, a cracker can gain access to it; our last defence is the log file system, so it is very
important to consider a method of being sure of the integrity of our log files.
If you have a printer installed on your server, or on a machine on your network, a good idea
would be to have actual physical hard copies of all-important logs. This can be easily
accomplished by using a continuous feed printer and having the syslog program sending all
logs you think are important out to /dev/lp0 (the printer device). Cracker can change the files,
programs, etc on your server, but can do nothing when you have a printer that prints a real paper
copy of all of your important logs.
As an example:
For logging of all telnet, mail, boot messages and ssh connections from your server to the
printer attached to THIS server, you would want to add the following line to the
/etc/syslog.conf file:
Step 1
• Edit the syslog.conf file (vi /etc/syslog.conf) and add at the end of this file the
following line:
authpriv.*;mail.*;local7.*;auth.*;daemon.info /dev/lp0
Step 2
• Now restart your syslog daemon for the change to take effect:
[root@deep /]# /etc/init.d/syslog restart
Shutting down kernel logger: [OK]
Shutting down system logger: [OK]
Starting system logger: [OK]
Starting kernel logger: [OK]
As an example:
For logging of all telnet, mail, boot messages and ssh connections from your server to the
printer attached to a REMOTE server in your local network, then you would want to add the
following line to /etc/syslog.conf file on the REMOTE server.
Step 1
• Edit the syslog.conf file (vi /etc/syslog.conf) on the REMOTE server (for
example: printer.openna.com) and add at the end of this file the following line:
authpriv.*;mail.*;local7.*;auth.*;daemon.info /dev/lp0
If you don’t have a printer in your network, you can also copy all the log files to another machine;
simply omit the above first step of adding /dev/lp0 to your syslog.conf file on remote and go
directly to the “-r” option second step on remote. Using the feature of copying all the log files to
another machine will give you the possibility to control all syslog messages on one host and will
tear down administration needs.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-97-320.jpg)
![98
Step 2
Since the default configuration of the syslog daemon is to not receive any messages from the
network, we must enable on the REMOTE server the facility to receive messages from the
network. To enable the facility to receive messages from the network on the REMOTE server,
add the following option “-r” to your syslog daemon script file (only on the REMOTE host):
• Edit the syslog daemon (vi +24 /etc/rc.d/init.d/syslog) and change:
daemon syslogd -m 0
To read:
daemon syslogd -r -m 0
Step 3
• Restart your syslog daemon on the remote host for the change to take effect:
[root@mail /]# /etc/init.d/syslog restart
Shutting down kernel logger: [OK]
Shutting down system logger: [OK]
Starting system logger: [OK]
Starting kernel logger: [OK]
Step 4
• Edit the syslog.conf file (vi /etc/syslog.conf) on the LOCAL server, and add at
the end of this file the following line:
authpriv.*;mail.*;local7.*;auth.*;daemon.info @printer
Where (printer) represent the hostname of the REMOTE server. Now if anyone ever hacks
your machine and attempts to erase vital system logs, you still have a hard copy of everything. It
should then be fairly simple to trace where they came from and deal with it accordingly.
Step 5
• Restart your syslog daemon on the LOCAL server for the change to take effect:
[root@deep /]# /etc/init.d/syslog restart
Shutting down kernel logger: [OK]
Shutting down system logger: [OK]
Starting system logger: [OK]
Starting kernel logger: [OK]
WARNING: Never use your Gateway Server as a host to control all syslog messages; this is a
very bad idea. More options and strategies exist with the sysklogd program, see the man pages
about sysklogd (8), syslog(2), and syslog.conf(5) for more information.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-98-320.jpg)
![General Security 0
CHAPTER 3
99
Getting some more security by removing manual pages
Here we have to think a little bit about manual pages installed on all Linux system. Manual pages
also known as man-pages are compressed files located under the /usr/share/man directory
on your system. These documentation files are very useful to get quick information on how
service, program, commands, and configuration files of specific software work. These files are
readable by the man program and depend of other installed software on Linux to work and display
the information.
On production servers where specific task are assigned and where we only run services to the
internal or external, does we really need to have these manual pages and related software
installed? Do we will connect to these production servers to read these manual pages? Does this
is really important to have them duplicated on all of our different servers? Personally, I don’t think
because we can have all of these useful documentation files available on our Linux workstation or
development server each time we need to consult them.
If you have made attention to what we have done previously to secure our server, you will
remember that most of all group and world-writable directories on our system comes from the
/var/cache directory which is owned by the man program associated with manual pages. By
removing manual pages and related software from our server, we can get some more security
and save some not negligible space which could help when we scan our server with integrity tool
like Tripwire. This also allow us to remove other software directly related to man program and
limit the number of installed component on our production server without scarifying in the
functionally of the server. If this is what you want to do, here are the steps to follow.
Step 1
First of all, we should remove the man software from our system. The man software is the
program we use to read manual pages. By removing this software we eliminate most of all group
and world-writable directories from our system.
• To remove the man software, use the following command:
[root@deep /]# rpm -e man
Step 2
Once the above software has been removed, we can continue with groff. Groff is a document
formatting system that takes standard text and formatting commands as input and produces
formatted output. This software is used by man to format man-pages.
• To remove the groff software, use the following command:
[root@deep /]# rpm -e groff
Step 3
Because we don’t use manual pages anymore on our production servers, we can remove all
man-pages that are already installed and available under the /usr/share/man directory.
• To remove all preinstalled man-pages from your server, use the following commands:
[root@deep /]# cd /usr/share/man/
[root@deep man]# rm -f man*/*.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-99-320.jpg)
![100
Step 4
Finally, it is important to note that any future installation and upgrade of RPM packages on the
system should be made with the “--excludedocs” option. This RPM option allow us to install or
upgrade the RPM package without the need to install the documentation part that may comes with
the software. For example, if I want to install or upgrade the bind package, I will use the
following RPM command.
• To install or upgrade RPM without documentation, use the following command:
[root@deep /]# rpm –Uvh --exculdedocs bind-version.i386.rpm
System is compromised!
If you believe that your system has been compromised, contact CERT ® Coordination Center or
your representative in FIRST (Forum of Incident Response and Security Teams).
Internet Email: cert@cert.org
CERT Hotline: (+1) 412-268-7090
Facsimile: (+1) 412-268-6989
CERT/CC personnel answer 8:00 a.m. – 8:00 p.m. EST (GMT –5)/EDT (GMT –4)) on working
days; they are on call for emergencies during other hours and on weekends and holidays.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-100-320.jpg)
![Pluggable Authentication Modules 0
CHAPTER 4
105
Disabling console program access
In a safe environment, where we are sure that console is secured because passwords for BIOS
and GRUB or LILO are set and all physical power and reset switches on the system are disabled,
it may be advantageous to entirely disable all console-equivalent access to programs like
poweroff, reboot, and halt for regular users on your server.
• To do this, run the following command:
[root@deep /]# rm -f /etc/security/console.apps/<servicename>
Where <servicename> is the name of the program to which you wish to disable console-
equivalent access.
• To disable console program access, use the following commands:
[root@deep /]# rm -f /etc/security/console.apps/halt
[root@deep /]# rm -f /etc/security/console.apps/poweroff
[root@deep /]# rm -f /etc/security/console.apps/reboot
This will disable console-equivalent access to programs halt, poweroff, and reboot.
Disabling all console access
The Linux-PAM library installed by default on your system allows the system administrator to
choose how applications authenticate users, such as for console access, program and file
access.
Step 1
In order to disable all these accesses for the users, you must comment out all lines that refer to
pam_console.so in the /etc/pam.d directory. This step is a continuation of the hack
“Disabling console program access”. The following script will do the trick automatically for you.
• As ‘root’ creates the disabling.sh script file (touch disabling.sh) and add the
following lines inside:
# !/bin/sh
cd /etc/pam.d
for i in * ; do
sed '/[^#].*pam_console.so/s/^/#/' < $i > foo && mv foo $i
done
Step 2
Now, we have to make the script executable and run it.
• This can be done with the following commands:
[root@deep /]# chmod 700 disabling.sh
[root@deep /]# ./disabling.sh
This will comment out all lines that refer to pam_console.so for all files located under
/etc/pam.d directory. Once the script has been executed, you can remove it from your system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-105-320.jpg)
![108
Step 1
The default console.perms configuration file of Linux is secure enough for regular use of the
system where an Xwindow interface is considered to be installed but in a highly secure
environment where the Graphical User Interface (GUI) is not installed or where some special
devices like sound, jaz, etc have no reason to exist, we can tighten the console.perms
security file of Linux to be more secure by eliminating non-existent or unneeded privileged users
to have capabilities that they would not otherwise have.
• Edit the console.perms file (vi /etc/security/console.perms), and change the
default lines inside this file:
# file classes -- these are regular expressions
<console>=tty[0-9][0-9]* vc/[0-9][0-9]* :[0-9].[0-9] :[0-9]
<xconsole>=:[0-9].[0-9] :[0-9]
# device classes -- these are shell-style globs
<floppy>=/dev/fd[0-1]*
/dev/floppy/* /mnt/floppy*
<sound>=/dev/dsp* /dev/audio* /dev/midi*
/dev/mixer* /dev/sequencer
/dev/sound/* /dev/beep
<cdrom>=/dev/cdrom* /dev/cdroms/* /dev/cdwriter* /mnt/cdrom*
<pilot>=/dev/pilot
<jaz>=/mnt/jaz*
<zip>=/mnt/pocketzip* /mnt/zip*
<ls120>=/dev/ls120 /mnt/ls120*
<scanner>=/dev/scanner /dev/usb/scanner*
<rio500>=/dev/usb/rio500
<camera>=/mnt/camera* /dev/usb/dc2xx* /dev/usb/mdc800*
<memstick>=/mnt/memstick*
<flash>=/mnt/flash*
<diskonkey>=/mnt/diskonkey*
<rem_ide>=/mnt/microdrive*
<fb>=/dev/fb /dev/fb[0-9]*
/dev/fb/*
<kbd>=/dev/kbd
<joystick>=/dev/js[0-9]*
<v4l>=/dev/video* /dev/radio* /dev/winradio* /dev/vtx* /dev/vbi*
/dev/video/*
<gpm>=/dev/gpmctl
<dri>=/dev/nvidia* /dev/3dfx*
<mainboard>=/dev/apm_bios
# permission definitions
<console> 0660 <floppy> 0660 root.floppy
<console> 0600 <sound> 0600 root
<console> 0600 <cdrom> 0660 root.disk
<console> 0600 <pilot> 0660 root.uucp
<console> 0600 <jaz> 0660 root.disk
<console> 0600 <zip> 0660 root.disk
<console> 0600 <ls120> 0660 root.disk
<console> 0600 <scanner> 0600 root
<console> 0600 <camera> 0600 root
<console> 0600 <memstick> 0600 root
<console> 0600 <flash> 0600 root
<console> 0600 <diskonkey> 0660 root.disk
<console> 0600 <rem_ide> 0660 root.disk
<console> 0600 <fb> 0600 root
<console> 0600 <kbd> 0600 root
<console> 0600 <joystick> 0600 root](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-108-320.jpg)
![Pluggable Authentication Modules 0
CHAPTER 4
109
<console> 0600 <v4l> 0600 root
<console> 0700 <gpm> 0700 root
<console> 0600 <mainboard> 0600 root
<console> 0600 <rio500> 0600 root
<xconsole> 0600 /dev/console 0600 root.root
<xconsole> 0600 <dri> 0600 root
To read :
# file classes -- these are regular expressions
<console>=tty[0-9][0-9]* vc/[0-9][0-9]* :[0-9].[0-9] :[0-9]
# device classes -- these are shell-style globs
<floppy>=/dev/fd[0-1]*
/dev/floppy/* /mnt/floppy*
<cdrom>=/dev/cdrom* /dev/cdroms/* /dev/cdwriter* /mnt/cdrom*
<pilot>=/dev/pilot
<fb>=/dev/fb /dev/fb[0-9]*
/dev/fb/*
<kbd>=/dev/kbd
<gpm>=/dev/gpmctl
<mainboard>=/dev/apm_bios
# permission definitions
<console> 0660 <floppy> 0660 root.floppy
<console> 0600 <cdrom> 0660 root.disk
<console> 0600 <pilot> 0660 root.uucp
<console> 0600 <fb> 0600 root
<console> 0600 <kbd> 0600 root
<console> 0700 <gpm> 0700 root
<console> 0600 <mainboard> 0600 root
Here we removed every privileged user related to the Graphical User Interface and others related
to sound, zip drive, jaz drive, scanner, joystick and video media at the physical console
on the server.
Putting limits on resource
The limits.conf file located under the /etc/security directory can be used to control and
limit resources for the users on your system. It is important to set resource limits on all your users
so they can't perform Denial of Service attacks (number of processes, amount of memory, etc) on
the server. These limits will have to be set up for the user when he or she logs in.
Step 1
For example, limits for all users on your system might look like this:
• Edit the limits.conf file (vi /etc/security/limits.conf) and change the lines:
#* soft core 0
#* hard rss 10000
#@student hard nproc 20
To read:
* hard core 0
* hard rss 5000
* hard nproc 35](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-109-320.jpg)
![110
This says to prohibit the creation of core files “core 0”, restrict the number of processes to 35
“nproc 35”, and restrict memory usage to 5M “rss 5000” for everyone except the super user
“root”. All of the above only concerns users who have entered through the login prompt on your
system. With this kind of quota, you have more control on the processes, core files, and memory
usage that users may have on your system. The asterisk “*” mean: all users that logs in on the
server.
Putting an asterisk “*” to cover all users can pose problem with daemon users account like “www”
for a web server, “mysql” for a SQL database server, etc. If we put an asterisk, then, these users
will be affected by the restriction and limitation of processes or memory usage.
To solve the problem, we can choose an existing group name in our system and add every
regular user to this group. In this manner, the restrictions and limitations will apply to all users
who are members of this group name only. A special group account named “users” can be used
for this purpose. This is the recommended method on putting limit on user resources.
• Edit the limits.conf file (vi /etc/security/limits.conf) and change the lines:
#* soft core 0
#* hard rss 10000
#@student hard nproc 20
To read:
@users hard core 0
@users hard rss 5000
@users hard nproc 35
If you decide to use a group name like “@users” to control and limit resources for the users on
your system, then it is important to not forget to change the GUI (Group User ID) of these users
to be “100”. “100” is the numeric value of the user’s ID “users”.
• The command to create a new user with group name which is set by default to users is:
[root@deep /]# useradd -g100 admin
The “-g100” option represents the number of the user’s initial login group and in our case “100”
is the group account name “users”. The “admin” parameter is the user name we want to add to
the group name “users”.
WARNING: Use the same command above for all users on your system you want to be member of
the “users” group account. It is also preferable to set this parameter first before adding users to
the system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-110-320.jpg)
![112
Blocking; su to root, by one and sundry
The su (Substitute User) command allows you to become other existing users on the system. For
example you can temporarily become ‘root’ and execute commands as the super-user ‘root’.
Step 1
If you don’t want anyone to su to root or want to restrict the su command to certain users then
uncomment the following line of your su configuration file in the /etc/pam.d directory. We
highly recommend that you limit the persons allowed to su to the root account.
• Edit the su file (vi /etc/pam.d/su) and uncomment the following line in the file:
auth required /lib/security/pam_wheel.so use_uid
After this line has been uncommented, the /etc/pam.d/su file should look like this:
#%PAM-1.0
auth sufficient /lib/security/pam_rootok.so
auth required /lib/security/pam_wheel.so use_uid
auth required /lib/security/pam_stack.so service=system-auth
account required /lib/security/pam_stack.so service=system-auth
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
session optional /lib/security/pam_xauth.so
Which means only those who are members of the “wheel” group can su to root; it also includes
logging. Note that the “wheel” group is a special account on your system that can be used for
this purpose. You cannot use any group name you want to make this hack. This hack combined
with specifying which TTY and VC devices super-user root is allowed to login on will improve
your security a lot on the system.
Step 2
Now that we have defined the “wheel” group in our /etc/pam.d/su file configuration, it is time
to add some users who will be allowed to su to super-user “root” account.
• If you want to make, for example, the user “admin” a member of the “wheel” group, and
thus be able to su to root, use the following command:
[root@deep /]# usermod -G10 admin
Which means “G” is a list of supplementary groups, where the user is also a member of. “10” is
the numeric value of the user’s ID “wheel”, and “admin” is the user we want to add to the
“wheel” group. Use the same command above for all users on your system you want to be able
to su to super-user “root” account.
NOTE: For Linux users, who use the Xwindow interface, it is important to note that if you can't su
in a GNOME terminal, it’s because you’ve used the wrong terminal. (So don't think that this advice
doesn't work simply because of a GNOME terminal problem!)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-112-320.jpg)
![Pluggable Authentication Modules 0
CHAPTER 4
113
Facultative:
A special line exists in the su file /etc/pam.d/su which allows you to implicitly trust users in the
“wheel” group (for security reasons, I don’t recommend using this option). This mean that all
users who are members of the “wheel” group can su to root without the need to enter the
super-user “root” password.
• To allow users who are members of the “wheel” group to su to root account without the
need to enter the “root” password, edit the su file (vi /etc/pam.d/su) and
uncomment the following line in the file:
auth sufficient /lib/security/pam_wheel.so trust use_uid
After this line has been uncommented, the /etc/pam.d/su file should look like this:
#%PAM-1.0
auth sufficient /lib/security/pam_rootok.so
auth sufficient /lib/security/pam_wheel.so trust use_uid
auth required /lib/security/pam_stack.so service=system-auth
account required /lib/security/pam_stack.so service=system-auth
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
session optional /lib/security/pam_xauth.so
Using sudo instead of su for logging as super-user
There is a security tool called “sudo” that we discuss into this book. This security software allow
us to archive the same result as using the su command to get root privilege on the server but in
a more secure and informative way. With sudo installed in our server, we can get information
about who is connected as super-user root as well as many other useful features. Please see
the chapter related to this security program into this book for more information about sudo.
If you want to use sudo to allow and control which is allowed to logging as super-user root on
your server, then you no longer need to use the su command of Linux to archive this task and we
can remove the SUID bit on this command to completely disable su and use sudo.
This let us remove one more SUID bit on our secure server and have a more complete and
powerful security software to control access to super-user root. This is the method I highly
recommend you to use instead of the su command of Linux.
Step 1
To archive this result, we have to remove the SUID bit of the su command and install the sudo
security software as explained further down in this book. This also implies that we don’t need to
modify the above su configuration file on our system. To recap, all we need to do is to remove
the SUID bit on the su command, and install sudo in our server.
• To remove the SUID bit on the su binary, use the following command:
[root@deep /]# chmod a-s /bin/su](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-113-320.jpg)
![General Optimization 0
CHAPTER 5
121
Some misunderstanding in the compiler flags options
At lot of discussions exist in the Linux community about the “-O” option and its level numbers.
Some Linux users try to convince that level number up to “-O3” like “-O9” will produce faster
program. The “-O9” flag doesn't do anything over “-O3”, if you don't believe me make a small file,
call it testO3.c and see:
Step 1
• Create the testO3.c file with the following command:
[root@deep tmp]# touch testO3.c
Step 2
• Run the GCC compiler with “-O3” flag through the testO3.c file with the command:
[root@deep tmp]# gcc -O3 -S -fverbose-asm testO3.c
Step 3
Look at testO3.s that it made, then run again with “-O9” and compare the output.
• Create the testO9.c file with the following command:
[root@deep tmp]# touch testO9.c
Step 4
• Run the GCC compiler again with “-O9” flag through the testO9.c file with command:
[root@deep tmp]# gcc -O9 -S -fverbose-asm testO9.c
Step 5
Now if you compare the output you will see no difference between the both files.
• To compare the output, use the following command:
[root@deep tmp]# diff testO3.s testO9.s > difference
WARNING: The “-O3” flag level number is the best and highest optimization flag you can use
during optimization of programs under Linux.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-121-320.jpg)
![General Optimization 0
CHAPTER 5
122
The gcc specs file
The /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file of Red Hat Linux is a set of
defines that the gcc compiler uses internally to set various aspects of the compile environment.
All customizations that you put in this file will apply for the entire variable environment on your
system, so putting optimization flags in this file is a good choice.
To squeeze the maximum performance from your x86 programs, you can use full optimization
when compiling with the “-O3” flag. Many programs contain “-O2” in the Makefile. The “-O3”
level number is the highest level of optimization. It will increase the size of what it produces, but it
runs faster in most case. You can also use the “-march=cpu_type” switch to optimize the
program for the CPU listed to the best of GCC’s ability. However, the resulting code will only be run
able on the indicated CPU or higher.
Below are the optimization flags that we recommend you to put in your /usr/lib/gcc-
lib/i386-redhat-linux/2.96/specs file depending on your CPU architecture. The
optimization options apply only when we compile and install a new program in our server. These
optimizations don’t play any role in our Linux base system; it just tells our compiler to optimize the
new programs that we will install with the optimization flags we have specified in the
/usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file. Adding options listed below
depending of your CPU architecture to the gcc 2.96 specs file will save you having to change
every CFLAGS in future Makefiles.
Step 1
The first thing to do is to verify the compiler version installed on your Linux server.
• To verify the compiler version installed on your system, use the command:
[root@deep /]# gcc -v
Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs
gcc version 2.96 20000731 (Red Hat Linux 7.3 2.96-110)
Step 2
For CPU i686 or PentiumPro, Pentium II, Pentium III, and Athlon
Edit the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file, scroll down a ways...
You'll see a section like the following:
*cpp_cpu_default:
-D__tune_i386__
*cpp_cpu:
-Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__
%{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__
%{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__
%{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro -
D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__
%{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:-
D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:-
D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__
}%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:-
D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__
}%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}}
*cc1_cpu:
%{!mcpu*: %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium}
%{mpentiumpro:-mcpu=pentiumpro}}](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-122-320.jpg)
![General Optimization 0
CHAPTER 5
126
Step3
Once our optimization flags have been applied to the gcc 2.96 specs file, it time to verify if the
modification work.
• To verify if the optimization work, use the following commands:
[root@deep tmp]# touch cpu.c
[root@deep tmp]# gcc cpu.c -S -fverbose-asm
[root@deep tmp]# less cpu.s
What you'll get is a file that contains depending of options you have chose, something like:
.file "ccnVPjeW.i"
.version "01.01"
# GNU C version 2.96 20000731 (Red Hat Linux 7.3 2.96-110) (i386-redhat-linux)
compiled by GNU C version 2.96 20000731 (Red Hat Linux 7.3 2.96-110).
# options passed: -O2 -march=i686 -funroll-loops -fverbose-asm
# options enabled: -fdefer-pop -foptimize-sibling-calls -fcse-follow-jumps
# -fcse-skip-blocks -fexpensive-optimizations -fthread-jumps
# -fstrength-reduce -funroll-loops -fpeephole -fforce-mem -ffunction-cse
# -finline -fkeep-static-consts -fcaller-saves -fpcc-struct-return -fgcse
# -frerun-cse-after-loop -frerun-loop-opt -fdelete-null-pointer-checks
# -fschedule-insns2 -fsched-interblock -fsched-spec -fbranch-count-reg
# -fnew-exceptions -fcommon -fverbose-asm -fgnu-linker -fregmove
# -foptimize-register-move -fargument-alias -fstrict-aliasing
# -fmerge-constants -fident -fpeephole2 -fmath-errno -m80387 -mhard-float
# -mno-soft-float -mieee-fp -mfp-ret-in-387 -march=i686
gcc2_compiled.:
.ident "GCC: (GNU) 2.96 20000731 (Red Hat Linux 7.3 2.96-110)"
WARNING: In our example we are optimized the specs file for a i686 CPU processor. It is important
to note that most of the “-f” options are automatically included when you use “-O2” and don't
need to be specified again. The changes that were shown were made so that a command like
"gcc" would really be the command "gcc -march=i686" without having to change every single
Makefile which can really be a pain.
Below is the explanation of the different optimization options we use:
• The “-march=cpu_type” optimization flag
The “-march=cpu_type” optimization option will set the default CPU to use for the
machine type when scheduling instructions.
• The “-funroll-loops” optimization flag
The “-funroll-loops” optimization option will perform the optimization of loop
unrolling and will do it only for loops whose number of iterations can be determined at
compile time or run time.
• The “-fomit-frame-pointer” optimization flag
The “-fomit-frame-pointer” optimization option, one of the most interesting, will
allow the program to not keep the frame pointer in a register for functions that don't need
one. This avoids the instructions to save, set up and restores frame pointers; it also
makes an extra register available in many functions and makes debugging impossible on
most machines.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-126-320.jpg)
![General Optimization 0
CHAPTER 5
127
WARNING: All future optimizations that we will describe in this book refer by default to a Pentium
PRO/II/III and higher i686 CPU family. So you must adjust the compilation flags for your specific
CPU processor type in the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file and
during your compilation time.
Striping all binaries and libraries files
When compiler builds program it add many comments and other stuffs inside the resulting binary
and library code which are used to debug application on the system or to make the code more
readable by developers. To get the latest bit of optimization, we can remove all comments and
other unneeded stuffs since there are only used for debugging purpose. This will make the
software runs a little bit faster because it will have fewer codes to read when executing.
I don’t know if it’s a good idea to talk about this hack because it’s really dangerous to apply and
can make your system unstable or completely unworkable if you don’t take care of what you do.
The process of eliminating all unneeded comments and other unneeded stuffs from your binaries
and libraries files is made by the use of the strip command of Linux. This command should be
used with care and in the good manner or you will certainly have a bad surprise.
Bellow, I will explain you how to apply it on your system and on which files or where you should
use it. It is very important to know that it’s NOT all binaries and especially libraries files that need
to be striped by this method but ONLY some of them. If you apply this hack on your entire
system, then something will inevitably break, you have been warned.
Finally, you should use this hack on servers where you DON’T compile software. If you compile
software on the server where you want to apply this hack, then nothing will work and you will not
be able to compile any software on it. Use this hack on server which doesn’t have any compiler
packages installed to make compilation.
Step 1
The first step in our procedure will be to be sure that the strip command is available on our
server; this command comes from the “binutils” RPM package. Therefore, if it is not installed,
install it from your CD-ROM.
Step 2
Once the strip program is installed on our server, it’s time to strip the required files. With the
commands below, we strip all binaries program available under the /bin, /sbin, /usr/bin and
/usr/sbin directories of your server.
• To strip all binaries program, use the following commands:
[root@deep /]# strip /bin/*
[root@deep /]# strip /sbin/*
[root@deep /]# strip /usr/bin/*
[root@deep /]# strip /usr/sbin/*
NOTE: When issuing the above commands, you will receive some error messages like “File
format not recognized” on your terminal. This is normal because some of the binaries files
are symbolic link pointing to other binaries on your system and the strip command generate the
warning because it cannot strip symbolic links.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-127-320.jpg)
![General Optimization 0
CHAPTER 5
128
Step 3
Now, it’s time to strip the libraries files. This is where the action can become dangerous if you
don’t take care or abuse the strip command. With the commands below, we strip all libraries
files available under the /lib and /usr/lib directory of the system.
• To strip all libraries files, use the following command:
[root@deep /]# strip -R .comment /usr/lib/*.so.*
[root@deep /]# strip -R .comment /lib/*.so.*
NOTE: Make attention to the above command, you can see here that I use the “-R” option with the
strip command. This option allows us to select a specific name to strip from the target libraries
files. With the “.comment” name, we inform the command to remove any lines inside the libraries
codes where this name appears. You can see that I don’t use the strip command without any
option as I do for the above step related to binaries program. This is very important and you
should never use the strip command without the above option to strip libraries files on your
system.
Tuning IDE Hard Disk Performance
Accessing a hard disk can be 50 to 100 times slower than reading data from RAM. File caches
using RAM can alleviate this. However, low memory conditions will reduce the amount of memory
available for the file-system cache slowing things down. File systems can also become heavily
fragmented, slowing down disk accesses. Heavy use of symbolic links on Unix systems can slow
down disk accesses too.
Default Linux installs are also notorious for setting hard disk default settings which are tuned for
compatibility and not for speed. Use the command hdparm to tune your Linux hard disk settings.
The hdparm is a tool, which can be used to tune and improve the performance of your IDE hard
disk. By default, any IDE drives you have in your Linux system are not optimized. Even if you
have an ULTRA DMA system you will not be able to take full advantage of its speed if you are not
using the hdparm tool to enable its features. This is because there is many different hard drive
makes and models and Linux cannot know every feature of each one.
Performance increases have been reported on massive disk I/O operations by setting the IDE
drivers to use DMA, 32-bit transfers and multiple sector modes. The kernel seems to use more
conservative settings unless told otherwise. The magic command to change the setting of your
drive is hdparm.
Before going into the optimization of your hard drive, it is important to verify that the hdparm
package is installed in your system. If you have followed every step during the installation of
Linux on your computer, then this package is not installed.
To verify if hdparm package is installed on your system, use the command:
[root@deep /]# rpm -q hdparm
package hdparm is not installed
If the hdparm package seems not to be installed, you’ll need to mount your CD-ROM drive
containing the Linux CD-ROM Part 1 and install it.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-128-320.jpg)
![General Optimization 0
CHAPTER 5
129
• To mount the CD-ROM drive, use the following commands:
[root@deep /]# mount /dev/cdrom /mnt/cdrom/
had: ATAPI 32X CD-ROM drive, 128kB Cache
mount: block device dev/cdrom is write-protected, mounting read-only
• To install the hdparm package on your Linux system, use the following command:
[root@deep /]# cd /mnt/cdrom/RedHat/RPMS/
[root@deep RPMS]# rpm -Uvh hdparm-version.i386.rpm
hdparm ##################################################
• To unmount your CD-ROM drive, use the following command:
[root@deep RPMS]# cd /; umount /mnt/cdrom/
Once hdparm package is installed on the system, it is time to go into the optimization of your
hard drive. It is important to note that depending on your model and make, there will be some
parameters that will apply and other that don’t. It is to your responsibility to know and understand
your disk drive before applying any optimization parameters as described below.
Finally, and especially for UltraDMA systems, it is vital to verify under your BIOS settings if the
parameters related to DMA support on your computer are enabled or you will inevitably break your
hard disk. You have been warned.
Step 1
The first parameter applies to the majority of all modern drives and models in the market and
enables 32-bit I/O over PCI buses. This option is one of the most important and will usually
double the speed of your drive.
• To enable 32-bit I/O over the PCI buses, use the following command:
[root@deep /]# /sbin/hdparm -c3 /dev/hda (or hdb, hdc etc).
This will usually, depending on your IDE Disk Drive model, cut the timing buffered disk reads time
by two. The hdparm (8) manpage says that you may need to use “-c3” for many chipsets since it
works with nearly all 32-bit IDE chipsets. All (E)IDE drives still have only a 16-bit connection
over the ribbon cable from the interface card.
Step 2
The second parameter applies only on standard DMA disk and will activate the simple DMA feature
of the disk. This feature is for old disk drives with DMA capabilities.
• To enable DMA, use the following command:
[root@deep /]# /sbin/hdparm -d1 /dev/hda (or hdb, hdc etc).
This may depend on support for your motherboard chipset being compiled into your kernel. Also,
this command will enable DMA support for your hard drive only for interfaces which support DMA, it
will cut the timing buffered disk reads time and will improve the performance by two.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-129-320.jpg)
![General Optimization 0
CHAPTER 5
130
Step 3
Multiword DMA mode 2, also known as ATA2 disk drive is the successor of the simple DMA drive. If
you have this kind of hard drive, then you must enable the parameter in your Linux system.
• To enable multiword DMA mode 2 transfers, use the following command:
[root@deep /]# /sbin/hdparm -d1 -X34 /dev/hda (or hdb, hdc etc).
This sets the IDE transfer mode for newer (E)IDE/ATA2 drives. (Check your hardware manual
to see if you have it).
Step 4
As for DMA mode 2, the UltraDMA mode 2 is an improvement of the DMA technology. If you have
this kind of drive in your system, then choose this mode.
• To enable UltraDMA mode 2 transfers, use the following command:
[root@deep /]# /sbin/hdparm -d1 -X66 /dev/hda (or hdb, hdc etc)
See your manual page about hdparm for more information. USE THIS OPTION WITH EXTREME
CAUTION!
Step 5
The UltraDMA mode 4 is one of the latest entries and one of the most popular at this time; it is
also known and referred as ATA/66. I guess that most of you have this kind of drive installed and
if it is the case then it is the one that you must choose for sure.
• To enable UltraDMA mode4 transfers, use the following command:
[root@deep /]# /sbin/hdparm -d1 -X12 -X68 /dev/hda (or hdb, hdc etc)
This will enable UltraDMA ATA/66 mode on your drive. See your manual page about hdparm
for more information. USE THIS OPTION WITH EXTREME CAUTION!
Step 6
Multiple sector mode (aka IDE Block Mode), is a feature of most modern IDE hard drives,
permitting the transfer of multiple sectors per I/O interrupt, rather than the usual one sector per
interrupt. When this feature is enabled, it typically reduces operating system overhead for disk I/O
by 30-50%. On many systems it also provides increased data throughput of anywhere from 5% to
50%.
• To set multiple sector mode I/O, use the following command:
[root@deep /]# /sbin/hdparm -mXX /dev/hda (or hdb, hdc etc)
Where “XX” represent the maximum setting supported by your drive. The “-i” flag can be used to
find the maximum setting supported by an installed drive: look for MaxMultSect in the output.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-130-320.jpg)
![General Optimization 0
CHAPTER 5
131
• To find the maximum setting of your drive, use the following command:
[root@deep /]# /sbin/hdparm -i /dev/hda (or hdb, hdc etc)
/dev/hda:
Model=QUANTUM FIREBALLP LM15, FwRev=A35.0700, SerialNo=883012661990
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
BuffType=3(DualPortCache), BuffSize=1900kB, MaxMultSect=16, MultSect=16
DblWordIO=no, OldPIO=2, DMA=yes, OldDMA=2
CurCHS=16383/16/63, CurSects=-66060037, LBA=yes, LBAsects=29336832
tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
UDMA modes: mode0 mode1 mode2 mode3 *mode4
Step 7
The get/set sector count is used to improve performance in sequential reads of large files! The
default setting is 8 sectors (4KB) and we will double and change it for 16. USE THIS OPTION
WITH EXTREME CAUTION!
• To improve the get/set sector count for file system read-ahead, use the command:
[root@deep /]# /sbin/hdparm -a16 /dev/hda (or hdb, hdc etc)
Step 8
The get/set interrupt-unmask flag will greatly improve Linux's responsiveness and eliminates
"serial port overrun" errors. USE THIS OPTION WITH EXTREME CAUTION!
• To improve and get/set interrupt-unmask flag for the drive, use the command:
[root@deep /]# /sbin/hdparm -u1 /dev/hda (or hdb, hdc etc)
Step 9
The IDE drive's write-caching feature will improve the performance of the hard disk. USE THIS
OPTION WITH EXTREME CAUTION!
• To enable the IDE drive's write-caching feature, use the following command:
[root@deep /]# /sbin/hdparm -W1 /dev/hda (or hdb, hdc etc)
Step 10
These options will allow the drive to retain your settings over a soft reset (as done during the error
recovery sequence). It is important to note that not all drives support this feature.
• To enables the drive to retain your settings, use the command:
[root@deep /]# /sbin/hdparm -K1 -k1 /dev/hda (or hdb, hdc etc)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-131-320.jpg)
![General Optimization 0
CHAPTER 5
132
Step 11
Once every tuning related to your specific drive have been set, you can test the results and see if
you want to keep them or not.
• You can test the results of your changes by running hdparm in performance test mode:
[root@deep /]# /sbin/hdparm -vtT /dev/hda (or hdb, hdc etc).
/dev/hda:
multcount = 16 (on)
I/O support = 3 (32-bit w/sync)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 1 (on)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 16 (on)
geometry = 1826/255/63, sectors = 29336832, start = 0
Timing buffer-cache reads: 128 MB in 0.85 seconds = 150.59 MB/sec
Timing buffered disk reads: 64 MB in 2.54 seconds = 25.20 MB/sec
Once you have a set of hdparm options, you can put the commands in your /etc/rc.local file
to run it every time you reboot the machine. When running from /etc/rc.local, you can add
the “-q” option for reducing screen clutter. In my case, I will put the following configuration in the
end of my rc.local file:
/sbin/hdparm -q -c3 -d1 -X12 -X68 -m16 -a16 -u1 -W1 -k1 -K1 /dev/had](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-132-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
136
A solution was found and this was the Modularized Kernel approach. A technique that allows
small pieces of compiled code to be inserted in or removed from the running kernel. In this way
the Kernel will only load and run drivers and features that your computer have and will forget
about the others. This practice is what all Linux vendors use to provide Linux kernels. They build
and link every driver and feature as a module (which keeps the binary kernel smaller) that can be
recognized and loaded if, and only if, they are needed by the kernel or the system.
Kernel developers provide the ability to build a Modularized kernel, through an option that asks
you during kernel configuration if you want to build the available drivers/features as a module.
This option appears at the beginning of the Kernel configuration in the following form "Enable
loadable module support (CONFIG_MODULES) [Y/n/?]". If you answer "Yes" here,
then the compiled Kernel will be a Modularized Kernel and all future questions appearing during
kernel configuration will give you the choice to compile the drivers/features into the Kernel code
as a module by answering "m" for module, "y" for yes includes the code, or "n" do not include the
code. Alternatively, if you answer "No" to the question "Enable loadable module support
(CONFIG_MODULES) [Y/n/?]", then the corresponding Kernel will be a Monolithic kernel and
all future questions appearing during kernel configuration will let you answer either "y" (yes,
include the driver/feature) or "n" (no, do not include the drivers/feature). This allows you to build a
Kernel where every driver/feature is compiled into it.
To recap, Modularized Kernels allow small pieces of compiled code, which reside under the
/lib/modules/2.4.x-x/ kernel directory to be inserted into or removed from the running
kernel and a Monolithic Kernel contains the drivers/features into its compiled code.
Some people will say that a loadable module is as good as hard-linked code. But what sort of
speed difference is seen when using loadable modules instead of hard-linked code? Well, here’s
an extract of the kernel mailing list archive:
The immediate response from some was "almost nothing," but further consideration has shown
this not to be true. There are, in fact, a number of costs associated with loadable modules. The
biggest, perhaps, relates to how loadable modules are placed in kernel memory. The code for a
module needs to live in a contiguous address space. The kernel sets up that address space with
a function called vmalloc, which allocates memory with virtual addresses. In other words, a
loadable module is in an address space that is visible to the kernel, but which is separate from
where the core kernel code goes. This difference is important. The core kernel address space is a
direct map of physical memory; it can be handled very efficiently in the processor's page table.
Indeed, on some processors, a single page table entry covers the entire kernel. Space obtained
from vmalloc, instead, uses one page table entry per memory page. A greater number of page
table entries mean more lookups, and more translation buffer misses.
One estimate is that the slowdown can be as much as 5%. Given this problem, why not load
modules into the regular kernel memory space? Module code requires a contiguous address
space. Since the standard kernel space is a direct map of physical memory, contiguous address
spaces must also be contiguous in physical memory. Once the system has been running for a
while, finding even two physically contiguous pages can be a challenge; finding enough to load a
large module can be almost impossible. Modules also seem to have endemic problems with race
conditions - it is possible, for example, for the kernel to attempt to access a newly-loaded module
before it is fully initialized. Modules can also, in some situations, be removed while still in use.
Such occurrences are obviously quite rare, but they can be catastrophic when they happen. The
race conditions can be fixed with enough work, but that may require changing some fundamental
kernel interfaces. In general, dealing with loadable modules is not an easy task; as one kernel
hacker told us in a private message: "Doing live surgery on the kernel is never going to be pretty."](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-136-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
138
Pristine source
If you don’t use the RPM package to install the kernel, it will be difficult for you to locate all the files
installed onto the system if you want to update your kernel in the future. To solve this problem, it’s
a good idea to make a list of files on the system before you install the kernel, and then one
afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the kernel:
[root@deep root]# find /* > Kernel1
• And the following one after you install the kernel:
[root@deep root]# find /* > Kernel2
• Then use this command to get a list of what changed:
[root@deep root]# diff Kernel1 Kernel2 > Kernel-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new kernel. In our example above, we use the /root directory of the system
to store all the generated file lists.
Making an emergency boot floppy
The first step before going into the configuration and compilation of our new kernel is to create an
emergency boot floppy in case something goes wrong during the build of your new Linux Kernel.
Here we create the boot floppy for a modularized kernel since the kernel that is presently installed
on our system should be a modularized kernel. We’ll see later that a method for creating the boot
disk for mololithic kernel exists and is different from what we use here. To create the emergency
boot floppy for a modularized kernel, follow these steps.
Step1
We have to find the present modularized kernel version used on our system. We need this
information to be able to create our emergency boot floppy disk.
• To know which kernel version is running on your system, use the command:
[root@dev /]# uname -a
Linux dev 2.4.18-3 #1 Thu Apr 18 07:37:53 EDT 2002 i686 unknown
From the above command, we know now that our kernel version is 2.4.18-3. Therefore we will
use this information in the next step to create the boot disk.
Step2
Once we know which kernel version we are currently running, we can use the command below to
create the boot disk.
• Put a floppy in your system and execute the following command as root:
[root@deep /]# mkbootdisk --device /dev/fd0H1440 2.4.18-3
Insert a disk in /dev/fd0. Any information on the disk will be lost.
Press <Enter> to continue or ^C to abort:](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-138-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
139
NOTE: In this example, the current kernel on our system is version 2.4.18-3 and this is why we
use “2.4.18-3” here. If you kernel version is different from what we use here, you just have to
change the example version number for the one that you have according to the result returned by
the “uname -a” command.
Following these guidelines, you will now have a boot floppy with a known working kernel in case
of problems with the upgrade. I recommend rebooting the system with the floppy to make sure
that the floppy works correctly before continuing.
Preparing the Kernel for the installation
We have to copy the new kernel tar archive to the appropriate location on our server /usr/src
and then we remove the old kernel from our system before installing a new one. Removing the
old kernel will not freeze your computer until you try to reboot it before installing the new one
because the Linux kernel resides in memory.
Step 1
We must copy the archive file of the kernel to the /usr/src directory and move to this directory.
• To copy the tar archive of the Linux kernel to the /usr/src directory, use the command:
[root@deep /]# cp linux-version.tar.gz /usr/src/
• To move to the /usr/src directory, use the following command:
[root@deep /]# cd /usr/src/
Step 2
Depending on how the Linux Kernel has been previously installed on your system, there are two
ways to uninstall it, these are shown below.
If you already have installed a Linux kernel with a tar archive before
These steps are required ONLY if you have previously installed a Linux kernel from a tar archive.
If it is a fresh, first install of the kernel, then uninstall the kernel-headers-version.
i386.rpm, kernel-version.i386.rpm packages that are on your system.
• Move to the /usr/src directory if you are not already in it with the following command:
[root@deep /]# cd /usr/src/
• Remove the Linux symbolic link with the following command:
[root@deep src]# rm -f linux
• Remove the Linux kernel headers directory with the following command:
[root@deep src]# rm -rf linux-2.4.x/
• Remove the Linux kernel with the following command:
[root@deep src]# rm -f /boot/vmlinuz-2.4.x
• Remove the Linux System.map file with the following command:
[root@deep src]# rm -f /boot/System.map-2.4.x
• Remove the Linux kernel modules directory (if available) with the following command:
[root@deep src]# rm -rf /lib/modules/2.4.x/](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-139-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
140
NOTE: Removing the old kernel modules is only required if you have installed a modularized
kernel version before. If the modules directory doesn’t exist under the /lib/modules directory,
it’s because your old kernel version is not a modularized kernel.
If the original kernel’s RPM packages are installed on your system
If the original kernel RPM packages are installed on your system, instead of the Linux kernel tar
archive, which they would be if you have just finished installing your new Linux system, or have
previously used an RPM package to upgrade your system, then use the following command to
uninstall the Linux kernel:
• Verify which kernel RPM packages are installed on your server with the command:
[root@deep src]# rpm -qa | grep kernel
kernel-2.4.18-3
• Also, verify if Kernel header package is installed on your server with the command:
[root@deep src]# rpm -q glibc-kernheaders
glibc-kernheaders-2.4-7.14
The above command shows us that kernel and glibc-kernheaders are the only kernel RPM
packages installed on our system. We uninstall them as show below.
• To uninstall the linux kernel RPM, use the following command:
[root@deep src]# rpm -e --nodeps kernel glibc-kernheaders
NOTE: If you receive an error message like: cannot remove /lib/modules/2.4.x directory,
directory not empty, then remove the directory manually with command like: rm –rf
/lib/modules/2.4.x/ form your system. This directory is related to the old kernel and it is not
required for the new kernel we want to install.
Step 3
Once we have uninstalled the old kernel and our new kernel tar archive has been copied to the
/usr/src directory, we must uncompress it and remove the tar archive (linux-version.
tar.gz) from the system to conserve disk space.
• To uncompress the kernel, use the following command:
[root@deep src]# tar xzpf linux-version.tar.gz
• To remove the kernel tar archive from the system, use the following command:
[root@deep src]# rm -f linux-version.tar.gz
WARNING: If kernel compilation is something new for you, then it is recommended to keep the
kernel tar archive (linux-version.tar.gz) until the end of the installation. This way, if you
make a mistake during compilation, you have the source available to try again.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-140-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
141
Applying the Grsecurity kernel patch
Grsecurity is a single patch file for the newest stable versions of Linux kernel that is an
attempt to greatly improve the security of a Linux system. It mainly accomplishes this by
physically patching the Linux kernel to make processes more restricted.
Many other projects like Grsecurity exist on the Internet. There are some well know like the
RSBAC (www.rsbac.org), LIDS (www.lids.org), SELinux (www.nsa.gov/selinux/), and OpenWall
(www.openwall.com/linux/), projects, all of which fulfills only one part of a complete security
system.
What makes Grsecurity so different and better than these other projects is mainly because
Grsecurity provides greatly needed additional security to Linux systems. In other words, it
covers all features of the other projects and adds additional security features that the other
projects do NOT cover. The types of added Grsecurity security are categorized as:
1. Complete Buffer Overflow Exploitation Protection.
2. File system Race Protection.
3. System Auditing.
4. Change-root Protections.
5. Portscan and OS Fingerprinting Protection.
6. Restricted Users.
7. Configurability Options.
8. Access Control.
Grsecurity patch may change from version to version, and some may contain various other
security features. It is important to use the Grsecurity patch that corresponds to the Linux
Kernel version that you compile on your server. If you compile kernel version 2.4.18, you have to
download Grsecurity patch for kernel version 2.4.18, etc.
WARNING: When applying the Grsecurity patch to the Kernel, a new “Grsecurity”
configuration section will be added at the end of your Linux Kernel configuration allowing you to
configure and enable the security features that you want.
Obtaining and Installing Grsecurity
To obtain the Grsecurity patch suitable for your Linux Kernel, simply follow the "download" link
on the Grsecurity home page, and download the file listed there. Grsecurity patch is
available directly via http://www.grsecurity.org/. Remember that Grsecurity is specific to one
kernel version, and as of this writing that version is 2.4.18.
• To apply the Grsecurity patch to the Linux kernel, use the commands:
[root@deep /]# cp grsecurity-1.9.4-2.4.18.patch /usr/src/
[root@deep /]# cd /usr/src/linux/
[root@deep linux]# patch -p1 < ../grsecurity-1.9.4-2.4.18.patch
[root@deep linux]# cd ../
[root@deep src]# rm -f grsecurity-1.9.4-2.4.18.patch
The step of patching your new kernel with Grsecurity patch is completed. Now follow the rest
of this Kernel installation guide to build the Linux kernel and reboot your system. Configuration
relating to Grsecurity will appears at the end of your Kernel configuration.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-141-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
143
• Edit the Makefile file (vi +20 /usr/src/linux/Makefile) and change the line:
HOSTCFLAGS = -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer
To read:
HOSTCFLAGS = -Wall -Wstrict-prototypes -O2 -march=i686 -funroll-loops -
fomit-frame- pointer
• Edit the Makefile file (vi +91 /usr/src/linux/Makefile) and change the line:
CFLAGS := $(CPPFLAGS) -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer
-fno-strict-aliasing
To read:
CFLAGS := $(CPPFLAGS) -Wall -Wstrict-prototypes -O2 -march=i686 -funroll-
loops -fomit-frame-pointer -fno-strict-aliasing
WARNING: In the last example, we optimize the code for an i686 CPU architecture, if you have a
different processor, you'll have to adjust the “-march=i686" options for your specific processor.
Never compile the Kernel with optimization code number superior to “-O2”, this do nothing more
and should produce an unstable kernel in some cases. Therefore use “-O2” at the maximum
optimization number but never something superior to it.
Cleaning up the Kernel
It is important to make sure that your /usr/include/asm, and /usr/include/linux
subdirectories are just symlinks to the kernel sources.
Step 1
The asm and linux subdirectories are soft links to the real include kernel source header
directories needed for our Linux architecture, for example /usr/src/linux/include/asm-
i386 for asm.
• To symlink the asm, and linux subdirectories to the kernel sources, type the following
commands on your terminal:
[root@deep src]# cd /usr/include/
[root@deep include]# rm -f asm linux
[root@deep include]# ln -s /usr/src/linux/include/asm-i386 asm
[root@deep include]# ln -s /usr/src/linux/include/linux linux
This is a very important part of the configuration: we remove the asm, and linux directories
under /usr/include then create new links that point to the same name directories under the
new kernel source version directory. The /usr/include directory contains important header
files needed by your kernel and programs to be able to compile on your system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-143-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
144
WARNING: If the previously installed kernel in your system was made by RPM packages, then the
asm and linux soft links will not exist since the uninstall of kernel-headers RPM package
removes them automatically for you. Don’t forget to create them.
Step 2
Make sure you have no stale .o files or dependencies lying around.
• To be sure that we have no stale .o files or dependencies lying around, type the
following commands on your terminal:
[root@deep include]# cd /usr/src/linux/
[root@deep linux]# make mrproper
WARNING: These two steps simply clean up anything that might have accidentally been left in the
source tree by the development team.
You should now have the source correctly installed. You can configure the kernel in one of three
ways. The first method is to use the make config command. It provides you with a text-based
interface for answering all the configuration options. You are prompted for all the options you
need to set up your kernel.
The second method is to use the make menuconfig command, which provides all the kernel
options in an easy-to-use menu. The third is to use the make xconfig command (only available
if the graphical interface of Linux is installed on the system), which provides a full graphical
interface to all the kernel options.
Step 3
For configuration in this guide, we will use the make config command because we have not
installed the XFree86 Window Interface on our Linux server or the necessary packages to use the
make menuconfig command.
• Type the following commands on your terminal to load the kernel configuration:
[root@deep /]# cd /usr/src/linux/ (if you are not already in this directory).
[root@deep linux]# make config
rm -f include/asm
( cd include ; ln -sf asm-i386 asm)
/bin/sh scripts/Configure arch/i386/config.in
#
# Using defaults found in arch/i386/defconfig
#](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-144-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
145
Configuring the Kernel
As soon as you enter make config at the prompt as described in the previous step, a list of
kernel configurable options will be displayed for you to choose to configure the kernel, you must
indicate what features and devices drivers you want to include on your system and select how to
include support for specific devices. Typically, for each configuration option, you have to respond
with one of the following choices:
[y] To compile into the kernel and always be loaded.
[m] To use a module for that feature and load that segment of code on demand.
[n] To skip and excludes the support for that specific device from the kernel.
WARNING: It is important to note that an [n] or [y] means the default choice. If a device does not
have a modular device driver or you have not compiled the Kernel as a Modularized Kernel, you
will not see the [m] option. Some time an [?] option will appear in the choices. This mean that
you can get more information about the feature when you type the ? + ENTER key. Choosing the
[?] help option opens another terminal describing the option.
A new Linux kernel is very specific to our computer hardware since we have to choose the right
drivers as well as features that we need to include and compile into the Kernel code. This implies
a good understanding and knowledge of your computer hardware. It is simply inconceivable to
build a Linux system if you don't know what hardware your computer has, especially if you spend
money to buy a computer and then take time to configure it. Therefore we assume that you know
all of your hardware and be aware that during the kernel configuration, you will be asked to
answer some important questions related to your specific computer hardware.
Be prepared to answer the following questions:
1. What type of processor do you have on your computer (i.e. Pentium III, AMD)?
2. How many processor do you have on your computer (i.e. 1, 2, 3)?
3. What kind of hard drive do you have on your computer (i.e. IDE, SCSI) ?
4. How many hard drives do you have on your computer? (i.e. want to make RAID)?
5. How much memories (RAM) do you have on your computer (i.e. 512 MB RAM)?
6. Do you have a network card? If so, who made it and what model is it?
7. Do you have a SCSI adapter? If so, who made it and what model is it?
8. Do you have a RAID system? If so, who made it and what model is it?
9. What type of mouse do you have (eg, PS/2, Microsoft, Logitech)?
10. If you have a serial mouse, what COM port is it connected to (eg, COM1)?
11. What is the make and model of your video card?
All of the above questions are very important and if you don't know the answers for all of them,
then it is recommended to get the information before going into Linux kernel configuration,
compilation and installation. If you have all of the information, then you can read the rest of this
guide.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-145-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
146
Monolithic kernel configuration
As we know now, they are two possible different configurations for the kernel. The first is called a
monolithic kernel the second is called a modularized kernel. Below we begin by
showing you the configuration of a monolithic kernel which is to compile the required code
and drivers directly into the kernel by answering the different kernel questions only by yes or no.
Don’t forget to only compile code that you need and use.
A new kernel is very specific to your computer hardware, in the monolithic kernel
configuration part below; we have the following hardware for our example. Of course you must
change them to fit whatever components you have in your system.
1 Pentium II 400 MHz (i686) processor
1 SCSI Motherboard
1 SCSI Hard Disk
1 SCSI Controler Adaptec AIC 7xxx
1 CD-ROM ATAPI IDE
1 Floppy Disk
2 Ethernet Cards Intel EtherExpressPro 10/100
1 Mouse PS/2
If you don’t want some of the options listed in the monolithic kernel configuration that I
enable by default, answer n (for no) instead of y (for yes) to the related questions. If you want
some other options that I disable, then answer y instead of n. Finally, the procedure of building a
new kernel is quite long, therefore I recommend you to take your time. Some coffees and
cigarettes will surely be welcome during these steps.
# Using defaults found in arch/i386/defconfig
#
*
* Code maturity level options
* Prompt for development and/or incomplete code/drivers (CONFIG_EXPERIMENTAL)
[N/y/?] Press Enter
This option relates to features or drivers that are currently considered to be in the alpha-test
phase and for a production system, it is highly recommended to answer by N to this question. Just
hit [Enter] to approve the default choice, which is NO (Do not prompt for development and/or
incomplete code/drivers).
*
* Loadable module support
* Enable loadable module support (CONFIG_MODULES) [Y/n/?] n
This option is very important since it asks us if we want to enable loadable module support into
the Kernel. Remember that our goal in this part of the guide is to build a Monolithic Linux Kernel
where all drivers and features are directly integrated and compiled into the Kernel code, therefore
our answer to this question must be No (Do not enable loadable module support). This means
that we have to enter n as the answer to this question to change the default value, which is Y for
Yes.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-146-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
147
*
* Processor type and features
* Processor family (386, 486, 586/K5/5x86/6x86/6x86MX, Pentium-Classic,
Pentium-MMX, Pentium-Pro/Celeron/Pentium-II, Pentium-III/Celeron(Coppermine),
Pentium-4, K6/K6-II/K6-III, Athlon/Duron/K7, Crusoe, Winchip-C6, Winchip-2,
Winchip-2A/Winchip-3, CyrixIII/C3) [Pentium-III/Celeron(Coppermine)] Pentium-4
This option asks you what type of processor you have on your computer and the default choice is
in brackets [Pentium-III/Celeron(Coppermine)]. Therefore, if you processor is not a
Pentium-III/Celerom(Coppermine) model, you have to enter your processor model here. The
available choices are listed in the question. In the above example, I changed the default value
[Pentium-III/Celeron(Coppermine)] and chose a Pentium-4 model.
Toshiba Laptop support (CONFIG_TOSHIBA) [N/y/?] Press Enter
This option asks you if you want to add a driver support for Toshiba Laptop. If you intend to run
this kernel on a Toshiba portable, then you have to answer Yes to this question and change the
default value which is No. In our example, we keep the default value of No by pressing the
[Enter] key.
Dell Inspiron 8000 support (CONFIG_I8K) [N/y/?] Press Enter
As for the previous option, this one asks you if you want to add a driver support for Dell Inspiron
8000 Laptop. If you intend to run this kernel on a Dell Inspiron 8000, then you have to answer
Yes to this question and change the default value which is No. In our example, we keep the
default value of No by pressing the [Enter] key again.
/dev/cpu/microcode - Intel IA32 CPU microcode support (CONFIG_MICROCODE)
[N/y/?] Press Enter
This option allows you to update the microcode on Intel processors in the IA32 family, e.g.
Pentium Pro, Pentium II, Pentium III, Pentium 4, Xeon etc. If you say Y here and change the
default value of N, then you'll need to say Y also to "/dev file system support" in the 'File
systems' section below. In most situations, this option is simply not needed and you can safety
keep the default setting of N by pressing the [Enter] key.
/dev/cpu/*/msr - Model-specific register support (CONFIG_X86_MSR) [N/y/?] y
This option allows you to enable a device that gives privileged processes access to the x86
Model-Specific Registers (MSRs). On multi-processor the MSR accesses are directed to a
specific CPU. It is a good idea to answer Y to this option to enable it, even if you only have one
processor on your system. Therefore answer Y to this question.
/dev/cpu/*/cpuid - CPU information support (CONFIG_X86_CPUID) [N/y/?] y
This option allows you to enable a device that gives processes access to the x86 CPUID
instructions to be executed on a specific processor. As for the previous option, it is a good idea to
change the default value of N to Y.
High Memory Support (off, 4GB, 64GB) [off] Press Enter
This option allows you to be able to use up to 64 Gigabytes of physical memory on x86 systems.
Usually, and it’s true for most of us, we have a system that will never run with more than 1
Gigabyte total physical RAM and if this is you case, then answer "off" here (default choice and
suitable for most users). In other part if you have a system that has between 1 and 4 Gigabytes
physical RAM, then answer "4GB" here. If more than 4 Gigabytes is used then answer "64GB"
here. In our example, we will configure the kernel to use a maximum of 1 Gigabyte total physical
RAM by pressing the [Enter] key to accept the default choice "off".](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-147-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
148
Math emulation (CONFIG_MATH_EMULATION) [N/y/?] Press Enter
This option allows Linux to emulate a math coprocessor (a feature used for floating point
operations). In general, only old 486SX and 386 processors do not have a a math coprocessor
built in. All modern Pentium I, II, III, IV and later, AMD, Cyrix, etc have a math coprocessor built in
and do not require this option to be turned on. If you have a very old 486SX or 386 processor in
your computer, then you will have to change the default value of N here to become Y, but in
general, everyone will keep the default value of N here since they have a math coprocessor built
in their processor chip.
MTRR (Memory Type Range Register) support (CONFIG_MTRR) [N/y/?] Press Enter
This option on Intel family processors (Pentium Pro, Pentium II and later) allows the Memory
Type Range Registers (MTRRs) to be used to control processor access to memory ranges. This
is most useful if you have XFree86 graphical interface installed on your computer or have more
than 1 processor on your system. Changing the default value of N here to become Y on a Linux
system where a graphical interface or multiple processors are present will increase the
performance of the system. If you don't use a graphical interface, or do not have more than one
processor installed on you system, you must keep the default value of N by pressing the
[Enter] key. Finally, it is important to note that this option is valid only if you have Intel family
processors (Pentium Pro, Pentium II and later) installed on your computer. It doesn't work with
AMD or Cyrix processors.
Symmetric multi-processing support (CONFIG_SMP) [Y/n/?] n
This option enables support for systems with more than one CPU. If you have a system with only
one CPU, like most personal computers, say N. If you have a system with more than one CPU,
say Y. In our example, we only have one processor installed on our computer and will change the
default value of Y to become N.
Local APIC support on uniprocessors (CONFIG_X86_UP_APIC) [N/y/?] (NEW) Press
Enter
This option appears only if you have answered N to the previous option and will let you enable
support for local APIC on system with single processor. Local APIC (Advanced Programmable
Interrupt Controller) is an integrated interrupt controller in the CPU that supports CPU-generated
self-interrupts (timer, performance counters), and the NMI watchdog, which detects hard lockups
on the server. On systems with single processor, you may have to answer Y here, but on systems
with multiple processors, you don't have to answer this question since the kernel will
automatically enable it. Usually we keep the default choice of N here since the feature is available
only on modern Pentium processors like the IV family. Pentium I, II, and III don’t support this
option, as well as AMD and Cyrix processors.
*
* General setup
* Networking support (CONFIG_NET) [Y/n/?] Press Enter
This option is very important and must be set to Y all the time, which is the default value. It allows
your Linux system to support and use networking. We use the default value of Y by pressing the
[Enter] key.
PCI support (CONFIG_PCI) [Y/n/?] Press Enter
This option allows us to enable or disable PCI support on Linux. In most cases, we have to say Y
here. PCI's are the white slots on your motherboard where you can add network cards, video
cards, etc. Since most of us use a PC and have this kind of slot available, it is important to say Y
here by pressing the [Enter] key to use the default choice.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-148-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
149
PCI access mode (BIOS, Direct, Any) [Any] Press Enter
On PCI systems, the BIOS can be used to detect the PCI devices and determine their
configuration. However, some old PCI motherboards have BIOS bugs and may crash if this is
done. Also, some embedded PCI-based systems don't have any BIOS at all. Linux can also try to
detect the PCI hardware directly without using the BIOS. With this option, you can specify how
Linux should detect the PCI devices. If you choose "BIOS", the BIOS will be used, if you choose
"Direct", the BIOS won't be used, and if you choose "Any", the kernel will try the direct access
method and falls back to the BIOS if that doesn't work. If unsure, go with the default, which is
"Any" by pressing the [Enter] key.
PCI device name database (CONFIG_PCI_NAMES) [Y/n/?] n
This option lets you enable a feature that allows the Kernel to contain a database of all known
PCI device names via different files under the /proc filesystem and make the information
comprehensible to the user or disable this feature and get device ID numbers instead of names.
Disabling this feature will save you about 80KB of kernel image size and will make the kernel
smaller in size, which is a good thing for increased performance. If you disable this feature, the
kernel will still run as usual, but will show you device ID numbers instead of names. Therefore, we
will change the default value of Y (get PCI device by names) to become N (get PCI device by ID
numbers) and will save 80KB of Kernel image size.
EISA support (CONFIG_EISA) [N/y/?] Press Enter
This option allows you to enable support for the Extended Industry Standard Architecture (EISA)
bus. EISA is now obsolete by the PCI bus and you may enable this option only if you use some
older ISA card on your system. Since most of us don't have these types of card, we can safety
keep the default value of N here by pressing the [Enter] key.
MCA support (CONFIG_MCA) [N/y/?] Press Enter
This option allows us to enable MCA support with the kernel. MCA (MicroChannel Architecture) is
found in some IBM PS/2 machines and laptops. It is a bus system similar to PCI or ISA. It is rare
that we have this kind of bus on our system and we can safety keep the default value of N (do not
support MCA on this kernel) by pressing the [Enter] key again.
Support for hot-pluggable devices (CONFIG_HOTPLUG) [Y/n/?] n
This option is often required only on laptop computer where you have a PCMCIA or PC-cards that
you can plug or unplug at any time. If the kernel that you compile is made to run on a standard
computer (not laptop, portable), then you have to say N here by changing the default option of Y
to become N.
System V IPC (CONFIG_SYSVIPC) [Y/n/?] Press Enter
This option allows us to enable IPC on Linux. IPC (Inter Process Communication) is a suite of
library functions and system calls which let processes (running programs) synchronize and
exchange information. It is generally considered to be a good thing, and some programs won't
run unless you say Y here. Therefore, we keep the default setting of Y by pressing the [Enter]
key.
BSD Process Accounting (CONFIG_BSD_PROCESS_ACCT) [N/y/?] Press Enter
This option allows us to enable a user level programs, which will be able to instruct the kernel (via
a special system call) to write process accounting information to a file. It is not vital or even really
necessary to enable this kind of option to get a working Linux kernel. Therefore, we keep the
default value of N here by pressing the [Enter] key.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-149-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
150
Sysctl support (CONFIG_SYSCTL) [Y/n/?] Press Enter
This option is very important and especially with Linux kernel 2.4.x generation. It is the feature
that allows us to dynamically change certain kernel parameters and variables on the fly through
the /proc filesystem with the use of the /etc/sysctl.conf file. We keep the default value of
Y by pressing the [Enter] key.
Kernel core (/proc/kcore) format (ELF, A.OUT) [ELF] Press Enter
This option allows a file under the /proc filesystem, which is called "kcore" to contain the
Kernel core image in two different formats, which are ELF or A.OUT. For newer Linux systems,
ELF (Executable and Linkable Format) is highly recommended and is the default option that we
choose by pressing the [Enter] key. A.OUT (Assembler.OUTput) was the old format method
used, but is now really obsolete.
Kernel support for a.out binaries (CONFIG_BINFMT_AOUT) [Y/n/?] n
This option is very important and allows us to provide support for A.OUT. A.OUT
(Assembler.OUTput) is a set of formats for libraries and executables used in the earliest versions
of UNIX and, as we said before, it is now really obsolete and was replaced with the ELF format.
To be sure none of our programs will use this older executable format, we will change the default
value of Y to become N.
Kernel support for ELF binaries (CONFIG_BINFMT_ELF) [Y/n/?] Press Enter
Here, it is important to answer Y to this question since this binary format is the one used now on
modern Linux systems. ELF (Executable and Linkable Format) is a format for libraries and
executables used across different architectures and operating systems. Many new executables
are distributed solely in ELF format and you definitely want to say Y to this option by pressing the
[Enter] key.
Kernel support for MISC binaries (CONFIG_BINFMT_MISC) [Y/n/?] Press Enter
This option enables wrapper-driven binary formats into the kernel. This feature is required by
programs that need an interpreter to run, like Java, Python or Emacs-Lisp. Since we're sure to
use one of these types of programs, it is safe to accept the default value of Y by pressing the
[Enter] key.
Power Management support (CONFIG_PM) [Y/n/?] n
This option allows Power Management Support for your computer. With this feature parts of your
computer are shut off or put into a power conserving "sleep" mode if they are not being used. In
general it is good to enable this option on portable computer (Laptop) only. Note that, even if you
say N here, Linux, on the x86 architecture, will issue the hlt instruction if nothing is to be done,
thereby sending the processor to sleep and saving power. Therefore our choice will be N here.
*
* Memory Technology Devices (MTD)
* Memory Technology Device (MTD) support (CONFIG_MTD) [N/y/?] Press Enter
This option enables MTD (Memory Technology Devices) on your computer. MTD are flash, RAM
and similar chips, often used for solid-state file systems on embedded devices. If you have some
of these devices in your system, then answer Y to the question. In most cases, the default choice
of N is recommended.
*
* Parallel port support
* Parallel port support (CONFIG_PARPORT) [N/y/?] Press Enter
If you want to use devices connected to your machine's parallel port like a printer, zip drive, etc,
then you need to say Y here otherwise the default value N is recommended.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-150-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
151
*
* Plug and Play configuration
* Plug and Play support (CONFIG_PNP) [Y/n/?] n
Plug and Play (PnP) is a standard for peripherals which allows those peripherals to be configured
by software. If you answer to this question by Y, then Linux will be able to configure your Plug and
Play devices. Under Linux, we really don't need to enable PNP support and our choice will be N
here.
*
* Block devices
* Normal PC floppy disk support (CONFIG_BLK_DEV_FD) [Y/n/?] Press Enter
This option allows us to use the floppy disk drive(s) in our PC under Linux. Since everyone has
and usually needs a floppy disk in their computer, the answer to this question will be Y. If you run
a Linux server in highly secure environment, you could answer to this question by N since we
never use floppy disk on this type of system.
XT hard disk support (CONFIG_BLK_DEV_XD) [N/y/?] Press Enter
This option allows us to enable the very old 8 bit hard disk controllers used in the IBM XT
computer and since it's pretty unlikely that you have one of these then you must answer to this
question by N. Therefore we simply press [Enter] because the default answer to this question
is N.
Compaq SMART2 support (CONFIG_BLK_CPQ_DA) [N/y/?] Press Enter
This option allows us to enable support for the Compaq Smart Array controllers. If you use these
kinds of boards in your system, then you should say Y here, otherwise the default value of N is
recommended.
Compaq Smart Array 5xxx support (CONFIG_BLK_CPQ_CISS_DA) [N/y/?] Press Enter
This option allows us to enable support for the Compaq Smart Array 5xxx controllers. If you use
these kinds of boards in your system, then you should say Y here, otherwise the default value of
N is recommended.
Mylex DAC960/DAC1100 PCI RAID Controller support (CONFIG_BLK_DEV_DAC960)
[N/y/?] Press Enter
This option enables support for the Mylex DAC960, AcceleRAID, and eXtremeRAID PCI RAID
controllers. If you use these kinds of boards in your system, then you should say Y here,
otherwise the default value of N is recommended.
Loopback device support (CONFIG_BLK_DEV_LOOP) [N/y/?] Press Enter
This option is a little different from other kernel options and if you enable it, you will be able to use
a regular file as a block device that let you have access to some special advanced features and
possibilities under Linux. The default value for this option will be ok for most of us and only
advanced users or user that know why they need these features will enable the "Loopback
device support" option. For SCSI systems with a modularized kernel, it is important to say Y
here since SCSI drivers will use this device.
Network block device support (CONFIG_BLK_DEV_NBD) [N/y/?] Press Enter
This option enables another advanced feature under Linux, the possibility for your system to be a
client for network block devices. The default value for this option would be ok for most of us and
only advanced users or user that know why they need this feature will enable it.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-151-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
152
RAM disk support (CONFIG_BLK_DEV_RAM) [N/y/?] Press Enter
This option will allow you to use a portion of your RAM memory as a block device, so that you can
make file systems on it, read and write to it and do all the other things that you can do with normal
block devices (such as hard drives). It is usually used to load and store a copy of a minimal root
file system off of a floppy into RAM during the initial install of Linux. Again, most normal users
won't need the RAM disk functionality and will answer N to this question. For SCSI systems with a
modularized kernel, it is important to say Y here since SCSI drivers will use this feature.
*
* Multi-device support (RAID and LVM)
* Multiple devices driver support (RAID and LVM) (CONFIG_MD) [N/y/?] Press
Enter
This option is required only for RAID and logical volume management (LVM). If you use them,
then change the default value of N to become Y.
*
* Networking options
* Packet socket (CONFIG_PACKET) [Y/n/?] Press Enter
This option allows you to enable applications, which communicate directly with network devices
without an intermediate network protocol implemented in the kernel like the tcpdump program. It
is a good idea to enable this feature for most of us.
Packet socket: mmapped IO (CONFIG_PACKET_MMAP) [N/y/?] y
This option allows packet protocol driver to use an IO (Input/Output) mechanism that results in
faster communication. Say Y here.
Kernel/User netlink socket (CONFIG_NETLINK) [N/y/?] y
This option allows us to enable two-way communication between the kernel and user processes.
Say Y here.
Routing messages (CONFIG_RTNETLINK) [N/y/?] (NEW) y
If we have said Y to the previous option, we must say Y here too or the previous option will not
work. Therefore our choice is Y.
Netlink device emulation (CONFIG_NETLINK_DEV) [N/y/?] (NEW) y
This option is a backward compatibility option, and we have to choose Y for now.
Network packet filtering (replaces ipchains) (CONFIG_NETFILTER) [N/y/?] y
This option enables support for a packet filter firewall on your system (Netfilter). It is very
important to answer to this question by Y if you want to support firewall and IPTables on your
computer. If you answer N to this question, then the firewalling features will not be available, even
if your have the IPTables software installed on your system.
Network packet filtering debugging (CONFIG_NETFILTER_DEBUG) [N/y/?] (NEW) y
This option turns on support for debugging the netfilter code. It is a good idea to enable it.
Socket Filtering (CONFIG_FILTER) [N/y/?] Press Enter
This option allows us to enable Linux Socket Filter, a feature needed by PPP packet filtering in
general. Therefore you only need to say Y here if you want to use PPP packet filtering on your
system. Since we use a network card to get a connection to the Internet and not PPP (modem
link), we keep the default value of N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-152-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
153
Unix domain sockets (CONFIG_UNIX) [Y/n/?] Press Enter
This option is very important and must always be set to Y. It allows us to include support for Unix
domain sockets; sockets are the standard Unix mechanism for establishing and accessing
network connections. It is vital to say Y to this option.
TCP/IP networking (CONFIG_INET) [Y/n/?] Press Enter
Another very important and vital option under Linux. This option allows us to enable support for
TCP/IP networking on the computer. TCP/IP are the protocols used on the Internet and on most
local Ethernets. It is highly recommended to say Y here even if you are not connected to the
Internet.
IP: multicasting (CONFIG_IP_MULTICAST) [Y/n/?] n
This option allows us to enable a code that addresses several networked computers at once. You
need it if you intend to participate in the MBONE, a high bandwidth network on top of the Internet
which carries audio and video broadcasts. For most people, it's safe to say N here.
IP: advanced router (CONFIG_IP_ADVANCED_ROUTER) [N/y/?] n
This option allows us to configure our Linux system to run mostly as a router. The answer to this
question won't directly affect the kernel: answering N will just cause the configuration to skip all
the questions about advanced routing. In many cases, we can safety keep the default value of N
here. Only users that want to run their Linux system primarily as a router will answer to this
question by Y to be presented a list of advanced routing features to enable or reject. If you want
to configure your Linux server as a Gateway server, you need to answer Y to this question and all
questions related to this option.
IP: kernel level autoconfiguration (CONFIG_IP_PNP) [N/y/?] Press Enter
This option must be set to Y only for diskless machines requiring network access to boot. For
most people, it's safe to say N here.
IP: tunneling (CONFIG_NET_IPIP) [N/y/?] Press Enter
This option will enable Tunneling on our system. Tunneling is a means of encapsulating data of
one protocol type within another protocol and sending it over a channel that understands the
encapsulating protocol. This is an advanced feature and only advanced users who know why they
need it must answer Y to this question.
IP: GRE tunnels over IP (CONFIG_NET_IPGRE) [N/y/?] Press Enter
This option will enable another kind of Tunneling feature on our system. This method is known as
GRE (Generic Routing Encapsulation). This is an advanced feature and only advanced users that
know why they need it must answer Y to this question.
IP: TCP Explicit Congestion Notification support (CONFIG_INET_ECN) [N/y/?]
Press Enter
This option enables Explicit Congestion Notification on your system. ECN allows routers to notify
clients about network congestion, resulting in fewer dropped packets and increased network
performance. This is a very good feature but, unfortunately, there are many broken firewalls on
the Internet, which refuse connections from ECN-enabled machines, and it may be a while before
these firewalls are fixed. Until then, to access a site behind such a firewall you will have to disable
this option by saying N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-153-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
154
IP: TCP syncookie support (disabled per default) (CONFIG_SYN_COOKIES) [N/y/?] y
This option is very important and every one must answer to this question by Y because normal
TCP/IP networking is open to an attack known as "SYN flooding". This denial-of-service attack
prevents legitimate remote users from being able to connect to your computer during an ongoing
attack and requires very little work from the attacker, who can operate from anywhere on the
Internet. SYN cookies provide protection against this type of attack. Therefore don't forget to
answer Y to this question.
*
* IP: Netfilter Configuration
All questions under the "IP: Netfilter Configuration" section of the Kernel configuration
are related to packet filter firewall support and features. We recommend you enable everything.
Below, we show you the answer for each question without any explanation on the features. If you
need to get more information about the features that you don't understand, you can simply type ?
[Enter] at the prompt to get help.
*
Connection tracking (required for masq/NAT) (CONFIG_IP_NF_CONNTRACK) [N/y/?]
(NEW) y
FTP protocol support (CONFIG_IP_NF_FTP) [N/y/?] (NEW) y
IRC protocol support (CONFIG_IP_NF_IRC) [N/y/?] (NEW) y
IP tables support (required for filtering/masq/NAT) (CONFIG_IP_NF_IPTABLES)
[N/y/?] (NEW) y
limit match support (CONFIG_IP_NF_MATCH_LIMIT) [N/y/?] (NEW) y
MAC address match support (CONFIG_IP_NF_MATCH_MAC) [N/y/?] (NEW) y
netfilter MARK match support (CONFIG_IP_NF_MATCH_MARK) [N/y/?] (NEW) y
Multiple port match support (CONFIG_IP_NF_MATCH_MULTIPORT) [N/y/?] (NEW) y
TOS match support (CONFIG_IP_NF_MATCH_TOS) [N/y/?] (NEW) y
LENGTH match support (CONFIG_IP_NF_MATCH_LENGTH) [N/y/?] (NEW) y
TTL match support (CONFIG_IP_NF_MATCH_TTL) [N/y/?] (NEW) y
tcpmss match support (CONFIG_IP_NF_MATCH_TCPMSS) [N/y/?] (NEW) y
Connection state match support (CONFIG_IP_NF_MATCH_STATE) [N/y/?] (NEW) y
Packet filtering (CONFIG_IP_NF_FILTER) [N/y/?] (NEW) y
REJECT target support (CONFIG_IP_NF_TARGET_REJECT) [N/y/?] (NEW) y
Full NAT (CONFIG_IP_NF_NAT) [N/y/?] (NEW) y
Packet mangling (CONFIG_IP_NF_MANGLE) [N/y/?] (NEW) y
TOS target support (CONFIG_IP_NF_TARGET_TOS) [N/y/?] (NEW) y
MARK target support (CONFIG_IP_NF_TARGET_MARK) [N/y/?] (NEW) y
LOG target support (CONFIG_IP_NF_TARGET_LOG) [N/y/?] (NEW) y
TCPMSS target support (CONFIG_IP_NF_TARGET_TCPMSS) [N/y/?] (NEW) y
WARNING: If you want to enable IPTables support into the kernel, the iptables program must
be installed first or you will receive error messages during kernel compilation. This is because
when IPTables support is enabled, the kernel will associate some part of the iptables
program with it configuration. Therefore don’t forget to install IPTables before configuring kernel
with IPTables support. Finally the same warning is true for quota support into the kernel.
*
*
* The IPX protocol (CONFIG_IPX) [N/y/?] Press Enter
This option allows us to enable Novell networking protocol support on Linux. You need it if you
want to access Novell NetWare file or print servers using Linux or if you want to configure your
Linux system to run as a Novell NetWare file or print server. In most cases, this is not required
and we can answer N to this question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-154-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
155
Appletalk protocol support (CONFIG_ATALK) [N/y/?] Press Enter
This option allows us to enable AppleTalk protocol support on Linux. You need it if you want to
access Apple computers using Linux. In most cases, this is not required and we can answer N to
this question.
DECnet Support (CONFIG_DECNET) [N/y/?] Press Enter
This option allows us to enable DECnet networking protocol support on Linux. The DECnet
networking protocol was used in many products made by Digital (now Compaq). In most cases,
this is not required and we can answer N to this question.
802.1d Ethernet Bridging (CONFIG_BRIDGE) [N/y/?] Press Enter
This option will allow your Linux system to act as an Ethernet bridge, which means that the
different Ethernet segments it is connected to will appear as one Ethernet to the participants.
Several such bridges can work together to create even larger networks of Ethernets using the
IEEE 802.1 spanning tree algorithm. In most cases, this is not required and we can answer N to
this question.
*
* QoS and/or fair queueing
* QoS and/or fair queueing (CONFIG_NET_SCHED) [N/y/?] Press Enter
This option allows us to enable QoS (Quality of Service) support on Linux. When the kernel has
several packets to send out over a network device, it has to decide which ones to send first,
which ones to delay, and which ones to drop. This is the job of the packet scheduler, and several
different algorithms for how to do this "fairly" have been proposed. If we answer N to this
question, the standard packet scheduler, which is a FIFO (first come, first served) will be used by
default. The standard packet scheduler is enough for most of us and if you are running a router
system or are an advanced user who wants to experiment in some new way with TCP/IP
networking, then you can say Y to this question and be able to choose from among several
alternative algorithms which can then be attached to different network devices. In most cases, we
say N to this question.
*
* Telephony Support
* Linux telephony support (CONFIG_PHONE) [N/y/?] Press Enter
This option allows us to use a regular phone for voice-over-IP applications. This also means that
you have to have a telephony card attached to your computer and you know what to do. Most
people will simply answer N to this question.
*
* ATA/IDE/MFM/RLL support
* ATA/IDE/MFM/RLL support (CONFIG_IDE) [Y/n/?] Press Enter
This option allows the kernel to manage low cost mass storage units such as ATA/(E)IDE and
ATAPI units. The most common cases are IDE hard drives and ATAPI CD-ROM drives and
since we all have one of these devices attached to our computer we can safety say Y to this
question. The only time that you can answer to this question by N is when you know that your
system is pure SCSI.
*
* IDE, ATA and ATAPI Block devices
* Enhanced IDE/MFM/RLL disk/cdrom/tape/floppy support (CONFIG_BLK_DEV_IDE)
[Y/n/?] Press Enter
This option allows us to enable the new enhanced driver with IDE/MFM/RLL
disk/cdrom/tape/floppy drives. If you have one or more IDE drives, it is required to answer Y to
this question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-155-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
156
Use old disk-only driver on primary interface (CONFIG_BLK_DEV_HD_IDE) [N/y/?]
Press Enter
This option allows us to enable the old hard disk driver to control IDE/MFM/RLL
disk/cdrom/tape/floppy drives. It is highly recommended not to enable it, since we've already used
the new enhanced driver option with IDE/MFM/RLL disk/cdrom/tape/floppy drives above. This
option may be useful for older systems.
Include IDE/ATA-2 DISK support (CONFIG_BLK_DEV_IDEDISK) [Y/n/?] Press Enter
This option allows us to enable another enhanced support for MFM/RLL/IDE hard disks. The only
time that you can answer N to this question is when you know that your system is pure SCSI.
Therefore, we'll enable support for this option by saying Y to the question.
Use multi-mode by default (CONFIG_IDEDISK_MULTI_MODE) [Y/n/?] n
This option allows us to fix possible error messages that can appear on IDE systems. This error
message may look like:
hda: set_multmode: status=0x51 { DriveReady SeekComplete Error }
hda: set_multmode: error=0x04 { DriveStatusError }
If you get this kind of error message on your system, then you have to say Y to this option. We
suppose that you have a good IDE disk drive and that this error will never appear for you, in this
case, we will change the default value of Y to become N.
Include IDE/ATAPI CDROM support (CONFIG_BLK_DEV_IDECD) [Y/n/?] n
This option allows us to instruct Linux to identify more than one CD-ROM drive along with other
IDE devices, as "hdb" or "hdc", or something similar at boot time. If you have only one CD-ROM
drive installed on your system, you can say N to this question even if it uses the ATAPI protocol
and save some KB into the kernel. You have to say Y to this option only if you handle more than
one CD-ROM drive in your computer. Since most people will usually have only one CD-ROM
installed into their computer, we will change the default value of Y to become N. If you have one
CD-ROM and another one CD-ROM for burning disk, then you have answer Y to this question.
Include IDE/ATAPI TAPE support (CONFIG_BLK_DEV_IDETAPE) [N/y/?] Press Enter
This option allows us to enable an IDE tape drive using the ATAPI protocol. If you have an IDE
tape drive installed on your system, you must answer Y to this question. Most people will say N
here.
Include IDE/ATAPI FLOPPY support (CONFIG_BLK_DEV_IDEFLOPPY) [N/y/?] Press Enter
This option allows us to enable an IDE floppy drive which uses the ATAPI protocol. If you have
this kind of floppy drive installed on your system, you must answer this question Y. Most people
will say N here.
SCSI emulation support (CONFIG_BLK_DEV_IDESCSI) [N/y/?] Press Enter
This option provides SCSI host adapter emulation for IDE ATAPI devices, and will allow us to
use a SCSI device driver instead of a native ATAPI driver. If you intend to install and use a CD-
RW drive on your computer, then you have to say Y here. Again, most people will say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-156-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
157
*
* IDE chipset support/bugfixes
* CMD640 chipset bugfix/support (CONFIG_BLK_DEV_CMD640) [Y/n/?] n
This option allows us to include code which tries to automatically detect and correct the CMD640
problems under Linux. The CMD-Technologies CMD640 IDE chip is used on many common 486
and Pentium motherboards, usually in combination with a "Neptune" or "SiS" chipset.
Unfortunately, it has a number of rather nasty design flaws that can cause severe data corruption
under many common conditions. To know if you need to enable this option for your system to
correct this bug, edit the /proc/cpuinfo file and see if the parameter "f00f_bug" is set to no
or yes. If the "f00f_bug" value is set to no, then you don't need to enable this option and can
say N to the question, otherwise you have to say Y here.
RZ1000 chipset bugfix/support (CONFIG_BLK_DEV_RZ1000) [Y/n/?] n
This option allows us to include code which tries to automatically detect and correct the RZ1000
problems under Linux. The PC-Technologies RZ1000 IDE chip is used on many common 486
and Pentium motherboards, usually along with the "Neptune" chipset. As for the CMD640 bug
above, it also has a number of rather nasty design flaws that can cause severe data corruption
under many common conditions. To know if you need to enable this option for your system to
correct this bug, edit the /proc/cpuinfo file and see if the parameter "coma_bug" is set to no
or yes. If the "coma_bug" value is set to no, then you don't need to enable this option and can
say N to the question, otherwise you have to say Y here.
Generic PCI IDE chipset support (CONFIG_BLK_DEV_IDEPCI) [Y/n/?] Press Enter
This option helps the IDE driver to automatically detect and configure all PCI-based IDE
interfaces in your system. If you have PCI systems which use IDE drive(s), then say Y to this
question. Most of us have PCI systems which use IDE and have answer Y this question.
Sharing PCI IDE interrupts support (CONFIG_IDEPCI_SHARE_IRQ) [Y/n/?] Press
Enter
This option allows us to enable support for sharing a single IRQ with other cards under ATA/IDE
chipsets. In general, everyone has answer Y this question.
Generic PCI bus-master DMA support (CONFIG_BLK_DEV_IDEDMA_PCI) [Y/n/?] Press
Enter
This option allows us to reduce CPU overhead with IDE drive(s) on PCI system capable of bus-
master DMA operation. If you have a modern IDE ATA/33/66/100 hard drive, then it is
recommended to answer this question Y.
Boot off-board chipsets first support (CONFIG_BLK_DEV_OFFBOARD) [N/y/?] Press
Enter
This option allows us to reverse the device scan order to improve the usability of some boot
managers such as lilo when booting from a drive on an off-board controller. In many cases, this
option is really not required and you can accept the default value of N here.
Use PCI DMA by default when available (CONFIG_IDEDMA_PCI_AUTO) [Y/n/?] Press
Enter
A very important option to enable on all modern IDE disk drives. This option allows us to use the
DMA feature of our disk drive under Linux to improve performance. Most people running a capable
DMA drive will answer to this question by Y.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-157-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
158
All of the following Kernel options are related to the special onboard chipsets that you may have
on your motherboard. Therefore, specific drivers are provided for each of them and you have to
choose from the list the one that matches you chipset. If you have an Intel onboard chipset, then
you can safety choose the default answer of N to all of the questions, since the kernel supports it
naturally.
Other chipset models must be selected from the list. In many cases, if your chipset is not listed,
this means that it is automatically supported by the Kernel. Note that two options have their
default answer set to Y (Intel PIIXn chipsets support (CONFIG_BLK_DEV_PIIX)
[Y/n/?] and PIIXn Tuning support (CONFIG_PIIX_TUNING) [Y/n/?]). If you have a
Pentium II or later processor, you must keep the default value of these two option to Y.
AEC62XX chipset support (CONFIG_BLK_DEV_AEC62XX) [N/y/?] Press Enter
ALI M15x3 chipset support (CONFIG_BLK_DEV_ALI15X3) [N/y/?] Press Enter
CMD64X chipset support (CONFIG_BLK_DEV_CMD64X) [N/y/?] Press Enter
CY82C693 chipset support (CONFIG_BLK_DEV_CY82C693) [N/y/?] Press Enter
Cyrix CS5530 MediaGX chipset support (CONFIG_BLK_DEV_CS5530) [N/y/?] Press
Enter
HPT34X chipset support (CONFIG_BLK_DEV_HPT34X) [N/y/?] Press Enter
HPT366 chipset support (CONFIG_BLK_DEV_HPT366) [N/y/?] Press Enter
Intel PIIXn chipsets support (CONFIG_BLK_DEV_PIIX) [Y/n/?] Press Enter
PIIXn Tuning support (CONFIG_PIIX_TUNING) [Y/n/?] Press Enter
NS87415 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_NS87415) [N/y/?] Press
Enter
PROMISE PDC202{46|62|65|67|68} support (CONFIG_BLK_DEV_PDC202XX) [N/y/?] Press
Enter
ServerWorks OSB4/CSB5 chipsets support (CONFIG_BLK_DEV_SVWKS) [N/y/?] Press
Enter
SiS5513 chipset support (CONFIG_BLK_DEV_SIS5513) [N/y/?] Press Enter
SLC90E66 chipset support (CONFIG_BLK_DEV_SLC90E66) [N/y/?] Press Enter
Tekram TRM290 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_TRM290) [N/y/?]
Press Enter
VIA82CXXX chipset support (CONFIG_BLK_DEV_VIA82CXXX) [N/y/?] Press Enter
Other IDE chipset support (CONFIG_IDE_CHIPSETS) [N/y/?] Press Enter
IGNORE word93 Validation BITS (CONFIG_IDEDMA_IVB) [N/y/?] Press Enter
*
* SCSI support
* SCSI support (CONFIG_SCSI) [Y/n/?] Press Enter
This option allows us to enable SCSI hard disks, SCSI tape drives, SCSI CD-ROM's or any other
SCSI devices under Linux. If you have a SCSI like system, you need to answer Y to this question.
If you don't have any SCSI devices on your system, you can safety answer N to the question. For
users that have a SCSI system, it is very important for you to know the name of your SCSI host
adapter (the card inside your computer that "speaks" the SCSI protocol, also called SCSI
controller), because you will be asked for it if you enable this option. Once again, if you don't have
a SCSI system, simply answer N to this question and skip this section of the Linux Kernel
configuration.
*
* SCSI support type (disk, tape, CD-ROM)
* SCSI disk support (CONFIG_BLK_DEV_SD) [Y/n/?] Press Enter
This option allows us to enable support for a SCSI hard disk under Linux. If you have enabled the
SCSI support feature above because you have a SCSI hard drive on your system, then it's here
that you have to specify it by answering Y to the question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-158-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
159
Maximum number of SCSI disks that can be loaded as modules
(CONFIG_SD_EXTRA_DEVS) [40] Press Enter
This option allows us to control the amount of additional space allocated in tables for drivers that
are loaded as modules after the kernel is booted. In the event that the SCSI core itself was
loaded as a module, this value is the number of additional disks that can be loaded after the first
host driver is loaded. Since we're compiling a Monolithic Kernel where no modules are available,
this option doesn't concern us and we can safety press the [Enter] key to accept the default
value.
SCSI tape support (CONFIG_CHR_DEV_ST) [N/y/?] Press Enter
This option allows us to enable support for a SCSI tape drive under Linux. If you have enabled
the SCSI support feature above because you have a SCSI tape drive on your system, then it's
here that you have to specify it by answering Y to the question. For most SCSI users, the answer
is N (no, we don't have a SCSI tape drive on this computer).
SCSI OnStream SC-x0 tape support (CONFIG_CHR_DEV_OSST) [N/y/?] Press Enter
This option allows us to enable support for the OnStream SC-x0 SCSI tape drives under Linux. If
you have this kind of SCSI tape drive installed on your computer, then you have to answer to this
question by Y. Most SCSI users will simply say N to this question.
SCSI CD-ROM support (CONFIG_BLK_DEV_SR) [N/y/?] Press Enter
This option allows us to enable support for a SCSI CD-ROM under Linux. If you have enabled the
SCSI support feature above because you have a SCSI CD-ROM on your system, then it's here
that you have to specify it by answering Y to the question. For most SCSI users, the answer is N
(no, we don't have a SCSI CD-ROM on this computer).
SCSI generic support (CONFIG_CHR_DEV_SG) [N/y/?] Press Enter
This option allows us to enable support to use SCSI scanners, synthesizers or CD-writers or just
about anything having "SCSI" in its name other than hard disks, CD-ROMs or tapes. If you have
one of these SCSI items installed on your computer, then you have to say Y here as well as for
the "SCSI disk support" option above to enable the driver.
VERY IMPORTANT NOTE: For users having IDE CD-writers, you have to say Y to this question too,
even if your CD-writers are not SCSI CD-writers. Most SCSI users will simply say N to his
question.
*
* Some SCSI devices (e.g. CD jukebox) support multiple LUNs
* Enable extra checks in new queueing code (CONFIG_SCSI_DEBUG_QUEUES) [Y/n/?]
Press Enter
This option turns on a lot of additional consistency checking for the new queuing code on SCSI
devices. It is a good idea to enable it by saying Y to the question.
Probe all LUNs on each SCSI device (CONFIG_SCSI_MULTI_LUN) [Y/n/?] n
This option force the SCSI driver to probe for multiple LUN's (Logical Unit Number) on your
system and will certainly affect the performance of the system. Most SCSI users will simply
disable this option by saying N to this question to improve performance. If you enable this option,
then we assume that you know what you're doing.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-159-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
160
Verbose SCSI error reporting (kernel size +=12K) (CONFIG_SCSI_CONSTANTS)
[Y/n/?] n
This option allows any error messages regarding your SCSI hardware to be more
understandable, this enlarges your kernel by about 12 KB. If performance is important to you, we
highly recommend you to disable this option by answering the question N.
SCSI logging facility (CONFIG_SCSI_LOGGING) [N/y/?] Press Enter
This option allows us to turns on a logging facility that can be used to debug a number of SCSI
related problems. Again, if performance is important to you, we highly recommend you to disable
this option by keeping the default value of N here.
*
* SCSI low-level drivers
Below you will be presented a list of available SCSI controllers to choose from, simply select the
SCSI controller that is installed on your system and disable all the others. As an example, we will
pretend that we have an Adaptec AIC7080 controller and will enable it further down. We chose
an Adaptec AIC7080 model for our example; don't forget to change our choice if you have
another kind of SCSI controller installed on your system.
*
3ware Hardware ATA-RAID support (CONFIG_BLK_DEV_3W_XXXX_RAID) [N/y/?] Press
Enter
7000FASST SCSI support (CONFIG_SCSI_7000FASST) [N/y/?] Press Enter
ACARD SCSI support (CONFIG_SCSI_ACARD) [N/y/?] Press Enter
Adaptec AHA152X/2825 support (CONFIG_SCSI_AHA152X) [N/y/?] Press Enter
Adaptec AHA1542 support (CONFIG_SCSI_AHA1542) [N/y/?] Press Enter
Adaptec AHA1740 support (CONFIG_SCSI_AHA1740) [N/y/?] Press Enter
Adaptec AIC7xxx support (CONFIG_SCSI_AIC7XXX) [N/y/?] y
Maximum number of TCQ commands per device (CONFIG_AIC7XXX_CMDS_PER_DEVICE)
[253] (NEW) Press Enter
Initial bus reset delay in milli-seconds (CONFIG_AIC7XXX_RESET_DELAY_MS)
[15000] (NEW) Press Enter
Build Adapter Firmware with Kernel Build (CONFIG_AIC7XXX_BUILD_FIRMWARE)
[N/y/?] (NEW) Press Enter
Adaptec I2O RAID support (CONFIG_SCSI_DPT_I2O) [N/y/?] Press Enter
AdvanSys SCSI support (CONFIG_SCSI_ADVANSYS) [N/y/?] Press Enter
Always IN2000 SCSI support (CONFIG_SCSI_IN2000) [N/y/?] Press Enter
AM53/79C974 PCI SCSI support (CONFIG_SCSI_AM53C974) [N/y/?] Press Enter
AMI MegaRAID support (CONFIG_SCSI_MEGARAID) [N/y/?] Press Enter
BusLogic SCSI support (CONFIG_SCSI_BUSLOGIC) [N/y/?] Press Enter
Compaq Fibre Channel 64-bit/66Mhz HBA support (CONFIG_SCSI_CPQFCTS) [N/y/?]
Press Enter
DMX3191D SCSI support (CONFIG_SCSI_DMX3191D) [N/y/?] Press Enter
DTC3180/3280 SCSI support (CONFIG_SCSI_DTC3280) [N/y/?] Press Enter
EATA ISA/EISA/PCI (DPT and generic EATA/DMA-compliant boards) support
(CONFIG_SCSI_EATA) [N/y/?] Press Enter
EATA-DMA [Obsolete] (DPT, NEC, AT&T, SNI, AST, Olivetti, Alphatronix) support
(CONFIG_SCSI_EATA_DMA) [N/y/?] Press Enter
EATA-PIO (old DPT PM2001, PM2012A) support (CONFIG_SCSI_EATA_PIO) [N/y/?] Press
Enter
Future Domain 16xx SCSI/AHA-2920A support (CONFIG_SCSI_FUTURE_DOMAIN) [N/y/?]
Press Enter
Intel/ICP (former GDT SCSI Disk Array) RAID Controller support
(CONFIG_SCSI_GDTH) [N/y/?] Press Enter
Generic NCR5380/53c400 SCSI support (CONFIG_SCSI_GENERIC_NCR5380) [N/y/?] Press
Enter
IBM ServeRAID support (CONFIG_SCSI_IPS) [N/y/?] Press Enter
Initio 9100U(W) support (CONFIG_SCSI_INITIO) [N/y/?] Press Enter
Initio INI-A100U2W support (CONFIG_SCSI_INIA100) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-160-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
161
NCR53c406a SCSI support (CONFIG_SCSI_NCR53C406A) [N/y/?] Press Enter
NCR53c7,8xx SCSI support (CONFIG_SCSI_NCR53C7xx) [N/y/?] Press Enter
SYM53C8XX Version 2 SCSI support (CONFIG_SCSI_SYM53C8XX_2) [N/y/?] Press Enter
NCR53C8XX SCSI support (CONFIG_SCSI_NCR53C8XX) [N/y/?] Press Enter
SYM53C8XX SCSI support (CONFIG_SCSI_SYM53C8XX) [Y/n/?] n
PAS16 SCSI support (CONFIG_SCSI_PAS16) [N/y/?] Press Enter
PCI2000 support (CONFIG_SCSI_PCI2000) [N/y/?] Press Enter
PCI2220i support (CONFIG_SCSI_PCI2220I) [N/y/?] Press Enter
PSI240i support (CONFIG_SCSI_PSI240I) [N/y/?] Press Enter
Qlogic FAS SCSI support (CONFIG_SCSI_QLOGIC_FAS) [N/y/?] Press Enter
Qlogic ISP SCSI support (CONFIG_SCSI_QLOGIC_ISP) [N/y/?] Press Enter
Qlogic ISP FC SCSI support (CONFIG_SCSI_QLOGIC_FC) [N/y/?] Press Enter
Qlogic QLA 1280 SCSI support (CONFIG_SCSI_QLOGIC_1280) [N/y/?] Press Enter
Seagate ST-02 and Future Domain TMC-8xx SCSI support (CONFIG_SCSI_SEAGATE)
[N/y/?] Press Enter
Simple 53c710 SCSI support (Compaq, NCR machines) (CONFIG_SCSI_SIM710) [N/y/?]
Press Enter
Symbios 53c416 SCSI support (CONFIG_SCSI_SYM53C416) [N/y/?] Press Enter
Tekram DC390(T) and Am53/79C974 SCSI support (CONFIG_SCSI_DC390T) [N/y/?] Press
Enter
Trantor T128/T128F/T228 SCSI support (CONFIG_SCSI_T128) [N/y/?] Press Enter
UltraStor 14F/34F support (CONFIG_SCSI_U14_34F) [N/y/?] Press Enter
UltraStor SCSI support (CONFIG_SCSI_ULTRASTOR) [N/y/?] Press Enter
*
* Fusion MPT device support
* Fusion MPT (base + ScsiHost) drivers (CONFIG_FUSION) [N/y/?] Press Enter
*
* I2O device support
* I2O support (CONFIG_I2O) [N/y/?] Press Enter
This option allows us to enable support for Intelligent Input/Output (I2O) architecture. In order for
this to work, you need to have an I2O interface adapter card in your computer. If you have this
kind of I2O interface adapter card installed on your system, then you can say Y to the question
and you will get a choice of interface adapter drivers and OSM's. Most users simply say N here.
*
* Network device support
* Network device support (CONFIG_NETDEVICES) [Y/n/?] Press Enter
This option is one of the most important and allows us to enable support and feature for network
cards under Linux. Therefore, we have to answer Y to this question.
*
* ARCnet devices
* ARCnet support (CONFIG_ARCNET) [N/y/?] Press Enter
This option allows us to enable ARCnet chipset support under Linux. If you have a network card
of this type installed on your system, then say Y here, otherwise and for most users, you have to
keep the default value of N here.
Dummy net driver support (CONFIG_DUMMY) [Y/n/?] n
This option is only useful for PPP dial up modem users on Linux. If you don't use your system to
make PPP connections, then you can safety answer N to this question. Only users who have a
modem and want to use it to establish a connection via PPP or SLIP need to say Y here. Since,
in our example, we use a network card to make a network connection, we will answer the
question N.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-161-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
162
Bonding driver support (CONFIG_BONDING) [N/y/?] Press Enter
This option allows us to 'bond' multiple Ethernet Channels together. This is called 'Etherchannel'
by Cisco, 'Trunking' by Sun, and 'Bonding' in Linux. It is a technique to merge Ethernet segments
together for doubling the speed of a connection. In most cases, we can safety choose the default
choice of N here. You must have two Ethernet cards installed on your system and on the remote
computer where you want to use this technique to be able to use it. Also, this is an advanced
feature and only experienced Linux users may need it. Therefore, we will answer to the question
by N.
EQL (serial line load balancing) support (CONFIG_EQUALIZER) [N/y/?] Press Enter
This option allows us to enable the same feature as the previous option, but this time for two
modems and two telephone lines. Therefore, we will simply say N to this question.
Universal TUN/TAP device driver support (CONFIG_TUN) [N/y/?] Press Enter
This option allows us to enable TUN/TAP support under Linux. TUN/TAP provides packet
reception and transmission for user space programs. It can be viewed as a simple Point-to-Point
or Ethernet device, which instead of receiving packets from a physical media, receives them from
user space program and instead of sending packets via physical media, writes them to the user
space program. It is rare that we need this feature, if you need it then simply say Y here. Most
users will say N here.
*
* Ethernet (10 or 100Mbit)
* Ethernet (10 or 100Mbit) (CONFIG_NET_ETHERNET) [Y/n/?] Press Enter
This option is very important and allows us to enable support for Ethernet Network Interface
Cards (NIC's) under Linux. Now, everyone has a NIC in their computer and if you want to be able
to use your network card, then you have to say Y here. Note that the answer to this question
won't directly affect the kernel: saying N will just cause the configuration to skip all the questions
about Ethernet network cards. We must say Y here to be able to select the network card that we
have in our computer from the list of supported network cards.
It is very important to know the name of the network card(s) installed in your system because you
will be asked for it. As an example we will pretend that we have an "EtherExpressPro/100"
network card in our computer and we will enable support for it. This is an example and don't
forget to change our default choice if you have another kind of network card installed in your
system. In general, we say Y for the network card that we have and N for all other network cards.
Sun Happy Meal 10/100baseT support (CONFIG_HAPPYMEAL) [N/y/?] Press Enter
Sun GEM support (CONFIG_SUNGEM) [N/y/?] Press Enter
3COM cards (CONFIG_NET_VENDOR_3COM) [N/y/?] Press Enter
AMD LANCE and PCnet (AT1500 and NE2100) support (CONFIG_LANCE) [N/y/?] Press
Enter
Western Digital/SMC cards (CONFIG_NET_VENDOR_SMC) [N/y/?] Press Enter
Racal-Interlan (Micom) NI cards (CONFIG_NET_VENDOR_RACAL) [N/y/?] Press Enter
DEPCA, DE10x, DE200, DE201, DE202, DE422 support (CONFIG_DEPCA) [N/y/?] Press
Enter
HP 10/100VG PCLAN (ISA, EISA, PCI) support (CONFIG_HP100) [N/y/?] Press Enter
Other ISA cards (CONFIG_NET_ISA) [N/y/?] Press Enter
EISA, VLB, PCI and on board controllers (CONFIG_NET_PCI) [Y/n/?] Press Enter
AMD PCnet32 PCI support (CONFIG_PCNET32) [N/y/?] Press Enter
Apricot Xen-II on board Ethernet (CONFIG_APRICOT) [N/y/?] Press Enter
CS89x0 support (CONFIG_CS89x0) [N/y/?] Press Enter
DECchip Tulip (dc21x4x) PCI support (CONFIG_TULIP) [N/y/?] Press Enter
Generic DECchip & DIGITAL EtherWORKS PCI/EISA (CONFIG_DE4X5) [N/y/?] Press
Enter
Digi Intl. RightSwitch SE-X support (CONFIG_DGRS) [N/y/?] Press Enter
Davicom DM910x/DM980x support (CONFIG_DM9102) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-162-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
163
EtherExpressPro/100 support (CONFIG_EEPRO100) [Y/n/?] Press Enter
Myson MTD-8xx PCI Ethernet support (CONFIG_FEALNX) [N/y/?] Press Enter
National Semiconductor DP8381x series PCI Ethernet support (CONFIG_NATSEMI)
[N/y/?] Press Enter
PCI NE2000 and clones support (see help) (CONFIG_NE2K_PCI) [N/y/?] Press Enter
RealTek RTL-8139 PCI Fast Ethernet Adapter support (CONFIG_8139TOO) [N/y/?]
Press Enter
SiS 900/7016 PCI Fast Ethernet Adapter support (CONFIG_SIS900) [N/y/?] Press
Enter
SMC EtherPower II (CONFIG_EPIC100) [N/y/?] Press Enter
Sundance Alta support (CONFIG_SUNDANCE) [N/y/?] Press Enter
TI ThunderLAN support (CONFIG_TLAN) [N/y/?] Press Enter
VIA Rhine support (CONFIG_VIA_RHINE) [N/y/?] Press Enter
Winbond W89c840 Ethernet support (CONFIG_WINBOND_840) [N/y/?] Press Enter
Pocket and portable adapters (CONFIG_NET_POCKET) [N/y/?] Press Enter
*
* Ethernet (1000 Mbit)
* Alteon AceNIC/3Com 3C985/NetGear GA620 Gigabit support (CONFIG_ACENIC)
[N/y/?] Press Enter
D-Link DL2000-based Gigabit Ethernet support (CONFIG_DL2K) [N/y/?] Press Enter
National Semiconduct DP83820 support (CONFIG_NS83820) [N/y/?] Press Enter
Packet Engines Hamachi GNIC-II support (CONFIG_HAMACHI) [N/y/?] Press Enter
SysKonnect SK-98xx support (CONFIG_SK98LIN) [N/y/?] Press Enter
FDDI driver support (CONFIG_FDDI) [N/y/?] Press Enter
This option allows us to enable FDDI card support under Linux. FDDI (Fiber Distributed Data
Interface) is a high speed local area network design to runs over copper or fiber. If you are
connected to such a network and want a driver for the FDDI card in your computer, say Y here.
Most users will simply say N here.
PPP (point-to-point protocol) support (CONFIG_PPP) [N/y/?] Press Enter
This option allows us to enable PPP support under Linux. PPP (Point to Point Protocol) is the
protocol used by modern modems to establish a remote connection with your ISP. If you have a
modem card installed on your system to make a remote connection with your ISP, then you need
to answer Y to this question. If you don't use PPP to connect on the Internet, then you can safety
say N here. In our example, we assume that you use another method, like a network interface, to
connect to the Internet and say N here.
SLIP (serial line) support (CONFIG_SLIP) [N/y/?] Press Enter
This option allows us to enable SLIP or CSLIP support under Linux. These protocols are really
old now and have been replaced by the PPP protocol (see the above option). If for any reason
you still use them, then say Y here. Most users will answer this question N.
*
* Wireless LAN (non-hamradio)
* Wireless LAN (non-hamradio) (CONFIG_NET_RADIO) [N/y/?] Press Enter
This option allows us to enable support for wireless LANs and everything having to do with radio,
but not with amateur radio or FM broadcasting. If you need such support on your system, then
say Y here, also if you need Wireless Extensions with wireless PCMCIA (PC-) cards on Linux, you
need to say Y here too. Most users will simply say N here.
*
* Token Ring devices
* Token Ring driver support (CONFIG_TR) [N/y/?] Press Enter
This option allows us to enable support for Token Ring under Linux. Token Ring is IBM's way of
communication on a local network; the rest of the world uses Ethernet. If you need Token Ring
support on your computer, then say Y here. Most people will select the default choice of N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-163-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
164
Fibre Channel driver support (CONFIG_NET_FC) [N/y/?] Press Enter
This option allows us to enable Fibre Channel support under Linux. Fibre Channel is a high speed
serial protocol mainly used to connect large storage devices to the computer. If you have a Fibre
channel adapter card in your computer, then you can say Y here. Most users will simply say N
here.
*
* Wan interfaces
* Wan interfaces support (CONFIG_WAN) [N/y/?] Press Enter
This option allows us to enable support for Wan interfaces under Linux. Wide Area Networks
(WANs), such as X.25, frame relay and leased lines, are used to interconnect Local Area
Networks (LANs) over vast distances. If you have these kinds of cards installed on your system,
then you can answer Y to the question. Most users will say N here.
*
* Amateur Radio support
* Amateur Radio support (CONFIG_HAMRADIO) [N/y/?] Press Enter
This option allows us to enable Amateur Radio support under Linux. If you want to connect your
Linux box to an amateur radio, answer Y here. Note that the answer to this question won't directly
affect the kernel: saying N will just cause the configuration to skip all the questions about amateur
radio. Most people will say N here.
*
* IrDA (infrared) support
* IrDA subsystem support (CONFIG_IRDA) [N/y/?] Press Enter
This option allows us to enable support for the IrDA (TM) protocols under Linux. IrDA (Infrared
Data Associations) is a support for wireless infrared communication. Laptops or computers that
use infrared and PDA's users will say Y here. Most users will say N here.
*
* ISDN subsystem
* ISDN support (CONFIG_ISDN) [N/y/?] Press Enter
This option allows us to enable ISDN support under Linux. ISDN (Integrated Services Digital
Networks) is a special type of fully digital telephone service; it's mostly used to connect to your
Internet service provider (with SLIP or PPP). If you have this type of card installed on your
computer (popular in Europe), then you have to say Y here otherwise, you will certainly keep the
default value of N here.
*
* Old CD-ROM drivers (not SCSI, not IDE)
* Support non-SCSI/IDE/ATAPI CDROM drives (CONFIG_CD_NO_IDESCSI) [N/y/?] Press
Enter
If you have a CD-ROM drive that is neither SCSI nor IDE/ATAPI, say Y here, otherwise N. Note
that the answer to this question doesn't directly affect the kernel: saying N will just cause the
configuration to skip all the questions about these CD-ROM drives. Most users will say N here.
*
* Input core support
* Input core support (CONFIG_INPUT) [N/y/?] Press Enter
This option allows us to enable any of the USB HID (Human Interface Device) options in the USB
support section which require Input core support. If you intended to use USB on Linux, say Y here
otherwise, you can safety say N here. It is rare that we have to use USB on server systems.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-164-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
165
*
* Character devices
* Virtual terminal (CONFIG_VT) [Y/n/?] Press Enter
This option is very important and allows us to enable support for terminal devices with display and
keyboard devices. On Linux, you need at least one virtual terminal device in order to make use of
your keyboard and monitor. Therefore, we have to say Y here.
Support for console on virtual terminal (CONFIG_VT_CONSOLE) [Y/n/?] Press Enter
This option allows us to enable support for consoles on virtual terminals. Most users will simply
keep the default value of Y here.
Standard/generic (8250/16550 and compatible UARTs) serial support
(CONFIG_SERIAL) [Y/n/?] n
This option allows us to use serial mice, modems and similar devices that connect to the standard
serial ports under Linux. If you use your Linux system as a workstation with graphical interface
installed, then you must answer Y to the question. If you use your Linux system for dedicated
Ethernet WWW/FTP servers, then you can say N here. In our example, we assume that your
Linux system is dedicated and configured to run as a server only and answer N to the question.
Non-standard serial port support (CONFIG_SERIAL_NONSTANDARD) [N/y/?] Press
Enter
This option allows us to enable support for any non-standard serial boards under Linux. These
are usually used for systems that need many serial ports, because they serve many terminals or
dial-in connections. If this is not true in your case, then you can safety say N here. Most people
will simply say N here.
Unix98 PTY support (CONFIG_UNIX98_PTYS) [Y/n/?] Press Enter
This option is important in modern Linux system and allows us to enable pseudo terminal (PTY)
support under Linux. Everyone must say Y here. Information about pseudo terminal (PTY) can be
found in the Kernel documentation.
Maximum number of Unix98 PTYs in use (0-2048) (CONFIG_UNIX98_PTY_COUNT) [256]
128
This option allows us to set the maximum number of Unix98 PTY's that can be used at any one
time. It is important to note that each additional set of 256 PTY's occupy approximately 8 KB of
kernel memory on 32-bit architectures and for security as well as performance reasons, we must
keep the number as low as possible. Therefore, we will change the default value of 256 to 128.
*
* I2C support
* I2C support (CONFIG_I2C) [N/y/?] Press Enter
This option allows us to enable I2C and SMBus support under Linux. IC2 is a slow serial bus
protocol used in many micro controller applications and developed by Philips. If you need this
feature, then say Y to the question otherwise say N here. Most people will say N here.
Bus Mouse Support (CONFIG_BUSMOUSE) [N/y/?] Press Enter
This option allows us to enable bus mouse support under Linux. All modern computer now use a
serial mouse. If you still continue to use a bus mouse on your system, then you have to say Y
here. Laptop users also need to say Y here. Most users will simply say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-165-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
166
Mouse Support (not serial and bus mice) (CONFIG_MOUSE) [Y/n/?] Press Enter
This option allows us to enable support for no serial or bus mice support under Linux. This is for
machines with a mouse, which is neither a serial, nor a bus mouse. Examples are PS/2 mice. If
you have a PS/2 mouse, then say Y here, otherwise say N here. Laptop and workstation users
also need to say Y here. If you use your Linux system for dedicated Ethernet WWW/FTP servers,
then you can say N here and save some space in your Kernel code.
PS/2 mouse (aka "auxiliary device") support (CONFIG_PSMOUSE) [Y/n/?] Press
Enter
This option enables support for a PS/2 mouse on your system. The PS/2 mouse connects to a
special mouse port that looks much like the keyboard port (small circular connector with 6 pins). If
you have this kind of mouse (most modern computers and laptops have and use it), then say Y
here otherwise say N.
C&T 82C710 mouse port support (as on TI Travelmate) (CONFIG_82C710_MOUSE)
[N/y/?] Press Enter
This option allows us to enable support for a certain kind of PS/2 mouse used on the TI
Travelmate. If you have this kind of mouse installed on your system, then say Y here. Most users
will say N to this question.
PC110 digitizer pad support (CONFIG_PC110_PAD) [N/y/?] Press Enter
This drives the digitizer pad on the IBM PC110 palmtop. It can turn the digitizer pad into PS/2
mouse emulation with tap gestures or into an absolute pad. Most users will answer this question
N.
QIC-02 tape support (CONFIG_QIC02_TAPE) [N/y/?] Press Enter
This option allows us to enable QIC-02 tape support under Linux. QIC-02 is a non-SCSI tape
drive and if you use it, then says Y to the question. Most users will say N here.
*
* Watchdog Cards
* Watchdog Timer Support (CONFIG_WATCHDOG) [N/y/?] Press Enter
This option enables Watchdog Timer support under Linux. For details about watchdog, please
read Documentation/watchdog.txt in the kernel source. Most users will say N here.
Intel i8x0 Random Number Generator support (CONFIG_INTEL_RNG) [N/y/?] Press
Enter
This option allows us to enable support for a driver that provides kernel-side support for the
Random Number Generator hardware found on Intel i8xx-based motherboards. If you have this
generator hardware installed on your computer, then you can say Y to the question. Most people
will say N here.
/dev/nvram support (CONFIG_NVRAM) [N/y/?] Press Enter
This option allows us to get read and write access to the 50 bytes of non-volatile memory in the
real time clock (RTC). This memory is conventionally called "CMOS RAM" on PCs and may be
used to view settings there, or to change them (with some utility). Most users will say N to this
question.
Enhanced Real Time Clock Support (CONFIG_RTC) [N/y/?] Press Enter
This option allows us to get access to the real time clock (or hardware clock) built into the
computer. If you run Linux on a multiprocessor machine and said Y to "Symmetric Multi
Processing" above, you should say Y here to read and set the RTC in an SMP compatible
fashion.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-166-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
167
Double Talk PC internal speech card support (CONFIG_DTLK) [N/y/?] Press Enter
This option allows us to enable support for the DoubleTalk PC under Linux. If you have a speech
card installed on your computer, then answer to this question by Y. Most users will say N here.
Siemens R3964 line discipline (CONFIG_R3964) [N/y/?] Press Enter
This option allows synchronous communication with devices using the Siemens R3964 packet
protocol. Unless you are dealing with special hardware like PLC's, you are unlikely to need this.
Most users will simply say N here.
Applicom intelligent fieldbus card support (CONFIG_APPLICOM) [N/y/?] Press
Enter
This option provides the kernel-side support for the intelligent fieldbus cards made by Applicom
International. Unless you are dealing with such a card, you are unlikely to need this. Most users
will simply say N here.
*
* Ftape, the floppy tape device driver
* Ftape (QIC-80/Travan) support (CONFIG_FTAPE) [N/y/?] Press Enter
This option allows us to enable support for some well know tape drives under Linux. If you enable
this option, then you'll be asked to select your make and model from the available list. If you don't
use tape drive on your computer, then you can say N to this option.
/dev/agpgart (AGP Support) (CONFIG_AGP) [Y/n/?] n
This option is only pertinent if you use XFree86 on your computer and want to use the GLX or
DRI features for better performance. Enable this option only if you use graphical interface on your
computer. If you system is a server, then you really don't need this option. In our example, we are
configuring the kernel for a server purpose, therefore, we will say N here.
Direct Rendering Manager (XFree86 4.1.0 and higher DRI support) (CONFIG_DRM)
[Y/n/?] n
This option is directly related to the use of XFree86 and graphical interface on your computer. It
allows us to enable Kernel-level support for the Direct Rendering Infrastructure (DRI) for better
performance of the system. If your computer doesn't have a graphical interface installed, and is
configured as a server, then you don't need to enable this option. If you say Y here because you
have and use a graphical interface, then you need to select the module that's right for your
graphics card from the available list. In our example, we are configuring the kernel for a server
purpose, therefore, we will say N here.
ACP Modem (Mwave) support (CONFIG_MWAVE) [N/y/?] Press Enter
This option allows us to enable ACP modem (Mwave) support under Linux. ACP is a WinModem
composed of a kernel driver and a user level application. Together these components support
direct attachment to public switched telephone networks (PSTNs) and support selected countries.
Most user will simply say N here.
*
* Multimedia devices
* Video For Linux (CONFIG_VIDEO_DEV) [N/y/?] Press Enter
This option allows us to enable support for audio/video capture and overlay devices and FM radio
cards under Linux. If you use graphical interface on your system, you need to say Y here. If your
system is configured as a server and you don't have graphical interface installed on it, then you
can safety say N here. In our example, we are configuring the kernel for a server purpose,
therefore, we will say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-167-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
168
*
* File systems
* Quota support (CONFIG_QUOTA) [N/y/?] y
This option is important to allow us to set per user/group limits for disk usage (also called disk
quotas) under Linux. It is sometimes required on server environments where such a purpose is
required. If you use Linux as a workstation with graphical interface, it is not useful to enable it.
Only enable this option on server environment when required.
Kernel automounter support (CONFIG_AUTOFS_FS) [N/y/?] Press Enter
This option allows us to automatically mount remote file systems on demand. A newer version of
the automounter with more features is now available; therefore we really don't need to enable this
option. Say N here.
Kernel automounter version 4 support (also supports v3) (CONFIG_AUTOFS4_FS)
[Y/n/?] n
This option allows us to enable support for the new automounter tool. If you are not a part of a
fairly large, distributed network or don't have a laptop which needs to dynamically reconfigure to
the local network, you probably do not need an automounter, and can say N here.
Ext3 journalling file system support (EXPERIMENTAL) (CONFIG_EXT3_FS) [N/y/?] y
This option is very important and allows us to enable support for the new EXT3 journaling file
system support under Linux. EXT3 is the journaling version of the Second extended file system
(often called ext3), the de facto standard Linux file system (method to organize files on a storage
device) for hard disks. Most users will say Y here to be able to get advantage of the new EXT3
journaling file system.
JBD (ext3) debugging support (CONFIG_JBD_DEBUG) [N/y/?] y
This option allows us to enable debugging output while the system is running EXT3 under Linux,
in order to help track down any problems you are having. It is a good idea to enable it, therefore
answer Y to the question.
DOS FAT fs support (CONFIG_FAT_FS) [N/y/?] Press Enter
If you want to use one of the FAT-based file systems (the MS-DOS, VFAT (Windows 95) and
UMSDOS (used to run Linux on top of an ordinary DOS partition) file systems, then you must say Y
here. In many case, we really don't need this kind of feature and can safety say N here.
Recommended choice is N and especially for a linux server system.
Compressed ROM file system support (CONFIG_CRAMFS) [N/y/?] Press Enter
This option allows us to enable CramFs support under Linux. CramFs (Compressed ROM File
System) is designed to be a simple, small, and compressed file system for ROM based embedded
systems. Most users will simply say N here.
Virtual memory file system support (former shm fs) (CONFIG_TMPFS) [Y/n/?] Press
Enter
This option allows us to enable support for Tmpfs under Linux. Tmpfs is a file system which
keeps all files in virtual memory. It is a good idea to enable it on your computer.
Simple RAM-based file system support (CONFIG_RAMFS) [N/y/?] Press Enter
This option allows us to enable support for Ramfs under Linux. Ramfs is a file system which
keeps all files in RAM. It allows read and write access. Most users will say N here.
ISO 9660 CDROM file system support (CONFIG_ISO9660_FS) [Y/n/?] Press Enter
This option is important and allows Linux to access and read your CD-ROM. Since everyone has
and uses a CD-ROM, it is vital to say Y to this question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-168-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
169
Microsoft Joliet CDROM extensions (CONFIG_JOLIET) [N/y/?] Press Enter
This option allows us to be able to read Joliet CD-ROM's under Linux. Joliet is a Microsoft
extension for the ISO 9660 CD-ROM file system, which allows for long filenames in unicode
format. If you think that you'll need to access some files written in this format under Linux with
your CD-ROM, then you need to say Y here. For workstations, you may need to say Y here but
for Linux servers, you have to say N here.
Transparent decompression extension (CONFIG_ZISOFS) [N/y/?] Press Enter
This is a Linux-specific extension to RockRidge which lets you store data in compressed form on
a CD-ROM and have it transparently decompressed when the CD-ROM is accessed. Again, for
workstations you may need it but for servers, you don't need it. Say N.
Minix fs support (CONFIG_MINIX_FS) [N/y/?] Press Enter
This option allows us to enable Minix fs support with Linux. The minix file system (method to
organize files on a hard disk partition or a floppy disk) was the original file system for Linux, but
has been superseded by the second extended file system (ext2fs). Simply say N here.
FreeVxFS file system support (VERITAS VxFS(TM) compatible) (CONFIG_VXFS_FS)
[N/y/?] Press Enter
FreeVxFS is a file system driver that support the VERITAS VxFS(TM) file system format.
VERITAS VxFS(TM) is the standard file system of SCO UnixWare (and possibly others) and
optionally available for Sunsoft Solaris, HP-UX and many other operating systems. If you want to
support such file system on your computer, then say Y here otherwise say N.
NTFS file system support (read only) (CONFIG_NTFS_FS) [N/y/?] Press Enter
NTFS is the file system of Microsoft Windows NT. Say Y if you want to get read access to files on
NTFS partitions of your hard drives otherwise say N here. Most users will say N.
OS/2 HPFS file system support (CONFIG_HPFS_FS) [N/y/?] Press Enter
OS/2 is IBM's operating system for PC's, the same as Warp, and HPFS is the file system used for
organizing files on OS/2 hard disk partitions. Say Y if you want to be able to read files from and
write files to an OS/2 HPFS partition on your hard drive. Most users will say N here.
/proc file system support (CONFIG_PROC_FS) [Y/n/?] Press Enter
This is a virtual file system providing information about the status of the system. Several
programs depend on this, so everyone should say Y here.
/dev/pts file system for Unix98 PTYs (CONFIG_DEVPTS_FS) [Y/n/?] Press Enter
This option allows us to get a virtual file system which can be mounted on /dev/pts with
"mount -t devpts". This, together with the pseudo terminal master multiplexer /dev/ptmx, is
used for pseudo terminal support as described in The Open Group's Unix98 standard. Again,
everyone should say Y here.
ROM file system support (CONFIG_ROMFS_FS) [N/y/?] Press Enter
This is a very small read-only file system mainly intended for initial ram disks of installation disks,
but it could be used for other read-only media as well. Most people will simply say N to this
question. If you want to run SCSI system on modularized kernel, you should say Y here.
Second extended fs support (CONFIG_EXT2_FS) [Y/n/?] Press Enter
This is the de facto standard Linux file system (method to organize files on a storage device) for
hard disks and you must say Y to this question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-169-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
170
System V/Xenix/V7/Coherent file system support (CONFIG_SYSV_FS) [N/y/?] Press
Enter
SCO, Xenix and Coherent are commercial Unix systems for Intel machines, and Version 7 was
used on the DEC PDP-11. Saying Y here would allow you to read from their floppies and hard
disk partitions. Most people will say N to this question.
UDF file system support (read only) (CONFIG_UDF_FS) [N/y/?] Press Enter
This is the new file system used on some CD-ROMs and DVD's. Say Y if you intend to mount
DVD discs or CD-RW's written in packet mode, or if written to by other UDF utilities, such as
DirectCD. Only enable this option if you have some such need.
UFS file system support (read only) (CONFIG_UFS_FS) [N/y/?] Press Enter
BSD and derivative versions of Unix (such as SunOS, FreeBSD, NetBSD, OpenBSD and
NeXTstep) use a file system called UFS. Some System V Unixes can create and mount hard disk
partitions and diskettes using this file system as well. Saying Y here will allow you to read from
these partitions. Most users will say N here.
*
* Network File Systems
* Coda file system support (advanced network fs) (CONFIG_CODA_FS) [N/y/?] Press
Enter
Coda is an advanced network file system, similar to NFS in that it enables you to mount file
systems of a remote server and access them with regular Unix commands as if they were sitting
on your hard disk. Enable this option only if you need it otherwise disable it.
NFS file system support (CONFIG_NFS_FS) [Y/n/?] n
If you are connected to some other (usually local) Unix computer (using SLIP, PLIP, PPP or
Ethernet) and want to mount files residing on that computer (the NFS server) using the Network
File Sharing protocol, say Y here.
NFS server support (CONFIG_NFSD) [Y/n/?] n
If you want your Linux box to act as an NFS *server*, so that other computers on your local
network which support NFS can access certain directories on your box transparently, you have
two options: you can use the self-contained user space program nfsd, in which case you should
say N here, or you can say Y and use the kernel based NFS server. The advantage of the kernel
based solution is that it is faster. Finally, if you don't want to support NFS server at all, simply say
N here.
SMB file system support (to mount Windows shares etc.) (CONFIG_SMB_FS) [N/y/?]
Press Enter
SMB (Server Message Block) is the protocol Windows for Workgroups (WfW), Windows 95/98,
Windows NT, 2000, XP and OS/2 Lan Manager use to share files and printers over local
networks. Saying Y here allows you to mount their file systems (often called "shares" in this
context) and access them just like any other Unix directory. Enable this option only if you need it.
In most cases the answer to this question will be N even if you install Samba on your system.
NCP file system support (to mount NetWare volumes) (CONFIG_NCP_FS) [N/y/?]
Press Enter
NCP (NetWare Core Protocol) is a protocol that runs over IPX and is used by Novell NetWare
clients to talk to file servers. It is to IPX what NFS is to TCP/IP, if that helps. Saying Y here allows
you to mount NetWare file server volumes and to access them just like any other Unix directory.
Enable this option only if you need it. In most cases the answer to this question will be N. You do
not have to say Y here if you want your Linux box to act as a file *server* for Novell NetWare
clients.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-170-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
171
*
* Partition Types
* Advanced partition selection (CONFIG_PARTITION_ADVANCED) [N/y/?] Press Enter
This option allows us to enable the use of hard disks under Linux which were partitioned under an
operating system running on a different architecture than the Linux system. Note that the answer
to this question won't directly affect the kernel: saying N will just cause the configuration to skip all
the questions about foreign partitioning schemes.
*
* Console drivers
* VGA text console (CONFIG_VGA_CONSOLE) [Y/n/?] Press Enter
This option allows us to use Linux in text mode through a display that complies with the generic
VGA standard. Virtually everyone wants that. Everyone should say Y here.
Video mode selection support (CONFIG_VIDEO_SELECT) [N/y/?] Press Enter
This option allows us to enable support for text mode selection on kernel startup. In general we
don't need to enable this option. Say N here and read the file Documentation/svga.txt for
more information about the Video mode selection support if you are curious.
*
* Sound
* Sound card support (CONFIG_SOUND) [Y/n/?] n
This option allows us to enable sound support under Linux. If you have a sound card in your
computer, then say Y here and select from the available list of sound card the one that you have.
If you run Linux as a workstation, you may need to say Y here, if you run Linux as a server, you
really don't need to enable this option and can safety say N here.
*
* USB support
* Support for USB (CONFIG_USB) [Y/n/?] n
This option allows us to enable USB support under Linux. If your computer has a USB port and
you want to use USB devices, then you have to say Y here. For servers you really don't need to
say Y here and can safety say N.
*
* Kernel hacking
* Kernel debugging (CONFIG_DEBUG_KERNEL) [N/y/?] Press Enter
You have to say Y here only if you are developing drivers or trying to debug and identify kernel
problems. Most users will simply say N here.
*
* Grsecurity
*
Grsecurity (CONFIG_GRKERNSEC) [N/y/?] y
This option allows us to enable Grsecurity support under Linux. If you say Y to this option, you
will be able to configure many Grsecurity features that will enhance the security of your Linux
system in many ways. This option is available ONLY if you have patched your Linux kernel with
the Grsecurity patch as discussed previously in this chapter. For best security of your Linux
server say Y here.
Security level (Low, Medium, High, Customized) [Customized] Press Enter
This Grsecurity option allows us to choose between three predefined Grsecurity
configurations or to customize the Grsecurity configuration as we want. For better control of
what the patch does and what we may or may not need, we chose to have full control about what
should or shouldn’t be enable with Grsecurity by pressing the [Enter] key to accept the
default choice of [Customized], which let us see all available security features and chose only
the one we need for our server and kernel security setup.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-171-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
172
*
* Buffer Overflow Protection
*
Openwall non-executable stack (CONFIG_GRKERNSEC_STACK) [N/y/?] y
This Grsecurity option allows us to enable the non-executable stack protection on the system.
If you say Y here, your system will not allow execution of code on the stack, making buffer
overflow exploitation more difficult. It’s a good idea to say Y here.
Gcc trampoline support (CONFIG_GRKERNSEC_STACK_GCC) [N/y/?] Press Enter
This Grsecurity option allows us to support trampoline code along with the above stack
protection. Trampolining is an action to use the ability of the stack to contain executable code,
which can improve program efficiency in a few cases. Since few programs and some version of
GCC use and need 'trampolining', it is preferable to NOT enable this option to avoid break of some
program on the system. Say N here by pressing the [Enter] key.
Read-only kernel memory (CONFIG_GRKERNSEC_KMEM) [N/y/?] y
This Grsecurity option allows us to enable the read-only kernel memory protection on the
system. If you say Y here, root will not be able to modify the contents of kernel memory. It’s a
good idea to say Y here. If you are building a Monolithic Kernel, then the ability of an attacker to
insert foreign code into a running kernel is completely removed. Yes, another good idea to build a
Monolithic Kernel instead of a Modularized Kernel. Say Y here.
*
* Access Control Lists
*
Grsecurity ACL system (CONFIG_GRKERNSEC_ACL) [N/y/?] y
This Grsecurity option allows us to enable the Access Control List system (ACL) for
Grsecurity. It’s a good idea to say Y here. ACL allows us to better control what program, files,
etc on the system are allowed to do. We use it to apply a security policy that will work for the
entire system. You can install and run Grsecurity without ACL but it is recommended for
optimum security to enable this feature and use it. Once properly implemented, it will become
impossible for a cracker to access and damage your Linux server. Personally, with Grsecurity
ACL, I don’t know how someone could break into a Linux system. Say Y here.
ACL Debugging Messages (CONFIG_GR_DEBUG) [N/y/?] y
This Grsecurity option allows the Grsecurity ACL system to print debugging messages as
an aid to finding problems in your ACL sets. It’s really a good idea to say Y here since it can
become very difficult to debug problems with ACL if this option is not enable. Say Y here.
Extra ACL Debugging Messages (CONFIG_GR_SUPERDEBUG) [N/y/?] Press Enter
This Grsecurity option allows us to enable additional debugging messages that can help in
finding problems in ACL sets or to gain a better understanding of the internal workings of the ACL
system. In many cases, it is not necessary to enable this additional debugging messages feature
for ACL and we will say N here.
Denied capability logging (CONFIG_GRKERNSEC_ACL_CAPLOG) [N/y/?] y
This Grsecurity option allows us to enable the denied capability logging support protection with
ACL. If you say Y here, logs will be produced when a root-owned process does not have a
needed capability raised in his set. This can help to debug the ACL of Grsecurity again. It’s a
good idea to say Y here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-172-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
173
Path to gradm (CONFIG_GRADM_PATH) [/sbin/gradm] Press Enter
This Grsecurity option allows us to specify the path of the gradm binary installed on the
system. gradm is the binary program used to manage Grsecurity ACL on the server. You have
to download and install it to be able to use ACL on your server. Please read the next chapter in
this book to get information about gradm and how to setup Grsecurity ACL on your system.
The default path for gradm as shown above is correct and we press the [Enter] key to accept
the default value for the path.
Maximum tries before password lockout (CONFIG_GR_MAXTRIES) [3] 2
Once the gradm program is installed on your server to control and manage Grsecurity ACL,
you will have to create a password for gradm to work. This option allows us to specify the number
of times a user can attempt to authorize them selves with the Grsecurity ACL system. Here we
change the default value of 3 to become 2, meaning that users are allowed to authorize
themselves with the Grsecurity ACL system twice.
Time to wait after max password tries, in seconds (CONFIG_GR_TIMEOUT) [30]
Press Enter
This option specifies the time the user must wait after attempting to authorize to the ACL system
with the maximum number of invalid passwords. Just press [Enter] to accept the default entry.
*
* Filesystem Protections
*
Proc restrictions (CONFIG_GRKERNSEC_PROC) [N/y/?] y
This Grsecurity option allows us to enable the proc restrictions protection on the system. If you
say Y here, the permissions of the /proc file system will be altered to enhance system security
and privacy. It’s a very good idea to say Y here.
Restrict to user only (CONFIG_GRKERNSEC_PROC_USER) [N/y/?] y
This Grsecurity option allows us to enable restrict to user only protection on the system. If you
say Y here, non-root users will only be able to view their own processes, restricts them from
viewing network-related information, viewing kernel symbol and module information. It’s a very
good idea to say Y here.
Additional restrictions (CONFIG_GRKERNSEC_PROC_ADD) [N/y/?] y
This Grsecurity option allows us to enable additional restrictions protection on the system. If
you say Y here, additional restrictions will be placed on the /proc file system that keep normal
users from viewing cpu and device information. Again, it’s a good idea to say Y here.
Linking restrictions (CONFIG_GRKERNSEC_LINK) [N/y/?] y
This Grsecurity option allows us to enable the linking restrictions protection on the system. If
you say Y here, /tmp race exploits will be prevented, since users will no longer be able to follow
symlinks owned by other users in world-writeable +t directories, unless the owner of the symlink
is the owner of the directory. Users will also not be able to hard link to files they do not own. It’s
really a good idea to say Y here.
FIFO restrictions (CONFIG_GRKERNSEC_FIFO) [N/y/?] y
This Grsecurity option allows us to enable FIFO restrictions protection on the system. If you
say Y here, users will not be able to write to FIFOs they don't own in world-writeable +t
directories, unless the owner of the FIFO is the same owner of the directory it's held in. It’s a
good idea to say Y here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-173-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
174
Secure file descriptors (CONFIG_GRKERNSEC_FD) [N/y/?] y
This Grsecurity option allows us to enable secure file descriptors protection on the system. If
you say Y here, binaries will be protected from data spoofing attacks. It’s a very good idea to say
Y here.
Chroot jail restrictions (CONFIG_GRKERNSEC_CHROOT) [N/y/?] y
This Grsecurity option allows us to enable chroot jail restrictions protection on the system. If
you say Y here, you will be able to choose several options that will make breaking out of a
chrooted jail much more difficult. It’s a very good idea to say Y here.
Restricted signals (CONFIG_GRKERNSEC_CHROOT_SIG) [N/y/?] y
This Grsecurity option allows us to enable the restricted signals protection on the system. If
you say Y here, processes inside a chroot will not be able to send signals outside of the chroot.
It’s a good idea to say Y here.
Deny mounts (CONFIG_GRKERNSEC_CHROOT_MOUNT) [N/y/?] y
This Grsecurity option allows us to enable deny mounts protection on the system. If you say Y
here, processes inside a chroot will not be able to mount or remount file systems. It’s a good idea
to say Y here.
Deny double-chroots (CONFIG_GRKERNSEC_CHROOT_DOUBLE) [N/y/?] y
This Grsecurity option allows us to enable the deny double-chroot protection on the system. If
you say Y here, processes inside a chroot will not be able to chroot again. This is a widely used
method of breaking out of a chroot jail and should not be allowed. It’s a good idea to say Y here.
Enforce chdir("/") on all chroots (CONFIG_GRKERNSEC_CHROOT_CHDIR) [N/y/?] y
This Grsecurity option allows us to enable the enforce chdir("/") on all chroots protection
on the system. If you say Y here, the current working directory of all newly-chrooted applications
will be set to the root directory of the chroot. It’s a good idea to say Y here.
Deny (f)chmod +s (CONFIG_GRKERNSEC_CHROOT_CHMOD) [N/y/?] y
This Grsecurity option allows us to enable the deny (f)chmod +s protection on the system.
If you say Y here, processes inside a chroot will not be able to chmod or fchmod files to make
them have suid or sgid bits. It’s a really good idea to say Y here.
Deny mknod (CONFIG_GRKERNSEC_CHROOT_MKNOD) [N/y/?] y
This Grsecurity option allows us to enable the deny mknod protection on the system. If you
say Y here, processes inside a chroot will not be allowed to mknod (create device on the system).
It’s a good idea to say Y here, unless you run into software incompatibilities.
Deny ptraces (CONFIG_GRKERNSEC_CHROOT_PTRACE) [N/y/?] y
This Grsecurity option allows us to enable the deny ptraces protection on the system. If you
say Y here, processes inside a chroot will not be able to ptrace other processes. It’s a good idea
to say Y here.
Restrict priority changes (CONFIG_GRKERNSEC_CHROOT_NICE) [N/y/?] y
This Grsecurity option allows us to enable the restrict priority changes protection on the
system. If you say Y here, processes inside a chroot will not be able to raise the priority of
processes in the chroot, or alter the priority of processes outside the chroot. It’s a good idea to
say Y here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-174-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
175
Capability restrictions within chroot (CONFIG_GRKERNSEC_CHROOT_CAPS) [N/y/?]
Press Enter
This Grsecurity option allows us to enable the capability restrictions within chroot protection on
the system. If you say Y here, the capabilities on all root processes within a chroot jail will be
lowered to stop module insertion, raw i/o, system and net admin tasks, rebooting the system,
modifying immutable files, and changing the system time. This option can break some
applications on the server and we disable it by answering to the question by N.
Secure keymap loading (CONFIG_GRKERNSEC_KBMAP) [N/y/?] y
This Grsecurity option allows us to enable the secure keymap loading protection on the
system. If you say Y here, KDSKBENT and KDSKBSENT ioctl calls being called by unprivileged
users will be denied. This means that unprivileged users with access to the console will NOT be
able to modify keyboard bindings. It’s a good idea to say Y here.
*
* Kernel Auditing
*
Single group for auditing (CONFIG_GRKERNSEC_AUDIT_GROUP) [N/y/?] Press Enter
This Grsecurity option allows us to enable single group for auditing protection on the system. If
you say Y here, the exec, chdir, (un)mount, and ipc logging features of Grsecurity will
only operate on a group you specify. By default Grsecurity produces a large amount of logs
from the entire system on a production server; we don’t really need the entire auditing feature
provided by Grsecurity even on specified group. Therefore we simply say N to this question
and enable later in this Grsecurity kernel configuration the auditing log that we really need for
production servers.
Exec logging (CONFIG_GRKERNSEC_EXECLOG) [N/y/?] Press Enter
This Grsecurity option allows us to enable the exec logging protection on the system. If you
say Y here, execution of any program on the server will be logged to syslog. This option when
enabled will produce a LOT of logs, especially on an active system. Therefore, we don’t
recommend you to enable this security option. If you are those people that like to spend all their
time reading log files, you could enable this option but be aware that it will take you a day to read
all the logs generated by this option on a production server. Be reasonable, and say N here.
Log execs within chroot (CONFIG_GRKERNSEC_CHROOT_EXECLOG) [N/y/?] y
This Grsecurity option allows us to enable the log execs within chroot protection on the
system. If you say Y here, all executions of any program inside a chroot jail environment will be
logged to syslog. Contrary to the previous option that logs execution of any program on the
system, this option ONLY logs the execution of program inside a chroot jail. In general, services
running in chroot jail environment are limited, meaning that your log file will not become too BIG
to read; therefore we can say Y here to enable this security option on the server.
Chdir logging (CONFIG_GRKERNSEC_AUDIT_CHDIR) [N/y/?] Press Enter
This Grsecurity option allows us to enable the chdir logging protection on the system. If you
say Y here, all chdir() calls will be logged. Chdir() calls is when you navigate via your console
on your server by using the ‘cd’ command. Imagine how many times you use the ‘cd’ command
on your server when you performing some administration tasks. I recommend you to NOT enable
this option if you want to avoid some BIG log files to read again. Say N here.
(Un)Mount logging (CONFIG_GRKERNSEC_AUDIT_MOUNT) [N/y/?] Press Enter
This Grsecurity option allows us to enable the (Un)Mount logging protection on the system. If
you say Y here, all mounts and unmounts calls will be logged to syslog. This means that each
time you mount or unmount drive on your server; the action will be logged to syslog. You can
enable this security option if you like, but personally, I prefer to disable it. Say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-175-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
176
IPC logging (CONFIG_GRKERNSEC_AUDIT_IPC) [N/y/?] y
This Grsecurity option allows us to enable the IPC logging protection on the system. If you say
Y here, creation and removal of message queues, semaphores, and shared memory will be
logged. It’s a good idea to say Y here.
Ptrace logging (CONFIG_GRKERNSEC_AUDIT_PTRACE) [N/y/?] Press Enter
This Grsecurity option allows us to enable the ptrace logging protection on the system. If you
say Y here, all successful ptraces will be logged. Ptraces are special operations performed
when programs like strace or gdb are run. In general we never install programs like strace
and gdb on a production server. These programs are only required on development server to
debug software. If you don’t have these kinds of programs installed on your server, you can
safety say N here, otherwise say Y to this security option.
Signal logging (CONFIG_GRKERNSEC_SIGNAL) [N/y/?] y
This Grsecurity option allows us to enable the signal logging protection on the system. If you
say Y here, certain important signals will be logged, such as SIGSEGV that will inform you of
when an error in a program occurred, which in some cases could mean a possible exploit
attempt. It’s a good idea to say Y here.
Fork failure logging (CONFIG_GRKERNSEC_FORKFAIL) [N/y/?] y
This Grsecurity option allows us to enable the fork failure logging protection on the system. If
you say Y here, all failed fork() attempts will be logged. This could suggest a fork bomb, or
someone attempting to overstep their process limit. It’s a good idea to say Y here.
Set*id logging (CONFIG_GRKERNSEC_SUID) [N/y/?] Press Enter
This Grsecurity option allows us to enable the set*id logging protection on the system. If you
say Y here, all set*id() calls will be logged. Enabling this security option could produce a lot of
logs on an active system that run some services that use set*id() calls to operate. Mailman,
Exim, Sendmail are known to software that uses many set*id() calls. Also, we already have
other security programs doing the same job like sXid, therefore we don’t really need to enable
this option. It is your’s to decide if you really need it or not. Personally, I disable this option by
saying N here and use sXid to achieve the same result.
Log set*ids to root (CONFIG_GRKERNSEC_SUID_ROOT) [N/y/?] y
This Grsecurity option allows us to enable the log set*ids to root protection on the system. If
you say Y here, only set*id() calls where a user is changing to the GID or UID of the root user
will be logged. Such information could be useful when detecting a possible intrusion attempt. This
option will produce smaller logs than logging all calls; therefore it’s a good idea to say Y here.
Time change logging (CONFIG_GRKERNSEC_TIME) [N/y/?] y
This Grsecurity option allows us to enable the time change logging protection on the system. If
you say Y here, any changes of the system clock will be logged. It’s a good idea to say Y here.
*
* Executable Protections
*
Exec process limiting (CONFIG_GRKERNSEC_EXECVE) [N/y/?] y
This Grsecurity option allows us to enable the exec process limiting protection on the system.
If you say Y here, users with a resource limit on processes will have the value checked during
execve() calls (execution of program). It’s really a good idea to say Y here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-176-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
177
Dmesg(8) restriction (CONFIG_GRKERNSEC_DMESG) [N/y/?] y
This Grsecurity option allows us to enable the dmesg restriction protection on the system. If
you say Y here, non-root users will not be able to use dmesg(8) to view up to the last 4kb of
messages in the kernel's log buffer. Again, it’s really a good idea to say Y here.
Randomized PIDs (CONFIG_GRKERNSEC_RANDPID) [N/y/?] y
This Grsecurity option allows us to enable the randomized PIDs protection on the system. If
you say Y here, all PIDs created on the system will be pseudo-randomly generated. This is
extremely effective to disallow an attacker from guessing pids of daemons, etc. It’s a good idea to
say Y here.
Altered default IPC permissions (CONFIG_GRKERNSEC_IPC) [N/y/?] Press Enter
This Grsecurity option allows us to enable the altered default IPC permissions protection on
the system. If you say Y here, the default permissions for IPC objects will be set based on the file
system umask of the user creating the object. This is a good security feature but unfortunately, it
is known to break software like Apache. Therefore we say N here.
Limit uid/gid changes to root (CONFIG_GRKERNSEC_TTYROOT) [N/y/?] y
This Grsecurity option allows us to enable the limit UID/GID changes to root protection on the
system. If you say Y here, you will be able choose from three options that will allow you to restrict
access to the root account by console type. Therefore we say Y here.
Deny physical consoles (tty) (CONFIG_GRKERNSEC_TTYROOT_PHYS) [N/y/?] Press
Enter
This Grsecurity option allows us to enable the deny physical consoles (tty) protection on the
system. If you say Y here, access to root from physical consoles will be denied. This is only
recommended for rare cases where you will never need to be physically at the machine. For most
of us, this is not the case and we have to say N here.
Deny serial consoles (ttyS) (CONFIG_GRKERNSEC_TTYROOT_SERIAL) [N/y/?] y
This Grsecurity option allows us to enable deny serial consoles (ttyS) protection on the
system. If you say Y here, access to root from serial consoles will be denied. Most people can say
Y here, since most don't use serial devices for their console access.
Deny pseudo consoles (pty) (CONFIG_GRKERNSEC_TTYROOT_PSEUDO) [N/y/?] Press
Enter
This Grsecurity option allows us to enable the deny pseudo consoles (pty) protection on the
system. If you say Y here, access to root from pseudo consoles will be denied. Pseudo consoles
include consoles from telnet, ssh, or any other kind of interactive shell initiated from the
network. In general, most of us use at least SSH to make a remote connection. Therefore we
have to say N here, if we want to continue to use SSH for secure remote connection.
Fork-bomb protection (CONFIG_GRKERNSEC_FORKBOMB) [N/y/?] y
This Grsecurity option allows us to enable fork-bomb protection on the system. If you say Y
here, you will be able to configure a group to add to users on your system that you want to be
unable to fork-bomb the system. It’s a good idea to say Y here and chose in the next security
option the GID to run this protection with.
GID for restricted users (CONFIG_GRKERNSEC_FORKBOMB_GID) [1006] Press Enter
This Grsecurity option allows us to enable the GID for restricted users’ protection on the
system. Here we have to enter a GID number as the value, the default value is 1006 and we can
keep it by pressing the [Enter] key. It is important to note that this GID should be added to any
user for which the feature should be activated. See the next chapter of this book for more
information about the procedures to follow. At this time you only need to accept the default value.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-177-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
178
Forks allowed per second (CONFIG_GRKERNSEC_FORKBOMB_SEC) [40] Press Enter
This Grsecurity option allows us to enable the forks allowed per second protection on the
system. This option is a continuation of the above forks options, here we have to enter the
maximum number of forks allowed per second, and the default setting should be fine for most
users.
Maximum processes allowed (CONFIG_GRKERNSEC_FORKBOMB_MAX) [20] 35
This Grsecurity option allows us to enable the maximum processes allowed on the system.
Here we have to enter the maximum number of processes users in the fork-bomb protected
group can run. We change the default value of 20 to become 35. 35 is the number you have set
into the /etc/security/limit.conf file. Please see what is set into your limit.conf file
and report it here. In general, 35 is a good value to go with.
Trusted path execution (CONFIG_GRKERNSEC_TPE) [N/y/?] y
This Grsecurity option allows us to enable trusted path execution protection on the system. If
you say Y here, you will be able to choose a GID to add to the supplementary groups of users
you want to mark as "untrusted." These users will not be able to execute any files that are not in
root-owned directories writeable only by root. It’s a good idea to say Y here.
Glibc protection (CONFIG_GRKERNSEC_TPE_GLIBC) [N/y/?] y
This Grsecurity option allows us to enable the glibc protection on the system. If you say Y
here, all non-root users executing any files while glibc specific environment variables such as
LD_PRELOAD are set, will have their environment cleared of these variables, since they could be
used to evade the trusted path execution protection. It also protects against evasion through
executing the dynamic linker to run a rogue binary. It is highly recommended you say Y here.
Partially restrict non-root users (CONFIG_GRKERNSEC_TPE_ALL) [N/y/?] y
This Grsecurity option allows us to enable the partially restrict non-root users protection on the
system. If you say Y here, All non-root users other than the ones in the group specified in the
main TPE option will only be allowed to execute files in directories they own that are not group or
world-writeable, or in directories owned by root and writeable only by root. It’s a good idea to say
Y here.
GID for untrusted users: (CONFIG_GRKERNSEC_TPE_GID) [1005] Press Enter
This Grsecurity option allows us to enable the GID for untrusted user’s protection on the
system. Here we have to enter a GID number as the value, the default value is 1005 and we can
keep it by pressing the [Enter] key. It is important to note that this GID should be added to any
user for which the feature should be activated. See the next chapter of this book for more
information about the procedures to follow. At this time you only need to accept the default value.
Restricted ptrace (CONFIG_GRKERNSEC_PTRACE) [N/y/?] y
This Grsecurity option allows us to enable the restricted ptrace protection on the system. If
you say Y here, no one but root will be able to ptrace processes. Tracing syscalls inside the
kernel will also be disabled. It’s a good idea to say Y here.
Allow ptrace for group (CONFIG_GRKERNSEC_PTRACE_GROUP) [N/y/?] Press Enter
This Grsecurity option allows us to enable the allow ptrace for group protection on the system.
If you say Y here, you will be able to choose a GID of users will be able to ptrace. Remember
that ptraces are special operations performed when programs like strace or gdb are run for
debugging purpose. Since these kind of program should in general be run on development server
and by a root user only, we can safety say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-178-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
179
*
* Network Protections
*
Randomized IP IDs (CONFIG_GRKERNSEC_RANDID) [N/y/?] y
This Grsecurity option allows us to enable the allow randomized IP IDs protection on the
system. If you say Y here, the entire ID field on all outgoing packets will be randomized. This
hinders OS fingerprinters and keeps your machine from being used as a bounce for an
untraceable portscan. It’s a good idea to say Y here.
Randomized TCP source ports (CONFIG_GRKERNSEC_RANDSRC) [N/y/?] y
This Grsecurity option allows us to enable the randomized TCP source ports protection on the
system. If you say Y here, situations where a source port is generated on the fly for the TCP
protocol (ie. with connect() ) will be altered so that the source port is generated at random,
instead of a simple incrementing algorithm. It’s a good idea to say Y here.
Randomized RPC XIDs (CONFIG_GRKERNSEC_RANDRPC) [N/y/?] y
This Grsecurity option allows us to enable the randomized RPC XIDs protection on the
system. If you say Y here, the method of determining XIDs for RPC requests will be randomized,
instead of using linux's default behavior of simply incrementing the XID. This allows us to have a
more secure RPC connection on the system. It’s a good idea to say Y here.
Altered Ping IDs (CONFIG_GRKERNSEC_RANDPING) [N/y/?] y
This Grsecurity option allows us to enable the altered Ping IDs protection on the system. If you
say Y here, the way Linux handles echo replies will be changed so that the reply uses an ID equal
to the ID of the echo request. This will help in confusing OS detection. It’s a good idea to say Y
here.
Randomized TTL (CONFIG_GRKERNSEC_RANDTTL) [N/y/?] y
This Grsecurity option allows us to enable the randomized TTL protection on the system. If
you say Y here, your TTL (time to live) for packets will be set at random, with a base of the
sysctl ttl default, to further confuse OS detection. It’s a good idea to say Y here.
Socket restrictions (CONFIG_GRKERNSEC_SOCKET) [N/y/?] y
This Grsecurity option allows us to enable the socket restrictions protection on the system. If
you say Y here, you will be able to choose from three options related to socket protection on the
server. From these three available security options, you’ll have to choose the one that best fits
your requirements. Therefore, it’s a good idea to say Y here since we have to choose one of the
three available security options for our needs.
Deny any sockets to group (CONFIG_GRKERNSEC_SOCKET_ALL) [N/y/?] y
This Grsecurity option allows us to enable the socket restrictions protection on the system. If
you say Y here, you will be able to choose a GID of whose users will be unable to connect to
other hosts from your machine or run server applications from your machine. This is the security
option that we’ll choose. Say Y here.
GID to deny all sockets for: (CONFIG_GRKERNSEC_SOCKET_ALL_GID) [1004] Press
Enter
This Grsecurity option allows us to enable the GID to deny all sockets protection on the
system. Here we have to enter a GID number as the value, the default value is 1004 and we can
keep it by pressing the [Enter] key. It is important to note that this GID should be added to any
user for which the feature should be activated. See the next chapter of this book for more
information about the procedures to follow. At this time you only need to accept the default value.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-179-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
180
Deny client sockets to group (CONFIG_GRKERNSEC_SOCKET_CLIENT) [N/y/?] Press
Enter
This Grsecurity option allows us to enable the deny client sockets to group protection on the
system. If you say Y here, you will be able to choose a GID of whose users will be unable to
connect to other hosts from your machine, but will be able to run servers. We have already
chosen the above option that enables socket protection on both ways (users cannot connect to
other hosts or run server applications from our machine), therefore we don’t need to say Y here.
Say N here.
Deny server sockets to group (CONFIG_GRKERNSEC_SOCKET_SERVER) [N/y/?] Press
Enter
This Grsecurity option allows us to enable the deny server sockets to group protection on the
system. If you say Y here, you will be able to choose a GID of whose users will be unable to run
server applications from your machine. As for the above option, we already have chosen the first
option that enable socket protection on both way (users cannot connect to other hosts or run
server applications from our machine), therefore we don’t need to say Y here. Say N here.
*
* Sysctl support
*
Sysctl support (CONFIG_GRKERNSEC_SYSCTL) [N/y/?] Press Enter
This Grsecurity option allows us to enable Grsecurity sysctl support protection on the
system. If you say Y here, you will be able to change the options that Grsecurity runs with at
bootup, without having to recompile your kernel. You can echo values to files in
/proc/sys/kernel/grsecurity to enable (1) or disable (0) various features. Enabling this
option will reduce the effectiveness of the added security of the Grsecurity patch, therefore we
say N here.
*
* Miscellaneous Features
*
Seconds in between log messages(minimum) (CONFIG_GRKERNSEC_FLOODTIME) [30]
Press Enter
This Grsecurity option allows us to enable the seconds in between log messages protection
on the system. This option allows you to enforce the number of seconds between Grsecurity
log messages. The default should be suitable for most people. Just press the [Enter] key here
to accept the default value.
BSD-style coredumps (CONFIG_GRKERNSEC_COREDUMP) [N/y/?] y
This Grsecurity option allows us to enable the BSD-style coredumps protection on the
system. If you say Y here, Linux will use a style similar to BSD for coredumps, core.processname.
Not a security feature, just a useful one. Say Y here.
*** End of Linux kernel configuration.
*** Check the top-level Makefile for additional configuration.
*** Next, you must run 'make dep'.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-180-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
181
Modularized kernel configuration
Building kernel with modules (Modularized kernel) has some advantages. It allows easy
portability between different Linux systems, since you can choose and build different parts of the
kernel as a module and load that segment of code on demand. Below we show you the
configuration of Modularized kernel, which is to compile some needed codes and drivers as
a module into the kernel by answering the different questions using y, n or m. As for the previous
Monolithic kernel configuration, don’t forget to only compile code that you need and use.
A new kernel is very specific to your computer hardware, in the Modularized kernel
configuration part below; we assume the following hardware for our example. Of course you must
change them to fit your system components.
1 Pentium-III 667 MHz (i686) processor
1 Motherboard Asus P3V4X Pro 133Mhz EIDE
1 Hard Disk Ultra ATA/100 EIDE
1 Chipset Apollo Pro133A
1 CD-ROM ATAPI IDE
1 Floppy Disk
2 Ethernet Cards 3COM 3c597 PCI 10/100
1 Mouse PS/2
If you don’t want some options listed in the Modularized kernel configuration that I enable by
default, answer n (for no) instead of y (for yes) or m (for modularized if possible) to the related
questions. If you want some other options that I disable, then answer y or m instead of n.
rm -f include/asm
( cd include ; ln -sf asm-i386 asm)
/bin/sh scripts/Configure arch/i386/config.in
#
# Using defaults found in arch/i386/defconfig
#
*
* Code maturity level options
*
Prompt for development and/or incomplete code/drivers (CONFIG_EXPERIMENTAL) [N/y/?] Press
Enter
*
* Loadable module support
*
Enable loadable module support (CONFIG_MODULES) [Y/n/?] Press Enter
Set version information on all module symbols (CONFIG_MODVERSIONS) [Y/n/?] n
Kernel module loader (CONFIG_KMOD) [Y/n/?] Press Enter
*
* Processor type and features
*
Processor family (386, 486, 586/K5/5x86/6x86/6x86MX, Pentium-Classic, Pentium-MMX,
Pentium-Pro/Celeron/Pentium-II, Pentium-III/Celeron(Coppermine), Pentium-4, K6/K6-II/K6-
III, Athlon/Duron/K7, Elan, Crusoe, Winchip-C6, Winchip-2, Winchip-2A/Winchip-3,
CyrixIII/C3) [Pentium-III/Celeron(Coppermine)] Press Enter
Toshiba Laptop support (CONFIG_TOSHIBA) [N/y/m/?] Press Enter
Dell laptop support (CONFIG_I8K) [N/y/m/?] Press Enter
/dev/cpu/microcode - Intel IA32 CPU microcode support (CONFIG_MICROCODE) [N/y/m/?] m
/dev/cpu/*/msr - Model-specific register support (CONFIG_X86_MSR) [N/y/m/?] m
/dev/cpu/*/cpuid - CPU information support (CONFIG_X86_CPUID) [N/y/m/?] m
High Memory Support (off, 4GB, 64GB) [off] Press Enter
Math emulation (CONFIG_MATH_EMULATION) [N/y/?] Press Enter
MTRR (Memory Type Range Register) support (CONFIG_MTRR) [N/y/?] Press Enter
Symmetric multi-processing support (CONFIG_SMP) [Y/n/?] n
Local APIC support on uniprocessors (CONFIG_X86_UP_APIC) [N/y/?] (NEW) y
IO-APIC support on uniprocessors (CONFIG_X86_UP_IOAPIC) [N/y/?] (NEW) y
*
* General setup](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-181-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
182
*
Networking support (CONFIG_NET) [Y/n/?] Press Enter
PCI support (CONFIG_PCI) [Y/n/?] Press Enter
PCI access mode (BIOS, Direct, Any) [Any] Press Enter
PCI device name database (CONFIG_PCI_NAMES) [Y/n/?] n
EISA support (CONFIG_EISA) [N/y/?] Press Enter
MCA support (CONFIG_MCA) [N/y/?] Press Enter
Support for hot-pluggable devices (CONFIG_HOTPLUG) [Y/n/?] n
System V IPC (CONFIG_SYSVIPC) [Y/n/?] Press Enter
BSD Process Accounting (CONFIG_BSD_PROCESS_ACCT) [N/y/?] Press Enter
Sysctl support (CONFIG_SYSCTL) [Y/n/?] Press Enter
Kernel core (/proc/kcore) format (ELF, A.OUT) [ELF] Press Enter
Kernel support for a.out binaries (CONFIG_BINFMT_AOUT) [Y/m/n/?] m
Kernel support for ELF binaries (CONFIG_BINFMT_ELF) [Y/m/n/?] Press Enter
Kernel support for MISC binaries (CONFIG_BINFMT_MISC) [Y/m/n/?] m
Power Management support (CONFIG_PM) [Y/n/?] n
*
* Memory Technology Devices (MTD)
*
Memory Technology Device (MTD) support (CONFIG_MTD) [N/y/m/?] Press Enter
*
* Parallel port support
*
Parallel port support (CONFIG_PARPORT) [N/y/m/?] Press Enter
*
* Plug and Play configuration
*
Plug and Play support (CONFIG_PNP) [Y/m/n/?] n
*
* Block devices
*
Normal PC floppy disk support (CONFIG_BLK_DEV_FD) [Y/m/n/?] Press Enter
XT hard disk support (CONFIG_BLK_DEV_XD) [N/y/m/?] Press Enter
Compaq SMART2 support (CONFIG_BLK_CPQ_DA) [N/y/m/?] Press Enter
Compaq Smart Array 5xxx support (CONFIG_BLK_CPQ_CISS_DA) [N/y/m/?] Press Enter
Mylex DAC960/DAC1100 PCI RAID Controller support (CONFIG_BLK_DEV_DAC960) [N/y/m/?] Press
Enter
Loopback device support (CONFIG_BLK_DEV_LOOP) [N/y/m/?] Press Enter
Network block device support (CONFIG_BLK_DEV_NBD) [N/y/m/?] Press Enter
RAM disk support (CONFIG_BLK_DEV_RAM) [N/y/m/?] Press Enter
*
* Multi-device support (RAID and LVM)
*
Multiple devices driver support (RAID and LVM) (CONFIG_MD) [N/y/?] Press Enter
*
* Networking options
*
Packet socket (CONFIG_PACKET) [Y/m/n/?] Press Enter
Packet socket: mmapped IO (CONFIG_PACKET_MMAP) [N/y/?] y
Netlink device emulation (CONFIG_NETLINK_DEV) [N/y/m/?] (NEW) m
Network packet filtering (replaces ipchains) (CONFIG_NETFILTER) [N/y/?] y
Network packet filtering debugging (CONFIG_NETFILTER_DEBUG) [N/y/?] (NEW) y
Socket Filtering (CONFIG_FILTER) [N/y/?] Press Enter
Unix domain sockets (CONFIG_UNIX) [Y/m/n/?] Press Enter
TCP/IP networking (CONFIG_INET) [Y/n/?] Press Enter
IP: multicasting (CONFIG_IP_MULTICAST) [Y/n/?] n
IP: advanced router (CONFIG_IP_ADVANCED_ROUTER) [N/y/?] Press Enter
IP: kernel level autoconfiguration (CONFIG_IP_PNP) [N/y/?] Press Enter
IP: tunneling (CONFIG_NET_IPIP) [N/y/m/?] Press Enter
IP: GRE tunnels over IP (CONFIG_NET_IPGRE) [N/y/m/?] Press Enter
IP: TCP Explicit Congestion Notification support (CONFIG_INET_ECN) [N/y/?] Press Enter
IP: TCP syncookie support (disabled default) (CONFIG_SYN_COOKIES) [N/y/?] y
*
* IP: Netfilter Configuration
*
Connection tracking (required for masq/NAT) (CONFIG_IP_NF_CONNTRACK) [N/y/m/?] (NEW) m
FTP protocol support (CONFIG_IP_NF_FTP) [N/m/?] (NEW) m
IRC protocol support (CONFIG_IP_NF_IRC) [N/m/?] (NEW) m
IP tables support (required for filtering/masq/NAT) (CONFIG_IP_NF_IPTABLES) [N/y/m/?]
(NEW) m
limit match support (CONFIG_IP_NF_MATCH_LIMIT) [N/m/?] (NEW) m](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-182-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
183
MAC address match support (CONFIG_IP_NF_MATCH_MAC) [N/m/?] (NEW) m
netfilter MARK match support (CONFIG_IP_NF_MATCH_MARK) [N/m/?] (NEW) m
Multiple port match support (CONFIG_IP_NF_MATCH_MULTIPORT) [N/m/?] (NEW) m
TOS match support (CONFIG_IP_NF_MATCH_TOS) [N/m/?] (NEW) m
AH/ESP match support (CONFIG_IP_NF_MATCH_AH_ESP) [N/m/?] (NEW) m
LENGTH match support (CONFIG_IP_NF_MATCH_LENGTH) [N/m/?] (NEW) m
TTL match support (CONFIG_IP_NF_MATCH_TTL) [N/m/?] (NEW) m
tcpmss match support (CONFIG_IP_NF_MATCH_TCPMSS) [N/m/?] (NEW) m
Connection state match support (CONFIG_IP_NF_MATCH_STATE) [N/m/?] (NEW) m
Packet filtering (CONFIG_IP_NF_FILTER) [N/m/?] (NEW) m
REJECT target support (CONFIG_IP_NF_TARGET_REJECT) [N/m/?] (NEW) m
Full NAT (CONFIG_IP_NF_NAT) [N/m/?] (NEW) m
MASQUERADE target support (CONFIG_IP_NF_TARGET_MASQUERADE) [N/m/?] (NEW) m
REDIRECT target support (CONFIG_IP_NF_TARGET_REDIRECT) [N/m/?] (NEW) m
Packet mangling (CONFIG_IP_NF_MANGLE) [N/m/?] (NEW) m
TOS target support (CONFIG_IP_NF_TARGET_TOS) [N/m/?] (NEW) m
MARK target support (CONFIG_IP_NF_TARGET_MARK) [N/m/?] (NEW) m
LOG target support (CONFIG_IP_NF_TARGET_LOG) [N/m/?] (NEW) m
TCPMSS target support (CONFIG_IP_NF_TARGET_TCPMSS) [N/m/?] (NEW) m
ipchains (2.2-style) support (CONFIG_IP_NF_COMPAT_IPCHAINS) [N/y/m/?] (NEW) Press Enter
ipfwadm (2.0-style) support (CONFIG_IP_NF_COMPAT_IPFWADM) [N/y/m/?] (NEW) Press Enter
*
*
*
The IPX protocol (CONFIG_IPX) [N/y/m/?] Press Enter
Appletalk protocol support (CONFIG_ATALK) [N/y/m/?] Press Enter
DECnet Support (CONFIG_DECNET) [N/y/m/?] Press Enter
802.1d Ethernet Bridging (CONFIG_BRIDGE) [N/y/m/?] Press Enter
*
* QoS and/or fair queueing
*
QoS and/or fair queueing (CONFIG_NET_SCHED) [N/y/?] Press Enter
*
* Telephony Support
*
Linux telephony support (CONFIG_PHONE) [N/y/m/?] Press Enter
*
* ATA/IDE/MFM/RLL support
*
ATA/IDE/MFM/RLL support (CONFIG_IDE) [Y/m/n/?] Press Enter
*
* IDE, ATA and ATAPI Block devices
*
Enhanced IDE/MFM/RLL disk/cdrom/tape/floppy support (CONFIG_BLK_DEV_IDE) [Y/m/n/?] Press
Enter
*
* Please see Documentation/ide.txt for help/info on IDE drives
*
Use old disk-only driver on primary interface (CONFIG_BLK_DEV_HD_IDE) [N/y/?] Press
Enter
Include IDE/ATA-2 DISK support (CONFIG_BLK_DEV_IDEDISK) [Y/m/n/?] Press Enter
Use multi-mode by default (CONFIG_IDEDISK_MULTI_MODE) [Y/n/?] n
Include IDE/ATAPI CDROM support (CONFIG_BLK_DEV_IDECD) [Y/m/n/?] Press Enter
Include IDE/ATAPI TAPE support (CONFIG_BLK_DEV_IDETAPE) [N/y/m/?] Press Enter
Include IDE/ATAPI FLOPPY support (CONFIG_BLK_DEV_IDEFLOPPY) [N/y/m/?] Press Enter
SCSI emulation support (CONFIG_BLK_DEV_IDESCSI) [N/y/m/?] Press Enter
*
* IDE chipset support/bugfixes
*
CMD640 chipset bugfix/support (CONFIG_BLK_DEV_CMD640) [Y/n/?] n
RZ1000 chipset bugfix/support (CONFIG_BLK_DEV_RZ1000) [Y/n/?] n
Generic PCI IDE chipset support (CONFIG_BLK_DEV_IDEPCI) [Y/n/?] Press Enter
Sharing PCI IDE interrupts support (CONFIG_IDEPCI_SHARE_IRQ) [Y/n/?] Press Enter
Generic PCI bus-master DMA support (CONFIG_BLK_DEV_IDEDMA_PCI) [Y/n/?] Press Enter
Boot off-board chipsets first support (CONFIG_BLK_DEV_OFFBOARD) [N/y/?] Press Enter
Use PCI DMA by default when available (CONFIG_IDEDMA_PCI_AUTO) [Y/n/?] Press Enter
AEC62XX chipset support (CONFIG_BLK_DEV_AEC62XX) [N/y/?] Press Enter
ALI M15x3 chipset support (CONFIG_BLK_DEV_ALI15X3) [N/y/?] Press Enter
AMD Viper support (CONFIG_BLK_DEV_AMD74XX) [N/y/?] Press Enter
CMD64X chipset support (CONFIG_BLK_DEV_CMD64X) [N/y/?] Press Enter
CY82C693 chipset support (CONFIG_BLK_DEV_CY82C693) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-183-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
184
Cyrix CS5530 MediaGX chipset support (CONFIG_BLK_DEV_CS5530) [N/y/?] Press Enter
HPT34X chipset support (CONFIG_BLK_DEV_HPT34X) [N/y/?] Press Enter
HPT366 chipset support (CONFIG_BLK_DEV_HPT366) [N/y/?] Press Enter
Intel PIIXn chipsets support (CONFIG_BLK_DEV_PIIX) [Y/n/?] Press Enter
PIIXn Tuning support (CONFIG_PIIX_TUNING) [Y/n/?] Press Enter
NS87415 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_NS87415) [N/y/?] Press Enter
PROMISE PDC202{46|62|65|67|68} support (CONFIG_BLK_DEV_PDC202XX) [N/y/?] Press Enter
ServerWorks OSB4/CSB5 chipsets support (CONFIG_BLK_DEV_SVWKS) [N/y/?] Press Enter
SiS5513 chipset support (CONFIG_BLK_DEV_SIS5513) [N/y/?] Press Enter
SLC90E66 chipset support (CONFIG_BLK_DEV_SLC90E66) [N/y/?] Press Enter
Tekram TRM290 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_TRM290) [N/y/?] Press
Enter
VIA82CXXX chipset support (CONFIG_BLK_DEV_VIA82CXXX) [N/y/?] y
Other IDE chipset support (CONFIG_IDE_CHIPSETS) [N/y/?] Press Enter
IGNORE word93 Validation BITS (CONFIG_IDEDMA_IVB) [N/y/?] Press Enter
*
* SCSI support
*
SCSI support (CONFIG_SCSI) [Y/m/n/?] n
*
* Fusion MPT device support
*
*
* I2O device support
*
I2O support (CONFIG_I2O) [N/y/m/?] Press Enter
*
* Network device support
*
Network device support (CONFIG_NETDEVICES) [Y/n/?] Press Enter
*
* ARCnet devices
*
ARCnet support (CONFIG_ARCNET) [N/y/m/?] Press Enter
Dummy net driver support (CONFIG_DUMMY) [M/n/y/?] Press Enter
Bonding driver support (CONFIG_BONDING) [N/y/m/?] Press Enter
EQL (serial line load balancing) support (CONFIG_EQUALIZER) [N/y/m/?] Press Enter
Universal TUN/TAP device driver support (CONFIG_TUN) [N/y/m/?] Press Enter
*
* Ethernet (10 or 100Mbit)
*
Ethernet (10 or 100Mbit) (CONFIG_NET_ETHERNET) [Y/n/?] Press Enter
Sun Happy Meal 10/100baseT support (CONFIG_HAPPYMEAL) [N/y/m/?] Press Enter
Sun GEM support (CONFIG_SUNGEM) [N/y/m/?] Press Enter
3COM cards (CONFIG_NET_VENDOR_3COM) [N/y/?] y
3c501 "EtherLink" support (CONFIG_EL1) [N/y/m/?] (NEW) Press Enter
3c503 "EtherLink II" support (CONFIG_EL2) [N/y/m/?] (NEW) Press Enter
3c505 "EtherLink Plus" support (CONFIG_ELPLUS) [N/y/m/?] (NEW) Press Enter
3c509/3c529 (MCA)/3c579 "EtherLink III" support (CONFIG_EL3) [N/y/m/?] (NEW) Press
Enter
3c515 ISA "Fast EtherLink" (CONFIG_3C515) [N/y/m/?] (NEW) Press Enter
3c590/3c900 series (592/595/597) "Vortex/Boomerang" support (CONFIG_VORTEX) [N/y/m/?]
(NEW) y
AMD LANCE and PCnet (AT1500 and NE2100) support (CONFIG_LANCE) [N/y/m/?] Press Enter
Western Digital/SMC cards (CONFIG_NET_VENDOR_SMC) [N/y/?] Press Enter
Racal-Interlan (Micom) NI cards (CONFIG_NET_VENDOR_RACAL) [N/y/?] Press Enter
DEPCA, DE10x, DE200, DE201, DE202, DE422 support (CONFIG_DEPCA) [N/y/m/?] Press Enter
HP 10/100VG PCLAN (ISA, EISA, PCI) support (CONFIG_HP100) [N/y/m/?] Press Enter
Other ISA cards (CONFIG_NET_ISA) [N/y/?] Press Enter
EISA, VLB, PCI and on board controllers (CONFIG_NET_PCI) [Y/n/?] n
Pocket and portable adapters (CONFIG_NET_POCKET) [N/y/?] Press Enter
*
* Ethernet (1000 Mbit)
*
Alteon AceNIC/3Com 3C985/NetGear GA620 Gigabit support (CONFIG_ACENIC) [N/y/m/?] Press
Enter
D-Link DL2000-based Gigabit Ethernet support (CONFIG_DL2K) [N/y/m/?] Press Enter
National Semiconduct DP83820 support (CONFIG_NS83820) [N/y/m/?] Press Enter
Packet Engines Hamachi GNIC-II support (CONFIG_HAMACHI) [N/y/m/?] Press Enter
SysKonnect SK-98xx support (CONFIG_SK98LIN) [N/y/m/?] Press Enter
FDDI driver support (CONFIG_FDDI) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-184-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
185
PPP (point-to-point protocol) support (CONFIG_PPP) [N/y/m/?] Press Enter
SLIP (serial line) support (CONFIG_SLIP) [N/y/m/?] Press Enter
*
* Wireless LAN (non-hamradio)
*
Wireless LAN (non-hamradio) (CONFIG_NET_RADIO) [N/y/?] Press Enter
*
* Token Ring devices
*
Token Ring driver support (CONFIG_TR) [N/y/?] Press Enter
Fibre Channel driver support (CONFIG_NET_FC) [N/y/?] Press Enter
*
* Wan interfaces
*
Wan interfaces support (CONFIG_WAN) [N/y/?] Press Enter
*
* Amateur Radio support
*
Amateur Radio support (CONFIG_HAMRADIO) [N/y/?] Press Enter
*
* IrDA (infrared) support
*
IrDA subsystem support (CONFIG_IRDA) [N/y/m/?] Press Enter
*
* ISDN subsystem
*
ISDN support (CONFIG_ISDN) [N/y/m/?] Press Enter
*
* Old CD-ROM drivers (not SCSI, not IDE)
*
Support non-SCSI/IDE/ATAPI CDROM drives (CONFIG_CD_NO_IDESCSI) [N/y/?] Press Enter
*
* Input core support
*
Input core support (CONFIG_INPUT) [N/y/m/?] Press Enter
*
* Character devices
*
Virtual terminal (CONFIG_VT) [Y/n/?] Press Enter
Support for console on virtual terminal (CONFIG_VT_CONSOLE) [Y/n/?] Press Enter
Standard/generic (8250/16550 and compatible UARTs) serial support (CONFIG_SERIAL)
[Y/m/n/?] Press Enter
Support for console on serial port (CONFIG_SERIAL_CONSOLE) [N/y/?] Press Enter
Extended dumb serial driver options (CONFIG_SERIAL_EXTENDED) [N/y/?] Press Enter
Non-standard serial port support (CONFIG_SERIAL_NONSTANDARD) [N/y/?] Press Enter
Unix98 PTY support (CONFIG_UNIX98_PTYS) [Y/n/?] Press Enter
Maximum number of Unix98 PTYs in use (0-2048) (CONFIG_UNIX98_PTY_COUNT) [256] 128
*
* I2C support
*
I2C support (CONFIG_I2C) [N/y/m/?] Press Enter
*
* Mice
*
Bus Mouse Support (CONFIG_BUSMOUSE) [N/y/m/?] Press Enter
Mouse Support (not serial and bus mice) (CONFIG_MOUSE) [Y/m/n/?] n
*
* Joysticks
*
*
* Input core support is needed for gameports
*
*
* Input core support is needed for joysticks
*
QIC-02 tape support (CONFIG_QIC02_TAPE) [N/y/m/?] Press Enter
*
* Watchdog Cards
*
Watchdog Timer Support (CONFIG_WATCHDOG) [N/y/?] Press Enter
Intel i8x0 Random Number Generator support (CONFIG_INTEL_RNG) [N/y/m/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-185-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
186
/dev/nvram support (CONFIG_NVRAM) [N/y/m/?] Press Enter
Enhanced Real Time Clock Support (CONFIG_RTC) [N/y/m/?] Press Enter
Double Talk PC internal speech card support (CONFIG_DTLK) [N/y/m/?] Press Enter
Siemens R3964 line discipline (CONFIG_R3964) [N/y/m/?] Press Enter
Applicom intelligent fieldbus card support (CONFIG_APPLICOM) [N/y/m/?] Press Enter
*
* Ftape, the floppy tape device driver
*
Ftape (QIC-80/Travan) support (CONFIG_FTAPE) [N/y/m/?] Press Enter
/dev/agpgart (AGP Support) (CONFIG_AGP) [Y/m/n/?] n
Direct Rendering Manager (XFree86 DRI support) (CONFIG_DRM) [Y/n/?] n
ACP Modem (Mwave) support (CONFIG_MWAVE) [N/y/m/?] Press Enter
*
* Multimedia devices
*
Video For Linux (CONFIG_VIDEO_DEV) [N/y/m/?] Press Enter
*
* File systems
*
Quota support (CONFIG_QUOTA) [N/y/?] y
Kernel automounter support (CONFIG_AUTOFS_FS) [N/y/m/?] Press Enter
Kernel automounter version 4 support (also supports v3) (CONFIG_AUTOFS4_FS) [Y/m/n/?] n
Reiserfs support (CONFIG_REISERFS_FS) [N/y/m/?] Press Enter
Ext3 journalling file system support (EXPERIMENTAL) (CONFIG_EXT3_FS) [N/y/m/?] y
JBD (ext3) debugging support (CONFIG_JBD_DEBUG) [N/y/?] y
DOS FAT fs support (CONFIG_FAT_FS) [N/y/m/?] m
MSDOS fs support (CONFIG_MSDOS_FS) [N/y/m/?] m
VFAT (Windows-95) fs support (CONFIG_VFAT_FS) [N/y/m/?] m
Compressed ROM file system support (CONFIG_CRAMFS) [N/y/m/?] Press Enter
Virtual memory file system support (former shm fs) (CONFIG_TMPFS) [Y/n/?] Press Enter
Simple RAM-based file system support (CONFIG_RAMFS) [N/y/m/?] Press Enter
ISO 9660 CDROM file system support (CONFIG_ISO9660_FS) [Y/m/n/?] m
Microsoft Joliet CDROM extensions (CONFIG_JOLIET) [N/y/?] y
Transparent decompression extension (CONFIG_ZISOFS) [N/y/?] Press Enter
Minix fs support (CONFIG_MINIX_FS) [N/y/m/?] Press Enter
FreeVxFS file system support (VERITAS VxFS(TM) compatible) (CONFIG_VXFS_FS) [N/y/m/?]
Press Enter
NTFS file system support (read only) (CONFIG_NTFS_FS) [N/y/m/?] Press Enter
OS/2 HPFS file system support (CONFIG_HPFS_FS) [N/y/m/?] Press Enter
/proc file system support (CONFIG_PROC_FS) [Y/n/?] Press Enter
/dev/pts file system for Unix98 PTYs (CONFIG_DEVPTS_FS) [Y/n/?] Press Enter
ROM file system support (CONFIG_ROMFS_FS) [N/y/m/?] Press Enter
Second extended fs support (CONFIG_EXT2_FS) [Y/m/n/?] Press Enter
System V/Xenix/V7/Coherent file system support (CONFIG_SYSV_FS) [N/y/m/?] Press Enter
UDF file system support (read only) (CONFIG_UDF_FS) [N/y/m/?] Press Enter
UFS file system support (read only) (CONFIG_UFS_FS) [N/y/m/?] Press Enter
*
* Network File Systems
*
Coda file system support (advanced network fs) (CONFIG_CODA_FS) [N/y/m/?] Press Enter
NFS file system support (CONFIG_NFS_FS) [Y/m/n/?] n
NFS server support (CONFIG_NFSD) [Y/m/n/?] n
SMB file system support (to mount Windows shares etc.) (CONFIG_SMB_FS) [N/y/m/?] Press
Enter
NCP file system support (to mount NetWare volumes) (CONFIG_NCP_FS) [N/y/m/?] Press Enter
*
* Partition Types
*
Advanced partition selection (CONFIG_PARTITION_ADVANCED) [N/y/?] Press Enter
*
* Native Language Support
*
Default NLS Option (CONFIG_NLS_DEFAULT) [iso8859-1] (NEW) Press Enter
Codepage 437 (United States, Canada) (CONFIG_NLS_CODEPAGE_437) [N/y/m/?] (NEW) Press
Enter
Codepage 737 (Greek) (CONFIG_NLS_CODEPAGE_737) [N/y/m/?] (NEW) Press Enter
Codepage 775 (Baltic Rim) (CONFIG_NLS_CODEPAGE_775) [N/y/m/?] (NEW) Press Enter
Codepage 850 (Europe) (CONFIG_NLS_CODEPAGE_850) [N/y/m/?] (NEW) Press Enter
Codepage 852 (Central/Eastern Europe) (CONFIG_NLS_CODEPAGE_852) [N/y/m/?] (NEW) Press
Enter
Codepage 855 (Cyrillic) (CONFIG_NLS_CODEPAGE_855) [N/y/m/?] (NEW) Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-186-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
187
Codepage 857 (Turkish) (CONFIG_NLS_CODEPAGE_857) [N/y/m/?] (NEW) Press Enter
Codepage 860 (Portuguese) (CONFIG_NLS_CODEPAGE_860) [N/y/m/?] (NEW) Press Enter
Codepage 861 (Icelandic) (CONFIG_NLS_CODEPAGE_861) [N/y/m/?] (NEW) Press Enter
Codepage 862 (Hebrew) (CONFIG_NLS_CODEPAGE_862) [N/y/m/?] (NEW) Press Enter
Codepage 863 (Canadian French) (CONFIG_NLS_CODEPAGE_863) [N/y/m/?] (NEW) Press Enter
Codepage 864 (Arabic) (CONFIG_NLS_CODEPAGE_864) [N/y/m/?] (NEW) Press Enter
Codepage 865 (Norwegian, Danish) (CONFIG_NLS_CODEPAGE_865) [N/y/m/?] (NEW) Press Enter
Codepage 866 (Cyrillic/Russian) (CONFIG_NLS_CODEPAGE_866) [N/y/m/?] (NEW) Press Enter
Codepage 869 (Greek) (CONFIG_NLS_CODEPAGE_869) [N/y/m/?] (NEW) Press Enter
Simplified Chinese charset (CP936, GB2312) (CONFIG_NLS_CODEPAGE_936) [N/y/m/?] (NEW)
Press Enter
Traditional Chinese charset (Big5) (CONFIG_NLS_CODEPAGE_950) [N/y/m/?] (NEW) Press Enter
Japanese charsets (Shift-JIS, EUC-JP) (CONFIG_NLS_CODEPAGE_932) [N/y/m/?] (NEW) Press
Enter
Korean charset (CP949, EUC-KR) (CONFIG_NLS_CODEPAGE_949) [N/y/m/?] (NEW) Press Enter
Thai charset (CP874, TIS-620) (CONFIG_NLS_CODEPAGE_874) [N/y/m/?] (NEW) Press Enter
Hebrew charsets (ISO-8859-8, CP1255) (CONFIG_NLS_ISO8859_8) [N/y/m/?] (NEW) Press Enter
Windows CP1250 (Slavic/Central European Languages) (CONFIG_NLS_CODEPAGE_1250) [N/y/m/?]
(NEW) Press Enter
Windows CP1251 (Bulgarian, Belarusian) (CONFIG_NLS_CODEPAGE_1251) [N/y/m/?] (NEW) Press
Enter
NLS ISO 8859-1 (Latin 1; Western European Languages) (CONFIG_NLS_ISO8859_1) [N/y/m/?]
(NEW) Press Enter
NLS ISO 8859-2 (Latin 2; Slavic/Central European Languages) (CONFIG_NLS_ISO8859_2)
[N/y/m/?] (NEW) Press Enter
NLS ISO 8859-3 (Latin 3; Esperanto, Galician, Maltese, Turkish) (CONFIG_NLS_ISO8859_3)
[N/y/m/?] (NEW) Press Enter
NLS ISO 8859-4 (Latin 4; old Baltic charset) (CONFIG_NLS_ISO8859_4) [N/y/m/?] (NEW)
Press Enter
NLS ISO 8859-5 (Cyrillic) (CONFIG_NLS_ISO8859_5) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-6 (Arabic) (CONFIG_NLS_ISO8859_6) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-7 (Modern Greek) (CONFIG_NLS_ISO8859_7) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-9 (Latin 5; Turkish) (CONFIG_NLS_ISO8859_9) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-13 (Latin 7; Baltic) (CONFIG_NLS_ISO8859_13) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-14 (Latin 8; Celtic) (CONFIG_NLS_ISO8859_14) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-15 (Latin 9; Western European Languages with Euro) (CONFIG_NLS_ISO8859_15)
[N/y/m/?] (NEW) Press Enter
NLS KOI8-R (Russian) (CONFIG_NLS_KOI8_R) [N/y/m/?] (NEW) Press Enter
NLS KOI8-U/RU (Ukrainian, Belarusian) (CONFIG_NLS_KOI8_U) [N/y/m/?] (NEW) Press Enter
NLS UTF8 (CONFIG_NLS_UTF8) [N/y/m/?] (NEW) Press Enter
*
* Console drivers
*
VGA text console (CONFIG_VGA_CONSOLE) [Y/n/?] Press Enter
Video mode selection support (CONFIG_VIDEO_SELECT) [N/y/?] Press Enter
*
* Sound
*
Sound card support (CONFIG_SOUND) [Y/m/n/?] n
*
* USB support
*
Support for USB (CONFIG_USB) [Y/m/n/?] n
*
* USB Controllers
*
*
* USB Device Class drivers
*
*
* SCSI support is needed for USB Storage
*
*
* USB Human Interface Devices (HID)
*
*
* Input core support is needed for USB HID
*
*
* USB Imaging devices
*](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-187-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
188
*
* USB Multimedia devices
*
*
* Video4Linux support is needed for USB Multimedia device support
*
*
* USB Network adaptors
*
*
* USB port drivers
*
*
* USB Serial Converter support
*
*
* USB Miscellaneous drivers
*
*
* Kernel hacking
*
Kernel debugging (CONFIG_DEBUG_KERNEL) [N/y/?] Press Enter
*
* Grsecurity
*
Grsecurity (CONFIG_GRKERNSEC) [N/y/?] y
Security level (Low, Medium, High, Customized) [Customized] Press Enter
*
* Buffer Overflow Protection
*
Openwall non-executable stack (CONFIG_GRKERNSEC_STACK) [N/y/?] y
Gcc trampoline support (CONFIG_GRKERNSEC_STACK_GCC) [N/y/?] Press Enter
Read-only kernel memory (CONFIG_GRKERNSEC_KMEM) [N/y/?] y
*
* Access Control Lists
*
Grsecurity ACL system (CONFIG_GRKERNSEC_ACL) [N/y/?] y
ACL Debugging Messages (CONFIG_GR_DEBUG) [N/y/?] y
Extra ACL Debugging Messages (CONFIG_GR_SUPERDEBUG) [N/y/?] Press Enter
Denied capability logging (CONFIG_GRKERNSEC_ACL_CAPLOG) [N/y/?] y
Path to gradm (CONFIG_GRADM_PATH) [/sbin/gradm] Press Enter
Maximum tries before password lockout (CONFIG_GR_MAXTRIES) [3] 2
Time to wait after max password tries, in seconds (CONFIG_GR_TIMEOUT) [30] Press Enter
*
* Filesystem Protections
*
Proc restrictions (CONFIG_GRKERNSEC_PROC) [N/y/?] y
Restrict to user only (CONFIG_GRKERNSEC_PROC_USER) [N/y/?] y
Additional restrictions (CONFIG_GRKERNSEC_PROC_ADD) [N/y/?] y
Linking restrictions (CONFIG_GRKERNSEC_LINK) [N/y/?] y
FIFO restrictions (CONFIG_GRKERNSEC_FIFO) [N/y/?] y
Secure file descriptors (CONFIG_GRKERNSEC_FD) [N/y/?] y
Chroot jail restrictions (CONFIG_GRKERNSEC_CHROOT) [N/y/?] y
Restricted signals (CONFIG_GRKERNSEC_CHROOT_SIG) [N/y/?] y
Deny mounts (CONFIG_GRKERNSEC_CHROOT_MOUNT) [N/y/?] y
Deny double-chroots (CONFIG_GRKERNSEC_CHROOT_DOUBLE) [N/y/?] y
Enforce chdir("/") on all chroots (CONFIG_GRKERNSEC_CHROOT_CHDIR) [N/y/?] y
Deny (f)chmod +s (CONFIG_GRKERNSEC_CHROOT_CHMOD) [N/y/?] y
Deny mknod (CONFIG_GRKERNSEC_CHROOT_MKNOD) [N/y/?] y
Deny ptraces (CONFIG_GRKERNSEC_CHROOT_PTRACE) [N/y/?] y
Restrict priority changes (CONFIG_GRKERNSEC_CHROOT_NICE) [N/y/?] y
Capability restrictions within chroot (CONFIG_GRKERNSEC_CHROOT_CAPS) [N/y/?] Press Enter
Secure keymap loading (CONFIG_GRKERNSEC_KBMAP) [N/y/?] y
*
* Kernel Auditing
*
Single group for auditing (CONFIG_GRKERNSEC_AUDIT_GROUP) [N/y/?] Press Enter
Exec logging (CONFIG_GRKERNSEC_EXECLOG) [N/y/?] Press Enter
Log execs within chroot (CONFIG_GRKERNSEC_CHROOT_EXECLOG) [N/y/?] y
Chdir logging (CONFIG_GRKERNSEC_AUDIT_CHDIR) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-188-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
189
(Un)Mount logging (CONFIG_GRKERNSEC_AUDIT_MOUNT) [N/y/?] Press Enter
IPC logging (CONFIG_GRKERNSEC_AUDIT_IPC) [N/y/?] y
Ptrace logging (CONFIG_GRKERNSEC_AUDIT_PTRACE) [N/y/?] Press Enter
Signal logging (CONFIG_GRKERNSEC_SIGNAL) [N/y/?] y
Fork failure logging (CONFIG_GRKERNSEC_FORKFAIL) [N/y/?] y
Set*id logging (CONFIG_GRKERNSEC_SUID) [N/y/?] Press Enter
Log set*ids to root (CONFIG_GRKERNSEC_SUID_ROOT) [N/y/?] y
Time change logging (CONFIG_GRKERNSEC_TIME) [N/y/?] y
*
* Executable Protections
*
Exec process limiting (CONFIG_GRKERNSEC_EXECVE) [N/y/?] y
Dmesg(8) restriction (CONFIG_GRKERNSEC_DMESG) [N/y/?] y
Randomized PIDs (CONFIG_GRKERNSEC_RANDPID) [N/y/?] y
Altered default IPC permissions (CONFIG_GRKERNSEC_IPC) [N/y/?] Press Enter
Limit uid/gid changes to root (CONFIG_GRKERNSEC_TTYROOT) [N/y/?] y
Deny physical consoles (tty) (CONFIG_GRKERNSEC_TTYROOT_PHYS) [N/y/?] Press Enter
Deny serial consoles (ttyS) (CONFIG_GRKERNSEC_TTYROOT_SERIAL) [N/y/?] y
Deny pseudo consoles (pty) (CONFIG_GRKERNSEC_TTYROOT_PSEUDO) [N/y/?] Press Enter
Fork-bomb protection (CONFIG_GRKERNSEC_FORKBOMB) [N/y/?] y
GID for restricted users (CONFIG_GRKERNSEC_FORKBOMB_GID) [1006] Press Enter
Forks allowed per second (CONFIG_GRKERNSEC_FORKBOMB_SEC) [40] Press Enter
Maximum processes allowed (CONFIG_GRKERNSEC_FORKBOMB_MAX) [20] 35
Trusted path execution (CONFIG_GRKERNSEC_TPE) [N/y/?] y
Glibc protection (CONFIG_GRKERNSEC_TPE_GLIBC) [N/y/?] y
Partially restrict non-root users (CONFIG_GRKERNSEC_TPE_ALL) [N/y/?] y
GID for untrusted users: (CONFIG_GRKERNSEC_TPE_GID) [1005] Press Enter
Restricted ptrace (CONFIG_GRKERNSEC_PTRACE) [N/y/?] y
Allow ptrace for group (CONFIG_GRKERNSEC_PTRACE_GROUP) [N/y/?] Press Enter
*
* Network Protections
*
Randomized IP IDs (CONFIG_GRKERNSEC_RANDID) [N/y/?] y
Randomized TCP source ports (CONFIG_GRKERNSEC_RANDSRC) [N/y/?] y
Randomized RPC XIDs (CONFIG_GRKERNSEC_RANDRPC) [N/y/?] y
Altered Ping IDs (CONFIG_GRKERNSEC_RANDPING) [N/y/?] y
Randomized TTL (CONFIG_GRKERNSEC_RANDTTL) [N/y/?] y
Socket restrictions (CONFIG_GRKERNSEC_SOCKET) [N/y/?] y
Deny any sockets to group (CONFIG_GRKERNSEC_SOCKET_ALL) [N/y/?] y
GID to deny all sockets for: (CONFIG_GRKERNSEC_SOCKET_ALL_GID) [1004] Press Enter
Deny client sockets to group (CONFIG_GRKERNSEC_SOCKET_CLIENT) [N/y/?] Press Enter
Deny server sockets to group (CONFIG_GRKERNSEC_SOCKET_SERVER) [N/y/?] Press Enter
*
* Sysctl support
*
Sysctl support (CONFIG_GRKERNSEC_SYSCTL) [N/y/?] Press Enter
*
* Miscellaneous Features
*
Seconds in between log messages(minimum) (CONFIG_GRKERNSEC_FLOODTIME) [30] Press Enter
BSD-style coredumps (CONFIG_GRKERNSEC_COREDUMP) [N/y/?] y
*** End of Linux kernel configuration.
*** Check the top-level Makefile for additional configuration.
*** Next, you must run 'make dep'.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-189-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
190
Compiling the Kernel
This section applies to both Monolithic and Modularized kernels. Once the kernel configuration
has been completed, return to the /usr/src/linux directory (if you are not already in it) and
compile the new kernel. You do so by using the following command:
• To compile the Kernel, use the following command:
[root@deep linux]# make dep; make clean; make bzImage
This line contains three commands in one. The first one, make dep, actually takes your
configuration and builds the corresponding dependency tree. This process determines what gets
compiled and what doesn’t. The next step, make clean, erases all previous traces of a
compilation so as to avoid any mistakes in which the wrong version of a feature gets tied into the
kernel. Finally, make bzImage does the full compilation of the kernel.
After the process is complete, the kernel is compressed and ready to be installed on your system.
Before we can install the new kernel, we must know if we need to compile the corresponding
modules. This is required ONLY if you said yes to “Enable loadable module support
(CONFIG_MODULES)” and have compiled some options in the kernel configuration above as a
module (See Modularized kernel configuration). In this case, you must execute the following
commands:
• To compile the corresponding modules for your kernel, use the following commands:
[root@deep linux]# make modules
[root@deep linux]# make modules_install
WARNING: The make modules and make modules_install commands are required ONLY if
you say yes to “Enable loadable module support (CONFIG_MODULES)” in your kernel
configurations (See Modularized kernel configuration) because you want to build a
modularized kernel.
Installing the Kernel
This section applies to both the Monolithic and Modularized kernel. Ok, the kernel has been
configured, compiled and is now ready to be installed your system. Below are the steps required
to install all the necessary kernel components in your system.
Step 1
Copy the file /usr/src/linux/arch/i386/boot/bzImage from the kernel source tree to the
/boot directory, and give it an appropriate new name.
• To copy the bzImage file to the /boot directory, use the following commands:
[root@deep /]# cd /usr/src/linux/ (if you are not already in it)
[root@deep linux]# cp arch/i386/boot/bzImage /boot/vmlinuz-2.4.18](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-190-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
191
Step 2
A new System.map file is generated when you compile a kernel, and is a list of all the addresses
in that kernel and their corresponding symbols. Every time that you create a new kernel, such a
file System.map is created and saved in /usr/src/linux. It's a text file, which is read by a
few programs to do address <-> symbol translation, and which you need if you ever get an Oops.
Certain commands, like klog, ps, and lsof, use the System.map file to get the name of kernel
symbols. Without it some commands like lsof will complain that they can't find a System.map
file to match the currently booted kernel.
Copy the file /usr/src/linux/System.map from the kernel source tree to the /boot
directory, and give it an appropriate new name.
• To copy the System.map file to the /boot directory, use the following commands:
[root@deep /]# cd /usr/src/linux/ (if you are not already in it)
[root@deep linux]# cp System.map /boot/System.map-2.4.18
Step 3
Move into the /boot directory and rebuild the links vmlinuz and System.map.
• To rebuild the vmlinuz and System.map files, use the following commands:
[root@deep linux]# cd /boot/
[root@deep /boot]# ln -fs vmlinuz-2.4.18 vmlinuz
[root@deep /boot]# ln -fs System.map-2.4.18 System.map
We must rebuild the links of vmlinuz and System.map to point them to the new installed kernel
version. Without the new links LILO or GRUB program will look, by default, for the old version of
your Linux kernel.
Step 4
Remove obsolete and unnecessary files under the /boot directory to increase disk space:
• To remove obsolete and unnecessary files under the /boot directory, use commands:
[root@deep /]# cd /boot/ (if you are not already in it)
[root@deep /boot]# rm -f module-info
[root@deep /boot]# rm -f initrd-2.4.x.img
The module-info is a link, which points to the old modules directory of your original kernel.
Since we have installed a brand new kernel, we don’t need to keep this broken link.
The initrd-2.4.x.img is a file that contains an initial RAM disk image that serves as a
system before the disk is available. This file is only available and is installed by the Linux initial
setup installation if your system has a SCSI adapter present and only if your system has a SCSI
adapter. If we use and have a SCSI system, the required driver now will have been incorporated
into our new Linux kernel since we have build it by answering Y to the question related to our
SCSI model during the configuration of the kernel, so we can safely remove this file (initrd-
2.4.x.img).](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-191-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
192
Step 5
Create a new Linux kernel directory that will handle all the header files related to Linux kernel for
future compilation of other programs on your system.
Recall, we had created two symlinks under the /usr/include directory that pointed to the Linux
kernel header files to be able to compile it without receiving errors and also be able to compile
future programs. The /usr/include directory is where all the header files for your Linux system
are kept for reference and dependencies when you compile and install new programs.
The asm, and linux links are used when programs need to know some functions which are
compile-time specific to the kernel installed on your system. Programs call other headers as well
in the /usr/include directory when they must know specific information, dependencies, etc of
your system.
• To create a new Linux kernel directory to handle all header files, use the commands:
[root@deep /]# mkdir -p /usr/src/linux-2.4.18/include
[root@deep /]# cd /usr/src/linux/
[root@deep linux]# cp -r include/asm-i386 ../linux-2.4.18/include/
[root@deep linux]# cp -r include/linux ../linux-2.4.18/include/
[root@deep linux]# cd ../
[root@deep src]# rm -rf /usr/src/linux
[root@deep src]# cd /usr/src/ (to be sure that we are into the src directory)
[root@deep src]# ln -s /usr/src/linux-2.4.18 linux
First we create a new directory named “linux-2.4.18” based on the version of the kernel we
have installed for easy interpretation, then we copy directories asm-i386, and linux from
/usr/src/linux/include to our new location /usr/src/linux-2.4.18/include. After
we remove the entire source directory where we compiled the new kernel, we create a new
symbolic link named “linux” under /usr/src that points to our new /usr/src/linux-
2.4.18 directory. With these steps, future compiled programs will know where to look for
headers related to the kernel on your server.
NOTE: This step will allow us to gain space on our hard drive and will reduce the security risks.
The Linux kernel source directory handles a lot files and is about 94M in size when
uncompressed. With the procedure described above, our Linux kernel directory began
approximately 4M in size so we save 90MB for the same functionalities.
Verifying or upgrading your boot loader
Once the new kernel image has been installed on your server, we have to inform our boot loader
about it. This is done inside the configuration file of your boot loader software. Below I show you
how to do it depending if you use GRUB or LILO as your boot loader.
LILO:
This step applies only if you use LILO as your boot loader on the system. If you use GRUB as
your boot loader instead of LILO (highly recommended), then you can skip this section and go
directly to the next one.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-192-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
193
Step 1
You need to edit the lilo.conf file to make your new kernel one of the boot time options:
• Edit the lilo.conf file (vi /etc/lilo.conf) and make the appropriate change on
the line that reads “image=/boot/vmlinuz-x.x.x”.
[root@deep /]# vi /etc/lilo.conf
boot=/dev/sda
map=/boot/map
install=/boot/boot.b
timeout=00
default=linux
restricted
password=somepasswd
image=/boot/vmlinuz
label=linux
read-only
root=/dev/sda6
Step 2
Once the necessary modifications have been made in the /etc/lilo.conf file as shown
above, we update our lilo.conf file for the change to take effect.
• This can be done with the following command:
[root@deep /]# /sbin/lilo
Added linux *
GRUB:
This step applies only if you use GRUB as your boot loader on the system. In most case, GRUB
does not need to be updated when you install a new kernel on your computer but here we have to
verify inside our grub.conf file if all default setting are still available and configured because
when we uninstall Red Hat kernel RPM package, the software automatically remove some needed
parameters from the GRUB configuration file.
Step 1
Edit your GRUB configuration file and be sure that everything is correct and look like the following.
Your setting should differ from the example below.
• Edit the grub.conf file (vi /etc/grub.conf) and check your setting.
[root@deep /]# vi /etc/grub.conf
default 0
timeout 00
title Red Hat Linux
kernel (hd0,0)/vmlinuz ro root=/dev/sda5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-193-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
195
Rebooting your system to load the new kernel
Whether you have installed a new Monolithic Kernel where codes and drivers are compiled into
the kernel and are always loaded or a Modularized Kernel where some segment of codes are
compiled into the kernel as a module and loaded on demand, it is time to Reboot your system
and test your results.
• To reboot your Linux system, use the following command:
[root@deep /]# reboot
When the system is rebooted and you are logged in, verify the new version of your kernel with the
following command:
• To verify the version of your new kernel, use the following command:
[root@deep /]# uname -a
Linux dev 2.4.18-grsec-1.9.4 #1 Wed Jun 19 15:14:55 EDT 2002 i686 unknown
Congratulations!
Delete programs, edit files pertaining to modules
This section applies only if you chose to install a Monolithic Kernel on your system. By default
when you install Linux for the first time (like we did), the kernel is built as a Modularized Kernel.
This means that each device or function we need exists as a module and is controlled by the
Kernel Daemon program named kmod. kmod automatically loads some modules and functions
into memory as they are needed, and unloads them when they’re no longer being used.
kmod and other module management programs included in the modutils RPM package use the
modules.conf file located in the /etc directory to know for example which Ethernet card you
have, if your Ethernet card requires special configuration and so on. If we don’t use any modules
in our newly compiled kernel because we have compiled the kernel as a Monolithic Kernel and
ONLY in this case, we can remove the modules.conf file and completely uninstall the
modutils RPM package.
• To remove the modules.conf file, use the following command:
[root@deep /]# rm -f /etc/modules.conf
• To uninstall the modutils package, use the following command:
[root@deep /]# rpm -e --nodeps modutils
WARNING: Once again, the above is required only if you said no to “Enable loadable module
support (CONFIG_MODULES)” in your kernel configuration because you have decided to build
a Monolithic Kernel.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-195-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
196
Making a new rescue floppy for Modularized Kernel
This section applies only if you chose to install a Modularized Kernel on your system.
Immediately after you successfully start your system and log in as root, you should create a new
emergency boot floppy disk. The procedure to achieve it is the same as shown at the beginning
of this chapter related to Linux Kernel.
Please go back to the beginning of this chapter and follow the procedures to recreate a new
emergency boot floppy disk suitable for the new install Linux kernel on your system. Don’t forget
to test the boot disk to be sure that it works.
The mkbootdisk program runs only on Modularized Kernel. So you can’t use it on a Monolithic
Kernel; instead create an emergency boot floppy disk for Monolithic kernel as shown below.
Making a emergency boot floppy disk for Monolithic Kernel
This section applies only if you chose to install a Monolithic Kernel in your system. Because it is
possible to create a rescue floppy only on modularized kernel, we must find another way to boot
our Linux system for a monolithic kernel if the Linux kernel on the hard disk is damaged.
This is possible with a Linux emergency boot floppy disk. You should create it immediately after
you successfully start your system and log in as root. To create the emergency boot floppy, follow
these steps:
Step 1
Insert a floppy disk and format it with the following command:
[root@deep /]# fdformat /dev/fd0H1440
Double-sided, 80 tracks, 18 sec/track. Total capacity 1440 kB.
Formatting ... done
Verifying ... done
Step 2
Copy the actual file “vmlinuz” from the /boot directory to the floppy disk:
[root@deep /]# cp /boot/vmlinuz /dev/fd0H1440
cp: overwrite `/dev/fd0H1440'? y
NOTE: The vmlinuz file is a symbolic link that points to the real Linux kernel.
Step 3
Determine the kernel’s root device with the following command:
[root@deep /]# rdev
/dev/sda6 /
NOTE: The kernel’s root device is the disk partition where the root file system is located. In this
example, the root device is /dev/sda6; the device name should be different on your system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-196-320.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
197
Step 4
Set the kernel’s root device with the following command:
[root@deep /]# rdev /dev/fd0H1440 /dev/sda6
NOTE: To set the kernel’s root device, use the device reported by the “rdev” command utility in
the previous step.
Step 5
Mark the root device as read-only with the following command:
[root@deep /]# rdev -R /dev/fd0H1440 1
NOTE: This causes Linux to initially mount the root file system as read-only. By setting the root
device as read-only, you avoid several warnings and error messages.
Step 6
Now put the boot floppy in the drive A: and reboot your system with the following command to be
sure that your new boot disk is working:
[root@something /]# reboot
Following these guidelines, you will now have a boot floppy with a known working kernel in case
of problems with the upgrade. I recommend rebooting the system with the floppy to make sure
that the floppy works correctly.
Step 7
Because the mkbootdisk and dosfstools program are required only when you have a
Modularized kernel installed in your Linux system, we can remove the unneeded mkbootdisk
and dosfstools packages from the system when we have a Monolithic kernel installed on our
server.
• To uninstall the mkbootdisk and dosfstools utility, use the following command:
[root@deep /]# rpm –e mkbootdisk dosfstools](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-197-320.jpg)
![Process file system management 0
CHAPTER 7
202
What is sysctl?
sysctl is an interface that allows you to make changes to a running Linux system. It serves two
functions: to read and to modify system settings.
• To view all readable variables, use the following command:
[root@deep /]# sysctl -a
• To read a particular variable, for example, fs.file-max, use the following command:
[root@deep /]# sysctl fs.file-max
fs.file-max = 8192
• To set a particular variable for fs.file-max, use the following command:
[root@deep /]# sysctl -w fs.file-max=5536
fs.file-max = 16384
Settings of sysctl variables are usually strings, numbers, or booleans (a boolean being 1 for
yes or a 0 for no). If you set and change variable manually with the sysctl command as show
above, your changes will not resists on the next reboot of the system. For this reason, we will use
and show you further down in this chapter how to make your changes permanent even on
possible reboot of the server by using the /etc/sysctl.conf file.
/proc/sys/vm: The virtual memory subsystem of Linux
All parameters described in this chapter reside under the /proc/sys/vm directory of the server
and can be used to tune the operation of the virtual memory (VM) subsystem of the Linux kernel.
Be very careful when attempting this. You can optimize your system, but you can also cause it to
crash. Since every system is different, you'll probably want some control over this piece of the
system.
Finally, these are advanced setting and if you don’t understand them, then don’t try to play in this
area or try to use all the examples below in your system. Remember that all systems are different
and require different settings and customizations. The majority of the following hacks will work
fine on a server with >= at 512MB of RAM or a minimum of 256MB of RAM. Below this amount of
memory, nothing is guaranteed and the default setting will just be fine for you.
Next I’ll show you parameters that can be optimized. All suggestions I make in this section are
valid for all kinds of servers. The only difference depends on the amount of RAM your machine
has and this is where the settings will change.
The above figure shows a snapshot of /proc/sys/vm directory on an OpenNA Linux & Red Hat
Linux system running kernel version 2.4. Please note that this picture may look different on your
system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-202-320.jpg)
![Process file system management 0
CHAPTER 7
204
The default kernel setup for the bdflush parameters is:
"40 64 64 256 500 3000 60 0 0"
The default setup for the bdflush parameters under OpenNA Linux is:
"60 64 64 256 500 3000 80 0 0"
The default setup for the bdflush parameters under Red Hat Linux is:
"30 64 64 256 500 3000 60 0 0"
Step 1
To change the values of bdflush, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following line:
# Improve file system performance
vm.bdflush = 60 64 64 256 500 3000 80 0 0
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command in your terminal screen:
[root@deep /]# sysctl -w vm.bdflush="60 64 64 256 500 3000 80 0 0"
The kswapd parameter:
The kswapd file is related to the kernel swapout daemon. This file /proc/sys/vm/kswapd
frees memory on the system when it gets fragmented or full. Its task is to keep the memory
management system operating efficiently. Since every system is different, you'll probably want
some control over this piece of the system.
There are three parameters to tune in this file and two of them (tries_base and
swap_cluster) have the largest influence on system performance. The kswapd file can be
used to tune the operation of the virtual memory (VM) subsystem of the Linux kernel.
Parameter 1 (tries_base):
The kswapd parameter 1 specifies the maximum number of pages kswapd tries to free in one
round. Usually this number will be divided by 4 or 8, so it isn't as big as it looks. Increase this
number to cause swap to be released faster, and increase overall swap throughput. The default
value is 512 pages. We keep the default value here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-204-320.jpg)
![Process file system management 0
CHAPTER 7
205
Parameter 2 (tries_min):
The kswapd parameter 2 specifies the minimum number of pages kswapd tries to free a least
each time it is called. Basically it's just there to make sure that kswapd frees some pages even
when it's being called with minimum priority. The default value is 32 pages. We keep the default
value here.
Parameter 3 (swap_cluster):
The kswapd parameter 3 specifies the number of pages kswapd writes in one iteration. You
want this large to increase performance so that kswapd does its I/O in large chunks and the disk
doesn't have to seek often, but you don't want it to be too large since that would flood the request
queue. The default value is 8 pages. We improve the default value here.
The default kernel setup for the kswapd parameters is:
"512 32 8"
The default setup for the kswapd parameters under OpenNA Linux is:
"512 32 32"
The default setup for the kswapd parameters under Red Hat Linux is:
"512 32 8"
Step 1
To change the values of kswapd, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Increase swap bandwidth system performance
vm.kswapd = 512 32 32
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w vm.kswapd=”512 32 32”](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-205-320.jpg)
![Process file system management 0
CHAPTER 7
206
The overcommit_memory parameter:
The overcommit_memory parameter is simply a flag that enables memory overcommitment.
Memory overcommitment is a procedure to check that a process has enough memory to allocate
a new virtual mapping. When this flag is 0, the kernel checks before each malloc() to see if
there's enough memory left. If the flag is 1, the system pretends there's always enough memory
and don't make the check on the system. This feature can be very useful ONLY on big servers
with a lot of pysical memories available (>= 2GB) because there are a lot of programs that
malloc() huge amounts of memory "just-in-case" and don't use much of it.
The default kernel setup for the overcommit_memory parameter is:
"0"
The default setup for the overcommit_memory parameter under OpenNA Linux is:
"0"
The default setup for the overcommit_memory parameter under Red Hat Linux is:
"0"
Step 1
To change the value of overcommit_memory, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enables/Disables memory overcommitment
vm.overcommit_memory = 0
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
WARNING: Only change the default value of 0 to become 1 on systems with more than 2GB of
RAM. Recall that on small systems the value must be set to 0 (overcommit_memory=0).
There is another way to update the entry without restarting the network by using the following
command into your terminal screen:
[root@deep /]# sysctl -w overcommit_memory=0](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-206-320.jpg)
![Process file system management 0
CHAPTER 7
207
The page-cluster parameter:
The Linux virtual memory subsystem avoids excessive disk seeks by reading multiple pages on a
page fault. The number of pages it reads is highly dependent on the amount of memory in your
machine. The number of pages the kernel reads in at once is equal to 2 ^ page-cluster. Values
above 2 ^ 5 don't make much sense for swap because we only cluster swap data in 32-page
groups. The page-cluster parameter is used to tune the operation of the virtual memory (VM)
subsystem of the Linux kernel.
The default kernel setup for the kswapd parameter is:
"3"
The default setup for the kswapd parameter under OpenNA Linux is:
"5"
The default setup for the kswapd parameter under Red Hat Linux is:
"4"
Step 1
To change the value of page-cluster, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Increase number of pages kernel reads in at once
vm.page-cluster = 5
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w vm.page-cluster=5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-207-320.jpg)
![Process file system management 0
CHAPTER 7
208
The pagetable_cache parameter:
The kernel keeps a number of page tables in a per-processor cache (this helps a lot on SMP
systems). The cache size for each processor will be between the low and the high value. On SMP
systems it is used so that the system can do fast pagetable allocations without having to acquire
the kernel memory lock.
For large systems, the settings are probably OK. For normal systems they won't hurt a bit. For
small systems (<16MB RAM) and on a low-memory, single CPU system it might be
advantageous to set both values to 0 so you don't waste the memory.
The default kernel setup for the kswapd parameters is:
"25 50"
The default setup for the kswapd parameters under OpenNA Linux is:
"25 50"
The default setup for the kswapd parameters under Red Hat Linux is:
"25 50"
Step 1
To change the values of pagetable_cache, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Improve number of page tables keeps in a per-processor cache
vm.pagetable_cache = 25 50
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
WARNING: Only change these values on systems with multiple processors (SMP) or on small
systems (single processor) with less than 16MB of RAM. Recall that on small systems the both
values must be set to 0 (vm.pagetable_cache = 0 0).
There is another way to update the entry without restarting the network by using the following
command into your terminal screen:
[root@deep /]# sysctl -w vm.pagetable_cache=”25 50”](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-208-320.jpg)
![Process file system management 0
CHAPTER 7
209
/proc/sys/fs: The file system data of Linux
All parameters described later in this chapter reside under the /proc/sys/fs directory of the
server and can be used to tune and monitor miscellaneous things in the operation of the Linux
kernel. Be very careful when attempting this. You can optimize your system, but you can also
cause it to crash. Since every system is different, you'll probably want some control over these
pieces of the system.
Finally, these are advanced settings and if you don’t understand them, then don’t play in this area
or try to use all the examples below in your system. Remember that all systems are different and
required different setting and customization.
Below I show you only parameters that can be optimized for the system. All suggestions I
enumerate in this section are valid for every kind of servers. The only difference depends on the
amount of MB of RAM your machines have and this is where settings will change.
The above figure shows a snapshot of /proc/sys/fs directory on a OpenNA Linux & Red Hat
Linux system running kernel version 2.4. Please note that this picture may look different on your
system.
The file-max & file-nr parameters:
The file-max and file-nr files work together on Linux, we use the file-max parameter to
sets the maximum number of file-handles that the Linux kernel will allocate and the file-nr file
to get information about the number of allocated file handles, the number of used file handles and
the maximum number of file handles presently on the system. A large-scale production server
may easily require many thousands of file-handles, depending on the kind and number of
services running concurrently on the server. On busy servers where many services are running
concurrently, we generally tune this file (file-max) to improve the number of open files available
on the system.
It is important to note that you need to increase the limit of open files available on your server
ONLY when you get lots of error messages about running out of file handles. If you don’t receive
this kind of error message, you really DON’T need to increase the default value.
• To know the number of allocated file handles, the number of used file handles and the
maximum number of file handles on your system, use the following command:
[root@deep /]# cat /proc/sys/fs/file-nr
405 137 8192](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-209-320.jpg)
![Process file system management 0
CHAPTER 7
210
The first value (405) in our result is the number of allocated file handles, the second value (137) is
the number of used file handles, and the last value (8192) is the maximum number of file handles.
When the allocated file handles (405) come close to the maximum (8192), but the number of
actually used ones (137) is far less, you've encountered a peak in your usage of file handles and
you don't need to increase the maximum. The default kernel setup is suitable for most of us.
The default kernel setup for the file-max parameter is:
"8192"
The default setup for the file-max parameter under OpenNA Linux is:
"8192"
The default setup for the file-max parameter under Red Hat Linux is:
"8192"
Step 1
To adjust the value of file-max, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Increase limit of file-handles
fs.file-max = 8192
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w fs.file-max=8192](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-210-320.jpg)
![Process file system management 0
CHAPTER 7
212
Prevent your system responding to ping request:
Preventing your system for responding to ping requests can make a big improvement in your
network security since no one can ping your server and receive an answer. The TCP/IP
protocol suite has a number of weaknesses that allows an attacker to leverage techniques in the
form of covert channels to surreptitiously pass data in otherwise benign packets.
Step 1
Preventing your server from responding to ping requests can help to minimize this problem. Not
responding to pings would at least keep most "crackers" out because they would never know it's
there. ICMP blocking can hurt the performance of long-duration TCP connections, and this is due
to the fact that MTU discovery relies on ICMP packets to work.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable ignoring ping request
net.ipv4.icmp_echo_ignore_all = 1
When this key is on (1), the computer will ignore all ICMP packets. It is not recommended to turn
on this key, except in special situations when you receive an ICMP packet based attack. In the
above parameter, we enable this option.
Step 2
Once the configuration has been set, you must restart your network for the change to take effect.
The command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.icmp_echo_ignore_all=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-212-320.jpg)
![Process file system management 0
CHAPTER 7
213
Refuse responding to broadcasts request:
As for the ping request, it’s also important to disable broadcast requests. When a packet is
sent to an IP broadcast address (i.e. 192.168.1.255) from a machine on the local network,
that packet is delivered to all machines on that network. Then all the machines on a network
respond to this ICMP echo request and the result can be severe network congestion or outages
(Denial-of-Service attacks). See the RFC 2644 for more information.
Step 1
To disable broadcast requests, type the following command on your terminal.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable ignoring broadcasts request
net.ipv4.icmp_echo_ignore_broadcasts = 1
When this key is on (1), the server will never answer to "ping" if its destination address is
multicast or broadcast. It is good to turn on (1) this key to avoid your server becoming an
involuntary partner of a DoS attack. In the above parameter, we enable this option.
Step 2
Once the configuration has been set, you must restart your network for the change to take effect.
The command to restart the network is the following:
• To restart all networks devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.icmp_echo_ignore_broadcasts=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-213-320.jpg)
![Process file system management 0
CHAPTER 7
214
Routing Protocols:
Routing and routing protocols can create several problems. IP source routing, where an IP packet
contains details of the path to its intended destination, is dangerous because according to RFC
1122 the destination host must respond along the same path. If an attacker was able to send a
source routed packet into your network, then he would be able to intercept the replies and fool
your host into thinking it is communicating with a trusted host.
Step 1
I strongly recommend that you disable IP source routing on all network interfaces on the system
to protect your server from this hole. If IP source routing is set to off (0), the server will not accept
source-routed frames. Remember that Source-routed frames have an embedded, explicit
description of the route between source and destination. Normally, IP packet routing is based
solely on destination's address.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable IP source routing
net.ipv4.conf.all.accept_source_route = 0
net.ipv4.conf.default.accept_source_route = 0
Source-routed packets are a powerful concept but were never used, and can bring security
problems because they allow a non-blind, REMOTE spoof. It is a very good idea to turn off (0)
these keys. In the above parameters, we disable these options.
Step 2
Once configurations have been set, you must restart your network for the change to take effect.
The command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.conf.all.accept_source_route=1
[root@deep /]# sysctl -w net.ipv4.conf.default.accept_source_route=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-214-320.jpg)
![Process file system management 0
CHAPTER 7
215
Enable TCP SYN Cookie Protection:
A "SYN Attack" is a Denial of Service (DoS) attack that consumes all the resources on your
machine, forcing you to reboot. Denials of Service attacks (attacks which incapacitate a server
due to high traffic volume or ones those tie-up system resources enough that the server cannot
respond to a legitimate connection request from a remote system) are easily achievable from
internal resources or external connections via extranets and Internet. Enabling TCP SYN Cookie
Protection will help to eliminate the problem. Below is a simple explanation of the concept.
Every TCP/IP connection begins with a 3-way handshake:
1. Client sends a packet (packet 1) to server with SYN bit on, and waits;
2. Server sends a confirmation packet (packet 2) to client, and waits;
3. Client sends a third packet (packet 3) that consolidates the connection.
Once the 3-way handshake is done, the server keeps data of packet 1 in a queue to compare it
with packet 3 and establish the connection. This queue is limited in size and a quite high timeout.
The SYN-flood attack exploits this fact and sends a lot of type-1 packets with random IP source
addresses; the phase 3 answers never arrive; and once the queue is full, the server cannot
receive more connections, be they legitimate or forged.
The SYN cookie "trick" is to embed a code in the header of phase 2 packets, so server DOES
NOT NEED TO KEEP any information about the client. If the phase 3 packet arrives someday,
the server will calculate the port and client initial sequence number based solely on that packet -
and will be able to establish the connection.
Step 1
Since this embedded codification reduces randomness of the server initial sequence number, and
thus can increase the "chance" of IP spoof family attacks, SYN cookies are used only in
emergency situations, that is, when the half-open connections queue is full.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable TCP SYN Cookie Protection
net.ipv4.tcp_syncookies = 1
If this key is on (1), the kernel will send "SYN cookies" ONLY and ONLY when the half-open
connections queue becomes full. This will mitigate the effects of SYN-flood DoS attacks. In the
above parameter, we enable this option.
Step 2
Once the configuration has been set, you must restart your network for the change to take effect.
The command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
WARNING: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.tcp_syncookies=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-215-320.jpg)
![Process file system management 0
CHAPTER 7
216
Disable ICMP Redirect Acceptance:
When hosts use a non-optimal or defunct route to a particular destination, an ICMP redirect
packet is used by routers to inform the hosts what the correct route should be. If an attacker is
able to forge ICMP redirect packets, he or she can alter the routing tables on the host and
possibly subvert the security of the host by causing traffic to flow via a path you didn't intend.
Step 1
A legitimate ICMP REDIRECT packet is a message from a router that says "router X is better
than me to reach network Y". Therefore, in complex networks, it will be highly recommended to
keep these keys activated. On simple networks, it’s strongly recommended to disable ICMP
Redirect Acceptance into all available interfaces on the server to protect it from this hole.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable ICMP Redirect Acceptance
net.ipv4.conf.all.accept_redirects = 0
net.ipv4.conf.default.accept_redirects = 0
When these keys are off (0), the kernel does not honor ICMP_REDIRECT packets, thus avoiding
a whole family of attacks based on forging of this type of packet. In the above parameters, we
disable these options.
Step 2
Once the configurations have been set, you must restart your network for the change to take
effect. The command to restart the network is the following:
• To restart all networks devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.conf.all.accept_redirects=0
[root@deep /]# sysctl -w net.ipv4.conf.default.accept_redirects=0](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-216-320.jpg)
![Process file system management 0
CHAPTER 7
217
Enable bad error message Protection:
This option will alert you about all bad error messages on your network.
Step 1
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following line:
# Enable/Disable bad error message Protection
net.ipv4.icmp_ignore_bogus_error_responses = 1
Step 2
Once configuration has been set, you must restart your network for the change to take effect. The
command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.icmp_ignore_bogus_error_responses=1
Enable IP spoofing protection:
The spoofing protection prevents your network from being the source of spoofed (i.e. forged)
communications that are often used in DoS Attacks.
Step 1
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable IP spoofing protection
net.ipv4.conf.all.rp_filter = 2
net.ipv4.conf.default.rp_filter = 2
These keys control IP Spoof detection and can have the following values:
0=absolutely no checking (the default)
1=simple checking (only obvious spoofs)
2=strong checking (positive source verification)
The recommended standard is level 2, which can bring "problems in complex (non loop free)
networks". In the above parameters, we enable the “strong checking” option.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-217-320.jpg)
![Process file system management 0
CHAPTER 7
218
Step 2
Once the configurations have been made, you must restart your network for the change to take
effect. The command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
NOTE: This parameter will prevent spoofing attacks against your internal networks but your
external addresses can still be spoofed.
There is another way to update the entry without restarting the network by using the following
command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.conf.all.rp_filter=2
[root@deep /]# sysctl -w net.ipv4.conf.default.rp_filter=2
Enable Log Spoofed, Source Routed and Redirect Packets:
This change will log all Spoofed Packets, Source Routed Packets, and Redirect Packets to your
log files.
Step 1
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable Log Spoofed, Source Routed, Redirect Packets
net.ipv4.conf.all.log_martians = 1
net.ipv4.conf.default.log_martians = 1
When this key is on (1), the kernel will log any "impossible" packets, where the IP source address
spoofing is obvious. Example: packet with source address equal to 127.0.0.1 coming from an
Ethernet interface. In the above parameter, we enable this option.
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.conf.all.log_martians=1
[root@deep /]# sysctl -w net.ipv4.conf.default.log_martians=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-218-320.jpg)
![Process file system management 0
CHAPTER 7
219
Other possible optimization of the system
All information described next relates to tuning we can make on the system. Be very careful when
attempting this. You can optimize your system, but you can also cause it to crash.
The shared memory limit parameters:
Shared memory is used for inter-process communication, and to store resources that are shared
between multiple processes such as cached data and code. If insufficient shared memory is
allocated any attempt to access resources that use shared memory such as database
connections or executable code will perform poorly.
Under UNIX world, shared memory is referred to as "System V IPC". Almost all modern operating
systems provide these features, but not all of them have them turned on or sufficiently sized by
default, especially systems with BSD heritage. With Linux the default shared memory limit (both
SHMMAX and SHMALL) is 32 MB in 2.4 kernels, but fortunately can be changed into the /proc file
system.
On system with small MB of RAM (>= 128MB), the default setting of 32MB for shared memory on
the system is enough and should not be changed. On system with lot of RAM, we can readjust
the default setting to better fit our machine and server performance.
Below, I show you an example, to allow 128MB of shared memory on the system. The new value
you enter for the shared memory should be four time less than what your total MB of RAM is. For
example if you have 512MB of RAM installed on your computer, then you can set the default
shared memory to 128MB as we do here. If you have less than what I use in this example, you
have to adjust the value to fit your needs.
Step 1
To change the default values of shared memory, type the following commands on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Improve shared memory size
kernel.shmall = 134217728
kernel.shmmax = 134217728
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following commands into your terminal screen:
[root@deep /]# sysctl -w kernel.shmall="134217728"
[root@deep /]# sysctl -w kernel.shmmax="134217728"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-219-320.jpg)
![Process file system management 0
CHAPTER 7
220
Tuning the default and maximum window size parameters:
The following hack allow us to raise network limits on Linux by requesting that the kernel provide
a larger send buffer for the server's connections to its clients. This is possible by tuning the
"default send window" and "maximum send window" parameters.
Default parameter under Linux is 64kb, this is correct for regular use of the OS but if you run your
system as a server, it is recommended to change the default parameter to a sufficiently-high
value like 2000kb. 64kb is equal to 65536 (64 * 1024 = 65536), to set the values to 2000kb, we
should enter new values of 2048000 (2000 * 1024 = 2048000).
Step 1
To change the default values of default and maximum window size, type the following
commands on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Improve default and maximum window size
net.core.wmem_max = 2048000
net.core.wmem_default = 2048000
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following commands into your terminal screen:
[root@deep /]# sysctl -w net.core.wmem_max="2048000"
[root@deep /]# sysctl -w net.core.wmem_default="2048000"
The ulimit parameter:
The ulimit command provides control over the resources available to the shell and to
processes started by it, on systems that allow such control. We can use it to change the default
resource set by the system to users or super user “root” but actually, we use it for changing
resources for the super user “root” only because resources for normal users are controlled and
managed through the /etc/security/limit.conf file.
It is not all default resources available through the ulimit parameter that need to be changed
but only those which can improve the performance of the server in a high load environment. Most
default values for the super user “root” are acceptable and should be kept unchanged. Linux itself
has a “Maximum Processes” per user limit. This feature allows us to control the number of
processes an existing user or the “root” super user on the server may be authorized to have. To
increase performance on highly loaded servers, we can safely set the limit of processes for the
super-user “root” to be unlimited. This is what we will do in the following steps.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-220-320.jpg)
![Process file system management 0
CHAPTER 7
221
One question remains, how can we change the default resources for a specific user on the
system? Each new user has a hidden file called “.bashrc” available in their home directory. It is
into this file that we can change the default value of resources for the specific user. In the
example below, we do it for the super user “root” but the procedure is the same for any users on
the system. Just edit their corresponding “.bashrc” file and make the change.
Step 1
• Edit the .bashrc file for super user “root” (vi /root/.bashrc) and add the line:
ulimit -u unlimited
NOTE: You must exit and re-login from your terminal for the change to take effect.
Step 2
To verify that you are ready to go, make sure that when you type as root the command ulimit
-a on your terminal, it shows "unlimited" next to max user processes.
[root@deep /]# ulimit -a
core file size (blocks, -c) 0
data seg size (kbytes, -d) unlimited
file size (blocks, -f) unlimited
max locked memory (kbytes, -l) unlimited
max memory size (kbytes, -m) unlimited
open files (-n) 1024
pipe size (512 bytes, -p) 8
stack size (kbytes, -s) 8192
cpu time (seconds, -t) unlimited
max user processes (-u) unlimited
virtual memory (kbytes, -v) unlimited
NOTE: You may also do ulimit -u unlimited at the command prompt instead of adding it to
the /root/.bashrc file but the value will not survive to a reboot.
The atime attribute:
Linux records information about when files were created and last modified as well as when it was
last accessed. There is a cost associated with recording the last access time. The ext2 file
system of Linux has an attribute that allows the super-user to mark individual files such that their
last access time is not recorded. This may lead to significant performance improvements on often
accessed, frequently changing files such as the contents of News Server, Web Server, Proxy
Server, Database Server among other directories.
• To set the attribute to a file, use:
[root@deep /]# chattr +A filename For a specific file
For a whole directory tree, do something like:
[root@deep /root]# chattr -R +A /var/spool For a News and Mail Server directory
[root@deep /root]# chattr -R +A /home/httpd/html For a Web Server directory
[root@deep /root]# chattr -R +A /var/lib/mysql For a SQL Database directory](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-221-320.jpg)
![Process file system management 0
CHAPTER 7
222
The noatime attribute:
Linux has a special mount option for file systems called noatime that can be added to each line
that addresses one file system in the /etc/fstab file. If a file system has been mounted with
this option, reading accesses to the file system will no longer result in an update to the atime
information associated with the file like we have explained previously. The importance of the
noatime setting is that it eliminates the need by the system to make writes to the file system for
files, which are simply being read. Since writes can be somewhat expensive, this can result in
measurable performance gains. Note that the write time information to a file will continue to be
updated anytime the file is written to. In our example below, we will set the noatime option to our
/chroot file system.
Step 1
• Edit the fstab file (vi /etc/fstab) and add in the line that refers to the /chroot file
system, the noatime option after the defaults option as show below:
LABEL=/chroot /chroot ext3 defaults,noatime 1 2
Step 2
Once you have made the necessary adjustments to the /etc/fstab file, it is time to inform the
system about the modification.
• This can be accomplished with the following commands:
[root@deep /]# mount /chroot -oremount
Each file system that has been modified must be remounted with the command shown above. In
our example we have modified the /chroot file system and it is for this reason that we remount
this file system with the above command.
Step 3
• You can verify if the modifications have been correctly applied to the Linux system with
the following command:
[root@deep /]# cat /proc/mounts
/dev/root / ext3 rw 0 0
/proc /proc proc rw 0 0
/dev/sda1 /boot ext3 rw 0 0
/dev/sda10 /cache ext3 rw,nodev 0 0
/dev/sda9 /chroot ext3 rw,noatime 0 0
/dev/sda8 /home ext3 rw,nosuid 0 0
/dev/sda13 /tmp ext3 rw,noexec,nosuid 0 0
/dev/sda7 /usr ext3 rw 0 0
/dev/sda11 /var ext3 rw 0 0
/dev/sda12 /var/lib ext3 rw 0 0
none /dev/pts devpts rw 0 0
This command will show you all file systems on your server with parameters applied to them. If
you see something like:
/dev/sda9 /chroot ext3 rw,noatime 0 0
Congratulations!](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-222-320.jpg)
![TCP/IP Network Management 0
CHAPTER 8
227
Linux TCP/IP
Abstract
This chapter has been inserted here because it is preferable not to be connected to the network if
the parts "Installation-Related Reference" and "Security and Optimization-Related Reference" of
the book have not been completed. It is not wise to apply new security configurations to your
system if you are online. Also, don’t forget that the firewall, which represents 50% of networking
security, is still not configured on the Linux server. Finally it is very important and I say VERY
IMPORTANT that you check all configuration files related to Linux networking to be sure that
everything is configured correctly. Please follow all recommendations and steps in this chapter
before continuing reading this book. This will allow us to be sure that if something goes wrong in
the other chapters, it will be not related to your networking configurations.
• To stop specific network device manually on your system, use the following command:
[root@deep /]# ifdown eth0
• To start specific network device manually on your system, use the following command:
[root@deep /]# ifup eth0
• To stop all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network stop
Shutting down interface eth0 [OK]
Disabling IPv4 packet forwarding [OK]
• To start all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network start
Enabling IPv4 packet forwarding [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Until now, we have not played with the networking capabilities of Linux. Linux is one of the best
operating systems in the world for networking features. Most Internet sites around the world
already know this, and have used it for some time. Understanding your network hardware and all
the files related to it is very important if you want to have a full control of what happens on your
server. Good knowledge of primary networking commands is vital. Network management covers
a wide variety of topics. In general, it includes gathering statistical data and monitoring the status
of parts of your network, and taking action as necessary to deal with failures and other changes.
The most primitive technique for network monitoring is periodic "pinging" of critical hosts. More
sophisticated network monitoring requires the ability to get specific status and statistical
information from a range of devices on the network. These should include various sorts of
datagram counts, as well as counts of errors of different kinds. For these reasons, in this chapter
we will try to answer fundamental questions about networking devices, files related to network
functionality, and essential networking commands.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-227-320.jpg)
![TCP/IP Network Management 0
CHAPTER 8
235
The following values may be used:
NETWORKING=answer, where answer is yes or no (Configure networking or not configure networking).
HOSTNAME=hostname, where hostname is the hostname of your server.
GATEWAY=gwip, where gwip is the IP address of the remote network gateway (if available).
GATEWAYDEV=gwdev, where gwdev is the device name (eth#) you use to access the remote gateway.
The /etc/sysctl.conf file:
With the new version of Red Hat Linux, all kernel parameters available under the /proc/sys/
subdirectory of Linux can be configured at runtime. You can use the new /etc/sysctl.conf
file to modify and set kernel parameters at runtime. The sysctl.conf file is read and loaded
each time the system reboots or when you restart your network. All settings are now stored in the
/etc/sysctl.conf file. All modifications to /proc/sys should be made through
/etc/sysctl.conf, because they are better for control, and are executed before rc.local or
any other "users" scripts.
Below, we’ll focus only on the kernel option for IPv4 forwarding support. See later in this chapter
the TCP/IP security parameters related to the sysctl.conf file.
To enable IPv4 forwarding on your Linux system, use the following command:
Step 1
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following line:
# Enable packet forwarding (required only for Gateway, VPN, Proxy, PPP)
net.ipv4.ip_forward = 1
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
WARNING: You must enable packet forwarding only on a machine that serves as a Gateway
Server, VPN Server, Proxy Server or with PPP connection. Forwarding allows packets that are
destined for another network interface (if you have another one) to pass through the network.
There is another way to update the entry without restarting the network by using the following
command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.ip_forward=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-235-320.jpg)
![TCP/IP Network Management 0
CHAPTER 8
236
The /etc/hosts file:
As your machine gets started, it will need to know the mapping of some hostnames to IP
addresses before DNS can be referenced. This mapping is kept in the /etc/hosts file. In the
absence of a name server, any network program on your system consults this file to determine
the IP address that corresponds to a host name.
Following is a sample /etc/hosts file:
IP Address Hostname Alias
127.0.0.1 localhost.localdomain localhost
208.164.186.1 deep.openna.com deep
208.164.186.2 mail.openna.com mail
208.164.186.3 web.openna.com web
The leftmost column is the IP address to be resolved. The next column is that host’s name. Any
subsequent columns are the aliases for that host. In the second line, for example, the IP address
208.164.186.1 if for the host deep.openna.com. Another name for deep.openna.com is
deep.
WARNING: Some people have reporting that a badly formed line in the /etc/hosts file may result
to a "Segmentation fault (core dumped)" with the syslogd daemon, therefore I recommend you
to double check your entry under this file and be sure that its respond to the example as shown
above. The “Alias” part of the line is important if you want to be able to use the FQDN (Fully
Qualified Domain Name) of the system reported by the hostname -f command.
After you are finished adding and configuring your networking files, don’t forget to restart your
network for the changes to take effect.
• To restart your network, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
WARNING: Time out problems for telnet or ftp connection are often caused by the server trying
to resolve the client IP address to a DNS name. Either DNS isn’t configured properly on your
server or the client machines aren’t known to the DNS server. If you intend to run telnet or ftp
services on your server, and aren’t using DNS, don’t forget to add the client machine name and
IP in your /etc/hosts file on the server or you can expect to wait several minutes for the DNS
lookup to time out, before you get a login prompt.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-236-320.jpg)
![TCP/IP Network Management 0
CHAPTER 8
237
Testing TCP/IP Networking
Once we have applied TCP/IP security and optimization parameters to our server and checked
or configured all files related to network functionality, we can run some tests to verify that
everything works as expected. It is important to note that at this stage every test must be
successful and not have any errors. It is to your responsibility to know and understand networking
architecture and basic TCP/IP protocols before testing any parts of your networking configuration
and topology.
Step 1
To begin, we can use the ifconfig utility to display all the network interfaces on the server.
• To display all the interfaces you have on your server, use the command:
[root@deep /]# ifconfig
The output should look something like this:
eth0 Link encap:Ethernet HWaddr 00:E0:18:90:1B:56
inet addr:208.164.186.2 Bcast:208.164.186.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:1295 errors:0 dropped:0 overruns:0 frame:0
TX packets:1163 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
Interrupt:11 Base address:0xa800
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:3924 Metric:1
RX packets:139 errors:0 dropped:0 overruns:0 frame:0
TX packets:139 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
NOTE: If the ifconfig tool is invoked without any parameters, it displays all interfaces you
configured. An option of “-a” shows the inactive one as well.
Step 2
If all network interfaces on the server look as you expect, then it is time to verify that you can
reach your hosts. Choose a host from your internal network, for instance 192.168.1.1
• To verify that you can reach your internal hosts, use the command:
[root@deep /]# ping 192.168.1.1
The output should look something like this:
PING 192.168.1.1 (192.168.1.1) from 192.168.1.1 : 56 data bytes
64 bytes from 192.168.1.1: icmp_seq=0 ttl=128 time=1.0 ms
64 bytes from 192.168.1.1: icmp_seq=1 ttl=128 time=1.0 ms
64 bytes from 192.168.1.1: icmp_seq=2 ttl=128 time=1.0 ms
64 bytes from 192.168.1.1: icmp_seq=3 ttl=128 time=1.0 ms
--- 192.168.1.1 ping statistics ---
4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 1.0/1.0/1.0 ms](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-237-320.jpg)
![TCP/IP Network Management 0
CHAPTER 8
238
WARNING: Do not try to ping a host in which you have applied the previous TCP/IP security
settings to prevent your system to respond to ping request. Instead try to ping another host
without this feature enable. Also if you don’t receive an answer from the internal host you try to
ping, verify if your hubs, routers, network cards, and network topology are correct.
If you are able to ping your internal host, congratulations! Now we must ping an external
network, for instance 216.148.218.195
• To verify that you can reach the external network, use the command:
[root@deep /]# ping 216.148.218.195
The output should look something like this:
PING 216.148.218.195 (216.148.218.195) from 216.148.218.195 :56 data byte
64 bytes from 216.148.218.195: icmp_seq=0 ttl=128 time=1.0 ms
64 bytes from 216.148.218.195: icmp_seq=1 ttl=128 time=1.0 ms
64 bytes from 216.148.218.195: icmp_seq=2 ttl=128 time=1.0 ms
64 bytes from 216.148.218.195: icmp_seq=3 ttl=128 time=1.0 ms
--- 216.148.218.195 ping statistics ---
4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 1.0/1.0/1.0 ms
Step 3
You should now display the routing information with the command route to see if the hosts have
the correct routing entries.
• To display the routing information, use the command:
[root@deep /]# route -n
The output should look something like this:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
208.164.186.2 0.0.0.0 255.255.255.255UH 0 0 0 eth0
208.164.186.0 208.164.186.2 255.255.255.0 UG 0 0 0 eth0
208.164.186.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-238-320.jpg)
![Step 4
Another useful option is “netstat -vat”, which shows all active and listen TCP connections.
• To shows all active and listen TCP connections, use the command:
[root@deep /]# netstat -vat
The output may look something similar to this example depending if the related services are
running. Be aware that your results will almost certainly vary from the ones shown below:
Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 0 deep.openna.co:domain *:* LISTEN
tcp 0 0 localhost:domain *:* LISTEN
tcp 0 0 deep.openna.com:ssh gate.openna.com:1682ESTABLISHED
tcp 0 0 *:webcache *:* LISTEN
tcp 0 0 deep.openar:netbios-ssn *:* LISTEN
tcp 0 0 localhost:netbios-ssn *:* LISTEN
tcp 0 0 localhost:1032 localhost:1033 ESTABLISHED
tcp 0 0 localhost:1033 localhost:1032 ESTABLISHED
tcp 0 0 localhost:1030 localhost:1031 ESTABLISHED
tcp 0 0 localhost:1031 localhost:1030 ESTABLISHED
tcp 0 0 localhost:1028 localhost:1029 ESTABLISHED
tcp 0 0 localhost:1029 localhost:1028 ESTABLISHED
tcp 0 0 localhost:1026 localhost:1027 ESTABLISHED
tcp 0 0 localhost:1027 localhost:1026 ESTABLISHED
tcp 0 0 localhost:1024 localhost:1025 ESTABLISHED
tcp 0 0 localhost:1025 localhost:1024 ESTABLISHED
tcp 0 0 deep.openna.com:www *:* LISTEN
tcp 0 0 deep.openna.com:https *:* LISTEN
tcp 0 0 *:389 *:* LISTEN
tcp 0 0 *:ssh *:* LISTEN
Step 5
Sometimes machines on your network will discard your IP packets and finding the offending
Gateway responsilbe can be difficult. Fortunately the tracepath utility attempts to trace the
route an IP packet would follow to some Internet host. Choose an Internet host, for instance
64.81.28.146
• To print the route packets take to network host, use the command:
[root@deep /]# tracepath 64.81.28.146
The output should look something like this:
1?: [LOCALHOST] pmtu 1500
1?: 207.35.78.1
2?: 10.70.1.1
3?: 206.47.228.178
4?: 206.108.97.149
5?: 206.108.103.214
6?: 206.108.103.228
7?: 208.51.134.9
8?: 208.48.234.189
9?: 206.132.41.78 asymm 10
10?: 204.246.213.226 asymm 13
11?: 206.253.192.217 asymm 13
12?: 206.253.195.218 asymm 14
13: 64.81.28.146 asymm 15 139ms reached
Resume: pmtu 1500 hops 13 back 15](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-239-320.jpg)
![TCP/IP Network Management 0
CHAPTER 8
240
Step 6
Finally, we will use the hostname command of Linux to show if our systems host name is
correct.
• To display and print the current host name of your server, use the command:
[root@deep /]# hostname
deep
The hostname command without any options will print the current host name of our system, in
this example “deep”.
Now, it’s important to verify if the Fully Qualified Domain Name (FQDN) of our server is reported
correctly.
• To display and print the FQDN of your server, use the command:
[root@deep /]# hostname -f
deep.openna.com
The last checkup
If you can answer, “Yes” to each of the questions below, then your network is working and you
can continue .
Parameters inside ifcfg-ethN files are corrects
The /etc/resolv.conf file contain your primary and secondary Domain Name Server
All parameters included in the /etc/host.conf file are corrects
All parameters included in the /etc/sysconfig/network file are corrects
The /etc/hosts file contain the mapping of your hostnames to IP addresses
All network interfaces on the server have the right parameter
You can reach the internal and external hosts
Your hosts have the correct routing entry
The status of the interfaces has been checked and looks fine
You are able to print the route packets take to network host](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-240-320.jpg)
![Firewall Basic Concept 0
CHAPTER 9
244
The Well Known Ports are assigned by the IANA and on most systems can only be used by
system (or root) processes or by programs executed by privileged users (our daemons running in
the background). Ports are used by the TCP protocol [RFC793] to name the ends of logical
connections, which carry long-term conversations. For the purpose of providing services to
unknown callers, a service contact port is defined. The contact port is sometimes called the "well-
known port". Wherever possible, the same port assignments are also used by UDP protocol
[RFC768]. For many years Well Known Ports were in the range 0-255. Recently, the range for
the Well Known Ports has been expanded from 0 to 1023 to respond to the exponential growth of
the Internet.
The Registered Ports are also listed by the IANA and, on most systems, can be used by ordinary
user processes or programs executed by ordinary users. The IANA registers the use of these
ports as a convenience to the community. Again, wherever possible, these same port
assignments are used with UDP [RFC768]. The Registered Ports are in the range 1024-49151.
Finally, the Dynamic and/or Private Ports are those ranging from 49152 through to 65535.
1
2
3
[1] The Well Known Ports represent 2% of all available ports [0-1023].
[2] The Registered Ports represent 73% of all available ports [1024-49151].
[3] The Dynamic and/or Private Ports represent 25% of all available ports [49152-65535].
Ports Numbers Graphical Representation](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-244-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
262
Pristine source
As we don’t use the RPM package to install the program, it would be difficult for us to locate all the
files installed on the system if in the future we want to upgrade. To solve this problem, it is a good
idea to make a list of files on the system before you install GIPTables, and then one afterwards,
we can then compare them using the diff utility to find out what files were installed and where
they were placed.
• Simply run the following command before installing the software:
[root@deep root]# find /* > GIPTables1
• And the following one after you install the software:
[root@deep root]# find /* > GIPTables2
• Then use the following command to get a list of what changed:
[root@deep root]# diff GIPTables1 GIPTables2 > GIPTables-Installed
With this procedure, if any future upgrade appears, all you have to do is to read the generated list
of what files were added or changed by the program and remove them manually from your
system before installing the new software. In the above example, we use the /root directory of
the system to store the generated list of files.
Compiling - Optimizing & Installing GIPTables
To install the GIPTables software on your system, just download the latest version of the
software from http://www.giptables.org/ site, and then as user ‘root’ expand the archive under
your /var/tmp directory.
• To accomplish this use the following commands:
[root@deep /]# cp giptables-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf giptables-version.tar.gz
Next, move into the newly created GIPTables source directory and perform the following steps
to install the software for your system.
• To move into the newly created GIPTables source directory use the command:
[root@deep tmp]# cd giptables-1.1/
• To install GIPTables enter the following command:
[root@deep giptables-1.1]# ./install.sh
The “install.sh” script file will simply install any GIPTables components on your system to
the right location.
Once the installation of GIPTables has been completed, we can free up some disk space by
deleting both the program tar archive and the related source directory since they are no longer
needed.
• To delete GIPTables and its related source directory, use the commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf giptables-version/
[root@deep tmp]# rm -f giptables-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-262-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
264
Step1
First of all, choose one of the default configuration files that can be found in the
/lib/giptables/conf directory. Find one that mostly meets your needs, make a backup copy
of it and open it with any kind of text editor to configure it. In our example, we will configure our
GIPTables firewall software for a Gateway/Proxy Server with two network interfaces since it is
the most complicated configuration we are likely to encounter.
• These procedures can be accomplished with the following commands:
[root@deep /]# cd /lib/giptables/conf/
[root@deep conf]# cp giptables.conf.gateway giptables.conf.mybox
[root@deep conf]# ln -sf /lib/giptables/conf/giptables.conf.mybox
/etc/giptables.conf
In the above steps, we make a copy of our original “giptables.conf.gateway” file and create
a symbolic link pointing to the copy.
NOTE: It is a good idea not to directly modify an example configuration file, because if it gets
damage, then you have to install the entire package again in order to get it back.
Step2
Now our giptables.conf file that is a symbolic link pointing to the original configuration file for
a Gateway set-up exists. It is time to edit it and provide or change some minimal values to make it
work for our system.
In the GIPTables configuration file below, we’ll ONLY explain how to configure and set
parameters that are the same for all types of GIPTables firewall configuration. Parts that differ
are associated with different available GIPTables modules that must be loaded by the firewall
configuration file to enable different services. All available modules with GIPTables firewall are
explained later in this document.
• Edit the giptables.conf file (vi /etc/giptables.conf) and configure it to fit in
with your system and networking setup. Text in bold is what you should change to make
the firewall work with your server:
The Debug Definition:
The first configurable option that is the same for all types of GIPTables firewall configuration is
“DEBUG” and by default it is set to “off”. This option is useful only for debugging purposes.
# ----------------------------------------------------------------------------
# DEBUG
#
DEBUG="off"
If you set this option to "on", the firewall will display all IPTables rules relating to the
GIPTables configuration file that you use to the screen, nothing will go to the kernel. The
displayed set of rules will be commented so that you will not end up with lots of rules on the
screen that you do not understand. This way you can see only what the firewall is generating, and
also you will be able to better understand which rule is for what.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-264-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
265
When this option is set to "off" (the default setting), the firewall will send all generated rules to
the kernel, nothing will be displayed on your screen and the firewall will run on your system.
Therefore is you want to run GIPTables on your system, you must be sure that this option is set
to ‘off’.
NOTE: When the “DEBUG” option is set to “on”, it is possible to redirect the output of the firewall
rules to a file, and use this file as learning example of how to set up IPTables rules for different
kind of services. This is possible with the following command:
[root@deep tmp]# /etc/init.d/giptables start > output-result.txt
The Monolithic Kernel Definition:
The second configurable option that is the same for all types of GIPTables configuration is the
“MONOLITIC_KERNEL” option and it is set to “no” by default. This option exists because the Linux
kernel can be compiled so that it contains all the required IPTables driver code directly in it.
This way we have a MONOLITHIC KERNEL and we would need to answer by “yes” to the option.
# ----------------------------------------------------------------------------
# Some definitions for easy maintenance
# Edit these to suit your system
#
MONOLITIC_KERNEL="no"
If you set this option to ‘yes’, then GIPTables will be informed that all native IPTables
modules are directly compiled into the kernel. It is important to say ‘yes’ here only if you have a
Monolithic Linux Kernel installed on your computer otherwise say ‘no’. Then the firewall will look
for and load all the IPTables modules that are required, depending on your configuration file.
NOTE: If you compile your kernel as Monolithic, you should know what IPTables modules you
need to compile directly into the kernel, since the firewall will not try to load them. If you missed
some modules, you will inevitably get errors, or the firewall might not work as expected. The best
solution for a Monolithic Kernel set-up is to compile all native iptables modules into the kernel.
Also, don’t forget to set MONOLITIC_KERNEL="yes" in the firewall configuration file.
The External Network Interface Definition:
The next configurable option that is the same for all type of GIPTables firewall configuration is
one of the most important settings in the configuration file.
# Interface 0: This is our external network interface
# It is directly connected to Internet
INTERFACE0="eth0"
INTERFACE0_IPADDR="x.x.x.x"
ANY_IPADDR="0/0"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-265-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
273
# Refuse incoming packets claiming to be from the IP addresses of our
interfaces
REFUSE_SPOOFING_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_IN_REFUSE_SPOOFING[0]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[0]="no"
NETWORK1_IN_REFUSE_SPOOFING[0]="yes"
REFUSE_SPOOFING_IPADDR[1]=$INTERFACE1_IPADDR
INTERFACE0_IN_REFUSE_SPOOFING[1]="no"
INTERFACE1_IN_REFUSE_SPOOFING[1]="yes"
NETWORK1_IN_REFUSE_SPOOFING[1]="no"
B) We log and drop all incoming packets claiming to be from the broadcast source address
range. We accept broadcast source packets only in one situation: when we have a DHCP
Server, and this, because a DHCP Client will request its IP address by sending out and
DHCP discovery packet that has source IP address "0.0.0.0" and destination IP
address "255.255.255.255". In this situation, the Gateway Server is also a DHCP
Server, so we will accept by default those broadcast source packets only on the internal
interface.
# Refuse incoming packets claiming to be from broadcast-src address range
REFUSE_SPOOFING_IPADDR[2]="0.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[2]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[2]="no"
NETWORK1_IN_REFUSE_SPOOFING[2]="yes"
C) We log and drop all incoming packets claiming to be from the reserved loopback IP
address range. This is so obvious. We should never have incoming packets with source
IP address from the loopback address range. We can refuse them safely on all our
interfaces.
# Refuse incoming packets claiming to be from reserved loopback address range
REFUSE_SPOOFING_IPADDR[3]="127.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[3]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[3]="yes"
NETWORK1_IN_REFUSE_SPOOFING[3]="yes"
E) We log and drop all incoming packets claiming to be from the well-known private
networks: A, B, C. We can safely refuse all packets claiming to be from those private
networks on all of our interfaces, and internal network.
# Refuse incoming packets claiming to be from class A private network
REFUSE_SPOOFING_IPADDR[4]="10.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[4]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[4]="yes"
NETWORK1_IN_REFUSE_SPOOFING[4]="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-273-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
274
# Refuse incoming packets claiming to be from class B private network
REFUSE_SPOOFING_IPADDR[5]="172.16.0.0/12"
INTERFACE0_IN_REFUSE_SPOOFING[5]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[5]="yes"
NETWORK1_IN_REFUSE_SPOOFING[5]="yes"
# Refuse incoming packets claiming to be from class C private network
REFUSE_SPOOFING_IPADDR[6]="192.168.0.0/16"
INTERFACE0_IN_REFUSE_SPOOFING[6]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[6]="no"
NETWORK1_IN_REFUSE_SPOOFING[6]="yes"
WARNING: There is only one exception in which case we do not refuse incoming packets on our
internal interface claiming to be from our internal private network. This appears only for a
Gateway Server when your internal network is from class C. You should not refuse incoming
packets on internal interface from your internal network.
F) We log and drop all incoming packets claiming to be from class D, E, and unallocated IP
addresses. These are classes that are not currently used or that are unallocated. There is
no reason for an incoming packet to have a source IP address from one of those classes.
# Refuse incoming packets claiming to be from class D, E, and unallocated
REFUSE_SPOOFING_IPADDR[7]="224.0.0.0/3"
INTERFACE0_IN_REFUSE_SPOOFING[7]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[7]="yes"
NETWORK1_IN_REFUSE_SPOOFING[7]="yes"
The above Spoofing and bad address protection assume that you have two network interfaces
installed on your system. This configuration is suitable for a Gateway Server. If you only have one
network interface on your server, then you can ignore, comment out or remove those options that
refer to internal interface and network (Interface1 and Network1).
If REFUSE_SPOOFING="no" then the entire spoofing protection section is ignored.
The above configuration closes our discussion about parameters that are the same for all types of
GIPTables firewall configurations. Once you have configured all of the customized values in this
part of the GIPTables configuration file, suitable for your type of system, you are ready to start
the software.
/etc/rc.d/rc.giptables.blocked: The GIPTables Blocked File
Sometimes you’ll know an address that you would like to block from having any access at all to
your server. Instead of entering the entire iptables line per IP address for those jerks on the
internet, you can write them into the rc.giptables.blocked file, that will take the IP
addresses, strip out any comments and run the resulting list through an iptables routine.
The net effect is the /etc/giptables.conf file increases no more than needed, especially
when one might have a large number of IP addresses to deny.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-274-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
276
• To start GIPTables on your system, use the following command:
[root@deep /]# /etc/init.d/giptables start
The GIPTables Firewall Module Files
Once you have chosen the right GIPTables configuration file suitable for the type of server that
you want to protect and all the parameters, which are the same for all type of GIPTables firewall
have been configured, GIPTables should have enough information to run properly on your
system. Really, your work is done and your firewall is well protected against attacks.
Everyone has a different set-up for his firewall design and sometimes we need to implement a
new service and then open and control the port associated with this service on our server.
GIPTables allows us to add, modify, delete, and customize any existing or expected services in
a simple manner through its modules feature.
With GIPTables, each service like DNS, FTP, HTTPD, etc have their own modules. Those
modules are loaded only when defined in the giptables.conf file, so that if there are no
options related to FTP for example, the FTP module will not be loaded. You can specify on which
interface or network the module will work, and what kind of requests (incoming or outgoing) can
go thought that interface or network.
All GIPTables modules are located under the /lib/giptables/modules directory and it’s in
these module files that we handle all rules relating to the specific service. When we configure,
customize and enable service parameters in the giptables.conf file, the parameter in
question get its information about IPTables rules that must be used through the modules files
available under the /lib/giptables/modules directory. If the parameter of the specific
service that you want to enable is not defined into the GIPTables configuration file, then this
service will not load its IPTables rules from its modules file and will not run with your
GIPTables Firewall software.
If you look in the /lib/giptables/modules directory, you’ll find the following modules for the
services that can be enabled with GIPTables Firewall.
giptables-ANY The ANY module, which refer to ANY services
giptables-AUTH The AUTH module, which refer to AUTH services
giptables-DHCP The DHCP module, which refer to DHCP services
giptables-DNS The DNS module, which refer to DNS services
giptables-FINGER The FINGER module, which refer to FINGER services
giptables-FTP The FTP module, which refer to FTP services
giptables-HTTP The HTTP module, which refer to HTTP services
giptables-HTTPS The HTTPS module, which refer to HTTPS services
giptables-ICMP The ICMP module, which refer to ICMP services
giptables-IMAP The IMAP module, which refer to IMAP services
giptables-IMAPS The IMAPS module, which refer to IMAPS services
giptables-LDAP The LDAP module, which refer to LDAP services
giptables-LDAPS The LDAPS module, which refer to LDAPS services
giptables-MYSQL The MYSQL module, which refer to MYSQL services
giptables-NETBIOS The NetBIOS module, which refer to NetBIOS services
giptables-NNTP The NNTP module, which refer to NNTP services
giptables-NNTPS The NNTPS module, which refer to NNTPS services
giptables-NTP The NTP module, which refer to NTP services
giptables-ORACLE The ORACLE module, which refer to ORACLE services
giptables-POP3 The POP3 module, which refer to POP3 services
giptables-POP3S The POP3S module, which refer to POP3S services
giptables-POSTGRES The POSTGRES module, which refer to POSTGRES services](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-276-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
278
Enabling outgoing client requests
In the example below, we define and enable MySQL outgoing client requests for a Gateway
Server. The difference about parameters with other type of servers is that we need to define
additional network interface (INTERFACE1) and network (NETWORK1) for a Gateway Server set-
up. All text in bold should be configured to define and enable the MySQL service.
# ----------------------------------------------------------------------------
# MYSQL outgoing client request
#
# Interface 0 MYSQL outgoing client request
INTERFACE0_MYSQL_CLIENT="yes"
INTERFACE0_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_OUT_DST_IPADDR[0]=$ANY_IPADDR
In the above example, we first enable MySQL outgoing client request on the external interface
(INTERFACE0_MYSQL_CLIENT="yes").
Next, we instruct the system that the parameters apply to interface 0 (INTERFACE0) for the
MySQL service (MYSQL) for outgoing requests (OUT) with the source IP address (SRC_IPADDR)
coming from our external interface IP address ($INTERFACE0_IPADDR). Which means, packets
having our external interface IP address, as a source IP address will be able to go out and/or start
a new connection.
Finally, we inform the system that the parameters also apply to interface 0 (INTERFACE0) for the
MySQL service (MYSQL) for outgoing requests (OUT) with the destination IP address
(DST_IPADDR) going to anywhere ($ANY_IPADDR). And this means, packets having our external
interface, as the destination IP address will be able to go out and/or start a new connection.
Using the connection tracking capability of IPTables, the related MySQL incoming packets are
automatically allowed back in by the firewall. In this case, our machine can be a MySQL client that
is allowed to access any MySQL server on the Internet.
If we want to restrict access to only one external MySQL server, the parameters should be
configured like in the example below:
# Interface 0 MYSQL outgoing client request
INTERFACE0_MYSQL_CLIENT="yes"
INTERFACE0_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_OUT_DST_IPADDR[0]="x.x.x.x"
In this case, "x.x.x.x" is the IP address of the external MySQL server that we want to access.
For a second MySQL server, another set of parameters should be added, like in the example
below:](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-278-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
279
# Interface 0 MYSQL outgoing client request
INTERFACE0_MYSQL_CLIENT="yes"
INTERFACE0_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_OUT_DST_IPADDR[0]="x.x.x.x"
INTERFACE0_MYSQL_OUT_SRC_IPADDR[1]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_OUT_DST_IPADDR[1]="y.y.y.y"
"x.x.x.x" is the IP address of the first external MySQL server that we want to access and
"y.y.y.y" is the IP address of the second external MySQL server that we want to access. Please
note that the index of parameters has been increased, so that the first set of parameters have the
index 0, and the second set of parameters have the index 1.
NOTE: This rule is the same for all GIPTables Firewall parameters that have an index. If you
would like to add a second set of parameters, just copy/paste them, make the required changes
and do not forget to increase the index.
On a Gateway Server or machines with two networks interfaces, we need to define the following
additional parameters for the firewall to recognize the other network interface and the private
network behind it.
# Interface 1 MYSQL outgoing client request
INTERFACE1_MYSQL_CLIENT="yes"
INTERFACE1_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_MYSQL_OUT_DST_IPADDR[0]=$NETWORK1
In the above example, we enable MySQL outgoing client request on the internal interface
(INTERFACE1_MYSQL_CLIENT="yes").
We instruct the system that the parameters apply to internal interface 1 (INTERFACE1) for the
MySQL service (MYSQL) to outgoing requests (OUT) with source IP address (SRC_IPADDR)
coming from our internal interface IP address ($INTERFACE1_IPADDR). Therefore, any packets
having our internal interface IP address, as source IP address will be able to go out and/or start a
new connection.
Next, we inform the system that the parameters also apply to internal interface 1 (INTERFACE1)
for the MySQL service (MYSQL) for outgoing requests (OUT) with a destination IP address
(DST_IPADDR) going from our internal subnet IP address range ($NETWORK1). Therefore, any
packets from our internal subnet will be able to go out and/or start new connections.
Using the connection tracking capability of IPTables, the related MySQL incoming packets are
automatically allowed back in by the firewall. In this case, our machine can be a MySQL client that
is allowed to access any MySQL server from our internal subnet.
# Network 1 MYSQL forwarded outgoing client request
NETWORK1_MYSQL_CLIENT="yes"
NETWORK1_MYSQL_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_MYSQL_OUT_DST_IPADDR[0]=$ANY_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-279-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
280
Here, we enable MySQL outgoing client requests on our internal subnet
(NETWORK1_MYSQL_CLIENT="yes").
We instruct the system that the parameters apply to our internal subnet (NETWORK1) for the
MySQL service (MYSQL) for outgoing requests (OUT) with the source IP address (SRC_IPADDR)
coming from our internal subnet IP address range ($NETWORK1).
In the second line, we inform the system that the parameters also apply to our internal subnet
(NETWORK1) for the MySQL service (MYSQL) to outgoing requests (OUT) with destination IP
address (DST_IPADDR) going to anywhere ($ANY_IPADDR).
Using the connection tracking capability of IPTables, the related MySQL incoming packets are
automatically allowed back in by the firewall. In this case, our machines from our internal subnet
are the MySQL clients and are allowed to access any MySQL server on the Internet.
NOTE: The requests are automatically SNATed (MASQUERADEd) by the GIPTables Firewall, so
that the MySQL server from the Internet thinks that talks with our Gateway server.
In general, you should only replace MYSQL with the name of the service that you want to define
for the parameters to work for other type of services. In our example, we use MYSQL; it is to you
to change it for the service of your choice.
Enabling incoming client requests
As we can see, all of the above parameters apply only to outgoing client requests for the MySQL
service on a Gateway Server. Now for incoming server requests, we should add the related lines
to the configuration file to allow them in. In the example below, we define and enable MySQL
incoming client requests for a Gateway Server. The difference with parameters for other types of
servers is that here we need to define an additional network interface (INTERFACE1) and a
network (NETWORK1) for a Gateway Server set-up.
# ----------------------------------------------------------------------------
# MYSQL incoming client request
#
# Interface 0 MYSQL incoming client request
INTERFACE0_MYSQL_SERVER="yes"
INTERFACE0_MYSQL_IN_SRC_IPADDR[0]=$ANY_IPADDR
INTERFACE0_MYSQL_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
In the above example, we first enable incoming client request for MySQL on the external interface
(INTERFACE0_MYSQL_SERVER="yes").
Next, we instruct the system that the parameters apply to external interface 0 (INTERFACE0) for
the MySQL service (MYSQL) for incoming requests (IN) with the source IP address (SRC_IPADDR)
coming from anywhere ($ANY_IPADDR). This mean that we permit the firewall to receive packets
coming from anywhere on our external interface to start a new connection.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-280-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
281
Finally, we inform the system that parameters also apply to external interface 0 (INTERFACE0)
for MySQL service (MYSQL) on incoming requests (IN) with destination IP address (DST_IPADDR)
coming from our external IP address ($INTERFACE0_IPADDR). In other terms, incoming packets
having our external interface, as destination IP address will be able to come in and/or start a new
connection.
Using the connection tracking capability of IPTables, the related MySQL outgoing packets are
automatically allowed back out by the firewall. In this case, our machine is a MySQL server that is
allowed to receive requests from any MySQL client from the Internet.
If we want to allow access to only one external client machine on the MySQL server, the
parameters should be configured like in the example below:
# Interface 0 MYSQL incoming client request
INTERFACE0_MYSQL_SERVER="yes"
INTERFACE0_MYSQL_IN_SRC_IPADDR[0]="x.x.x.x"
INTERFACE0_MYSQL_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
In this case, "x.x.x.x" is the IP address of the external client machine that is allowed to access
our MySQL server. For a second external client machine allowed, another set of parameters
should be added, like in the example below:
# Interface 0 MYSQL incoming client request
INTERFACE0_MYSQL_SERVER="yes"
INTERFACE0_MYSQL_IN_SRC_IPADDR[0]="x.x.x.x"
INTERFACE0_MYSQL_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_IN_SRC_IPADDR[1]="y.y.y.y"
INTERFACE0_MYSQL_IN_DST_IPADDR[1]=$INTERFACE0_IPADDR
"x.x.x.x" is the IP address of the first external client machine that is allowed to access our
MySQL server and "y.y.y.y" is the IP address of the second external client machine that is
allowed to access our MySQL server. Please note that the index of parameters has been
increased, so that the first set of parameters have the index 0, and the second set of parameters
have the index 1.
NOTE: This rule is the same for all GIPTables Firewall parameters that have an index. If you
would like to add a second set of parameters, just copy/paste them, make the required changes
and do not forget to increase the index.
Don’t forget that we need to add all of the lines below for a Gateway Server set-up for the firewall
to recognize the second network interface and our internal subnet. The definitions and
explanations are the same as for outgoing client requests explained earlier.
# Interface 1 MYSQL incoming client request
INTERFACE1_MYSQL_SERVER="yes"
INTERFACE1_MYSQL_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_MYSQL_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-281-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
282
In the above example, we enable MySQL incoming client request on the internal interface
(INTERFACE1_MYSQL_SERVER="yes").
Next, we instruct the firewall that on the internal interface (INTERFACE1), all MySQL (MYSQL)
incoming packets (IN) with source IP address (SRC_IPADDR) from our internal subnet IP address
range ($NETWORK1) and with destination IP address (DST_IPADDR) coming from our internal
interface IP address ($INTERFACE1_IPADDR) will be allowed to come in and/or start a new
connection.
In other terms, any incoming MySQL packets with source IP address from our internal subnet IP
address range and with our internal interface IP address as destination IP address will be allowed
to come in and/or start a new connection.
Using the connection tracking capability of IPTables, the related MySQL outgoing packets are
automatically allowed back out by the firewall. In this case, our machine is a MySQL server that is
allowed to receive requests from any MySQL client from our internal subnet.
There might be a situation when we would like to access the MySQL server from our internal
subnet using the external interface IP address ($INTERFACE0_IPADDR) as destination IP
address (DST_IPADDR). This is the case when we connect to the MySQL server using its host
name instead of the IP address. Our DNS server might resolve the MySQL server's IP address as
the external interface IP address. In this case, the parameters should be configured like in the
example below:
# Interface 1 MYSQL incoming client request
INTERFACE1_MYSQL_SERVER="yes"
INTERFACE1_MYSQL_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_MYSQL_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_MYSQL_IN_SRC_IPADDR[1]=$NETWORK1
INTERFACE1_MYSQL_IN_DST_IPADDR[1]=$INTERFACE0_IPADDR
As you can see, we have copy/paste the first set of parameters, then changes the destination IP
address (DST_IPADDR) to our external interface IP address ($INTERFACE0_IPADDR) and also
increase the index number.
# Network 1 MYSQL forwarded incoming server request
NETWORK1_MYSQL_SERVER="yes"
NETWORK1_MYSQL_IN_CLI_IPADDR[0]=$ANY_IPADDR
NETWORK1_MYSQL_IN_SRV_IPADDR[0]="192.168.1.1"
In the above example, we enable MySQL incoming client request on our internal subnet
(NETWORK1_MYSQL_SERVER="yes").
Next, we instruct the firewall that in our internal subnet (NETWORK1), all MySQL (MYSQL) incoming
packets (IN) with source IP address (SRC_IPADDR) of any IP address ($ANY_IPADDR) and with
destination IP address (DST_IPADDR) "192.168.1.1" will be allowed to come in and/or start a
new connection.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-282-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
283
In other terms, any incoming MySQL packets with any IP address as source IP address and with
192.168.1.1 as destination IP address will be allowed to come in and/or start a new
connection.
Using the connection tracking capability of IPTables, the related MySQL outgoing packets are
automatically allowed back out by the firewall. In this case, our machine from our internal subnet
that has the IP address 192.168.1.1 is the MySQL server and it is allowed to receive requests
from any MySQL client from the Internet.
NOTE: The MySQL client from the Internet thinks that it talks to our Gateway server, so the actual
destination IP address of the packet is our external interface IP address
($INTERFACE1_IPADDR), but the packet is automatically DNATed to 192.168.1.1.
Pay special attention to the above parameters. We noted that IP address “192.168.1.1” is used
as the value for the incoming client requests with the forwarding feature. This is important, if your
internal workstation IP address is different, you will have to adjust the setting to fit your own IP
address for each of the forwarding definitions.
Step3
Now that our parameters for MySQL service have been correctly entered in the GIPTables
configuration file, we need to restart our GIPTables firewall for the changes to take effect.
• To restart GIPTables on your system, use the following command:
[root@deep /]# /etc/init.d/giptables restart
Well, now we have a better idea about what these cryptic definitions do and how to change them
to fit our needs depending of the type of firewall that we need for our server. Human error is
inevitable and if we entered all the additional parameters into GIPTables by hand, we could in
inadvertently make some errors. To avoid this risk, GIPTables provides through it’s
“giptables.conf.README” file all the possible definitions for available services that can be
used with it.
Therefore, if you need to add some additional services, which do not exist by default in the
giptables.conf file, you can refer to this file to get the parameters to make your service run
with GIPTables Firewall. All you’ll need to do is to cut and paste the required lines into your
GIPTables configuration file and set up each parameter by answering “yes” or “no” to the
questions.
Running the type of GIPTables firewall that you need
All servers should be configured to block the unused ports, even if they are not a firewall
server. This is required for increased security. Imagine that someone gains access to your main
firewall server: if your other servers are not configured to block unused ports, this can result a
serious network security risk. The same is true for local connections; unauthorized employees
can gain access to your other servers from inside the network.
As you should know now, before running GIPTables in your system, you must create a symbolic
link under the /etc directory that points to the GIPTables configuration file suitable for your
system. Once this configuration file exists under your /etc directory, all you have to do is to edit
it and set-up your networking configuration to make it work for you. This is true with all of server
types except for a Gateway Server which differs as explained below.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-283-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
285
Step1
The IP masquerading code will only work if IP forwarding is enabled on your system. This feature
is by default disabled and you can enable it with the following command:
• To enable IPv4 forwarding on your Linux system, use the following command:
Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable packet forwarding (required only for Gateway, VPN, Proxy, PPP)
net.ipv4.ip_forward = 1
You must restart your network for the change to take effect. The command to restart the network
is:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
Step 2
Create the symbolic link to the giptables.conf file that points to the right GIPTables
configuration file suitable for our setup of a Gateway Server.
• This can be accomplished with the following command:
[root@deep /]# ln -s /lib/giptables/conf/giptables.conf.gateway
/etc/giptables.conf
Step3
Once the symbolic link is created, we will edit it to suit our requirements. The text in bold are the
parts of the configuration that must be modified to satisfy your needs.
This is the configuration script file for a Gateway/Proxy Server, it will:
1 Log and limit the amount of incoming dropped packets.
2 Implement the Syn-flood protection.
3 Make sure that all NEW incoming tcp connections are SYN packets.
4 Protect from Spoofing and bad addresses.
5 Allow DNS client requests on internal & external interfaces.
6 Allow FTP client requests on internal & external interfaces.
7 Allow SSH client requests on internal & external interfaces.
8 Allow SSH server requests on the external interface.
9 Allow SMTP client requests on internal & external interfaces.
10 Allow POP3 client requests on the internal interface.
11 Allow POP3S client requests on the internal interface.
12 Allow HTTP client requests on internal & external interfaces.
13 Allow HTTPS client requests on internal & external interfaces.
14 Allow SQUID client requests on internal & external interfaces.
15 Allow SQUID server requests on the external interface.
16 Allow NETBIOS client requests on the internal interface.
17 Allow NETBIOS server requests on the internal interface.
18 Allow TRACEROUTE client requests on internal & external interfaces.
19 Allow TRACEROUTE server requests on internal & external interfaces.
20 Allow ICMP client requests on internal & external interfaces.
21 Allow DHCP client requests on the internal interface.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-285-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
286
If you don’t want some of the services listed in the firewall rules files for the Gateway/Proxy
Server, disable them by saying “no” to the questions. If you want some other services that are not
enabled, simply say, “yes” to the questions. If the service does not exist, add it to your
configuration based on the available examples from the giptables.conf.README file.
• To edit the giptales.conf file, use the following command:
[root@deep /]# vi /etc/giptables.conf
# ----------------------------------------------------------------------------
# GIPTables Firewall v1.1 http://www.giptables.org
# Copyright (C) 2002 Adrian Pascalau <apascalau@openna.com>
# GATEWAY main configuration file
#
# ----------------------------------------------------------------------------
# This file is part of GIPTables Firewall
#
# GIPTables Firewall is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
# ----------------------------------------------------------------------------
# DEBUG
#
DEBUG="off"
# ----------------------------------------------------------------------------
# Some definitions for easy maintenance
# Edit these to suit your system
#
MONOLITIC_KERNEL="no"
# Interface 0: This is our external network interface
# It is directly connected to Internet
INTERFACE0="eth0"
INTERFACE0_IPADDR="x.x.x.x"
ANY_IPADDR="0/0"
# Interface 1: This is our internal network interface
# It is directly connected to our internal Network 1
INTERFACE1="eth1"
INTERFACE1_IPADDR="192.168.1.254"
NETWORK1="192.168.1.0/24"
# Do you need Network Address Translation for your internal network?
NETWORK1_NAT="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-286-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
289
# ----------------------------------------------------------------------------
# Spoofing and bad addresses
#
REFUSE_SPOOFING="yes"
# Refuse incoming packets claiming to be from the ip addresses of our
interfaces
REFUSE_SPOOFING_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_IN_REFUSE_SPOOFING[0]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[0]="no"
NETWORK1_IN_REFUSE_SPOOFING[0]="yes"
REFUSE_SPOOFING_IPADDR[1]=$INTERFACE1_IPADDR
INTERFACE0_IN_REFUSE_SPOOFING[1]="no"
INTERFACE1_IN_REFUSE_SPOOFING[1]="yes"
NETWORK1_IN_REFUSE_SPOOFING[1]="no"
# Refuse incoming packets claiming to be from broadcast-src address range
REFUSE_SPOOFING_IPADDR[2]="0.0.0.0/8"
# If you provide DHCP services on one of your interfaces, do not forget to
# set the following option related to that interface to "no"
INTERFACE0_IN_REFUSE_SPOOFING[2]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[2]="no"
NETWORK1_IN_REFUSE_SPOOFING[2]="yes"
# Refuse incoming packets claiming to be from reserved loopback address range
REFUSE_SPOOFING_IPADDR[3]="127.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[3]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[3]="yes"
NETWORK1_IN_REFUSE_SPOOFING[3]="yes"
# Refuse incoming packets claiming to be from class A private network
REFUSE_SPOOFING_IPADDR[4]="10.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[4]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[4]="yes"
NETWORK1_IN_REFUSE_SPOOFING[4]="yes"
# Refuse incoming packets claiming to be from class B private network
REFUSE_SPOOFING_IPADDR[5]="172.16.0.0/12"
INTERFACE0_IN_REFUSE_SPOOFING[5]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[5]="yes"
NETWORK1_IN_REFUSE_SPOOFING[5]="yes"
# Refuse incoming packets claiming to be from class C private network
REFUSE_SPOOFING_IPADDR[6]="192.168.0.0/16"
INTERFACE0_IN_REFUSE_SPOOFING[6]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[6]="no"
NETWORK1_IN_REFUSE_SPOOFING[6]="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-289-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
290
# Refuse incoming packets claiming to be from class D, E, and unallocated
REFUSE_SPOOFING_IPADDR[7]="224.0.0.0/3"
INTERFACE0_IN_REFUSE_SPOOFING[7]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[7]="yes"
NETWORK1_IN_REFUSE_SPOOFING[7]="yes"
# ****************************************************************************
# *
# A N Y *
# *
# ****************************************************************************
ACCEPT_ANY="no"
# ****************************************************************************
# *
# D N S *
# *
# ****************************************************************************
ACCEPT_DNS="yes"
# ----------------------------------------------------------------------------
# DNS outgoing client request
#
# Interface 0 DNS outgoing client request
INTERFACE0_DNS_CLIENT="yes"
INTERFACE0_DNS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_DNS_OUT_DST_IPADDR[0]=$ISP_PRIMARY_DNS_SERVER
INTERFACE0_DNS_OUT_UDP_REQUEST[0]="yes"
INTERFACE0_DNS_OUT_TCP_REQUEST[0]="yes"
INTERFACE0_DNS_OUT_SPORT53_REQUEST[0]="no"
INTERFACE0_DNS_OUT_SRC_IPADDR[1]=$INTERFACE0_IPADDR
INTERFACE0_DNS_OUT_DST_IPADDR[1]=$ISP_SECONDARY_DNS_SERVER
INTERFACE0_DNS_OUT_UDP_REQUEST[1]="yes"
INTERFACE0_DNS_OUT_TCP_REQUEST[1]="yes"
INTERFACE0_DNS_OUT_SPORT53_REQUEST[1]="no"
# Network 1 DNS forwarded outgoing client request
NETWORK1_DNS_CLIENT="yes"
NETWORK1_DNS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_DNS_OUT_DST_IPADDR[0]=$ISP_PRIMARY_DNS_SERVER
NETWORK1_DNS_OUT_UDP_REQUEST[0]="yes"
NETWORK1_DNS_OUT_TCP_REQUEST[0]="yes"
NETWORK1_DNS_OUT_SPORT53_REQUEST[0]="no"
# ****************************************************************************
# *
# F T P *
# *
# ****************************************************************************
ACCEPT_FTP="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-290-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
291
# ----------------------------------------------------------------------------
# FTP outgoing client request
#
# Interface 0 FTP outgoing client request
INTERFACE0_FTP_CLIENT="yes"
INTERFACE0_FTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_FTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
INTERFACE0_FTP_OUT_PASIVE[0]="yes"
INTERFACE0_FTP_OUT_ACTIVE[0]="no"
# Network 1 FTP forwarded outgoing client request
NETWORK1_FTP_CLIENT="yes"
NETWORK1_FTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_FTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
NETWORK1_FTP_OUT_PASIVE[0]="yes"
NETWORK1_FTP_OUT_ACTIVE[0]="yes"
# ****************************************************************************
# *
# S S H *
# *
# ****************************************************************************
ACCEPT_SSH="yes"
# ----------------------------------------------------------------------------
# SSH outgoing client request
#
# Interface 0 SSH outgoing client request
INTERFACE0_SSH_CLIENT="yes"
INTERFACE0_SSH_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_SSH_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 SSH forwarded outgoing client request
NETWORK1_SSH_CLIENT="yes"
NETWORK1_SSH_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_SSH_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# SSH incoming client request
#
# Interface 0 SSH incoming client request
INTERFACE0_SSH_SERVER="yes"
INTERFACE0_SSH_IN_SRC_IPADDR[0]=$ANY_IPADDR
INTERFACE0_SSH_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-291-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
292
# Interface 1 SSH incoming client request
INTERFACE1_SSH_SERVER="yes"
INTERFACE1_SSH_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_SSH_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
# ****************************************************************************
# *
# T E L N E T *
# *
# ****************************************************************************
ACCEPT_TELNET="no"
# ----------------------------------------------------------------------------
# TELNET outgoing client request
#
# Interface 0 TELNET outgoing client request
INTERFACE0_TELNET_CLIENT="yes"
INTERFACE0_TELNET_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_TELNET_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 TELNET forwarded outgoing client request
NETWORK1_TELNET_CLIENT="yes"
NETWORK1_TELNET_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_TELNET_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# TELNET incoming client request
#
# Interface 1 TELNET incoming client request
INTERFACE1_TELNET_SERVER="no"
INTERFACE1_TELNET_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_TELNET_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
# ****************************************************************************
# *
# T E L N E T S *
# *
# ****************************************************************************
ACCEPT_TELNETS="no"
# ****************************************************************************
# *
# S M T P *
# *
# ****************************************************************************
ACCEPT_SMTP="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-292-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
293
# ----------------------------------------------------------------------------
# SMTP outgoing client request
#
# Interface 0 SMTP outgoing client request
INTERFACE0_SMTP_CLIENT="yes"
INTERFACE0_SMTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_SMTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 SMTP forwarded outgoing client request
NETWORK1_SMTP_CLIENT="yes"
NETWORK1_SMTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_SMTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# S M T P S *
# *
# ****************************************************************************
ACCEPT_SMTPS="no"
# ----------------------------------------------------------------------------
# SMTPS outgoing client request
#
# Interface 0 SMTPS outgoing client request
INTERFACE0_SMTPS_CLIENT="yes"
INTERFACE0_SMTPS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_SMTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 SMTPS forwarded outgoing client request
NETWORK1_SMTPS_CLIENT="yes"
NETWORK1_SMTPS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_SMTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# P O P 3 *
# *
# ****************************************************************************
ACCEPT_POP3="yes"
# ----------------------------------------------------------------------------
# POP3 outgoing client request
#
# Network 1 POP3 forwarded outgoing client request
NETWORK1_POP3_CLIENT="yes"
NETWORK1_POP3_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_POP3_OUT_DST_IPADDR[0]=$ANY_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-293-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
294
# ****************************************************************************
# *
# P O P 3 S *
# *
# ****************************************************************************
ACCEPT_POP3S="yes"
# ----------------------------------------------------------------------------
# POP3S outging client request
#
# Network 1 POP3S forwarded outging client request
NETWORK1_POP3S_CLIENT="yes"
NETWORK1_POP3S_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_POP3S_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# I M A P *
# *
# ****************************************************************************
ACCEPT_IMAP="no"
# ----------------------------------------------------------------------------
# IMAP outgoing client request
#
# Network 1 IMAP forwarded outgoing client request
NETWORK1_IMAP_CLIENT="yes"
NETWORK1_IMAP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_IMAP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# I M A P S *
# *
# ****************************************************************************
ACCEPT_IMAPS="no"
# ----------------------------------------------------------------------------
# IMAPS outgoing client request
#
# Network 1 IMAPS forwarded outgoing client request
NETWORK1_IMAPS_CLIENT="yes"
NETWORK1_IMAPS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_IMAPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# H T T P *
# *
# ****************************************************************************](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-294-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
295
ACCEPT_HTTP="yes"
# ----------------------------------------------------------------------------
# HTTP outgoing client request
#
# Interface 0 HTTP outgoing client request
INTERFACE0_HTTP_CLIENT="yes"
INTERFACE0_HTTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_HTTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 HTTP forwarded outgoing client request
NETWORK1_HTTP_CLIENT="yes"
NETWORK1_HTTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_HTTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# H T T P S *
# *
# ****************************************************************************
ACCEPT_HTTPS="yes"
# ----------------------------------------------------------------------------
# HTTPS outgoing client request
#
# Interface 0 HTTPS outgoing client request
INTERFACE0_HTTPS_CLIENT="yes"
INTERFACE0_HTTPS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_HTTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 HTTPS forwarded outgoing client request
NETWORK1_HTTPS_CLIENT="yes"
NETWORK1_HTTPS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_HTTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# S Q U I D *
# *
# ****************************************************************************
ACCEPT_SQUID="yes" # Squid in Proxy-Caching Mode
# ----------------------------------------------------------------------------
# SQUID outgoing client request
#
# Interface 0 SQUID outgoing client request
INTERFACE0_SQUID_CLIENT="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-295-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
296
INTERFACE0_SQUID_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_SQUID_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Interface 1 SQUID outgoing client request
INTERFACE1_SQUID_CLIENT="yes"
INTERFACE1_SQUID_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_SQUID_OUT_DST_IPADDR[0]=$NETWORK1
# Network 1 SQUID forwarded outgoing client request
NETWORK1_SQUID_CLIENT="yes"
NETWORK1_SQUID_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_SQUID_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# SQUID incoming client request
#
# Interface 0 SQUID incoming client request
INTERFACE0_SQUID_SERVER="yes"
INTERFACE0_SQUID_IN_SRC_IPADDR[0]=$ANY_IPADDR
INTERFACE0_SQUID_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
# Interface 1 SQUID incoming client request
INTERFACE1_SQUID_SERVER="yes"
INTERFACE1_SQUID_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_SQUID_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR
# ****************************************************************************
# *
# W E B C A C H E *
# *
# ****************************************************************************
ACCEPT_WEBCACHE="no" # Squid in HTTPD-Accelerator Mode
# ****************************************************************************
# *
# N N T P *
# *
# ****************************************************************************
ACCEPT_NNTP="no"
# ----------------------------------------------------------------------------
# NNTP outgoing client request
#
# Network 1 NNTP forwarded outgoing client request
NETWORK1_NNTP_CLIENT="yes"
NETWORK1_NNTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_NNTP_OUT_DST_IPADDR[0]=$ANY_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-296-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
298
INTERFACE0_AUTH_CLIENT="yes"
INTERFACE0_AUTH_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_AUTH_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 AUTH forwarded outgoing client request
NETWORK1_AUTH_CLIENT="yes"
NETWORK1_AUTH_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_AUTH_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# W H O I S *
# *
# ****************************************************************************
ACCEPT_WHOIS="no"
# ----------------------------------------------------------------------------
# WHOIS outgoing client request
#
# Interface 0 WHOIS outgoing client request
INTERFACE0_WHOIS_CLIENT="yes"
INTERFACE0_WHOIS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_WHOIS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 WHOIS forwarded outgoing client request
NETWORK1_WHOIS_CLIENT="yes"
NETWORK1_WHOIS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_WHOIS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# F I N G E R *
# *
# ****************************************************************************
ACCEPT_FINGER="no"
# ----------------------------------------------------------------------------
# FINGER outgoing client request
#
# Interface 0 FINGER outgoing client request
INTERFACE0_FINGER_CLIENT="yes"
INTERFACE0_FINGER_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_FINGER_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 FINGER forwarded outgoing client request
NETWORK1_FINGER_CLIENT="yes"
NETWORK1_FINGER_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_FINGER_OUT_DST_IPADDR[0]=$ANY_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-298-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
299
# ****************************************************************************
# *
# N T P *
# *
# ****************************************************************************
ACCEPT_NTP="no"
# ----------------------------------------------------------------------------
# NTP outgoing client request
#
# Interface 0 NTP outgoing client request
INTERFACE0_NTP_CLIENT="yes"
INTERFACE0_NTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_NTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 NTP forwarded outgoing client request
NETWORK1_NTP_CLIENT="no"
NETWORK1_NTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_NTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# N E T B I O S *
# *
# ****************************************************************************
ACCEPT_NETBIOS="yes"
# ----------------------------------------------------------------------------
# NETBIOS outgoing client request
#
# Interface 1 NETBIOS outgoing client request
INTERFACE1_NETBIOS_CLIENT="yes"
INTERFACE1_NETBIOS_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_NETBIOS_OUT_DST_IPADDR[0]=$NETWORK1
# ----------------------------------------------------------------------------
# NETBIOS incoming client request
#
# Interface 1 NETBIOS incoming client request
INTERFACE1_NETBIOS_SERVER="yes"
INTERFACE1_NETBIOS_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_NETBIOS_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR
# ****************************************************************************
# *
# S Y S L O G *
# *
# ****************************************************************************](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-299-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
300
ACCEPT_SYSLOG="no"
# ----------------------------------------------------------------------------
# SYSLOG outgoing client request
#
# Interface 1 SYSLOG outgoing client request
INTERFACE1_SYSLOG_CLIENT="yes"
INTERFACE1_SYSLOG_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_SYSLOG_OUT_DST_IPADDR[0]=$SYSLOG_SERVER
# ****************************************************************************
# *
# T R A C E R O U T E *
# *
# ****************************************************************************
ACCEPT_TRACEROUTE="yes"
# ----------------------------------------------------------------------------
# TRACEROUTE outgoing client request
#
# Interface 0 TRACEROUTE outgoing client request
INTERFACE0_TRACEROUTE_CLIENT="yes"
INTERFACE0_TRACEROUTE_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_TRACEROUTE_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 TRACEROUTE forwarded outgoing client request
NETWORK1_TRACEROUTE_CLIENT="yes"
NETWORK1_TRACEROUTE_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_TRACEROUTE_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# TRACEROUTE incoming client request
#
# Interface 1 TRACEROUTE incoming client request
INTERFACE1_TRACEROUTE_SERVER="no"
INTERFACE1_TRACEROUTE_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_TRACEROUTE_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
# ****************************************************************************
# *
# I C M P *
# *
# ****************************************************************************
ACCEPT_ICMP="yes"
# ----------------------------------------------------------------------------
# ICMP outgoing client request
#
# Interface 0 ICMP outgoing client request](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-300-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
301
INTERFACE0_ICMP_CLIENT="yes"
INTERFACE0_ICMP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_ICMP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 ICMP forwarded outgoing client request
NETWORK1_ICMP_CLIENT="yes"
NETWORK1_ICMP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_ICMP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# ICMP incoming client request
#
# Interface 1 ICMP incoming client request
INTERFACE1_ICMP_SERVER="no"
INTERFACE1_ICMP_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_ICMP_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE1_ICMP_IN_SRC_IPADDR[1]=$NETWORK1
INTERFACE1_ICMP_IN_DST_IPADDR[1]=$INTERFACE1_IPADDR
# ****************************************************************************
# *
# D H C P *
# *
# ****************************************************************************
ACCEPT_DHCP="yes"
# ----------------------------------------------------------------------------
# DHCP incoming client request
#
# Interface 1 DHCP incoming client request
INTERFACE1_DHCP_SERVER="yes"
# If above option is "yes", do not forget to configure the following
# lines in the "Spoofing and bad addresses" section
# REFUSE_SPOOFING_IPADDR[2]="0.0.0.0/8"
# INTERFACE1_IN_REFUSE_SPOOFING[2]="no"
INTERFACE1_DHCP_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_DHCP_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR
# ****************************************************************************
# *
# E N D *
# *
# ****************************************************************************
DROP_EVERYTHING_FROM_HERE="yes"
# ----------------------------------------------------------------------------
# LOG & DROP everything from here... just in case.
#](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-301-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
302
INTERFACE0_IN_DROP_EVERYTHING_FROM_HERE="yes"
INTERFACE1_IN_DROP_EVERYTHING_FROM_HERE="yes"
NETWORK1_IN_DROP_EVERYTHING_FROM_HERE="yes"
# ----------------------------------------------------------------------------
# End of file
Step 4
Once the configuration file has been configured, it is time to start the firewall on your system.
• To start the firewall on your system, use the following command:
[root@deep /]# /etc/init.d/giptables start
Starting Firewalling Services: [OK]
GIPTables-Firewall Administrative Tools
The commands listed below are some that we use often, but many more exist. Check the manual
page and documentation for more information.
IPTables
The iptables tool is used for the firewall packet filter administration of the system. We can use
it to set up a firewall rules file, as we are doing in this book. Once firewall rules have been created
we can play with its many commands to maintain, and inspect the rules in the kernel.
• To list all rules in the selected chain, use the command:
[root@deep /]# iptables –L
Chain INPUT (policy DROP)
target prot opt source destination
ACCEPT all -- anywhere anywhere
Chain FORWARD (policy DROP)
target prot opt source destination
Chain OUTPUT (policy DROP)
target prot opt source destination
ACCEPT all -- anywhere anywhere
This command will list all rules in the selected chain. If no chain is selected, all chains are listed.
• To list all INPUT rules in the selected chain, use the command:
[root@deep /]# iptables -L INPUT
Chain INPUT (policy DROP)
target prot opt source destination
ACCEPT all -- anywhere anywhere
ACCEPT all -- 192.168.1.0/24 anywhere
DROP all -- 204.254.45.9 anywhere
DROP all -- 187.231.11.5 anywhere
DROP all -- 207.35.78.5 anywhere](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-302-320.jpg)
![GIPTables Firewall 1
CHAPTER 0
303
• To list all OUTPUT rules in the selected chain, use the command:
[root@deep /]# iptables -L OUTPUT
Chain OUTPUT (policy DROP)
target prot opt source destination
ACCEPT all -- anywhere anywhere
ACCEPT all -- anywhere 192.168.1.0/24
ACCEPT udp -- 207.35.78.5 207.35.78.3 udp
spt:domain dpt:domain
ACCEPT tcp -- 207.35.78.5 207.35.78.3 tcp
spts:1024:65535 dpt:domain
• To list all FORWARD rules in the selected chain, use the command:
[root@deep /]# iptables -L FORWARD
Chain FORWARD (policy DROP)
target prot opt source destination
DROP tcp -- anywhere anywhere tcp
DROP tcp -- anywhere anywhere tcp
DROP all -- !192.168.0.0/24 anywhere
ACCEPT all -- 192.168.0.0/24 anywhere state NEW
ACCEPT all -- !192.168.0.0/24 anywhere state
This of course works only if you have configured Masquerading on your server (for Gateway
servers in general).
• To list all rules in numeric OUTPUT in the selected chain, use the command:
[root@deep /]# iptables –nL
Chain INPUT (policy DROP)
target prot opt source destination
ACCEPT all -- 0.0.0.0/0 0.0.0.0/0
ACCEPT all -- 192.168.1.0/24 0.0.0.0/0
DROP all -- 204.254.45.9 0.0.0.0/0
Chain FORWARD (policy DROP)
target prot opt source destination
Chain OUTPUT (policy DROP)
target prot opt source destination
ACCEPT all -- 0.0.0.0/0 0.0.0.0/0
ACCEPT all -- 0.0.0.0/0 192.168.1.0/24
ACCEPT udp -- 207.35.78.5 207.35.78.5
All the IP addresses and port numbers will be printed in numeric format.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-303-320.jpg)
![Squid 1
CHAPTER 1
308
These installation instructions assume
Commands are Unix-compatible.
The source path is /var/tmp (note that other paths are possible, as personal discretion).
Installations were tested on OpenNA Linux & Red Hat Linux 7.3.
All steps in the installation will happen using the super-user account “root”.
Whether kernel recompilation may be required: No
Latest Squid version number is 2.4.STABLE6
Packages
The following are based on information listed by Squid as of 2002/03/20. Please regularly check
at www.squid-cache.org for the latest status. We chose to install the required component from
source file because it provides the facility to fine tune the installation.
Source code is available from:
Squid Homepage: http://www.squid-cache.org/
Squid FTP Site: 206.168.0.9
You must be sure to download: squid-2.4.STABLE7-src.tar.gz
Though the procedures given in this chapter are likely to work on all Linux platforms, we have
only tested it on OpenNA Linux and Red Hat Linux.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all files
installed into the system if the package is updated in the future. To solve this problem, it is a good
idea to make a list of files on the system before you install Squid, and one afterwards, and then
compares them using the diff utility of Linux to find out what files are placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Squid1
• And the following one after you install the software:
[root@deep root]# find /* > Squid2
• Then use the following command to get a list of what changed:
[root@deep root]# diff Squid1 Squid2 > Squid-Installed
By doing this, if in the future any upgrade appears, all you have to do is to read the generated list
of what files were added or changed by the program and remove them manually from your
system before installing the new software. We use the /root directory of the system to store all
generated list files.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-308-320.jpg)
![Squid 1
CHAPTER 1
309
Compiling - Optimizing & Installing Squid
Below are the steps that you must make to configure, compile and optimize the Squid server
software before installing it on your system. First off, we install the program as user 'root' so as to
avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp squid-version-src.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf squid-version-src.tar.gz
Step 2
To avoid security risks, we must create a new user account called “squid” to be the owner of the
Squid database cache files and daemon.
• To create this special Squid user on OpenNA Linux, use the following command:
[root@deep tmp]# groupadd -g 23 squid > /dev/null 2>&1 || :
[root@deep tmp]# useradd -c "Squid Server" -d /var/spool/squid -g 23 -s
/bin/false -u 23 squid > /dev/null 2>&1 || :
• To create this special Squid user on Red Hat Linux, use the following command:
[root@deep tmp]# useradd -c "Squid Server" -u 23 -s /bin/false -r -d
/var/spool/squid squid 2>/dev/null || :
The above command will create a null account, with no password, no valid shell, no files owned-
nothing but a UID and a GID for the program. Remember that Squid daemon does not need to
have a shell account on the server.
Step 3
After that, move into the newly created Squid source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created Squid source directory use the command:
[root@deep tmp]# cd squid-2.4.STABLE7/
Step 4
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to fix some problems and to configure the program for our PATH
environment variable under Linux.
• Edit the acl.c file (vi +651 src/acl.c) and change the line:
*Top = splay_insert(xstrdup(t), *Top, aclDomainCompare);
To read:
*Top = splay_insert(xstrdup(t), *Top, aclHostDomainCompare);](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-309-320.jpg)
![Squid 1
CHAPTER 1
312
Step 6
Now, we must make a list of all existing files on the system before installing the software, and one
afterwards, then compare them using the diff utility tool of Linux to find out what files are placed
where and finally install the Squid Proxy Server:
[root@deep squid-2.4.STABLE7]# make all
[root@deep squid-2.4.STABLE7]# cd auth_modules
[root@deep auth_modules]# cd NCSA
[root@deep NCSA]# make
[root@deep NCSA]# cd ../PAM
[root@deep PAM]# make
[root@deep PAM]# cd ../SMB
[root@deep SMB]# make SAMBAPREFIX=/usr
[root@deep SMB]# cd ../getpwnam
[root@deep getpwnam]# make
[root@deep getpwnam]# cd
[root@deep root]# find /* > Squid1
[root@deep root]# cd /var/tmp/squid-2.4.STABLE7/
[root@deep squid-2.4.STABLE7]# make install
[root@deep squid-2.4.STABLE7]# cd auth_modules
[root@deep auth_modules]# install -m 4511 PAM/pam_auth /usr/lib/squid/
[root@deep auth_modules]# install -m 0511 NCSA/ncsa_auth /usr/lib/squid/
[root@deep auth_modules]# install -m 0511 SMB/smb_auth /usr/lib/squid/
[root@deep auth_modules]# install –m 0511 getpwnam/getpwnam_auth
/usr/lib/squid/
[root@deep auth_modules]# mkdir -p /var/spool/squid
[root@deep auth_modules]# mkdir -p /var/log/squid
[root@deep auth_modules]# chown -R squid.squid /var/spool/squid/
[root@deep auth_modules]# chown -R squid.squid /var/log/squid/
[root@deep auth_modules]# chmod 0750 /var/spool/squid/
[root@deep auth_modules]# chmod 0750 /var/log/squid/
[root@deep auth_modules]# rm -rf /var/logs/
[root@deep auth_modules]# rm -f /usr/sbin/RunCache
[root@deep auth_modules]# rm -f /usr/sbin/RunAccel
[root@deep auth_modules]# strip /usr/sbin/squid
[root@deep auth_modules]# strip /usr/sbin/client
[root@deep auth_modules]# strip /usr/lib/squid/*
[root@deep auth_modules]# /sbin/ldconfig
[root@deep auth_modules]# cd
[root@deep root]# find /* > Squid2
[root@deep root]# diff Squid1 Squid2 > Squid-Installed
The make all command will compile all source files into executable binaries that can be
installed, and make install will install the binaries and any supporting files into the appropriate
locations. Pay special attention to the authenticator module directory of Squid, we move into this
directory (auth_modules) and compile all authenticator modules that may be needed with
Squid.
Squid authenticator module is required when you want to authorize and authenticate users
before allowing them an access to the Internet or the network. Different authenticator modules
using different techniques are available with Squid. In our compilation, we compile Squid
authenticator modules for PAM, NCSA, SMB, and getpwnam. You don’t need to compile all of them
but only the one that you want to use or nothing if you are not intending to provide user
authentication for Proxy access.
The mkdir command will create two new directories named “squid” under /var/spool and
/var/log directory.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-312-320.jpg)
![Squid 1
CHAPTER 1
313
The rm command will remove the /var/logs directory since it has been created to handle the
log files for Squid that we have relocated during compile time into the /var/log/squid
directory.
The chown command will change the owner of the /var/spool/squid and /var/log/squid
directories to be owned by the user squid, and the chmod command will make the mode of both
squid directories (0750/drwxr-x---) for security reasons. This means that only squid owner
and group will be able to access these directories and others will not.
Note that we remove the small scripts named “RunCache” and “RunAccel” which take care of
starting Squid in either caching mode or accelerator mode, since we use a better script named
“squid” located under /etc/init.d directory that takes advantage of Linux system V.
Finally, the strip command will reduce the size of the specified binaries for optimum
performance.
Step 7
Once we’ve configured, optimized, compiled, and installed the Squid Proxy Server software, we
can free up some disk space by deleting the program tar archive and the related source directory
since they are no longer needed.
• This is done by using the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf squid-version/
[root@deep tmp]# rm -f squid-version-src.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install Squid. It will also remove the Squid compressed archive from the /var/tmp directory.
Configuring Squid
After Squid has been built and installed successfully on your system, your next step is to
configure and customize all the required parameters in the different Squid configuration files.
Parameters entered into the Squid configuration file (squid.conf) will decide how the Squid
software will run on the server and in which mode (either httpd-accelerator mode or in proxy-
caching mode). This shows us that the installation of Squid under Linux does not care and that
only the configuration of the squid.conf file will decide whether Squid will run in httpd-
accelerator or proxy-caching mode.
/etc/squid/squid.conf: (The Squid Configuration File)
/etc/sysconfig/squid: (The Squid System Configuration File)
/etc/logrotate.d/squid: (The Squid Log Rotation File)
/etc/init.d/squid: (The Squid Initialization File)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-313-320.jpg)
![Squid 1
CHAPTER 1
324
# via the HTTP, FTP, and gopher protocols) on a system closer to the
# requesting site than to the source. Web browsers can then use the
# local Squid cache as a proxy HTTP server, reducing access time as
# well as bandwidth consumption.
#
# processname: squid
# pidfile: /var/run/squid.pid
# config: /etc/squid/squid.conf
PATH=/usr/bin:/sbin:/bin:/usr/sbin
export PATH
# Source function library.
. /etc/rc.d/init.d/functions
# Source networking configuration.
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
# Check if the squid.conf file is present.
[ -f /etc/squid/squid.conf ] || exit 0
# Source Squid configureation.
if [ -f /etc/sysconfig/squid ]; then
. /etc/sysconfig/squid
else
SQUID_OPTS="-D"
SQUID_SHUTDOWN_TIMEOUT=100
fi
# Determine the name of the squid binary.
[ -f /usr/sbin/squid ] && SQUID=squid
[ -z "$SQUID" ] && exit 0
prog="$SQUID"
# Determine which one is the cache_swap directory
CACHE_SWAP=`sed -e 's/#.*//g' /etc/squid/squid.conf |
grep cache_dir | awk '{ print $3 }'`
[ -z "$CACHE_SWAP" ] && CACHE_SWAP=/var/spool/squid
RETVAL=0
start() {
for adir in $CACHE_SWAP; do
if [ ! -d $adir/00 ]; then
echo -n "init_cache_dir $adir... "
$SQUID -z -F 2>/dev/null
fi
done
echo -n $"Starting $prog: "
$SQUID $SQUID_OPTS 2> /dev/null &
# Trap and prevent certain signals from being sent to the Squid process.
trap '' 1 2 3 18
RETVAL=$?
[ $RETVAL -eq 0 ] && touch /var/lock/subsys/$SQUID
[ $RETVAL -eq 0 ] && echo_success
[ $RETVAL -ne 0 ] && echo_failure
echo
return $RETVAL](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-324-320.jpg)
![Squid 1
CHAPTER 1
325
}
stop() {
echo -n $"Stopping $prog: "
$SQUID -k check >/dev/null 2>&1
RETVAL=$?
if [ $RETVAL -eq 0 ] ; then
$SQUID -k shutdown &
rm -f /var/lock/subsys/$SQUID
timeout=0
while : ; do
[ -f /var/run/squid.pid ] || break
if [ $timeout -ge $SQUID_SHUTDOWN_TIMEOUT ]; then
echo
return 1
fi
sleep 2 && echo -n "."
timeout=$((timeout+2))
done
echo_success
echo
else
echo_failure
echo
fi
return $RETVAL
}
reload() {
$SQUID $SQUID_OPTS -k reconfigure
}
restart() {
stop
start
}
condrestart() {
[ -e /var/lock/subsys/squid ] && restart || :
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;
reload)
reload
;;
restart)
restart
;;
condrestart)
condrestart
;;
*)
echo $"Usage: $0 {start|stop|reload|restart|condrestart}"
exit 1
esac
exit $?](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-325-320.jpg)
![Squid 1
CHAPTER 1
326
Step 2
Once the /etc/init.d/squid script file has been created, it is important to make it
executable, change its default permissions, create the necessary links and start it. Making this file
executable will allow the system to run it, changing its default permission is to allow only the root
user to change this file for security reason, and creation of the symbolic links will let the process
control initialization of Linux which is in charge of starting all the normal and authorized processes
that need to run at boot time on your system to start the program automatically for you at each
system reboot.
• To make this script executable and to change its default permissions, use the commands:
[root@deep /]# chmod 700 /etc/init.d/squid
[root@deep /]# chown 0.0 /etc/init.d/squid
• To create the symbolic rc.d links for Squid, use the following commands:
[root@deep /]# chkconfig --add squid
[root@deep /]# chkconfig --level 345 squid on
• To start Squid software manually, use the following command:
[root@deep /]# /etc/init.d/squid start
Starting squid: [OK]
Running Squid with Users Authentication Support
Squid is a complete Proxy software solution that provides many useful features for the
administrator and it is up to us to use and configure them. In this section, we’ll discuss the Proxy
Authentication mechanism that provides a way to authenticate internal users who can use it to
access the Internet. This allows us to add an additional layer of security on which users need to
have rights (the authorization) to access the Internet via the Gateway server in the enterprise.
The Squid source code comes with a few authentication processes. These include:
LDAP: Uses the Lightweight Directory Access Protocol.
NCSA: Uses an NCSA-style username and password file.
MSNT: Uses a Windows NT authentication domain.
PAM: Uses the Linux Pluggable Authentication Modules scheme.
SMB: Uses a SMB server like Windows NT or Samba.
getpwam: Uses the old-fashioned Unix password file.
In order to authenticate users, you need to compile and install one of the above supplied
authentication modules. In our compilation of Squid, we have already included the most
interesting authentication modules, which were NCSA, PAM, SMB, and getpwam.
One problem with all of these authentication modules is the fact that the supplied username and
password are essentially sent in clear text between the browser and the proxy. Therefore,
administrators should not set-up the same username and password that users would use for
account login on the server (if they are allowed) or for email accounts.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-326-320.jpg)
![Squid 1
CHAPTER 1
329
Step 4
One of the last steps is to create accounts for all users who will be allowed to access the Internet
with Squid after proper authentication with a username and password. Remember, we have to
create null account, with no valid shell for our users.
• To create null user account, use the following command:
[root@deep /]# useradd -s /bin/false gmourani
The above command will create a null account, with no password, no valid shell, no files owned-
nothing but a UID and a GID.
• To set a password for this new user, use the following command:
[root@deep /]# passwd gmourani
Changing password for user gmourani
New UNIX password:
Retype new UNIX password:
passwd: all authentication tokens updated successfully
NOTE: It is NOT important to create a home directory for the users (i.e. /home/gmourani).
Squid with Users Authentication Support can run even if home directories are not created for the
users. All we need for authentication is a username and password. Therefore, a home directory is
futile and since we do not give shell access, there is really no reason for users to have a home
directory on the system.
Step 5
Finally, open your favorite browser and enter the username and password to access the Internet
with Squid as your Proxy Caching Gateway Server.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-329-320.jpg)
![Squid 1
CHAPTER 1
330
Securing Squid
This section deals specifically with actions we can take to improve and tighten security under
Squid. As with the other chapters, the interesting points here are that we refer to the features
available within the base installed program and not any additional software.
More control on mounting the cache directory of Squid
If you have created the cache directory of Squid in a separate partition your system (i.e.
/var/spool), like we have done during the initial set-up of Linux, then you can use the noexec,
nodev, and nosuid features of Linux to improve and consolidate the cache security.
These features can be set up in the /etc/fstab file to inform the system to not allow execution
of any binaries (noexec), to not interpret character or block special devices (nodev), and to not
allow set-user-identifier or set-group-identifier bits to take effect (nosuid) on the mounted file
system (/var/spool in our example).
Applying this procedure on the partition where the Squid Cache resides will help to eliminate the
possibility of DEV, SUID/SGID, and execution of any binaries that may be in the Squid cache.
Step 1
• Edit the fstab file (vi /etc/fstab) and add in the line that refer to /var/spool file
system the following options after the defaults option as show below:
LABEL=/var/spool /var/spool ext3 defaults,noexec,nodev,nosuid 1 2
Step 2
Once you have made the necessary adjustments to the /etc/fstab file, it is time to inform the
system about the modification.
• This can be accomplished with the following commands:
[root@deep /]# mount /var/spool -oremount
Each file system that has been modified must be remounted with the command as shown
previously. In our example we have modified the /var/spool file system and it is for this reason
that we remount this file system with the above command.
NOTE: If /var/spool is not a file system but just a directory, then the above command obviously
will not work. The ‘-oremount’ option of the Linux ‘mount’ command is used to remount a file
system, which resides on its own partition on your computer.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-330-320.jpg)
![Squid 1
CHAPTER 1
331
Step 3
• You can verify if the modifications have been correctly applied to the Linux system with
the following command:
[root@deep /]# cat /proc/mounts
/dev/root / ext2 rw 0 0
/proc /proc proc rw 0 0
/dev/sda1 /boot ext3 rw 0 0
/dev/sda9 /chroot ext3 rw 0 0
/dev/sda8 /home ext3 rw 0 0
/dev/sda13 /tmp ext3 rw 0 0
/dev/sda7 /usr ext3 rw 0 0
/dev/sda11 /var ext3 rw 0 0
/dev/sda12 /var/spool ext3 rw,noexec,nodev,nosuid 0 0
none /dev/pts devpts rw 0 0
This command will show you all file system in your Linux server with parameters applied to them.
If you see something like the following, congratulations!
/var/spool /var/spool ext3 rw,noexec,nodev,nosuid 0 0
Immunize the Squid configuration file
As we already know, the immutable bit can be used to prevent deletion, overwriting, or creation of
a symbolic link to a file. Once your squid.conf file has been configured, it’s a good idea to
immunize it with the following command:
[root@deep /]# chattr +i /etc/squid/squid.conf
Banning all Destination addresses except one
In the university libraries, we can often see computers available to students that want to, for their
studies, search for a specific author. Administrators have the task of configuring the computer to
allow only searches on one site where the database of all books and authors are stored.
Therefore, they don’t want to give students access to other sites on the Internet but just to the
database site.
With Squid as the Proxy Server, this can be accomplished easily by adding the right ACL to its
existing configuration file. In the next example, we introduce new ACL rules to our Squid
example configuration file to do just this.
• Edit the squid.conf file (vi /etc/squid/squid.conf) and add/change the
following options. Text in bold are what we have added/changed to our default Squid
example configuration file. Below is what we recommend you enter:
icp_port 0
hierarchy_stoplist cgi-bin ?
acl QUERY urlpath_regex cgi-bin ?
no_cache deny QUERY
cache_mem 128 MB
cache_replacement_policy heap GDSF
memory_replacement_policy heap GDSF
cache_dir diskd /var/spool/squid 2000 16 256
cache_store_log none
acl localhost src 127.0.0.1/255.255.255.255
acl Safe_ports port 80 443 1025-65535
acl CONNECT method CONNECT
acl DATABASE dst 207.78.0.1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-331-320.jpg)
![Squid 1
CHAPTER 1
333
http_access deny all
cache_mgr sysadmin@openna.com
cache_effective_user squid
cache_effective_group squid
logfile_rotate 0
log_icp_queries off
cachemgr_passwd my-secret-pass all
buffered_logs on
This new ACL configuration allows all internal clients from the private class C 192.168.1.0 to
access the Internet between 08:30 and 17:30. In this way, we limit the time when our staff can
connect to the Internet to the working hours of the company only.
NOTE: Don’t forget to restart your Squid Proxy Server for the changes to take effect. The order in
which each line appears in the Squid configuration file is important and you have to respect
them. You can’t just add ‘acl’ or ‘http_access’ parameters, wherever you want. Because the
program reads and interprets each access line in the order that they appear.
Optimizing Squid
This section deals specifically with the actions we can take to improve and tighten the
performance of Squid. Note that we refer to the features available within the base installed
program only.
The atime and noatime attributes
The atime and noatime attributes of Linux can be used to get a measurable performance gain
in the Squid cache directory (/var/spool/squid). See the chapter related to the kernel in this
book for more information on this issue.
Physical memory
The most important resource for Squid is physical memory. Your processor does not need to be
ultra-fast. Your disk system will be the major bottleneck, so fast disks are also important for high-
volume caches. Therefore, our recommendation is to use a SCSI disk with at least 512 MB of
physical memory.
Squid Administrative Tools
Now you’ve configured Squid, tightened its security and optimized it for maximum performance,
we can start to play with its utilities.
Stopping Squid process immediately
There are some interesting command line options, especially when we want to stop Squid on the
server. Unlike other services that run as daemons on the system, Squid cannot be stopped
directly and we have to wait for existing connections to terminate before Squid shutdowns.
Sometimes this is not appropriate and we have to stop Squid immediately. This is possible with
the following command:
• To stop Squid process immediately, use the following command:
[root@deep /]# /usr/sbin/squid -k kill
This command sends a KILL signal, which causes the Squid process to exit immediately, without
closing any connections or log files.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-333-320.jpg)
![Squid 1
CHAPTER 1
335
Step 2
Once the correct ACL have been added to your squid.conf file to allow purge requests on the
Proxy Server, we have to use the client program that comes with Squid to purge an object.
• This procedure can be accomplished with the following commands:
[root@deep /]# client -m PURGE http://www.mydomain.com/
Where <www.mydomain.com> is the object that you want to purge. If the purge was successful,
you will see a “200 OK” response. If the object was not found in the cache, you will see a "404
Not Found" response.
NOTE: The PURGE feature of Squid works only when Users Authentication Support is disabled in
the Squid configuration file. The client program of Squid is not capable of using User
Authentication because it doesn’t have the option to specify a username or password through its
command line.
The cachemgr.cgi program utility of Squid
The cachemgr.cgi utility program, which is available by default when you compile and install
Squid on your system, is designed to run through a web interface, and outputs various statistics
about Squid configuration and performance.
Personally, I don’t recommend you use it. The cachemgr.cgi is a buggy utility, which provides
incomprehensible and cryptic results. Connection to its web interface is not always guaranteed
even if you have the proper configuration. I think that more development and a complete revision
of its functionality is required. Especially when we want to make a remote connection to its web
interface. If you really want to use it, then here are the correct steps you must follow.
This program, by default, is located under the /usr/lib/squid directory, and you have to put it
in your “cgi-bin” directory (eg, /home/httpd/cgi-bin) to be able to use it. Follow the simple
steps below to use this program.
Step 1
The first step will be to move the “cachemgr.cgi” CGI file from the /usr/lib/squid directory
to your /home/httpd/cgi-bin directory.
• This procedure can be accomplished with the following command:
[root@deep /]# mv /usr/lib/squid/cachemgr.cgi /home/httpd/cgi-bin/
Step 2
Once you’ve put the “cachemgr.cgi” program into your /cgi-bin directory, it is time to
change its default mode permission and owner.
• These procedures can be accomplished with the following commands:
[root@deep /]# cd /home/httpd/cgi-bin/
[root@deep cgi-bin]# chown 0.0 cachemgr.cgi
[root@deep cgi-bin]# chmod 0511 cachemgr.cgi](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-335-320.jpg)
![SquidGuard 1
CHAPTER 2
340
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install SquidGuard, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > SquidGuard1
• And the following one after you install the software:
[root@deep root]# find /* > SquidGuard2
• Then use this command to get a list of what changed:
[root@deep root]# diff SquidGuard1 SquidGuard2 > SquidGuard-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing SquidGuard
Below are the steps that you must take to configure, compile and optimize the SquidGuard
software before installing it onto your system.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• These procedures can be accomplished with the following commands:
[root@deep /]# cp squidGuard-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf squidGuard-version.tar.gz
Step 2
After that, move into the newly created SquidGuard source directory and perform the following
steps to configure and optimize the software for your system.
• To move into the newly created SquidGuard source directory use the command:
[root@deep tmp]# cd squidGuard-1.2.0/
• To configure and optimize SquidGuard use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"
./configure
--prefix=/usr
--sysconfdir=/etc
--localstatedir=/var
--with-sg-config=/etc/squid/squidGuard.conf
--with-sg-logdir=/var/log/squid/squidGuard
--with-sg-dbhome=/var/spool/squid/squidGuard
--with-db-inc=/usr/include
--with-db-lib=/usr/lib](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-340-320.jpg)
![SquidGuard 1
CHAPTER 2
341
This tells SquidGuard to set itself up for this particular configuration setup with:
- The location of where the squidGuard configuration file must be installed.
- The location of where the squidGuard log file must be installed.
- The location of where the squidGuard database directory must be installed.
- The location of the Berkley DB includes files that SquidGuard need.
- The location of the Berkley DB library that SquidGuard need.
Step 3
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the SquidGuard software:
[root@deep squidGuard-1.2.0]# make
[root@deep squidGuard-1.2.0]# cd
[root@deep root]# find /* > SquidGuard1
[root@deep root]# cd /var/tmp/squidGuard-1.2.0/
[root@deep squidGuard-1.2.0]# make install
[root@deep squidGuard-1.2.0]# cd samples/
[root@deep samples]# install –m 511 squidGuard.cgi /home/httpd/cgi-bin/
[root@deep samples]# cd dest/
[root@deep dest]# mkdir -p /var/spool/squid/squidGuard
[root@deep dest]# chown -R squid.squid /var/spool/squid/squidGuard/
[root@deep dest]# chmod 0750 /var/spool/squid/squidGuard/
[root@deep dest]# chown -R squid.squid /var/log/squid/squidGuard/
[root@deep dest]# chmod 0750 /var/log/squid/squidGuard/
[root@deep dest]# cp blacklists.tar.gz /var/spool/squid/squidGuard/
[root@deep dest]# cd /var/spool/squid/squidGuard/
[root@deep squidGuard]# mkdir -p aggressive
[root@deep squidGuard]# mkdir -p gambling
[root@deep squidGuard]# mkdir -p hacking
[root@deep squidGuard]# mkdir -p porn
[root@deep squidGuard]# chown -R squid.squid aggressive/
[root@deep squidGuard]# chown -R squid.squid gambling/
[root@deep squidGuard]# chown -R squid.squid hacking/
[root@deep squidGuard]# chown -R squid.squid porn/
[root@deep squidGuard]# tar xzpf blacklists.tar.gz
[root@deep squidGuard]# cd blacklists
[root@deep blacklists]# install -m 644 aggressive/domains
../aggressive/
[root@deep blacklists]# install -m 644 aggressive/urls ../aggressive/
[root@deep blacklists]# install -m 644 gambling/domains ../gambling/
[root@deep blacklists]# install -m 644 gambling/urls ../gambling/
[root@deep blacklists]# install -m 644 hacking/domains ../hacking/
[root@deep blacklists]# install -m 644 hacking/urls ../hacking/
[root@deep blacklists]# install -m 644 porn/domains ../porn/
[root@deep blacklists]# install -m 644 porn/urls ../porn/
[root@deep blacklists]# install -m 644 porn/expressions ../porn/
[root@deep blacklists]# cd ..
[root@deep squidGuard]# chown -R squid.squid *
[root@deep squidGuard]# strip /usr/bin/squidGuard
[root@deep squidGuard]# /sbin/ldconfig
[root@deep squidGuard]# rm -rf blacklists blacklists.tar.gz
[root@deep squidGuard]# cd
[root@deep root]# find /* > SquidGuard2
[root@deep root]# diff SquidGuard1 SquidGuard2 > SquidGuard-Installed](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-341-320.jpg)
![SquidGuard 1
CHAPTER 2
342
The make command will compile all source files into executable binaries that can be installed,
and make install will install the binaries and any supporting files into the appropriate locations.
The “install -m 0511” command will install the CGI program called squidGuard.cgi
(squidGuard.cgi is a small script, which is used to explain to the user that the URL is blocked
and by which rule set) into your cgi-bin directory.
The “mkdir -p” command will create the SquidGuard directory and subdirectories to store
database filter files to run with squidGuard, the “chown and chmod” commands will set the
appropriate mode and ownership permissions to the squidGuard directory and it’s
subdirectories. The “tar” command will untar the blacklists.tar.gz compressed archive
containing all the filter files and the “install -m 644” commands will install the entire filter files
to their appropriate directories.
Finally, the strip command will reduce the size of the specified binaries for optimum
performance and the “rm -rf” commands will remove the blacklists directory and archive
file that we no longer need on our system.
Step 4
Once the configuration, optimization, compilation, and installation of the SquidGuard software
have been accomplished, we can free up some disk space by deleting the program tar archive
and the related source directory since they are no longer needed.
• To delete SquidGuard and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf squidGuard-version/
[root@deep tmp]# rm -f squidGuard-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install SquidGuard. It will also remove the SquidGuard compressed archive from the
/var/tmp directory.
Configuring SquidGuard
After SquidGuard has been built and installed successfully on your system, your next step is to
configure and customize the squidGuard.conf file to suit your needs.
The parameters entered into the squidGuard configuration file (squidGuard.conf) will decide
how the ACL should be applied and on which users, hosts, IP addresses, times, dates,
destination, source, etc.
/etc/squid/squidGuard.conf (The SquidGuard Configuration File)
/home/httpd/cgi-bin/squidGuard.cgi (The squidGuard.cgi File)
/etc/squid/squidGuard.conf: The SquidGuard Configuration File
The /etc/squid/squidGuard.conf file is the main and only configuration file for
squidGuard. It is not difficult to configure or understand but we have to understand the rules
order if we want to have a working SquidGuard configuration file.
The SquidGuard configuration file (squidGuard.conf) has a structure that must be followed
during its configuration. Next we show you the recommended structure for the configuration file
and the order in which declarations are supposed to appear. There are six different possible
declarations where five are optional and one required.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-342-320.jpg)
![SquidGuard 1
CHAPTER 2
351
NOTE: If you receive an error message here, it is likely because you have forgot to configure the
squidguard.cgi program to fit your domain name information. Please edit the
“/home/httpd/cgi-bin/suidguard.cgi” program and make the appropriate changes.
Optimizing SquidGuard
This section deals specifically with the actions we can take to improve and tighten the
performance of SquidGuard. Note that we refer to the features available within the base
installed program only.
Creating a prebuilt database
Ok, your SquidGuard program is running and you’re happy to see that it works, but wait a
minute; it is possible to make SquidGuard runs faster simply by converting its filter files (where
blacklisted domains and urls reside) into a db filter file.
The default filter files used with SquidGuard are in plain text format, and SquidGuard needs to
parse all the lines inside the filter file to decide if domains/url’s can be allowed or not. There is a
better method that gives the same result and also runs faster by converting all of its filter files into
a db file.
Step1
The first step to accomplish this conversion will be to use the “-C” command of SquidGuard.
This command simply converts the text file into a db file.
• To convert your filter text files into a db file,use the following commands:
[root@deep /]# cd /var/spool/squid/squidGuard/
[root@deep /]# squidGuard -C aggressive/domains
[root@deep /]# squidGuard -C aggressive/urls
[root@deep /]# squidGuard -C gambling/domains
[root@deep /]# squidGuard -C gambling/urls
[root@deep /]# squidGuard -C hacking/domains
[root@deep /]# squidGuard -C hacking/urls
[root@deep /]# squidGuard -C porn/domains
[root@deep /]# squidGuard -C porn/urls
The above commands, will convert a domainlist or urllist from plain text file to a prebuilt database.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-351-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
359
These installation instructions assume
Commands are Unix-compatible.
The source path is /usr/src (note that other paths are possible, at personal discretion).
Installations were tested on OpenNA Linux and Red Hat Linux 7.3.
All steps in the installation will happen using the super-user account “root”.
Whether kernel recompilation may be required: Yes
Latest FreeS/WAN version number is 1.95
The procedures given in this chapter are likely to work on all Linux platforms, but we have only
tested it on OpenNA Linux and Red Hat Linux.
Packages
The following is based on information listed by FreeS/WAN as of 2001/02/04. Please check
http://www.freeswan.org/ regularly for the latest status. We chose to install from source because it
provides the facility to fine tune the installation.
Source code is available from:
FreeS/WAN Homepage Site: http://www.freeswan.org/
FreeS/WAN FTP Site: 194.109.6.26
You must be sure to download: freeswan-1.95.tar.gz
Prerequisites
Linux FreeS/WAN requires that the software below is already installed on your system to be able
to run and work successfully. If this is not the case, you must install it from your Linux CD-ROM or
source archive file. Please make sure you have this program installed on your machine before
you proceed with this chapter.
gmp is required to compile and make FreeS/WAN works in your system.
NOTE: Not installing the GMP library will make pluto fail to compile on your server.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install FreeS/WAN, and
then one afterwards, and then compares them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Freeswan1
• And the following one after you install the software:
[root@deep root]# find /* > Freeswan2
• Then use this command to get a list of what changed:
[root@deep root]# diff Freeswan1 Freeswan2 > Freeswan-Installed](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-359-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
360
Compiling - Optimizing & Installing FreeS/WAN
Below are the required steps that you must make to compile and optimize the FreeS/WAN
software before installing it into your Linux system.
Step 1
The installation of IPSEC FreeS/WAN software requires some modification of your original
kernel since some parts of FreeS/WAN must be included and incorporated in your kernel before
you can use it. For this reason the first step in installing FreeS/WAN is to go to the Linux Kernel
section in this book and follow the instructions on how to install the Linux Kernel on your system
(even if you have already done this before) and come back to this section after you have
executed the “make dep; make clean” commands, but before the “make bzImage”
command in the Linux Kernel chapter.
Step 2
Once your kernel is configured and you download the FreeS/WAN program from the main
software site you must copy it to the /usr/src directory and change to this location before
expanding the archive. Putting FreeS/WAN under /usr/src/linux will confuse the links,
therefore, expand the software under /usr/src and never under /usr/src/linux directory.
• These procedures can be accomplished with the following commands:
[root@deep /]# cp freeswan-version.tar.gz /usr/src/
[root@deep /]# cd /usr/src/
[root@deep src]# tar xzpf freeswan-version.tar.gz
Step 3
After that, move into the newly created FreeS/WAN directory then configure, compile and
optimize it.
• To move into the top-level directory of FreeS/WAN distribution use the command:
[root@deep src]# cd freeswan-1.95/
Step 4
You must modify the Makefile.inc under the FreeS/WAN source directory to specify
installation paths and optimization parameters. We must modify this file to be compliant with
Linux file system structure, add optimization parameters for our specific processor architecture
and install FreeS/WAN files under our PATH environment variable.
• Edit the Makefile.inc file (vi Makefile.inc) and change all of the targeted lines in
the order shown below:
PUBDIR=$(DESTDIR)/usr/local/sbin
To read:
INC_USRLOCAL=/usr
REALPRIVDIR=/usr/local/lib/ipsec
To read:
INC_MANDIR=/share/man](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-360-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
361
MANTREE=$(DESTDIR)/usr/local/man
To read:
USERCOMPILE=-O2
CONFDIR=$(DESTDIR)/etc
To read:
KLIPSCOMPILE=-O2
All of the above changes, will relocate all files related to the FreeS/WAN software to the
destination target directories we have chosen. We also add optimization parameters related to
the type of processor that we use in our system for better performance.
Step 5
Once the modifications have been made to the source file of FreeS/WAN as described in step 4,
we need to patch the pre-configured Linux Kernel to include FreeS/WAN support.
• This procedure can be accomplished with the following command:
[root@deep freeswan-1.95]# make ogo
echo "===============" >>out.kpatch
echo "`date` `cd /usr/src/linux ; pwd`" >>out.kpatch
make _patches2.3 >>out.kpatch
…………
The make ogo command is what we use to patch the kernel. It will automatically start the kernel
configuration part for the second time and will let you answer all kernel configuration questions
before compilation and integration of its component into the kernel.
During the second kernel configuration, be sure that your kernel has been built with FreeS/WAN
support enabled. A new section related to Frees/WAN support named “IPSec options
(FreeS/WAN)” should appear in your kernel configuration after you have patched the kernel with
the FreeS/WAN program as described above. You need ensure that you have answered Y to the
following questions under the new kernel section: IPSec options (FreeS/WAN).
IP Security Protocol (FreeS/WAN IPSEC) (CONFIG_IPSEC) [Y/n/?]
*
* IPSec options (FreeS/WAN)
*
IPSEC: IP-in-IP encapsulation (tunnel mode) (CONFIG_IPSEC_IPIP) [Y/n/?]
IPSEC: Authentication Header (CONFIG_IPSEC_AH) [Y/n/?]
HMAC-MD5 authentication algorithm (CONFIG_IPSEC_AUTH_HMAC_MD5) [Y/n/?]
HMAC-SHA1 authentication algorithm (CONFIG_IPSEC_AUTH_HMAC_SHA1) [Y/n/?]
IPSEC: Encapsulating Security Payload (CONFIG_IPSEC_ESP) [Y/n/?]
3DES encryption algorithm (CONFIG_IPSEC_ENC_3DES) [Y/n/?]
IPSEC: IP Compression (CONFIG_IPSEC_IPCOMP) [Y/n/?]
IPSEC Debugging Option (CONFIG_IPSEC_DEBUG) [Y/n/?]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-361-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
362
NOTE: All the customization you made to your kernel the first time you ran the make config,
make dep, and make clean commands will be preserved, so you don’t need to reconfigure every
part of your kernel; Just the new section added by FreeS/WAN named “IPSec options
(FreeS/WAN)” is required, as shown above.
Some networking options will get turned on automatically, even if you previously turned them off;
This is because IPSEC needs them. Whichever configuration program you are using, you should
pay careful attention to a few issues: in particular, do NOT disable any of the following under the
“Networking Options” of your kernel configuration:
Kernel/User netlink socket (CONFIG_NETLINK) [Y/n/?]
Netlink device emulation (CONFIG_NETLINK_DEV) [Y/n/?]
Step 6
Once the make ogo command is completed, your FreeS/WAN software and Linux kernel with
FreeS/WAN support is ready to be installed on your server. We must make a list of files on the
system before you install the software and the kernel, and one afterwards, then compare them
using the diff utility to find out what files are placed where and finally install FreeS/WAN and
the new kernel with FreeS/WAN support in your server:
[root@deep freeswan-1.95]# cd
[root@deep root]# find /* > Freeswan1
[root@deep root]# cd /usr/src/freeswan-1.95/
[root@deep freeswan-1.95]# make install
[root@deep freeswan-1.95]# cd
[root@deep root]# find /* > Freeswan2
[root@deep root]# diff Freeswan1 Freeswan2 > Freeswan-Installed
The make install command will install all FreeS/WAN and kernel components together to the
appropriated location on your server.
Step 7
At this stage of your installation of FreeS/WAN, you must follow the rest of the instructions in the
Linux Kernel chapter of this book as normal to install the kernel. At this point, after you have
copied and installed your new kernel image, system.map, or modules (if necessary), and set the
lilo.conf file to load the new kernel, you must edit and customize the configuration files
related to FreeS/WAN “ipsec.conf” and “ipsec.secrets” before rebooting your system.
Step 8
Once the compilation, optimization and installation of the software have been finished, we can
free up some disk space by deleting the program tar archive and the related source directory
since they are no longer needed.
• To delete FreeS/WAN and its related source directory, use the following commands:
[root@deep /]# cd /usr/src/
[root@deep src]# rm -rf freeswan-version/
[root@deep src]# rm -f freeswan-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-362-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
366
/etc/ipsec.secrets: The FreeS/WAN File to store Secret Keys
The file ipsec.secrets stores the secrets used by the pluto daemon to authenticate
communication between both gateways. Two different kinds of secrets can be configured in this
file, preshared secrets and RSA private keys. You must check the permissions of this file to be
sure that the super-user “root” owns the file, and its permissions are set to block all access by
others.
Step 1
An example secret is supplied in the ipsec.secrets file by default. You should change it by
creating your own. With automatic keying you may have a shared secret up to 256 bits, which is
then used during the key exchanges to make sure a man in the middle attack does not occur.
• To create a new shared secret, use the following commands:
[root@deep /]# ipsec ranbits 256 > temp
New, random keys are created with the ranbits(8) utility in the file named “temp”. The ranbits
utility may pause for a few seconds if not enough entropy is available immediately. Don’t forget to
delete the temporary file as soon as you are done with it.
Step 2
Now that our new shared secret key has been created in the “temp” file, we must put it in the
/etc/ipsec.secrets file. When editing the ipsec.secrets file, you should see something
like the following appearing in your text editor. Each line has the IP addresses of the two
gateways plus the secret. It should look something like this:
# This file holds shared secrets which are currently the only inter-Pluto
# authentication mechanism. See ipsec_pluto(8) manpage. Each secret is
# (oversimplifying slightly) for one pair of negotiating hosts.
# The shared secrets are arbitrary character strings and should be both
# long and hard to guess.
# Note that all secrets must now be enclosed in quotes, even if they have
# no white space inside them.
10.0.0.1 11.0.0.1 "jxVS1kVUTTulkVRRTnTujSm444jRuU1mlkklku2nkW3nnVu
V2WjjRRnulmlkmU1Run5VSnnRT"
• Edit the ipsec.secrets file (vi /etc/ipsec.secrets) and change the default
secrets keys:
10.0.0.1 11.0.0.1 " jxVS1kVUTTulkVRRTnTujSm444jRuU1mlkklku2nkW3nnVu
V2WjjRRnulmlkmU1Run5VSnnRT "
To read:
208.164.186.1 208.164.186.2
"0x9748cc31_2e99194f_d230589b_cd846b57_dc070b01_74b66f34_19c40a1a_804906ed"
Where “208.164.186.1" and “208.164.186.2" are the IP addresses of the two gateways and
"0x9748cc31_2e99194f_d230589b_cd846b57_dc070b01_74b66f34_19c40a1a_804906
ed" (note that the quotes are required) is the shared secret we have generated above with the
command “ipsec ranbits 256 > temp” in the “temp” file.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-366-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
367
Step 3
The files ipsec.conf, and ipsec.secrets must be copied to the second gateway machine so
as to be identical on both ends. The only exception to this is the ipsec.conf file, which must
have in it a section labeled by the line config setup with the correct interface settings for the
second gateway, if they differ from the first. The ipsec.secrets file, contrary to the RSA
private key, should have the same-shared secrets on the two gateways.
WARNING: The file /etc/ipsec.secrets should have permissions rw------- (600) and be
owned by the super-user “root”. The file /etc/ipsec.conf is installed with permissions rw-r-
-r— (644) and must be owned also by “root”.
Configuring RSA private keys secrets
Recall that currently with FreeSWAN software there are two kinds of secrets: preshared secrets
and RSA private keys. The preshared secrets are what we have configured in our ipsec.conf
and ipsec.secrets example, above. Some people may prefer to use RSA private keys for
authentication by the Pluto daemon of the other hosts. If you are in this situation, you will have
to make some minor modifications to your ipsec.conf and ipsec.secrets files as described
in the following steps:
You need to create a separate RSA key for *each* gateway. Each one gets its private key in its
own ipsec.secrets file, and the public keys go in leftrsasigkey and rightrsasigkey
parameters in the conn description of ipsec.conf file, which goes to both.
Step 1
Create a separate RSA key for *each* gateway:
• On the first gateway (e.i. deep), use the following commands:
[root@deep /]# cd /
[root@deep /]# ipsec rsasigkey --verbose 1024 > deep-keys
computing primes and modulus...
getting 64 random bytes from /dev/random
looking for a prime starting there
found it after 30 tries
getting 64 random bytes from /dev/random
looking for a prime starting there
found it after 230 tries
swapping primes so p is the larger
computing (p-1)*(q-1)...
computing d...
computing exp1, exp1, coeff...
output...
• On the second gateway (e.i. mail), use the following commands:
[root@mail /]# cd /
[root@mail /]# ipsec rsasigkey --verbose 1024 > mail-keys
computing primes and modulus...
getting 64 random bytes from /dev/random
looking for a prime starting there
found it after 30 tries
getting 64 random bytes from /dev/random
looking for a prime starting there
found it after 230 tries
swapping primes so p is the larger
computing (p-1)*(q-1)...](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-367-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
369
RSA keys for gateway deep (deep-keys):
[root@deep /]# cd /
[root@deep /]# vi deep-keys
# 1024 bits, Fri Feb 4 05:05:19 2000
# for signatures only, UNSAFE FOR ENCRYPTION
#pubkey=0x010395daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74
bcfb51a6ecc08890d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb
4d7689f2da7a69199e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16
a76bab772888f1fd71aa08f08502a141b611f
Modulus:
0x95daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74bcfb51a6ecc08890
d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb4d7689f2da7a6919
9e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16a76bab772888f1fd
71aa08f08502a141b611f
PublicExponent: 0x03
# everything after this point is secret
PrivateExponent:
0x63e74967eaea2025c98c69f6ef0753a6a3ff6764157dbdf1f50013471324dd352366f48805b0b
37f232384b2b52ce2ee85d173468b62eaa052381a9588a317b3a1324d01a531a41fa7add6c5efbd
d88f4718feed2bc0246be924e81bb90f03e49ceedf7af0dd48f06f265b519600bd082c6e6bd27ea
a71cc0288df1ecc3b062b
Prime1:
0xc5b471a88b025dd09d4bd7b61840f20d182d9b75bb7c11eb4bd78312209e3aee7ebfe632304db
6df5e211d21af7fee79c5d45546bea3ccc7b744254f6f0b847f
Prime2:
0xc20a99feeafe79767122409b693be75f15e1aef76d098ab12579624aec708e85e2c5dd62080c3
a64363f2f45b0e96cb4aef8918ca333a326d3f6dc2c72b75361
Exponent1:
0x83cda11b0756e935be328fcebad5f6b36573bcf927a80bf2328facb6c0697c9eff2a9976cade7
9ea3ec0be1674fff4512e8d8e2f29c2888524d818df9f5d02ff
Exponent2:
0x815c66a9f1fefba44b6c2b124627ef94b9411f4f9e065c7618fb96dc9da05f03ec83e8ec055d7
c42ced4ca2e75f0f3231f5061086ccd176f37f9e81da1cf8ceb
Coefficient:
0x10d954c9e2b8d11f4db1b233ef37ff0a3cecfffad89ba5d515449b007803f577e3bd7f0183ced
dfd805466d62f767f3f5a5731a73875d30186520f1753a7e325
RSA keys for gateway mail (mail-keys):
[root@mail /]# cd /
[root@mail /]# vi mail-keys
# 1024 bits, Fri Feb 4 04:46:59 2000
# for signatures only, UNSAFE FOR ENCRYPTION
#pubkey=0x01037631b81f00d5e6f888c542d44dbb784cd3646f084ed96f942d341c7c4686c
bd405b805dc728f8697475f11e8b1dd797550153a3f0d4ff0f2b274b70a2ebc88f073748d1c1c88
21dc6be6a2f0064f3be7f8e4549f8ab9af64944f829b014788dd202cf7d2e320cab666f5e7a197e
64efe0bfee94e92ce4dad82d5230c57b89edf
Modulus:
0x7631b81f00d5e6f888c542d44dbb784cd3646f084ed96f942d341c7c4686cbd405b805dc728f8
697475f11e8b1dd797550153a3f0d4ff0f2b274b70a2ebc88f073748d1c1c8821dc6be6a2f0064f
3be7f8e4549f8ab9af64944f829b014788dd202cf7d2e320cab666f5e7a197e64efe0bfee94e92c
e4dad82d5230c57b89edf
PublicExponent: 0x03
# everything after this point is secret
PrivateExponent:
0x4ecbd014ab3944a5b08381e2de7cfadde242f4b03490f50d737812fd8459dd3803d003e84c5fa
f0f84ea0bf07693a64e35637c2a08dff5f721a324b1747db09f62c871d5e11711251b845ae76753
d4ef967c494b0def4f5d0762f65da603bc04c41b4c6cab4c413a72c633b608267ae2889c162a3d5
bc07ee083b1c6e038400b](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-369-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
370
Prime1:
0xc7f7cc8feaaac65039c39333b878bffd8f95b0dc22995c553402a5b287f341012253e9f25b839
83c936f6ca512926bebee3d5403bf9f4557206c6bbfd9aac899
Prime2:
0x975015cb603ac1d488dc876132d8bc83079435d2d3395c03d5386b5c004eadd4d7b01b3d86aad
0a2275d2d6b791a2abe50d7740b7725679811a32ca22db97637
Exponent1:
0x854fddb5471c84357bd7b777d0507ffe5fb92092c1bb92e37801c3cc5aa22b5616e29bf6e7ad1
028624a486e0c619d47f428e2ad2a6a2e3a159d9d2a911c85bb
Exponent2:
0x64e00e87957c81385b3daf9621e5d302050d7937377b92ad38d04792aadf1e8de52012290471e
06c1a3e1e47a61171d435e4f807a4c39a6561177316c9264ecf
Coefficient:
0x6f087591becddc210c2ee0480e30beeb25615a3615203cd3cef65e5a1d476fd9602ca0ef10d9b
858edb22db42c975fb71883a470b43433a7be57df7ace4a0a3f
Extract and copy the public RSA key files of deep and mail to your ipsec.conf files as shown
below. You can locate the line related to the public key by a sentence beginning with the
commented-out: “#pubkey=” line.
# sample connection
conn deep-mail
left=208.164.186.1
leftsubnet=192.168.1.0/24
leftnexthop=205.151.222.250
right=208.164.186.2
rightsubnet=192.168.1.0/24
rightnexthop=205.151.222.251
keyingtries=0
auth=ah
authby=rsasig
leftrsasigkey=0x010395daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801ce
a9cb74bcfb51a6ecc08890d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef9
8a7f29edcb4d7689f2da7a69199e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5d
eac3b19d561c16a76bab772888f1fd71aa08f08502a141b611f
rightrsasigkey=0x01037631b81f00d5e6f888c542d44dbb784cd3646f084ed96f942d341c
7c4686cbd405b805dc728f8697475f11e8b1dd797550153a3f0d4ff0f2b274b70a2ebc88f07
3748d1c1c8821dc6be6a2f0064f3be7f8e4549f8ab9af64944f829b014788dd202cf7d2e320
cab666f5e7a197e64efe0bfee94e92ce4dad82d5230c57b89edf
auto=start
NOTE: Don’t forget that, in this example, the “leftrsasigkey=” parameter contains the public
key of deep and the “rightrsasigkey=” parameter contains the public key of mail.
Step 3
Modify your /etc/ipsec.secrets files to use RSA private keys in *each* gateway:
Edit your original ipsec.secrets file (vi /etc/ipsec.secrets) and add the RSA private
key for authentication on both gateways:
The ipsec.secrets file for gateway deep:
[root@deep /]# vi /etc/ipsec.secrets
208.164.186.1 208.164.186.2
"0x9748cc31_2e99194f_d230589b_cd846b57_dc070b01_74b66f34_19c40a1a_804906ed"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-370-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
371
You must change your original ipsec.secrets file as shown above to look like the following on
both gateways. It is important to note that the private keys are not the same on both gateways,
deep and mail. The private key for deep comes from the RSA key file “deep-keys”, while the
private key for mail comes from the RSA key file “mail-keys”:
208.164.186.1 208.164.186.2: RSA {
Modulus:
0x95daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74bcfb51a6ecc08890
d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb4d7689f2da7a6919
9e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16a76bab772888f1fd
71aa08f08502a141b611f
PublicExponent: 0x03
# everything after this point is secret
PrivateExponent:
0x63e74967eaea2025c98c69f6ef0753a6a3ff6764157dbdf1f50013471324dd352366f48805b0b
37f232384b2b52ce2ee85d173468b62eaa052381a9588a317b3a1324d01a531a41fa7add6c5efbd
d88f4718feed2bc0246be924e81bb90f03e49ceedf7af0dd48f06f265b519600bd082c6e6bd27ea
a71cc0288df1ecc3b062b
Prime1:
0xc5b471a88b025dd09d4bd7b61840f20d182d9b75bb7c11eb4bd78312209e3aee7ebfe632304db
6df5e211d21af7fee79c5d45546bea3ccc7b744254f6f0b847f
Prime2:
0xc20a99feeafe79767122409b693be75f15e1aef76d098ab12579624aec708e85e2c5dd62080c3
a64363f2f45b0e96cb4aef8918ca333a326d3f6dc2c72b75361
Exponent1:
0x83cda11b0756e935be328fcebad5f6b36573bcf927a80bf2328facb6c0697c9eff2a9976cade7
9ea3ec0be1674fff4512e8d8e2f29c2888524d818df9f5d02ff
Exponent2:
0x815c66a9f1fefba44b6c2b124627ef94b9411f4f9e065c7618fb96dc9da05f03ec83e8ec055d7
c42ced4ca2e75f0f3231f5061086ccd176f37f9e81da1cf8ceb
Coefficient:
0x10d954c9e2b8d11f4db1b233ef37ff0a3cecfffad89ba5d515449b007803f577e3bd7f0183ced
dfd805466d62f767f3f5a5731a73875d30186520f1753a7e325
}
The ipsec.secrets file for gateway mail:
[root@mail /]# vi /etc/ipsec.secrets
208.164.186.1 208.164.186.2: RSA {
Modulus:
0x95daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74bcfb51a6ecc08890
d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb4d7689f2da7a6919
9e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16a76bab772888f1fd
71aa08f08502a141b611f
PublicExponent: 0x03
# everything after this point is secret
PrivateExponent:
0x63e74967eaea2025c98c69f6ef0753a6a3ff6764157dbdf1f50013471324dd352366f48805b0b
37f232384b2b52ce2ee85d173468b62eaa052381a9588a317b3a1324d01a531a41fa7add6c5efbd
d88f4718feed2bc0246be924e81bb90f03e49ceedf7af0dd48f06f265b519600bd082c6e6bd27ea
a71cc0288df1ecc3b062b
Prime1:
0xc5b471a88b025dd09d4bd7b61840f20d182d9b75bb7c11eb4bd78312209e3aee7ebfe632304db
6df5e211d21af7fee79c5d45546bea3ccc7b744254f6f0b847f
Prime2:
0xc20a99feeafe79767122409b693be75f15e1aef76d098ab12579624aec708e85e2c5dd62080c3
a64363f2f45b0e96cb4aef8918ca333a326d3f6dc2c72b75361
Exponent1:
0x83cda11b0756e935be328fcebad5f6b36573bcf927a80bf2328facb6c0697c9eff2a9976cade7
9ea3ec0be1674fff4512e8d8e2f29c2888524d818df9f5d02ff](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-371-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
372
Exponent2:
0x815c66a9f1fefba44b6c2b124627ef94b9411f4f9e065c7618fb96dc9da05f03ec83e8ec055d7
c42ced4ca2e75f0f3231f5061086ccd176f37f9e81da1cf8ceb
Coefficient:
0x10d954c9e2b8d11f4db1b233ef37ff0a3cecfffad89ba5d515449b007803f577e3bd7f0183ced
dfd805466d62f767f3f5a5731a73875d30186520f1753a7e325
}
Authentication by RSA Signatures requires that each host have its own private key. The key part
of an entry may start with a token indicating the kind of key. “RSA” signifies RSA private key and
“PSK” (which is the default) signifies PreShared Key. Since “PSK” is the default, we must specify
“RSA”, so that we’ll be able to use RSA private keys in this file (ipsec.secrets). The super-user
“root” should own the file ipsec.secrets, and its permissions should be set to block all access
by others.
Requiring network setup for IPSec
There are some considerations you must ensure are correct before running FreeS/WAN
software. These considerations are important if you don’t want to receive error messages during
start up of your VPN. The following are the steps to follow:
Step1
You will need to enable TCP/IP forwarding on the both gateway servers. In Linux, this is
accomplished by adding the following line:
• To enable IPv4 forwarding on your Linux system, edit the /etc/sysctl.conf file (vi
/etc/sysctl.conf) and add the following line:
# Enable/Disable packet forwarding
net.ipv4.ip_forward = 1
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
Step 2
Recall that automatically keyed connections use keys automatically generated by the Pluto key
negotiation daemon. The pluto daemon will start up, try to connect to the Pluto daemon at the
other end of the tunnel, and establish a connection. For this reason, an IPSEC gateway should
have packet filters rules (in the firewall script file) permitting the following protocols to traverse the
gateway when talking to other IPSEC gateway:
UDP port 500 for IKE implemented by the Pluto daemon
Protocol 50 for ESP encryption and/or authentication
Protocol 51 for AH packet-level authentication
See the GIPTables chapter in this book and the GIPTables manual for the correct rules to add
to your firewall on both gateway machines to allow IPSEC packets to traverse the remote network
gateway to your network gateway and vice versa.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-372-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
373
Step 3
The rp_filter subsystem (related to IP spoofing protection) must be turned off on both
gateways for IPSEC to work properly. This is accomplished by checking if the value 0 (off) is set
in the /proc/sys/net/ipv4/conf/ipsec0/rp_filter and
/proc/sys/net/ipv4/conf/eth0/rp_filter files respectively:
• To check if the value 0 (off) is set in the rp_filter files, use the commands:
[root@deep /]# cat /proc/sys/net/ipv4/conf/ipsec0/rp_filter
0
[root@deep /]# cat /proc/sys/net/ipv4/conf/eth0/rp_filter
0
NOTE: The subdirectory “ipsec0” in our example will be created only after the reboot of your
system. So you may check the value of the “rp_filter” file in the “ipsec0” directory after your
system has been restarted.
• To set the value 0 (off) in the both rp_filter files manually, use the commands:
[root@deep /]# echo 0 > /proc/sys/net/ipv4/conf/ipsec0/rp_filter
[root@deep /]# echo 0 > /proc/sys/net/ipv4/conf/eth0/rp_filter
Also you can put lines like the following in your firewall script files
/etc/rc.d/init.d/iptables on both gateways to automatically set these values to 0 (off)
and avoid making them manually:
# Disable IP spoofing protection to allow IPSEC to work properly
echo 0 > /proc/sys/net/ipv4/conf/ipsec0/rp_filter
echo 0 > /proc/sys/net/ipv4/conf/eth0/rp_filter
NOTE: In the example of the firewall script file above, we assume that eth0 is the interface you
use for your connection. Of course if you use eth1 you must change eth0 to eth1, and so on.
If you forget this step you will receive error messages on your terminal such as the following
during the start up of FreeSWAN IPSEC:
ipsec_setup: WARNING: ipsec0 has route filtering turned on, KLIPS may not work
ipsec_setup: (/proc/sys/net/ipv4/conf/ipsec0/rp_filter = `1', should be 0)
ipsec_setup: WARNING: eth0 has route filtering turned on, KLIPS may not work
ipsec_setup: (/proc/sys/net/ipv4/conf/eth0/rp_filter = `1', should be 0)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-373-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
374
Testing the FreeS/WAN installation
• Reboot the both gateways to get FreeS/WAN started.
• Examine the /var/log/messages file for any signs of trouble. If all goes well you
should see something like this in the /var/log/messages file:
Feb 2 05:22:35 deep ipsec_setup: Starting FreeS/WAN IPSEC
snap2000jan31b...
Feb 2 05:22:35 deep ipsec_setup: KLIPS debug `none'
Feb 2 05:22:35 deep ipsec_setup: KLIPS ipsec0 on eth0
192.168.1.1/255.255.255.0 broadcast 192.168.1.255
Feb 2 05:22:36 deep ipsec_setup: Disabling core dumps:
Feb 2 05:22:36 deep ipsec_setup: Starting Pluto (debug `none'):
Feb 2 05:22:37 deep ipsec_setup: Loading Pluto database `deep-mail':
Feb 2 05:22:37 deep ipsec_setup: Enabling Pluto negotiation:
Feb 2 05:22:37 deep ipsec_setup: Routing for Pluto conns `deep-mail':
Feb 2 05:22:37 deep ipsec_setup: Initiating Pluto tunnel `deep-mail':
Feb 2 05:22:39 deep ipsec_setup: 102 "deep-mail" #1: STATE_MAIN_I1:
initiate
Feb 2 05:22:39 deep ipsec_setup: 104 "deep-mail" #1: STATE_MAIN_I2: from
STATE_MAIN_I1; sent MI2, expecting MR2
Feb 2 05:22:39 deep ipsec_setup: 106 "deep-mail" #1: STATE_MAIN_I3: from
STATE_MAIN_I2; sent MI3, expecting MR3
Feb 2 05:22:39 deep ipsec_setup: 004 "deep-mail" #1: STATE_MAIN_I4: SA
established
Feb 2 05:22:39 deep ipsec_setup: 110 "deep-mail" #2: STATE_QUICK_I1:
initiate
Feb 2 05:22:39 deep ipsec_setup: 004 "deep-mail" #2: STATE_QUICK_I2: SA
established
Feb 2 05:22:39 deep ipsec_setup: ...FreeS/WAN IPSEC started
• Examine the /var/log/secure file for any signs of trouble. If all goes well you should
see something like the following:
Feb 21 14:45:42 deep Pluto[432]: Starting Pluto (FreeS/WAN Version 1.3)
Feb 21 14:45:43 deep Pluto[432]: added connection description "deep-mail"
Feb 21 14:45:43 deep Pluto[432]: listening for IKE messages
Feb 21 14:45:43 deep Pluto[432]: adding interface ipsec0/eth0 192.168.1.1
Feb 21 14:45:43 deep Pluto[432]: loading secrets from
"/etc/ipsec.secrets"
Feb 21 14:45:43 deep Pluto[432]: "deep-mail" #1: initiating Main Mode
Feb 21 14:45:44 deep Pluto[432]: "deep-mail" #1: ISAKMP SA established
Feb 21 14:45:44 deep Pluto[432]: "deep-mail" #2: initiating Quick Mode
POLICY_RSASIG+POLICY_ENCRYPT+POLICY_AUTHENTICATE+POLICY_TUNNEL+POLICY_PFS
Feb 21 14:45:46 deep Pluto[432]: "deep-mail" #2: sent QI2, IPsec SA
established
Feb 21 14:45:47 deep Pluto[432]: "deep-mail" #3: responding to Main Mode
Feb 21 14:45:49 deep Pluto[432]: "deep-mail" #3: sent MR3, ISAKMP SA
established
Feb 21 14:45:49 deep Pluto[432]: "deep-mail" #4: responding to Quick Mode
Feb 21 14:45:50 deep Pluto[432]: "deep-mail" #4: IPsec SA established](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-374-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
375
• On both gateways, the following entries should now exist in the /proc/net/ directory:
[root@deep /]# ls -l /proc/net/ipsec_*
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_eroute
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_klipsdebug
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_spi
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_spigrp
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_spinew
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_tncfg
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_version
• The IPSEC interfaces should be attached on top of the specified physical interfaces.
Confirm that with:
[root@deep /]# cat /proc/net/ipsec_tncfg
ipsec0 -> eth0 mtu=16260 -> 1500
ipsec1 -> NULL mtu=0 -> 0
ipsec2 -> NULL mtu=0 -> 0
ipsec3 -> NULL mtu=0 -> 0
• Now execute the following command to show minimal debugging information and see if
the output looks something like this:
[root@deep /]# ipsec look
deep.openna.com Fri Feb 4 17:25:17 EST 2000
============-============
192.168.1.1/32 -> 192.168.1.2/32 => tun0x106@192.168.1.2
esp0x4450894d@192.168.1.2 ah0x4450894c@192.168.1.2
------------=------------
ah0x3350f551@192.168.1.1 AH_HMAC_MD5: dir=in ooowin=32 seq=115
bit=0xffffffff alen=128 aklen=16
life(c,s,h)=bytes(16140,0,0)add(51656,0,0)use(54068,0,0)packets(115,0,0)
idle=499
ah0x4450894c@192.168.1.2 AH_HMAC_MD5: dir=out ooowin=32 seq=2828 alen=128
aklen=16
life(c,s,h)=bytes(449488,0,0)add(51656,0,0)use(51656,0,0)packets(2828,0,0
) idle=6
esp0x3350f552@192.168.1.1 ESP_3DES: dir=in ooowin=32 seq=115
bit=0xffffffff eklen=24
life(c,s,h)=bytes(13380,0,0)add(51656,0,0)use(54068,0,0)packets(115,0,0)
idle=499
esp0x4450894d@192.168.1.2 ESP_3DES: dir=out ooowin=32 seq=2828 eklen=24
life(c,s,h)=bytes(381616,0,0)add(51656,0,0)use(51656,0,0)packets(2828,0,0
) idle=6
tun0x105@192.168.1.1 IPIP: dir=in 192.168.1.2 -> 192.168.1.1
life(c,s,h)=add(51656,0,0)
tun0x106@192.168.1.2 IPIP: dir=out 192.168.1.1 -> 192.168.1.2
life(c,s,h)=bytes(327581,0,0)add(51656,0,0)use(51656,0,0)packets(2828,0,0
) idle=6
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 ipsec0
192.168.1.1 0.0.0.0 255.255.255.255 UH 0 0 0 eth0
192.168.1.2 192.168.1.2 255.255.255.255 UGH 0 0 0 ipsec0
Destination Gateway Genmask Flags MSS Window irtt Iface](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-375-320.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
376
• Try pinging 192.168.1.2 from the 192.168.1.1 client. If this works then you have set
it up correctly. If it does not work check your network to make sure 208.164.186.1 can
reach 208.164.186.2, and that TCP-IP forwarding is enabled, and make sure that no
firewall rules are blocking the packets, or trying to masquerade them before the rules
allowing IPSec related traffic. For this test to work, it is important to use pings that go
from one subnet to the other.
208.164.186.1 ---- 205.151.222.250 ---- 205.151.222.251 ---- 208.164.186.2
| |
192.168.1.0/24 192.168.1.0/24
| |
192.168.1.1 192.168.1.2
A last note about testing the installation of FreeSWAN IPSEC, if you encounter a problem that
you are unable to resolve, you can use the following command to view a collection of debugging
information (contents of files, selections from logs, etc.) related to the IPSEC
encryption/authentication system that you should send to the Linux-IPSEC Mailing List (linux-
ipsec@clinet.fi) to help you.
• Use the following command to make an output of a collection of debugging information:
[root@deep /]# ipsec barf > result
This command is primarily provided as a convenience for remote debugging; A single command
which packages up (and labels) all information that might be relevant to diagnosing a problem in
IPSEC.
Further documentation
For more details, there are several manual pages about FreeS/WAN that you could read:
$ man ipsec (8) - invoke IPSEC utilities.
$ man ipsec atoaddr, addrtoa (3) - convert Internet addresses to and from ASCII.
$ man ipsec atoasr (3) - convert ASCII to Internet address, subnet, or range.
$ man ipsec atobytes, bytestoa (3) - convert binary data bytes from and to ASCII formats.
$ man ipsec atodata, datatoa (3) - convert binary data from and to ASCII formats.
$ man ipsec atosa, satoa (3) - convert IPSEC Security Association IDs to and from ASCII.
$ man ipsec atosubnet, subnettoa (3) - convert subnet/mask ASCII form to and from addresses.
$ man ipsec atoul, ultoa (3) - convert unsigned-long numbers to and from ASCII.
$ man ipsec auto (8) - control automatically-keyed IPSEC connections.
$ man ipsec barf (8) - spew out collected IPSEC debugging information.
$ man ipsec bitstomask (3) - convert bit count to Internet subnet mask.
$ man ipsec eroute (8) - manipulate IPSEC extended routing tables.
$ man ipsec goodmask (3) - is this Internet subnet mask a valid one?
$ man ipsec hostof (3) - given Internet address and subnet mask, return host part.
$ man ipsec klipsdebug (8) - set Klips (kernel IPSEC support) debug features and level.
$ man ipsec look (8) - show minimal debugging information.
$ man ipsec manual (8) - take manually-keyed IPSEC connections up and down.
$ man ipsec masktobits (3) - convert Internet subnet mask to bit count.
$ man ipsec optionsfrom (3) - read additional ``command-line'' options from file.
$ man ipsec pluto (8) - IPsec IKE keying daemon.
$ man ipsec ranbits (8) - generate random bits in ASCII form.
$ man ipsec rangetoa (3) - convert Internet address range to ASCII.
$ man ipsec rsasigkey (8) - generate RSA signature key.
$ man ipsec setup (8) - control IPSEC subsystem.
$ man ipsec spi (8) - manage IPSEC Security Associations.
$ man ipsec spigrp (8) - group/ungroup IPSEC Security Associations.
$ man ipsec subnetof (3) - given Internet address and subnet mask, return subnet number.
$ man ipsec tncfg (8) - associate IPSEC virtual interface with real interface.
$ man ipsec whack (8) - control interface for IPSEC keying daemon.
$ man ipsec.conf (5) - IPSEC configuration and connections.
$ man ipsec.secrets (5) - secrets for IKE/IPsec authentication.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-376-320.jpg)
![GnuPG 1
CHAPTER 4
382
Prerequisites
GnuPG requires that the listed software below be already installed on your system to be able to
compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source
archive files. Please make sure you have this program installed on your machine before you
proceed with this chapter.
gettext is required to run GnuPG on your system.
python-1.5 is required to run GnuPG on your system.
expat is required to run GnuPG on your system.
gmp is required to run GnuPG on your system.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed on the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install GnuPG, and then
one afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > GnuPG1
• And the following one after you install the software:
[root@deep root]# find /* > GnuPG2
• Then use the following command to get a list of what changed:
[root@deep root]# diff GnuPG1 GnuPG2 > GnuPG-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing GnuPG
Below are the required steps that you must make to configure, compile and optimize the GnuPG
software before installing it into your Linux system. First off, we install the program as user 'root'
so as to avoid authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp gnupg-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf gnupg-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-382-320.jpg)
![GnuPG 1
CHAPTER 4
383
Step 2
In order to check that the version of GnuPG, which you are going to install, is an original and
unmodified one, use the commands described below and check the supplied signature. Since we
don’t have GnuPG already installed in the system, we have to verify the MD5 checksum of the
program.
• To verify the MD5 checksum of GnuPG, use the following command:
[root@deep /]# md5sum gnupg-version.tar.gz
This should yield an output similar to this:
d8b36d4dfd213a1a1027b1877acbc897 gnupg-1.0.7.tar.gz
Now check that this checksum is exactly the same as the one published on the GnuPG website at
the following URL: http://www.gnupg.org/download.html
Step 3
Next, move into the newly created GnuPG source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created GnuPG directory use the following command:
[root@deep tmp]# cd gnupg-1.0.7/
• To configure and optimize GnuPG use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--mandir=/usr/share/man
--infodir=/usr/share/info
--enable-shared
--disable-nls
WARNING: Pay special attention to the compile CFLAGS line above. We optimize GnuPG for an
i686 CPU architecture with the parameter “-march=i686”. Please don’t forget to adjust the
CFLAGS line to reflect your own system.
Step 4
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the GnuPG software:
[root@deep gnupg-1.0.7]# make
[root@deep gnupg-1.0.7]# make check
[root@deep gnupg-1.0.7]# cd
[root@deep root]# find /* > GnuPG1
[root@deep root]# cd /var/tmp/gnupg-1.0.7/
[root@deep gnupg-1.0.7]# make install
[root@deep gnupg-1.0.7]# strip /usr/bin/gpg
[root@deep gnupg-1.0.7]# strip /usr/bin/gpgv
[root@deep gnupg-1.0.7]# cd
[root@deep root]# find /* > GnuPG2
[root@deep root]# diff GnuPG1 GnuPG2 > GnuPG-Installed](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-383-320.jpg)
![GnuPG 1
CHAPTER 4
384
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 5
Once the configuration, optimization, compilation, and installation of the GnuPG software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete GnuPG and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf gnupg-version/
[root@deep tmp]# rm -f gnupg-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install GnuPG. It will also remove the GnuPG compressed archive from the /var/tmp/ directory.
Using GnuPG under Linux terminal
Here we show you how to use GnuPG using a terminal to manage GPG keys. The commands
listed below are ones that we use often, but many more exist. Check the manual page gpg (1)
for more information.
Creating a key-pair:
First of all, we must create a new key-pair (public and private) if this is a first use of the GnuPG
software to be able to use its encryption features.
Step 1
The “--gen-key” option of GnuPG is used to generate a new (public and private) key, we have to
use it every time we need to create a new GnuPG key on the system. When we issue this
command for the first time, GnuPG will create the required directory and options file for us.
• To create a new key-pair, use the following command:
[root@deep /]# gpg --gen-key
gpg (GnuPG) 1.0.7; Copyright (C) 2000 Free Software Foundation, Inc.
This program comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions. See the file COPYING for details.
gpg: /root/.gnupg: directory created
gpg: /root/.gnupg/options: new options file created
gpg: you have to start GnuPG again, so it can read the new options file](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-384-320.jpg)
![GnuPG 1
CHAPTER 4
385
Step 2
Once the command has been executed, we have to run it again for a second time to create our
public and private keys, because on first utilization, it just creates the required directory and
options file for us. Therefore, it will now create the keys.
• We start GnuPG again with the same command:
[root@deep /]# gpg --gen-key
gpg (GnuPG) 1.0.7; Copyright (C) 2002 Free Software Foundation, Inc.
This program comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions. See the file COPYING for details.
gpg: keyring `/root/.gnupg/secring.gpg' created
gpg: keyring `/root/.gnupg/pubring.gpg' created
Please select what kind of key you want:
(1) DSA and ElGamal (default)
(2) DSA (sign only)
(4) ElGamal (sign and encrypt)
(5) RSA (sign only)
Your selection? 1
DSA keypair will have 1024 bits.
About to generate a new ELG-E keypair.
minimum keysize is 768 bits
default keysize is 1024 bits
highest suggested keysize is 2048 bits
What keysize do you want? (1024) Press Enter
Requested keysize is 1024 bits
Please specify how long the key should be valid.
0 = key does not expire
<n> = key expires in n days
<n>w = key expires in n weeks
<n>m = key expires in n months
<n>y = key expires in n years
Key is valid for? (0) Press Enter
Key does not expire at all
Is this correct (y/n)? y
You need a User-ID to identify your key; the software constructs the user
id from Real Name, Comment and Email Address in this form:
"Heinrich Heine (Der Dichter) <heinrichh@duesseldorf.de>"
Real name: Gerhard Mourani
Email address: gmourani@openna.com
Comment: Press Enter
You selected this USER-ID:
"Gerhard Mourani <gmourani@openna.com>"
Change (N)ame, (C)omment, (E)mail or (O)kay/(Q)uit? o
You need a Passphrase to protect your secret key.
Enter passphrase: mypassphrase
Repeat passphrase: mypassphrase
We need to generate a lot of random bytes. It is a good idea to perform
some other action (type on the keyboard, move the mouse, utilize the
disks) during the prime generation; this gives the random number
generator a better chance to gain enough entropy.
++++++++++++++++++++.+++++.+++++.+++++++++++++++.+++++++++++++++..+++++++
++++++++...+++++++++++++++.+++++.++++++++++++++++++++++++++++++++++++++++
..+++++](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-385-320.jpg)
![GnuPG 1
CHAPTER 4
386
<+++++..>+++++.................<..+++++..................................
..........+++++^^^^
public and secret key created and signed.
NOTE: A new key-pair is created (secret and public key) in the “root” home directory ~/root
under the .gnupg subdirectory because we issued this GnuPG command as user “root”. If you
run the above command under other user into the system, then the generated keys will be
located under its home directory on the server.
Exporting GPG key/s for a user:
Once your own key-pair is created, you can expand your horizons by exporting and distributing
your public key over the world (NEVER export you private key). This can be done by publishing it
on your homepage, through an available key server on the Internet, or any other available
method. GnuPG has some useful options to help you publish your public key.
Step 1
First off, we have to extract our public key in ASCII text to be able to distribute it. ASCII text is a
good format to use because it allows people to get it easily. In this way, anyone can just cut and
past your public key and use it when they want to securely communicate with you.
• To extract your public key in ASCII armored output, use the following command:
[root@deep /]# gpg --export –ao UID
As an example:
[root@deep /]# gpg --export –ao Gerhard Mourani
Where “--export” is for extracting Public-key from your pubring encrypted file, “a” is to create
ASCII armored output that you can mail, publish or put it on a web page, “o” to put the result in a
file and UID represents the user key you want to export, which is in our example the user
“Gerhard Mourani” key that we have create previously.
Step 2
Once your public key has been extracted, the resulting output will be a file called “Gerhard”
under the directory where you are issuing the above command, representing the First name of
the user key to extract. In our example, the file is called “Gerhard” because it is the name of the
key we want to export in ASCII text format. Note that the file name will be different for your public
ASCII text format key.
• Edit your public key in ASCII armored output format, and distribute it:
[root@deep /]# vi Gerhard
-----BEGIN PGP PUBLIC KEY BLOCK-----
Version: GnuPG v1.0.7 (GNU/Linux)
Comment: For info see http://www.gnupg.org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](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-386-320.jpg)
![GnuPG 1
CHAPTER 4
387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=7bvq
-----END PGP PUBLIC KEY BLOCK-----
WARNING: Never export or distribute your private key to the world. I know, this seem to be a stupid
warning, but I’ve been informed that some people do it.
Importing GPG key/s from a user:
When you receive someone's public key (or some trusted third partly keys) you have to add them
to your key database in order to be able to use his/her keys for future encryption, verification and
authentication. This is often the case, when we install software that has a GPG key available for
verification. Therefore, here is what you should do before installing software that has a GPG key to
your disposal for authenticity.
Step 1
First off, we have to retrieve the GPG public key of the company, organization, etc that we want to
import into our keyring database. In our example, we will retrieve the GPG key that OpenNA uses
to sign RPM packages and other software.
This GPG public key is available from: http://www.openna.com/about/openna.asc. Cut and past it
into a file called “openna.asc” on your server machine where GnuPG is installed.
Step 2
Now, we have to import the OpenNA GPG public key into our database. This procedure should be
done for any GPG public keys that you want to use to verify authenticity of software you want to
install on your server. Most organizations have GPG public keys for you to download.
• To import Public Keys to your keyring database, use the following command:
[root@deep /]# gpg --import filename
As an example:
[root@deep /]# gpg --import openna.asc
gpg: key 3487965A: public key imported
gpg: /root/.gnupg/trustdb.gpg: trustdb created
gpg: Total number processed: 1
gpg: imported: 1
The above command will append the new key “filename” into the keyring database and will
update all already existing keys. It is important to note that GnuPG does not import keys that are
not self-signed (asc).](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-387-320.jpg)
![GnuPG 1
CHAPTER 4
388
Signing GPG key/s from a user:
When you import keys into your public keyring database and are sure that the trusted third party
is really the person they claim, you can start signing his/her keys. Signing a key certifies that you
know the owner of the keys and this leads to the situation where the signature acknowledges that
the user ID mentioned in the key is actually the owner of that key.
• To sign the key for company OpenNA that we have added into our keyring database
above, use the following command:
[root@deep /]# gpg --sign-key UID
As an example:
[root@deep /]# gpg --sign-key OpenNA
pub 1024D/3487965A created: 2001-07-02 expires: never trust: -/q
sub 1024g/0146F594 created: 2001-07-02 expires: never
(1). OpenNA Inc. <noc@openna.com>
pub 1024D/3487965A created: 2001-07-02 expires: never trust: -/q
Fingerprint: 7A3D 6871 2DF1 9210 8ABE AF36 D460 86D5 3487
965A
OpenNA Inc. <noc@openna.com>
Are you really sure that you want to sign this key
with your key: "Gerhard Mourani <gmourani@openna.com>"
Really sign? y
You need a passphrase to unlock the secret key for
user: "Gerhard Mourani <gmourani@openna.com>"
1024-bit DSA key, ID 2E5378F7, created 2002-04-24
Enter passphrase:
WARNING: You should only sign a key as being authentic when you are ABSOLUTELY SURE that
the key is really authentic! You should never sign a key based on any kind of assumption.
Checking GPG signature:
We have shown above how to sign a key, now we will explain how people can verify if the
signature is really the good one. Once you have extracted your public key and exported it,
everyone who knows or gets your public key should be able to check whether encrypted data
from you is also really signed by you.
• To check the signature of encrypted data, use the following command:
[root@deep /]# gpg --verify Data
The “--verify” option will check the signature where Data is the encrypted data/file you want
to verify.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-388-320.jpg)
![GnuPG 1
CHAPTER 4
389
Encrypting and decrypting GPG files:
After installing, importing, signing and configuring everything in the way that we want, we can
start encrypting and decrypting our files, software, etc.
• To encrypt and sign data for the user OpenNA that we have added on our keyring
database above, use the following command:
[root@deep /]# gpg -sear OpenNA file
As an example:
[root@deep /]# gpg -sear OpenNA Message-to-OpenNA.txt
You need a passphrase to unlock the secret key for
user: "Gerhard Mourani <gmourani@openna.com>"
1024-bit DSA key, ID 2E5378F7, created 2002-04-24
Enter passphrase:
Of the arguments passed, the “s” is for signing (To avoid the risk that somebody else claims to be
you, it is very useful to sign everything you encrypt), “e” for encrypting, “a” to create ASCII
armored output (“.asc” ready for sending by mail), “r” to encrypt the UID name and “file” is the
message you want to encrypt.
• To decrypt data, use the following command:
[root@deep /]# gpg -d file
For example:
[root@deep /]# gpg -d Message-from-GerhardMourani.asc
You need a passphrase to unlock the secret key for
user: "Gerhard Mourani (Open Network Architecture) <gmourani@openna.com>"
1024-bit DSA key, ID 2E5378F7, created 2002-04-24
Enter passphrase:
Where “d” is for decrypting and “file” is the message you want to decrypt. It is important that
the public key of the sender of the message we want to decrypt be in our public keyring database
or of course nothing will work.
Further documentation
For more details, there are some manual pages about GnuPG that you could read:
$ man gpg (1) - GPG encryption and signing tool.
$ man gpgv (1) - GPGV signature verification tool.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-389-320.jpg)
![OpenSSL 1
CHAPTER 5
395
These installation instructions assume
Commands are Unix-compatible.
The source path is /var/tmp (note that other paths are possible, at personal discretion).
Installations were tested on OpenNA Linux & Red Hat Linux 7.3.
All steps in the installation will happen using the super-user account “root”.
Whether kernel recompilation may be required: No
Latest OpenSSL version number is 0.9.6d
The procedures given in this chapter are likely to work on all Linux platforms, but we have only
tested it on OpenNA Linux and Red Hat Linux.
Packages
The following is based on information listed by OpenSSL as of 2002/05/09. Please check
http://www.openssl.org/ regularly for the latest status. We chose to install from source because it
provides the facility to fine tune the installation.
Source code is available from:
OpenSSL Homepage: http://www.openssl.org/
OpenSSL FTP Site: 129.132.7.170
You must be sure to download: openssl-0.9.6d.tar.gz
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install OpenSSL, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > OpenSSL1
• And the following one after you install the software:
[root@deep root]# find /* > OpenSSL2
• Then use the following command to get a list of what changed:
[root@deep root]# diff OpenSSL1 OpenSSL2 > OpenSSL-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-395-320.jpg)
![OpenSSL 1
CHAPTER 5
396
Compiling - Optimizing & Installing OpenSSL
Below are the steps that you must make to configure, compile and optimize the OpenSSL
software before installing it into your Linux system. First off, we install the program as user 'root'
so as to avoid authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp openssl-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf openssl-version.tar.gz
Step 2
Next, move into the newly created OpenSSL source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created OpenSSL directory use the following command:
[root@deep tmp]# cd openssl-0.9.6d/
Step 3
With OpenSSL, the optimization FLAGS should be changed in the “Configure” file of the
program. It is in this file that we define the GCC optimizations we want to use related to the type of
processor running in our system. OpenSSL is cryptography software and there are some
optimization hacks that we can make that can significantly increase the performance of the
program, therefore take the time to modify the “Configure” file of the software. This will be a
benefit for you.
a) Edit the Configure file (vi +337 Configure) and change the following lines:
"linux-elf", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -m486 -Wall::-
D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des}
${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-
fPIC:.so.$(SHLIB_MAJOR).$(SHLIB_MINOR)",
To read:
"linux-elf", "gcc:-DL_ENDIAN -DTERMIO -O3 -march=i686 -funroll-loops -fomit-
frame-pointer -Wall::-D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des}
${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-
fPIC:.so.$(SHLIB_MAJOR).$(SHLIB_MINOR)",
b) Edit the Configure file (vi +338 Configure) and change the following lines:
"debug-linux-elf","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -
DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -m486 -Wall::-D_REENTRANT:-lefence -
ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn",
To read:
"debug-linux-elf","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -
DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -O3 -march=i686 -funroll-loops -fomit-frame-
pointer -Wall::-D_REENTRANT:-lefence -ldl:BN_LLONG ${x86_gcc_des}
${x86_gcc_opts}:${x86_elf_asm}:dlfcn",](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-396-320.jpg)
![OpenSSL 1
CHAPTER 5
397
c) Edit the Configure file (vi +339 Configure) and change the following lines:
"debug-linux-elf-noefence","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG
-DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -m486 -Wall::-D_REENTRANT:-ldl:BN_LLONG
${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn",
To read:
"debug-linux-elf-noefence","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG
-DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -O3 -march=i686 -funroll-loops -fomit-frame-
pointer -Wall::-D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des}
${x86_gcc_opts}:${x86_elf_asm}:dlfcn",
Step 4
By default, OpenSSL source files assume that our “perl” binary program is located under
/usr/local/bin/perl. We must change this to reflect our environment variable.
• To point all OpenSSL script files to our “perl” binary, use the following command:
[root@deep openssl-0.9.6d]# perl util/perlpath.pl /usr/bin/perl
Step 5
At this stage, it is time to configure and compile OpenSSL for our system.
• To configure and optimize OpenSSL use the following compilation lines:
./Configure linux-elf no-asm shared
--prefix=/usr
--openssldir=/usr/share/ssl
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the OpenSSL software:
[root@deep openssl-0.9.6d]# LD_LIBRARY_PATH=`pwd` make all build-shared
[root@deep openssl-0.9.6d]# LD_LIBRARY_PATH=`pwd` make test apps tests
[root@deep openssl-0.9.6d]# cd
[root@deep root]# find /* > OpenSSL1
[root@deep root]# cd /var/tmp/openssl-0.9.6d/
[root@deep openssl-0.9.6d]# make install build-shared
[root@deep openssl-0.9.6d]# cd /usr/lib/
[root@deep lib]# mv libcrypto.so.0.9.6 ../../lib/
[root@deep lib]# mv libssl.so.0.9.6 ../../lib/
[root@deep lib]# ln -sf ../../lib/libcrypto.so.0.9.6 libcrypto.so
[root@deep lib]# ln -sf ../../lib/libcrypto.so.0.9.6 libcrypto.so.0
[root@deep lib]# ln -sf ../../lib/libssl.so.0.9.6 libssl.so
[root@deep lib]# ln -sf ../../lib/libssl.so.0.9.6 libssl.so.0
[root@deep lib]# mv /usr/share/ssl/man/man1/* /usr/share/man/man1/
[root@deep lib]# mv /usr/share/ssl/man/man3/* /usr/share/man/man3/
[root@deep lib]# mv /usr/share/ssl/man/man5/* /usr/share/man/man5/
[root@deep lib]# mv /usr/share/ssl/man/man7/* /usr/share/man/man7/
[root@deep lib]# rm -rf /usr/share/ssl/man/
[root@deep lib]# rm -rf /usr/share/ssl/lib/
[root@deep lib]# strip /usr/bin/openssl
[root@deep lib]# mkdir -p /usr/share/ssl/crl
[root@deep lib]# cd
[root@deep root]# find /* > OpenSSL2
[root@deep root]# diff OpenSSL1 OpenSSL2 > OpenSSL-Installed](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-397-320.jpg)
![OpenSSL 1
CHAPTER 5
398
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then test the OpenSSL libraries to finally install the binaries and any supporting files into the
appropriate locations.
Step 7
Once the configuration, optimization, compilation, and installation of the OpenSSL software has
completed, we can free up some disk space by deleting the program tar archive and the related
source directory since they are no longer needed.
• To delete OpenSSL and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf openssl-version/
[root@deep tmp]# rm -f openssl-version.tar.gz
Configuring OpenSSL
After OpenSSL has been built and installed successfully on your system, your next step is to
configure and customize the openssl.cnf and sign files to suit your needs.
/usr/shared/ssl/openssl.cnf (The OpenSSL Configuration File)
/usr/shared/ssl/misc/sign (A CA script file to sign certificates)
/usr/shared/ssl/openssl.cnf: The OpenSSL Configuration File
This is the general configuration file for OpenSSL, where you can configure the expiration date of
your keys, the name of your organization, address and so on.
The most important parameters you may need to change will be in the [ CA_default ] and
especially the [ req_distinguished_name ] sections of the file. We must change the
default one to fit our requirements and operating system. The text in bold is the parts of the
configuration file that must be customized and adjusted to satisfy our needs.
• Edit the openssl.cnf file (vi /usr/share/ssl/openssl.cnf) and set your needs.
#
# OpenSSL example configuration file.
# This is mostly being used for generation of certificate requests.
#
# This definition stops the following lines choking if HOME isn't
# defined.
HOME = .
RANDFILE = $ENV::HOME/.rnd
# Extra OBJECT IDENTIFIER info:
#oid_file = $ENV::HOME/.oid
oid_section = new_oids
# To use this configuration file with the "-extfile" option of the
# "openssl x509" utility, name here the section containing the
# X.509v3 extensions to use:
# extensions =
# (Alternatively, use a configuration file that has only
# X.509v3 extensions in its main [= default] section.)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-398-320.jpg)
![OpenSSL 1
CHAPTER 5
399
[ new_oids ]
# We can add new OIDs in here for use by 'ca' and 'req'.
# Add a simple OID like this:
# testoid1=1.2.3.4
# Or use config file substitution like this:
# testoid2=${testoid1}.5.6
####################################################################
[ ca ]
default_ca = CA_default # The default ca section
####################################################################
[ CA_default ]
dir = /usr/share/ssl # Where everything is kept
certs = $dir/certs # Where the issued certs are kept
crl_dir = $dir/crl # Where the issued crl are kept
database = $dir/ca.db.index # database index file.
new_certs_dir = $dir/ca.db.certs # default place for new certs.
certificate = $dir/certs/ca.crt # The CA certificate
serial = $dir/ca.db.serial # The current serial number
crl = $dir/crl.pem # The current CRL
private_key = $dir/private/ca.key # The private key
RANDFILE = $dir/ca.db.rand # private random number file
x509_extensions = usr_cert # The extentions to add to the cert
# Extensions to add to a CRL. Note: Netscape communicator chokes on V2 CRLs
# so this is commented out by default to leave a V1 CRL.
# crl_extensions = crl_ext
default_days = 365 # how long to certify for
default_crl_days= 30 # how long before next CRL
default_md = md5 # which md to use.
preserve = no # keep passed DN ordering
# A few difference way of specifying how similar the request should look
# For type CA, the listed attributes must be the same, and the optional
# and supplied fields are just that :-)
policy = policy_match
# For the CA policy
[ policy_match ]
countryName = match
stateOrProvinceName = match
organizationName = match
organizationalUnitName = optional
commonName = supplied
emailAddress = optional
# For the 'anything' policy
# At this point in time, you must list all acceptable 'object'
# types.
[ policy_anything ]
countryName = optional
stateOrProvinceName = optional
localityName = optional
organizationName = optional
organizationalUnitName = optional
commonName = supplied
emailAddress = optional](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-399-320.jpg)
![OpenSSL 1
CHAPTER 5
400
####################################################################
[ req ]
default_bits = 1024
default_keyfile = privkey.pem
distinguished_name = req_distinguished_name
attributes = req_attributes
x509_extensions = v3_ca # The extentions to add to the self signed cert
# Passwords for private keys if not present they will be prompted for
# input_password = secret
# output_password = secret
# This sets a mask for permitted string types. There are several options.
# default: PrintableString, T61String, BMPString.
# pkix : PrintableString, BMPString.
# utf8only: only UTF8Strings.
# nombstr : PrintableString, T61String (no BMPStrings or UTF8Strings).
# MASK:XXXX a literal mask value.
# WARNING: current versions of Netscape crash on BMPStrings or UTF8Strings
# so use this option with caution!
string_mask = nombstr
# req_extensions = v3_req # The extensions to add to a certificate request
[ req_distinguished_name ]
countryName = Country Name (2 letter code)
countryName_default = CA
countryName_min = 2
countryName_max = 2
stateOrProvinceName = State or Province Name (full name)
stateOrProvinceName_default = Quebec
localityName = Locality Name (eg, city)
localityName_default = Montreal
0.organizationName = Organization Name (eg, company)
0.organizationName_default = OpenNA, Inc.
# we can do this but it is not needed normally :-)
#1.organizationName = Second Organization Name (eg, company)
#1.organizationName_default = World Wide Web Pty Ltd
organizationalUnitName = Organizational Unit Name (eg, section)
organizationalUnitName_default = Open Network Architecture
commonName = Common Name (eg, YOUR name)
commonName_default = www.openna.com
commonName_max = 64
emailAddress = Email Address
emailAddress_default = noc@openna.com
emailAddress_max = 40
# SET-ex3 = SET extension number 3
[ req_attributes ]
challengePassword = A challenge password
challengePassword_min = 8
challengePassword_max = 20
unstructuredName = An optional company name](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-400-320.jpg)
![OpenSSL 1
CHAPTER 5
401
[ usr_cert ]
# These extensions are added when 'ca' signs a request.
# This goes against PKIX guidelines but some CAs do it and some software
# requires this to avoid interpreting an end user certificate as a CA.
basicConstraints=CA:FALSE
# Here are some examples of the usage of nsCertType. If it is omitted
# the certificate can be used for anything *except* object signing.
# This is OK for an SSL server.
# nsCertType = server
# For an object signing certificate this would be used.
# nsCertType = objsign
# For normal client use this is typical
# nsCertType = client, email
# and for everything including object signing:
# nsCertType = client, email, objsign
# This is typical in keyUsage for a client certificate.
# keyUsage = nonRepudiation, digitalSignature, keyEncipherment
# This will be displayed in Netscape's comment listbox.
nsComment = "OpenSSL Generated Certificate"
# PKIX recommendations harmless if included in all certificates.
subjectKeyIdentifier=hash
authorityKeyIdentifier=keyid,issuer:always
# This stuff is for subjectAltName and issuerAltname.
# Import the email address.
# subjectAltName=email:copy
# Copy subject details
# issuerAltName=issuer:copy
#nsCaRevocationUrl = http://www.domain.dom/ca-crl.pem
#nsBaseUrl
#nsRevocationUrl
#nsRenewalUrl
#nsCaPolicyUrl
#nsSslServerName
[ v3_req ]
# Extensions to add to a certificate request
basicConstraints = CA:FALSE
keyUsage = nonRepudiation, digitalSignature, keyEncipherment
[ v3_ca ]
# Extensions for a typical CA
# PKIX recommendation.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-401-320.jpg)
![OpenSSL 1
CHAPTER 5
402
subjectKeyIdentifier=hash
authorityKeyIdentifier=keyid:always,issuer:always
# This is what PKIX recommends but some broken software chokes on critical
# extensions.
#basicConstraints = critical,CA:true
# So we do this instead.
basicConstraints = CA:true
# Key usage: this is typical for a CA certificate. However since it will
# prevent it being used as an test self-signed certificate it is best
# left out by default.
# keyUsage = cRLSign, keyCertSign
# Some might want this also
# nsCertType = sslCA, emailCA
# Include email address in subject alt name: another PKIX recommendation
# subjectAltName=email:copy
# Copy issuer details
# issuerAltName=issuer:copy
# DER hex encoding of an extension: beware experts only!
# obj=DER:02:03
# Where 'obj' is a standard or added object
# You can even override a supported extension:
# basicConstraints= critical, DER:30:03:01:01:FF
[ crl_ext ]
# CRL extensions.
# Only issuerAltName and authorityKeyIdentifier make any sense in a CRL.
# issuerAltName=issuer:copy
authorityKeyIdentifier=keyid:always,issuer:always
WARNING: You don’t need to change all of the default options set in the file openssl.cnf; The
configurations you usually change will be into the [ CA_default ] and [
req_distinguished_name ] sections of the file.
/usr/share/ssl/misc/sign: The CA Script File to Sign Certificates
OpenSSL CA command has some strange requirements and the default OpenSSL config doesn't
allow one to easily use OpenSSL CA directly. It is for this reason that we don’t use the files CA.pl
or CA.sh to sign certificates.
Step 1
To solve the problem, we’ll create and customize the sign script file below to replace them. Text
in bold are the parts of the script that must be customized and adjusted to satisfy our needs.
• Create the sign script file (touch /usr/share/ssl/misc/sign) and add the
following lines:
#!/bin/sh
##](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-402-320.jpg)
![OpenSSL 1
CHAPTER 5
403
## sign.sh -- Sign a SSL Certificate Request (CSR)
## Copyright (c) 1998-1999 Ralf S. Engelschall, All Rights Reserved.
##
# argument line handling
CSR=$1
if [ $# -ne 1 ]; then
echo "Usage: sign.sign <whatever>.csr"; exit 1
fi
if [ ! -f $CSR ]; then
echo "CSR not found: $CSR"; exit 1
fi
case $CSR in
*.csr ) CERT="`echo $CSR | sed -e 's/.csr/.crt/'`" ;;
* ) CERT="$CSR.crt" ;;
esac
# make sure environment exists
if [ ! -d ca.db.certs ]; then
mkdir ca.db.certs
fi
if [ ! -f ca.db.serial ]; then
echo '01' >ca.db.serial
fi
if [ ! -f ca.db.index ]; then
cp /dev/null ca.db.index
fi
# create an own SSLeay config
cat >ca.config <<EOT
[ ca ]
default_ca = CA_own
[ CA_own ]
dir = /usr/share/ssl
certs = /usr/share/ssl/certs
new_certs_dir = /usr/share/ssl/ca.db.certs
database = /usr/share/ssl/ca.db.index
serial = /usr/share/ssl/ca.db.serial
RANDFILE = /usr/share/ssl/ca.db.rand
certificate = /usr/share/ssl/certs/ca.crt
private_key = /usr/share/ssl/private/ca.key
default_days = 365
default_crl_days = 30
default_md = md5
preserve = no
policy = policy_anything
[ policy_anything ]
countryName = optional
stateOrProvinceName = optional
localityName = optional
organizationName = optional
organizationalUnitName = optional
commonName = supplied
emailAddress = optional
EOT
# sign the certificate
echo "CA signing: $CSR -> $CERT:"
openssl ca -config ca.config -out $CERT -infiles $CSR
echo "CA verifying: $CERT <-> CA cert"
openssl verify -CAfile /usr/share/ssl/certs/ca.crt $CERT
# cleanup after SSLeay](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-403-320.jpg)
![OpenSSL 1
CHAPTER 5
404
rm -f ca.config
rm -f ca.db.serial.old
rm -f ca.db.index.old
# die gracefully
exit 0
Step 2
Once the script file has been created, it is important to make it executable and change its default
permissions. Making this file executable will allow the system to run it, changing its default
permission is to allow only the root user to change this file for security reason.
• To make this script executable and to change its default permissions, use the commands:
[root@deep /]# chmod 700 /usr/share/ssl/misc/sign
[root@deep /]# chown 0.0 /usr/share/ssl/misc/sign
OpenSSL Administrative Tools
Once your configuration options have been set in the openssl.cnf file, we can play with the
OpenSSL utility. As an example, we’ll show you how to create certificates for the Apache web
server and your own CA (Certifying Authority) to sign your “Certificate Signing Request” yourself.
All commands listed below are to be made in the /usr/share/ssl directory.
The Apache Key & CSR Generation:
The utility “openssl” that you use to generate RSA Private Keys (Key) and Certificate Signing
Requests (CSR) comes with OpenSSL and is usually installed under the directory /usr/bin on
our Linux distribution. Below are the steps to create certificates for Apache.
Step 1
First you have to know the Fully Qualified Domain Name (FQDN) of the website/server for which
you want to request a certificate. When you want to access your website/server through
https://www.mydomain.com/ then the FQDN of your website is www.mydomain.com.
Step 2
Second, select five large and relatively random files from your hard drive (compressed log files
are a good start) and put them under your /usr/share/ssl directory. These will act as your
random seed enhancers. We refer to them as random1: random2:...: random5 below.
• To select five random files and put them under /usr/share/ssl, use the commands:
[root@deep /]# cp /var/log/boot.log /usr/share/ssl/random1
[root@deep /]# cp /var/log/cron /usr/share/ssl/random2
[root@deep /]# cp /var/log/dmesg /usr/share/ssl/random3
[root@deep /]# cp /var/log/messages /usr/share/ssl/random4
[root@deep /]# cp /var/log/secure /usr/share/ssl/random5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-404-320.jpg)
![OpenSSL 1
CHAPTER 5
405
Step 3
Third, create the RSA private key protected with a pass-phrase for your web server. The
command below will generate 1024 bit RSA Private Key and store it in
www.mydomain.com.key.
It will ask you for a pass-phrase: use something secure and remember it. Your certificate will be
useless without the key. If you don't want to protect your key with a pass-phrase (only if you
absolutely trust that server machine, and you make sure the permissions are carefully set so only
you can read that key) you can leave out the -des3 option below.
• To generate the Key, use the following command:
[root@deep /]# cd /usr/share/ssl/
[root@deep ssl]# openssl genrsa -des3 –rand
random1:random2:random3:random4:random5 -out www.mydomain.com.key 1024
123600 semi-random bytes loaded
Generating RSA private key, 1024 bit long modulus
......................+++++
.....+++++
e is 65537 (0x10001)
Enter PEM pass phrase:
Verifying password - Enter PEM pass phrase:
WARNING: Please backup your www.mydomain.com.key file and remember the pass-phrase you
had to enter at a secure location. A good choice is to backup this information onto a diskette or
other removable media.
Step 4
Finally, generate a Certificate Signing Request (CSR) with the server RSA private key. The
command below will prompt you for the X.509 attributes of your certificate. Remember to give
the name “www.mydomain.com” when prompted for ‘Common Name'. Do not enter your personal
name here. We are requesting a certificate for a web server, so the ‘Common Name’ has to match
the FQDN of your website (a requirement of the browsers).
• To generate the CSR, use the following command:
[root@deep ssl]# openssl req -new -key www.mydomain.com.key -out
www.mydomain.com.csr
Using configuration from /usr/share/ssl/openssl.cnf
Enter PEM pass phrase:
You are about to be asked to enter information that will be incorporated
into your certificate request.
What you are about to enter is what is called a Distinguished Name or a
DN.
There are quite a few fields but you can leave some blank
For some fields there will be a default value,
If you enter '.', the field will be left blank.
-----
Country Name (2 letter code) [CA]:
State or Province Name (full name) [Quebec]:
Locality Name (eg, city) [Montreal]:
Organization Name (eg, company) [OpenNA, Inc.]:
Organizational Unit Name (eg, section) [Open Network Architecture]:
Common Name (eg, YOUR name) [www.openna.com]:
Email Address [noc@openna.com]:
Please enter the following 'extra' attributes](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-405-320.jpg)
![OpenSSL 1
CHAPTER 5
406
to be sent with your certificate request
A challenge password []:.
An optional company name []:.
WARNING: Make sure you enter the FQDN (Fully Qualified Domain Name) of the server when
OpenSSL prompts you for the “Common Name” (i.e. when you generate a CSR for a website which
will be later accessed via https://www.mydomain.com/, enter “www.mydomain.com” here).
After the generation of your Certificate Signing Request (CSR), you must send this certificate to a
commercial Certifying Authority (CA) like Thawte or Verisign for signing. You usually have to post
the CSR into a web form, pay for the signing, await the signed certificate and store it into a
“www.mydomain.com.crt” file. The result is then a real certificate, which can be used with
Apache.
The CA Key & CRT Generation:
If you don’t want to pay a commercial Certifying Authority (CA) to sign you certificates, you can
use your own CA and now have to sign the CSR yourself by this CA. This solution is economical,
and allows an organization to host their own CA server and generate as many certificates as they
need for internal use without paying a cent to a commercial CA.
Unfortunately using your own CA to generate certificates causes problems in electronic
commerce, because customers need to have some trust in your organization by the use of a
recognized commercial CA. See below on how to sign a CSR with your CA yourself.
Step 1
As for the Apache web server above, the first step is to create the RSA private key protected with
a pass-phrase for your CA. The command below will generate 1024 bit RSA Private Key and
stores it in the file “ca.key”. It will ask you for a pass-phrase: use something secure and
remember it. Your certificate will be useless without the key.
• To create the RSA private key for your (CA), use the following command:
[root@deep /]# cd /usr/share/ssl/
[root@deep ssl]# openssl genrsa -des3 -out ca.key 1024
Generating RSA private key, 1024 bit long modulus
...........................+++++
............................................+++++
e is 65537 (0x10001)
Enter PEM pass phrase:
Verifying password - Enter PEM pass phrase:
WARNING: Please backup your “ca.key” file and remember the pass-phrase you had to enter at a
secure location. A good choice is to backup this information onto a diskette or other removable
media.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-406-320.jpg)
![OpenSSL 1
CHAPTER 5
407
Step 2
Now, we must create a self-signed (CA) certificate (x509 structure) with the RSA key of the CA.
The req command creates a self-signed certificate when the -x509 switch is used.
• To create a self-signed (CA) certificate, use the following command:
[root@deep ssl]# openssl req -new -x509 -days 365 -key ca.key -out ca.crt
Using configuration from /usr/share/ssl/openssl.cnf
Enter PEM pass phrase:
You are about to be asked to enter information that will be incorporated
into your certificate request.
What you are about to enter is what is called a Distinguished Name or a
DN.
There are quite a few fields but you can leave some blank
For some fields there will be a default value,
If you enter '.', the field will be left blank.
-----
Country Name (2 letter code) [CA]:
State or Province Name (full name) [Quebec]:
Locality Name (eg, city) [Montreal]:
Organization Name (eg, company) [OpenNA, Inc.]:
Organizational Unit Name (eg, section) [Open Network]:Sales Dept
Common Name (eg, YOUR name) [www.openna.com]:
Email Address [noc@openna.com]:sales@openna.com
Step 3
Once the self-signed (CA) certificate has been created, we must place all certificates and CA files
into their appropriate directories.
• To place the files into their appropriate directories, use the following commands:
[root@deep ssl]# mv www.mydomain.com.key private/
[root@deep ssl]# mv ca.key private/
[root@deep ssl]# mv ca.crt certs/
Step 4
Finally, you can use this CA to sign all the servers CSR's in order to create real SSL Certificates
for use inside the web server (assuming you already have a www.mydomain.com.csr at hand).
We must also prepare the script “sign” for signing.
• To sign server CSR's in order to create real SSL Certificates, use the following command:
[root@deep ssl]# /usr/share/ssl/misc/sign www.mydomain.com.csr
CA signing: www.mydomain.com.csr -> www.mydomain.com.crt:
Using configuration from ca.config
Enter PEM pass phrase:
Check that the request matches the signature
Signature ok
The Subjects Distinguished Name is as follows
countryName :PRINTABLE:'CA'
stateOrProvinceName :PRINTABLE:'Quebec'
localityName :PRINTABLE:'Montreal'
organizationName :PRINTABLE:'OpenNA, Inc.'
organizationalUnitName :PRINTABLE:'Open Network Architecture'
commonName :PRINTABLE:'www.openna.com'
emailAddress :IA5STRING:'noc@openna.com'
Certificate is to be certified until Oct 18 14:59:29 2001 GMT (365 days)
Sign the certificate? [y/n]:y](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-407-320.jpg)
![OpenSSL 1
CHAPTER 5
408
1 out of 1 certificate requests certified, commit? [y/n]y
Write out database with 1 new entries
Data Base Updated
CA verifying: www.mydomain.com.crt <-> CA cert
www.mydomain.com.crt: OK
This signs the CSR and results in a “www.mydomain.com.crt” file. Move this file to its
appropriate directory as follows.
• To move the CRT file to its appropriate directory, use the following command:
[root@deep ssl]# mv www.mydomain.com.crt certs/
Now you have two files: “www.mydomain.com.key” and “www.mydomain.com.crt”. These
can now, for example, be used as follows, inside the virtual host section of your Apache server's
httpd.conf file:
SSLCertificateFile /usr/share/ssl/certs/www.mydomain.com.crt
SSLCertificateKeyFile /usr/share/ssl/private/www.mydomain.com.key
In this example, www.mydomain.com.crt is our web server Certificate Signing Request Public
Key, and www.mydomain.com.key is our web server RSA Private Key.
The www.mydomain.com.csr file is no longer needed; we can remove it from the system.
• To remove this file from the system, use the following command:
[root@deep ssl]# rm -f www.mydomain.com.csr
WARNING: If you receive an error message during the signing of the certificate, it’s probably
because you’ve entered the wrong FQDN (Fully Qualified Domain Name) for the server when
OpenSSL prompted you for the “Common Name”; the “Common Name” must be something like
“www.mydomain.com” and not “mydomain.com”. Also, since you generate both the certificate
and the CA certificate, it’s important that at least ONE piece of information differs between both
files, or you may encounter problems during the signature of the certificate request.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-408-320.jpg)
![OpenSSL 1
CHAPTER 5
409
Securing OpenSSL
This small section deals specifically with actions we can take to improve and tighten security
under OpenSSL. It’s important to note that we refer to the features available within the base
installed program and not to any additional software.
Changing the default mode of OpenSSL keys:
Make your keys “Read and Write” only by the super-user “root”. This is important because no
one needs to touch these files.
• To make your keys “read and Write” only by “root”, use the following commands:
[root@deep /]# chmod 750 /usr/share/ssl/private/
[root@deep /]# chmod 400 /usr/share/ssl/certs/ca.crt
[root@deep /]# chmod 400 /usr/share/ssl/certs/www.mydomain.com.crt
[root@deep /]# chmod 400 /usr/share/ssl/private/ca.key
[root@deep /]# chmod 400 /usr/share/ssl/private/www.mydomain.com.key
Some possible uses of OpenSSL software
OpenSSL can be used to:
1. Creation of your own Certifying Authority Server.
2. Creation of RSA, DH and DSA key parameters.
3. Creation of X.509 certificates, CSRs and CRLs.
4. Calculation of Message Digest.
5. Encryption and Descryptiion with Ciphers.
6. SSL/TLS Client and Server Tests.
7. Handling of S/MIME signed or encrypted mail.
8. Provide data confidentiality, integrity, authentication, and electronic signature in
transmission for the users.
9. Secure electronic commerce transactions.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-409-320.jpg)
![OpenSSH 1
CHAPTER 6
414
Prerequisites
OpenSSH requires that the listed software below be already installed on your system to be able to
compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source
archive files. Please make sure you have this program installed on your machine before you
proceed with this chapter.
OpenSSL is required to run OpenSSH on your system.
NOTE: For more information on OpenSSL software, see its related chapter in this book. Even if
you don’t need to use OpenSSL software to create or hold encrypted key files, it’s important to
note that OpenSSH requires its libraries files to be able to work.
Pristine source
As we don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install OpenSSH, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > OpenSSH1
• And the following one after you install the software:
[root@deep root]# find /* > OpenSSH2
• Then use the following command to get a list of what changed:
[root@deep root]# diff OpenSSH1 OpenSSH2 > OpenSSH-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing OpenSSH
Below are the steps that you must make to configure, compile and optimize the OpenSSH server
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp openssh-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf openssh-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-414-320.jpg)
![OpenSSH 1
CHAPTER 6
415
Step 2
In order to check that the version of OpenSSH, which you are, going to install, is an original and
unmodified one, please check the supplied signature with the GPG key of OpenSSH available on
the OpenSSH website.
To get a GPG key copy of OpenSSH, please point your browser to the following URL:
ftp://ftp.openbsd.org/pub/OpenBSD/OpenSSH/portable/openssh-3.4p1.tar.gz.sig. For more
information about how to use this key for verification, see the GnuPG chapter in this book.
Step 3
OpenSSH needs a UID and GID to properly run on the system but this UID/GID cannot run as
super-user root; for this reason we must create a special user with no shell privileges on the
system for running sshd daemon. This is required by the privilege separation feature of OpenSSH
by which operations that require root privilege are performed by a separate privileged monitor
process.
• To create this special OpenSSH user on OpenNA Linux, use the following command:
[root@deep tmp]# groupadd -g 39 sshd > /dev/null 2>&1 || :
[root@deep tmp]# useradd -c "SSH Server" -d /var/empty -g 39 -s
/bin/false -u 39 sshd > /dev/null 2>&1 || :
• To create this special OpenSSH user on Red Hat Linux, use the following command:
[root@deep tmp]# groupadd -g 39 sshd > /dev/null 2>&1 || :
[root@deep tmp]# useradd -u 39 -g 39 -s /bin/false -M -r -d /var/empty
sshd > /dev/null 2>&1 || :
The above command will create a null account, with no password, no valid shell, no files owned-
nothing but a UID and a GID for the program. Remember that OpenSSH daemon does not need
to have a shell account on the server.
Step 4
Now, edit the shells file (vi /etc/shells) and add a non-existent shell name
“/bin/false”, which is the one we used in the useradd command above.
[root@deep tmp]# vi /etc/shells
/bin/bash2
/bin/bash
/bin/sh
/bin/false This is our added no-existent shell
Patching OpenSSH to run in chroot jail mode for some users:
There is an external patch available for OpenSSH that allow us to compile OpenSSH with chroot
jail support. If you are interested in compiling OpenSSH to run in chroot jail environment for some
of your users, then I recommend you to follow these steps. If you don’t want to compile OpenSSH
with chroot jail support, you can simply skip these steps and go directly to next section where we
will compile the software for our system.
For OpenSSH to run and work in chroot jail mode, you have to be sure that you have recompiled
your Linux kernel without the Grsecurity option that allows us to enable chroot jail restrictions
protection on the system. You should be sure that “Chroot jail restrictions
(CONFIG_GRKERNSEC_CHROOT) [N/y/?]” is NOT enable or nothing will work.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-415-320.jpg)
![OpenSSH 1
CHAPTER 6
416
Step 1
First of, we have to retrieve the OpenSSH chroot patch available on the Internet. This patch can
be downloaded from the following location: http://chrootssh.sourceforge.net/
Step 2
Once you have a copy of this patch, you should move it under the /var/tmp directory and patch
your OpenSSH source files.
• This can be done with the following commands:
[root@deep /]# mv osshChroot-3.4.diff /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# patch -p0 < osshChroot-3.4.diff
NOTE: It’s important to note that the version number of the OpenSSH chroot patch that you have to
download from the Internet must match the version number of the OpenSSH software you
intended to install. For example, if the version number of OpenSSH is 3.4p1, you should
download the newer OpenSSH chroot patch that matches this number.
Step 3
After that, move into the newly created OpenSSH source directory and perform the following steps
to configure and optimize the software for your system.
• To move into the newly created OpenSSH directory use the following command:
[root@deep tmp]# cd openssh-3.4p1/
• To configure and optimize OpenSSH use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--sysconfdir=/etc/ssh
--libexecdir=/usr/libexec/openssh
--mandir=/usr/share/man
--with-pam
--with-ipaddr-display
--with-ipv4-default
--with-md5-passwords
--with-zlib
This tells OpenSSH to set itself up for this particular configuration setup with:
- Enable PAM support.
- Use the IP address instead of hostname.
- Use IPv4 by connections.
- Enable use of MD5 passwords.
- Use zlib for transport compression.
NOTE: Pay special attention to the compile CFLAGS line above. We optimize OpenSSH for an i686
CPU architecture with the parameter “-march=i686”. Please don’t forget to adjust this CFLAGS
line to reflect your own system and architecture.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-416-320.jpg)
![OpenSSH 1
CHAPTER 6
417
Step 4
Now, we must make a list of all existing files on the system before installing the software and one
afterwards then compare them using the diff utility to find out what files are placed where and
finally install the OpenSSH Server:
[root@deep openssh-3.4p1]# make
[root@deep openssh-3.4p1]# cd
[root@deep root]# find /* > OpenSSH1
[root@deep root]# cd /var/tmp/openssh-3.4p1/
[root@deep openssh-3.4p1]# make install
[root@deep openssh-3.4p1]# mkdir /var/empty
[root@deep openssh-3.4p1]# chown root.sys /var/empty
[root@deep openssh-3.4p1]# cd
[root@deep root]# find /* > OpenSSH2
[root@deep root]# diff OpenSSH1 OpenSSH2 > OpenSSH-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 5
Once the configuration, optimization, compilation, and installation of the OpenSSH software has
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete OpenSSH and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf openssh-version/
[root@deep tmp]# rm -f openssh-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install OpenSSH. It will also remove the OpenSSH compressed archive from the /var/tmp
directory.
Configuring OpenSSH
After OpenSSH has been built and installed successfully in your system, your next step is to
configure and customize its configuration files to fit your needs.
/etc/ssh/sshd_config: (The OpenSSH Server Configuration File)
/etc/ssh/ssh_config: (The OpenSSH Client Configuration File)
/etc/pam.d/sshd: (The OpenSSH PAM Support Configuration File)
/etc/init.d/sshd: (The OpenSSH Initialization File)
/etc/ssh/sshd_config: The OpenSSH Server Configuration File
The sshd_config file is the system-wide server configuration file for OpenSSH which allows you
to set options that modify the operation of the sshd daemon. This file contains keyword-value
pairs, one per line, with keywords being case insensitive.
Here are the most important keywords to configure your sshd server for maximum security; a
complete listing and/or special requirements are available in the manual page for sshd (8). We
must change the default one to fit our requirements and operating system. The text in bold are
the parts of the configuration file that must be customized and adjusted to satisfy our needs.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-417-320.jpg)
![OpenSSH 1
CHAPTER 6
423
IdentityFile ~/.ssh/identity
IdentityFile ~/.ssh/id_dsa
IdentityFile ~/.ssh/id_rsa
These options specify alternate multiple authentication identity files to read.
Port 22
This option “Port” specifies on which port number ssh connects to on the remote host. The
default port is 22.
Protocol 2,1
This option “Protocol” specifies the protocol versions ssh should support in order of
preference. In our configuration the default is “2,1”. This means that ssh tries version 2 and falls
back to version 1 if version 2 is not available. Depending of the ssh client version you use to
connect, you may need to invert this order but you can connect with ssh client version 1 even if
the order is “2,1”.
Cipher blowfish
This option “Cipher” specifies what cipher should be used for encrypting sessions. The
blowfish use 64-bit blocks and keys of up to 448 bits.
EscapeChar ~
This option “EscapeChar” specifies the session escape character for suspension.
/etc/pam.d/sshd: The OpenSSH PAM Support Configuration File
For increased security of OpenSSH, we have compiled it to use the PAM mechanism for password
authentication.
Step 1
To be able to use this feature, we must create the /etc/pam.d/sshd file and add the following
parameters inside it.
• Create the sshd file (touch /etc/pam.d/sshd) and add the following lines:
#%PAM-1.0
auth required /lib/security/pam_stack.so service=system-auth
auth required /lib/security/pam_nologin.so
account required /lib/security/pam_stack.so service=system-auth
account required /lib/security/pam_access.so
account required /lib/security/pam_time.so
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_limits.so
Step2
Now, set the permissions of the sshd file to be (0640/-rw-r-----) and owned by the super-
user ‘root’ for security reasons.
• To change the permissions and ownership of the sshd file, use the commands:
[root@deep /]# chmod 640 /etc/pam.d/sshd
[root@deep /]# chown 0.0 /etc/pam.d/sshd](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-423-320.jpg)
![OpenSSH 1
CHAPTER 6
424
/etc/init.d/sshd: The OpenSSH Initialization File
The /etc/init.d/sshd script file is responsible to automatically starting and stopping the
OpenSSH server on your Linux system. Loading the sshd daemon as a standalone daemon will
eliminate load time and will even reduce swapping since non-library code will be shared.
Please note that the following script is suitable for Linux operating systems that use SystemV. If
you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to
make it work for you.
Step 1
Create the sshd script file (touch /etc/init.d/sshd) and add the following lines:
#!/bin/bash
# This shell script takes care of starting and stopping OpenSSH.
#
# chkconfig: 2345 55 25
# description: OpenSSH is a program thas allows to establish a secure remote
# connection to a server.
#
# processname: sshd
# config: /etc/ssh/ssh_host_key
# config: /etc/ssh/ssh_host_key.pub
# config: /etc/ssh/ssh_random_seed
# config: /etc/ssh/sshd_config
# pidfile: /var/run/sshd.pid
# Source function library.
. /etc/init.d/functions
# Source networking configuration.
. /etc/sysconfig/network
# Source OpenSSH configureation.
if [ -f /etc/sysconfig/sshd ] ; then
. /etc/sysconfig/sshd
fi
RETVAL=0
# Some functions to make the below more readable.
KEYGEN=/usr/bin/ssh-keygen
RSA1_KEY=/etc/ssh/ssh_host_key
RSA_KEY=/etc/ssh/ssh_host_rsa_key
DSA_KEY=/etc/ssh/ssh_host_dsa_key
PID_FILE=/var/run/sshd.pid
my_success() {
local msg
if [ $# -gt 1 ]; then
msg="$2"
else
msg="done"
fi
case "`type -type success`" in
function)
success "$1"
;;
*)
echo -n "${msg}"
;;
esac](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-424-320.jpg)
![OpenSSH 1
CHAPTER 6
425
}
my_failure() {
local msg
if [ $# -gt 1 ]; then
msg="$2"
else
msg="FAILED"
fi
case "`type -type failure`" in
function)
failure "$1"
;;
*)
echo -n "${msg}"
;;
esac
}
do_rsa1_keygen() {
if ! test -f $RSA1_KEY ; then
echo -n "Generating SSH1 RSA host key: "
if $KEYGEN -q -t rsa1 -f $RSA1_KEY -C '' -N '' >&/dev/null;
then
my_success "RSA1 key generation"
echo
else
my_failure "RSA1 key generation"
echo
exit 1
fi
fi
}
do_rsa_keygen() {
if ! test -f $RSA_KEY ; then
echo -n "Generating SSH2 RSA host key: "
if $KEYGEN -q -t rsa -f $RSA_KEY -C '' -N '' >&/dev/null; then
my_success "RSA key generation"
echo
else
my_failure "RSA key generation"
echo
exit 1
fi
fi
}
do_dsa_keygen() {
if ! test -f $DSA_KEY ; then
echo -n "Generating SSH2 DSA host key: "
if $KEYGEN -q -t dsa -f $DSA_KEY -C '' -N '' >&/dev/null; then
my_success "DSA key generation"
echo
else
my_failure "DSA key generation"
echo
exit 1
fi
fi
}
do_restart_sanity_check() {
sshd -t
RETVAL=$?
if [ ! "$RETVAL" = 0 ]; then
my_failure "Configuration file or keys"
echo](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-425-320.jpg)
![OpenSSH 1
CHAPTER 6
426
exit $RETVAL
fi
}
case "$1" in
start)
# Create keys if necessary
do_rsa1_keygen;
do_rsa_keygen;
do_dsa_keygen;
echo -n "Starting SSHD: "
if [ ! -f $PID_FILE ] ; then
sshd $OPTIONS
RETVAL=$?
if [ "$RETVAL" = "0" ] ; then
my_success "sshd startup" "sshd"
touch /var/lock/subsys/sshd
else
my_failure "sshd startup" ""
fi
fi
echo
;;
stop)
echo -n "Shutting down SSHD: "
if [ -f $PID_FILE ] ; then
killproc sshd
RETVAL=$?
[ $RETVAL -eq 0 ] && rm -f /var/lock/subsys/sshd
fi
echo
;;
restart)
do_restart_sanity_check
$0 stop
$0 start
RETVAL=$?
;;
condrestart)
if [ -f /var/lock/subsys/sshd ] ; then
do_restart_sanity_check
$0 stop
$0 start
RETVAL=$?
fi
;;
status)
status sshd
RETVAL=$?
;;
*)
echo "Usage: sshd {start|stop|restart|status|condrestart}"
exit 1
;;
esac
exit $RETVAL](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-426-320.jpg)
![OpenSSH 1
CHAPTER 6
427
Step 2
Once the /etc/init.d/sshd script file has been created, it is important to make it executable,
change its default permissions, create the necessary links and start it. Making this file executable
will allow the system to run it, changing its default permission is to allow only the root user to
change this file for security reason, and creation of the symbolic links will let the process control
initialization of Linux to start the program automatically for you at each system boot.
• To make this script executable and to change its default permissions, use the commands:
[root@deep /]# chmod 700 /etc/init.d/sshd
[root@deep /]# chown 0.0 /etc/init.d/sshd
• To create the symbolic rc.d links for OpenSSH, use the following commands:
[root@deep /]# chkconfig --add sshd
[root@deep /]# chkconfig --level 2345 sshd on
• To start OpenSSH software manually, use the following command:
[root@deep /]# /etc/init.d/sshd start
Starting SSHD: [OK]
Running OpenSSH in a chroot jail
This section applies only if you want to run OpenSSH in chroot jail environment for some of your
users. This kind of setup is useful for web hosting companies that want to provide shell access
via remote secure connection with OpenSSH but don’t want to allow full access to the server and
just limit users to their own web directory. By default, OpenSSH does not support the chroot jail
mode and we have to compile it with an external patch to enable the chroot mode extensions.
The patch is available from the following site: http://chrootssh.sourceforge.net/
Remember that you have to download the version number equal to the OpenSSH version number
you use in order for chroot jail support to work. At the beginning of this chapter, we have already
patched the software with the chroot jail mode extensions patch, therefore, we only need to
create the required skeleton environment and copy the necessary tools into this chroot jail
directory to enable chroot jail support. Below are the steps to follow if you want to run OpenSSH
with chroot jail support for the specified users.
The main benefit of a chroot jail is that the jail will limit the portion of the file system the daemon
can see to the root directory of the jail. Additionally, since the jail only needs to support OpenSSH,
the programs available in the jail can be extremely limited. More importantly, there is no need for
setuid-root programs, which can be used to gain root access and break out of the jail.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-427-320.jpg)
![OpenSSH 1
CHAPTER 6
429
Necessary steps to run OpenSSH in a chroot jail:
What you're essentially doing is creating a skeleton root file system with enough components
(binaries, libraries, etc.) to allow Unix to do a chroot when the user logs in.
Step 1
With OpenSSH, it’s important to give to your strictly SSH users a real shell account on the Linux
system because we want to allow remote shell access, even if limited to running just a few
commands on the server.
First, create the new users for this purpose; these users will be the users allowed to connect to
your OpenSSH server running in chroot jail mode. These have to be separate from regular user
accounts with unlimited access because of how the "chroot" environment works. Chroot makes it
appear from the user's perspective as if the level of the file system you've placed them in is the
top level of the file system.
Here we create a new SSH user called “gmourani” as an example and set it’s the home directory
under the /home/httpd/gmourani directory since it is the place where it’s the users web
directory and web pages will be located.
• Use the following command to create a new SSH user. This step must be done for each
additional new user you allow to access your OpenSSH server on OpenNA Linux.
[root@deep /]# useradd -m -d /home/httpd/gmourani gmourani
[root@deep /]# passwd gmourani
Changing password for user gmourani
New UNIX password:
Retype new UNIX password:
passwd: all authentication tokens updated successfully
• Use the following command to create a new SSH user. This step must be done for each
additional new user you allow to access your OpenSSH server on Red Hat Linux.
[root@deep /]# useradd -g users -d /home/httpd/gmourani gmourani
[root@deep /]# passwd gmourani
Changing password for user gmourani
New UNIX password:
Retype new UNIX password:
passwd: all authentication tokens updated successfully
The useradd command will add the new SSH user called “gmourani” to our Linux server and
will set it’s the home directory under the /home/httpd/gmourani directory on the system since
it is a useful location for remote clients to maintain their web accounts. The passwd command
will set the password for this user “gmourani”.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-429-320.jpg)
![OpenSSH 1
CHAPTER 6
430
Step 2
Once the new SSH users have been created, we must edit the /etc/passwd file and make the
appropriated changes to the accounts to allow OpenSSH to chroot when the users login on the
system. In general, the sshd daemon will chroot when it encounters the magic token '/./' in a
users home directory. Therefore this is what we will add to the passwd file for the SSH user in
question.
• Edit the passwd file (vi /etc/passwd) and change the following line:
gmourani:x:501:100::/home/httpd/gmourani:/bin/bash
To read:
gmourani:x:501:100::/home/httpd/gmourani/./:/bin/bash
NOTE: Don’t forget to make the same modification for each additional SSH user for whom you want
to chroot.
Step 3
Now, we have to create all the necessary chrooted environment subdirectories where we will
copy tools we want to allow this SSH user to use on the system.
• Use the following command to create all the necessary chroot subdirectories.
[root@deep /]# mkdir /home/httpd/gmourani/bin
[root@deep /]# mkdir /home/httpd/gmourani/dev
[root@deep /]# mkdir /home/httpd/gmourani/etc
[root@deep /]# mkdir /home/httpd/gmourani/lib
[root@deep /]# mkdir /home/httpd/gmourani/usr
[root@deep /]# mkdir /home/httpd/gmourani/usr/bin
[root@deep /]# mkdir /home/httpd/gmourani/usr/lib
• For Red Hat Linux 7.3 users, you should create the following additional directory:
[root@deep /]# mkdir /home/httpd/gmourani/lib/i686
Step 4
Next, we must change the permissions on all the chroot glue subdirectories to mode (0111/d--
x--x--x) for security reasons.
• Use the following command to change the permissions of all the subdirectories.
[root@deep /]# chmod -R 0111 /home/httpd/gmourani/*](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-430-320.jpg)
![OpenSSH 1
CHAPTER 6
431
Step 5
Once all permissions of the supporting glues have been changed, it is time to copy the required
binary programs to the related subdirectories in the chroot area for OpenSSH to work.
These programs are necessary to allow the SSH users to list, create directory, copy, remove, and
edit files on the SSH chroot jail directory. If there are features you don’t want the user to be able to
use, then don’t copy them to the chroot area.
• Use the following commands to copy the require binaries programs into the chroot area.
[root@deep /]# cp /bin/bash /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/cp /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/ls /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/mkdir /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/grep /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/rm /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/vi /home/httpd/gmourani/bin/
[root@deep /]# cp /usr/bin/dircolors /home/httpd/gmourani/usr/bin/
[root@deep /]# chmod 0111 /home/httpd/gmourani/bin/*
[root@deep /]# chmod 0111 /home/httpd/gmourani/usr/bin/*
NOTE: The above chmod commands will change default permissions of those programs under the
/bin directories of the chroot jail area to be (0111 ---x--x—x) because we don’t want users to
be able to modify or read binaries in the chroot area but just to execute them if necessary.
Step 6
The binaries we have copied into the chroot area have been compiled with shared libraries by
default and for this reason it is important to find the shared libraries dependencies associated with
them and copy them into the /lib subdirectory in the chroot jail area that we created earlier.
To find the shared library dependencies of binaries, you have to use the ldd command of Linux.
You must copy all the libraries below to the /home/httpd/gmourani/lib directory of the
chroot area. These libraries are part of libc, and needed by various programs.
• Use the following commands to copy the require libraries into the chroot area.
[root@deep /]# cp /lib/libtermcap.so.2 /home/httpd/gmourani/lib/
[root@deep /]# cp /lib/libdl.so.2 /home/httpd/gmourani/lib/
[root@deep /]# cp /lib/libc.so.6 /home/httpd/gmourani/lib/
[root@deep /]# cp /lib/libgcc_s.so.1 /home/httpd/gmourani/lib/
[root@deep /]# cp /lib/ld-linux.so.2 /home/httpd/gmourani/lib/
[root@deep /]# cp /usr/lib/libncurses.so.5 /home/httpd/gmourani/usr/lib/
[root@deep /]# strip -R .comment /home/httpd/gmourani/lib/*
• For Red Hat Linux 7.3 users, you should copy the following additional library:
[root@deep /]# cp /lib/i686/libc.so.6 /home/httpd/gmourani/lib/i686/](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-431-320.jpg)
![OpenSSH 1
CHAPTER 6
432
WARNING: Depending on what has been compiled, the required shared libraries may be different
than the ones illustrated above. Please use the ldd command on each binary under /bin
subdirectories of the chroot jail to find out the ones you need and copy them to the /lib
subdirectory of the chroot area.
The “strip -R .comment” command will remove all the named section “.comment” from the
libraries files under the /lib subdirectory and will make them smaller in size and can help in the
performance of them.
Step 7
One of the last step to do, is to make a copy of the “DIR_COLORS” and “passwd” files located
under the /etc directory to our chroot jail for SSH to be able to find it.
• Use the following commands to copy the file into the chroot area.
[root@deep /]# cp /etc/DIR_COLORS /home/httpd/gmourani/etc/
[root@deep /]# cp /etc/passwd /home/httpd/gmourani/etc/
Step 8
Finally, we have to create the /home/httpd/gmourani/dev/null device file and set its mode
appropriately.
• Use the following commands to create the null device into the chroot area.
[root@deep /]# mknod /home/httpd/gmourani/dev/null c 1 3
[root@deep /]# chmod 666 /home/httpd/gmourani/dev/null
Creating OpenSSH private & public keys
This section deals with actions that need to be performed to create new private/public keys for
our users to establish a secure connection on the server. There are cryptosystems where
encryption and decryption are done using separate keys, and it is not possible to derive the
decryption key from the encryption key. The idea is that each user creates a public/private key
pair for authentication purposes. The server knows the public key, and only the user knows the
private key.
The file $HOME/.ssh/authorized_keys lists the public keys that are permitted for logging in.
When the user logs in, the ssh program tells the server which key pair it would like to use for
authentication. The server checks if this key is permitted, and if so, sends the user (actually the
ssh program running on behalf of the user) a challenge, a random number, encrypted by the
user's public key. The challenge can only be decrypted using the proper private key. The user's
client then decrypts the challenge using the private key, proving that he/she knows the private
key but without disclosing it to the server.
Step 1
Below, are the steps to follow to create a new SSH private & public key for one user. This
example assumes that secure encrypted connections will be made between Linux servers.
• To create your (RSA) private & public keys for SSH2 of LOCAL, use the commands:
[root@deep /]# su gmourani
[gmourani@deep /]$ ssh-keygen -t rsa
Generating public/private rsa key pair.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-432-320.jpg)
![OpenSSH 1
CHAPTER 6
433
Enter file in which to save the key (/home/gmourani/.ssh/id_rsa):
Created directory '/home/gmourani/.ssh'.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /home/gmourani/.ssh/id_rsa.
Your public key has been saved in /home/gmourani/.ssh/id_rsa.pub.
The key fingerprint is:
ba:0c:08:6d:9d:51:4f:b3:32:68:9b:0d:83:ce:be:bd gmourani@deep
WARNING: The above example assumes that you want to generate (RSA) private & public keys for
SSH protocol 2 (highly recommended). If you want to generate (RSA) private & public keys for SSH
protocol 1, then you must use the ‘-t rsa1’ option to the key generation command as follows:
[root@deep /]# su gmourani
[gmourani@deep /]$ ssh-keygen -t rsa1
Using the ‘-t rsa1’ option will generate SSH1 instead of SSH2 private & public keys. The SSH1
private key will be named ”identity” and the public key will be “identity.pub”. The '-t'
option is used to specify the type of the key to create. The possible values are "rsa1" for protocol
version 1 and "rsa" or "dsa" for protocol version 2.
If you have multiple accounts you might want to create a separate key on each of them. You may
want to have separate keys for:
• Your server (1)
• Your server (2)
• Your server (3)
This allows you to limit access between these servers, e.g. not allowing the first server (1)
account to access your second server (2) account or the third server (3). This enhances the
overall security in the case any of your authentication keys are compromised for any reason.
Step 2
Copy your local public key id_rsa.pub for SSH2 or identity.pub for SSH1 from the
/home/gmourani/.ssh directory remotely under the name, say, “authorized_keys”. One
way to copy the file is to use the ftp command or you might need to send your public key in
electronic mail to the administrator of the other system. Just include the contents of the
~/.ssh/id_rsa.pub or ~/.ssh/identity.pub file in the message.
To resume the required steps:
1) The user creates his/her RSA keys pair by running ssh-keygen. This stores the private
key in id_rsa (SSH2) or in identity (SSH1) and the public key in id_rsa.pub (SSH2)
or in identity.pub (SSH1) into the user's home directory on the LOCAL machine.
2) The user should then copy the id_rsa.pub key (SSH2) or identity.pub key (SSH1)
to $HOME/.ssh/authorized_keys into his/her home directory on the REMOTE
machine.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-433-320.jpg)
![OpenSSH 1
CHAPTER 6
434
------------------- -------------------
| | | |
| | | |
| Server 1 | | Server 2 |
| | | |
| | | |
------------------- -------------------
User: gmourani User: gmourani
Pass-phrase: qwerty1 Pass-phrase: qwerty2
Private key: id_rsa Private key: id_rsa
Public key: id_rsa.pub --------------- authorized_keys
authorized_keys ----------------------- Public key: id_rsa.pub
Public key of user gmourani on the first server (1) is sending to the second server (2) under the
$HOME directory of user gmourani and become ‘authorized_keys’; the same action is made
on the second server (2). The public key of user gmourani on server (2) is sending to server (1)
under the $HOME directory of user gmourani and become ‘authorized_keys’.
NOTE: OpenSSH's public key is a one-line string. Adding public keys from commercial SSH tools
which stretch the public key over several lines, will not be recognized by OpenSSH.
OpenSSH Users Tools
The commands listed below are some that we use regularly, but many more exist, and you
should check the manual pages and documentation of OpenSSH for more details.
ssh
The ssh (Secure Shell) command provides secure encrypted communications between two
untrusted hosts over an insecure network. It is a program for securely logging into a remote
machine and executing commands from there. It is a suitable replacement for insecure programs
like telnet, rlogin, rcp, rdist, and rsh.
• To login to a remote machine, use the following command:
[root@deep /]# ssh -l <login_name> <hostname>
For example:
[root@deep /]# ssh -l gmourani deep.openna.com
gmourani@deep.openna.com’s password:
Last login: Tue Oct 19 1999 18:13:00 -0400 from deep.openna.com
No mail.
[gmourani@deep gmourani]$
Where <login_name> is the name you use to connect to the ssh server and <hostname> is
the remote address (you can use IP address here) of your ssh server.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-434-320.jpg)
![OpenSSH 1
CHAPTER 6
435
scp
The scp (Secure Copy) utility copies files from the local system to a remote system or vice versa,
or even between two remote systems using the scp command.
• To copy files from remote to local system, use the following commands:
[root@deep /]# su gmourani
[gmourani@deep /]$ scp -p <login_name@hostname>:/dir/for/file
localdir/to/filelocation
For example:
[gmourani@deep /]$ scp -p gmourani@mail:/etc/test1 /tmp
Enter passphrase for RSA key 'gmourani@mail.openna.com':
test1 | 2 KB | 2.0 kB/s | ETA: 00:00:00 | 100%
• To copy files from local to remote system, use the following commands:
[root@deep /]# su gmourani
[gmourani@deep /]$ scp -p localdir/to/filelocation
<username@hostname>:/dir/for/file
For example:
[gmourani@deep /]$ scp -p /usr/bin/test2 gmourani@mail:/var/tmp
gmourani@mail's password:
test2 | 7 KB | 7.9 kB/s | ETA: 00:00:00 | 100%
WARNING: The “-p” option indicates that the modification and access times, as well as modes of
the source file, should be preserved on the copy. Usually this is desirable. Please check the
chapter related to backups in this book for more information about other possible uses of SSH
technology with Linux.
Changing your pass-phrase
You can change the pass-phrase at any time by using the -p option of ssh-keygen.
• To change the pass-phrase, use the following commands:
[root@deep /]# su gmourani
[gmourani@deep /]$ ssh-keygen -p
Enter file in which the key is (/home/gmourani/.ssh/id_rsa):
Enter old passphrase:
Key has comment '/home/gmourani/.ssh/id_rsa'
Enter new passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved with the new passphrase.
Further documentation
For more details, there are several manual pages about OpenSSH that you can read:
$ man ssh (1) - OpenSSH secure shell client (remote login program).
$ man ssh [slogin] (1) - OpenSSH secure shell client (remote login program).
$ man ssh-add (1) - Adds identities for the authentication agent.
$ man ssh-agent (1) - Authentication agent.
$ man ssh-keygen (1) - Authentication key generation.
$ man sshd (8) - Secure shell daemon.
$ sftp-server (8) - SFTP server subsystem.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-435-320.jpg)
![Sudo 1
CHAPTER 7
440
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install Sudo, and then
one afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Sudo1
• And the following one after you install the software:
[root@deep root]# find /* > Sudo2
• Then use this command to get a list of what changed:
[root@deep root]# diff Sudo1 Sudo2 > Sudo-Installed
Compiling - Optimizing & Installing Sudo
Below are the steps that you must make to configure, compile and optimize the Sudo software
before installing it on your system. First off, we install the program as user 'root' so as to avoid
any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp sudo-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf sudo-version.tar.gz
Step 2
Next, move into the newly created Sudo source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created Sudo source directory use the command:
[root@deep tmp]# cd sudo-1.6.6/
• To configure and optimize Sudo use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--sbindir=/usr/sbin
--with-logging=syslog
--with-logfac=authpriv
--with-pam
--with-env-editor
--with-ignore-dot
--with-tty-tickets
--disable-root-mailer
--disable-root-sudo
--disable-path-info
--with-mail-if-noperms
--with-mailsubject="*** Sudo SECURITY information for %h ***"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-440-320.jpg)
![Sudo 1
CHAPTER 7
441
This tells Sudo to set itself up for this particular configuration setup with:
- Log sudo messages via syslog.
- The syslog facility to log with is authpriv.
- Enable PAM support.
- Use the environment variable EDITOR for visudo.
- Ignore '.' in the PATH.
- Use a different ticket file for each user/tty combo.
- Don't run the mailer as root, run as the user since it’s safer.
- Don't allow root to run sudo command.
- Print 'command not allowed' instead of 'command not found'.
- Send mail to sysadmin if user not allowed to runs command.
- Change subject of sudo mail result.
Step 3
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the Sudo software:
[root@deep sudo-1.6.6]# make
[root@deep sudo-1.6.6]# cd
[root@deep root]# find /* > Sudo1
[root@deep root]# cd /var/tmp/sudo-1.6.6/
[root@deep sudo-1.6.6]# make install
[root@deep sudo-1.6.6]# strip /usr/bin/sudo
[root@deep sudo-1.6.6]# strip /usr/sbin/visudo
[root@deep sudo-1.6.6]# mkdir –p –m0700 /var/run/sudo
[root@deep sudo-1.6.6]# cd
[root@deep root]# find /* > Sudo2
[root@deep root]# diff Sudo1 Sudo2 > Sudo-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 4
Once the configuration, optimization, compilation, and installation of the Sudo software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete Sudo and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf sudo-version/
[root@deep tmp]# rm -f sudo-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-441-320.jpg)
![Sudo 1
CHAPTER 7
443
This tells the sudoers file to set itself up for this particular configuration with:
Defaults rootpw
With sudo, certain configuration options may be changed from their default values at runtime via
one or more “Default_Entry” options defined in the sudoers file. This is what we do here. In
our case, we inform sudo with the “Defaults rootpw” option that we want it to prompt any
allowed user who wants to sudo to super-user “root” to enter the “root” password instead of
the password of the invoking user.
By default sudo asks for the users password instead of super-user password when someone
wants to sudo to “root” user. Because in this configuration file we want to allow full “root”
access for users that are members of the “wheel” group and because we trust these users, we
decide to change the default sudo setting and ask for “root” password before having access to
“root” privilege on the server.
This setting is useful when you make secure remote connections on the server with SSH software
and want to sudo to “root” user.
root ALL=(ALL) ALL
Here we inform sudo that the super-user “root” can run anything as any user on the server. This
option is required for the software to work or there is no need to use sudo. The “ALL=(ALL)
ALL” parameter means that everything is allowed for the super-user “root”.
%wheel ALL=(ALL) ALL
Here we allow every user who is a member of the group “wheel” to run all commands as super-
user “root” on the system. This is the equivalence of what we achieve with the PAM security
feature, but in a most efficient and secure way.
Step 2
Once the sudoers file has been configured, it is time to add some users who will be allowed to
sudo to “root” account.
• If you want to make, for example, the user “sysadmin” a member of the “wheel” group,
and thus be able to sudo to “root”, use the following command:
[root@deep /]# usermod -G10 sysadmin
This means “G” is a list of supplementary groups that the user is also a member of. “10” is the
numeric value of the user’s ID “wheel”, and “sysadmin” is the user we want to add to the
“wheel” group. Use the same command above for all users on your system you want to be able
to sudo to “root” account.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-443-320.jpg)
![Sudo 1
CHAPTER 7
444
/etc/pam.d/sudo: The Sudo PAM Support Configuration File
For better security of Sudo, we have compiled it to use the PAM mechanism for password
authentication.
Step 1
To be able to use this feature, we must create the /etc/pam.d/sudo file and add the following
parameters inside it.
• Create the sudo file (touch /etc/pam.d/sudo) and add the following lines:
#%PAM-1.0
auth required /lib/security/pam_stack.so service=system-auth
account required /lib/security/pam_stack.so service=system-auth
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
Step2
Now, set the permissions of the sudo file to be (0640/-rw-r-----) and owned by the super-
user ‘root’ for security reasons.
• To change the permissions and ownership of the sudo file, use the commands:
[root@deep /]# chmod 640 /etc/pam.d/sudo
[root@deep /]# chown 0.0 /etc/pam.d/sudo
A more complex sudoers configuration file
For those who want to have complete control on who can use the “root” account on the server
when there are many users with shell access doing different tasks, here is the sudoers
configuration file to go with. As you’ll see it is more complex and covers some basic definitions.
It is important to understand how the sudo policy works. To be as clear as possible, I will simply
say that when you allow full access to user with the “ALL” definition, you cannot deny other
access or privileges. Therefore, the best way is to allow only what you want user to be able to run
with “root” privilege through the “Cmnd alias specification” part of the configuration file
and use the defined aliases rules under the “User privilege specification” part of the
configuration file. Here is a working sudoers configuration file to better understand what I mean.
• Edit the sudoers file (visudo) and set your needs. Below is what we recommend you
use for servers that have many users with shell access:
# /etc/sudoers: OpenNA, Inc. (last updated 2002 Apr 19)
# This file MUST be edited with the 'visudo' command as root.
# User alias specification
User_Alias FULLTIME_USERS = sysadmin, gmourani
User_Alias PARTTIME_USERS = zeljko, mary
# Cmnd alias specification
Cmnd_Alias HTTP = /etc/init.d/httpd, /bin/vi /etc/httpd/*
Cmnd_Alias FTP = /etc/init.d/proftpd, /bin/vi /etc/proftpd.conf
Cmnd_Alias SMTP = /etc/init.d/exim, /bin/vi /etc/mail/*
Cmnd_Alias SQL = /etc/init.d/mysqld, /usr/bin/mysql,
/usr/bin/mysqladmin
Cmnd_Alias BIND = /etc/init.d/named, /bin/vi /chroot/named/*](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-444-320.jpg)
![Sudo 1
CHAPTER 7
445
# Defaults specification
Defaults:FULLTIME_USERS rootpw
Defaults:FULLTIME_USERS !lecture
# User privilege specification
# Super-user root can run anything as any user.
root ALL=(ALL) ALL
# Every users member of the group wheel will be allowed
# to run all commands as super-user root.
%wheel ALL=(ALL) ALL
# Full time users may run anything but need a password.
FULLTIME_USERS ALL = ALL
# Part time users may administrate httpd, ftpd, smtp, sql, and bind
servers.
PARTTIME_USERS ALL = HTTP, FTP, SMTP, SQL, BIND
This tells the sudoers file to set itself up for this particular configuration with:
User_Alias FULLTIME_USERS = sysadmin, gmourani
User_Alias PARTTIME_USERS = zeljko, mary
The “User_Alias” definition lines are used to define local users who may sudo to the “root”
account on the server. The definition works on the following way:
Alias_Type NAME = item1, item2, ...
Where “Alias_Type” is the type of alias to use, in our case, we use “User_Alias” to define
local users aliases on the system. A NAME is a string of uppercase letters, numbers, or
underscores characters ('_'). A NAME must always start with an uppercase letter. You can use as
any name as you like to define the NAME.
In our example, we use “FULLTIME_USERS” to define local users on the system who have full
time access to the server and “PARTTIME_USERS” to define local users on the server who have
part time access to the server for different reason. Item represents usernames to add in each
category separated by [,].
NOTE: It is important to note that users added to the “User_Alias” definition will be able to sudo
to super-user “root” account even if their names do not appear under the group “wheel”.
Cmnd_Alias HTTP = /etc/init.d/httpd, /bin/vi /etc/httpd/*
Cmnd_Alias FTP = /etc/init.d/proftpd, /bin/vi /etc/proftpd.conf
Cmnd_Alias SMTP = /etc/init.d/exim, /bin/vi /etc/mail/*
Cmnd_Alias SQL = /etc/init.d/mysqld, /usr/bin/mysql, /usr/bin/mysqladmin
Cmnd_Alias BIND = /etc/init.d/named, /bin/vi /chroot/named/*
Here we use another “Alias_Type” definition. This type of alias is used to define commands,
programs, files, and directories that we want to allow certain users to be able to use when issuing
the sudo command on the system. The definition works on the following way:
Alias_Type NAME = item1, item2, ...](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-445-320.jpg)
![Sudo 1
CHAPTER 7
446
Where “Alias_Type” is the type of alias to use, in our case, we use “Cmnd_Alias” to define
command aliases on the system. A NAME is a string of uppercase letters, numbers, or
underscores characters ('_'). A NAME must always start with an uppercase letter. You can use any
name you like to define the NAME.
In our example, we use “HTTP, FTP, SMTP, SQL, and BIND” to define command names aliases
associated with the commands we will allow local users to run when issuing the sudo command
on the system. Item represent the commands, files, programs, or directories to add in each
category separated by [,].
Defaults:FULLTIME_USERS rootpw
Defaults:FULLTIME_USERS !lecture
With sudo, certain configuration options may be changed from their default values at runtime via
one or more “Default_Entry” options. Again, this is what we do here. In our case, we inform
sudo with the “Defaults:FULLTIME_USERS rootpw” option that we want it to prompt any
users with full time access on the server “FULLTIME_USERS” to enter the “root” password
instead of their own password.
Remember that by default sudo asks for the users password instead of super-user password
when someone wants to sudo to “root”. Because we allow full “root” access for users under
the “FULLTIME_USERS” alias, we decide to change the default sudo setting and ask for the
“root” password before giving access to “root” privileges on the server. This also means that
users under the “PARTTIME_USERS” alias will have to enter their own password and not the
“root” password.
This is a security feature to separate trusted users with full time access on the server from semi-
trusted users with part time access on the system. Users having part time access could be
students, limited administrators, or anything else that you think about. In this way, users with part
time access do not know the “root” password since they enter their own passwords to get some
“root” privileges and we don’t need to change “root” password every time these users leave
the company.
The second “Default_Entry” option “Defaults:FULLTIME_USERS !lecture”, simply
informs sudo to not send a short lecture message about the policy use of the program to users
with full time access on the server.
FULLTIME_USERS ALL = ALL
Here we allow every user who is listed in the “FULLTIME_USERS” alias of the configuration file to
run all commands as super-user “root” on the system. This is the equivalence of what we
achieve with the PAM security feature.
PARTTIME_USERS ALL = HTTP, FTP, SMTP, SQL, BIND
Here we allow every user who is listed in the “PARTTIME_USERS” alias of the configuration file to
only run the specified commands as super-user “root”. These users will be able to edit, run and
restart the specified services and nothing else. Therefore, they only have limited “root” access
on the server to accomplish their tasks.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-446-320.jpg)
![Sudo 1
CHAPTER 7
447
Securing Sudo
This section deals specifically with actions we can take to improve and tighten security under
Sudo. Sudo is very good and well-written software with high security in mind. Once properly
compiled, installed, and configured, there are only some little things that we can do to better
secure it. Most important of all the security measures are already made within the software.
Disabling su access on the server
Once sudo is configured and running on your server, you really don’t need to continue to use ‘su’
to get “root” access on the system. Here, I show you how to disable ‘su’ command to provide
“root” access. The simpler approach will be to remove the SUID bit on the ‘su’ program. In this
way, no one will be able to use it to gain “root” anymore.
• To remove the SUID bit on the ‘su’ command, use the following command:
[root@deep /]# chmod 0511 /bin/su
Sudo Users Tools
The commands and options listed bellows are some that we use regularly, but many more exist,
and you should check the manual pages and documentation of Sudo for more details.
Running sudo for user with full “root” access
Basically, sudo is used by prepending "sudo" (followed by a space) to your command. It then
prompts you for your personal password or root password depending of the configuration and
then checks the /etc/sudoers configuration file to make sure you have "sudo" permission to
run that command on the server. Sudo runs the command as root (or another user) and logs the
details to syslog. It also logs problems and other invalid uses.
When you have full “root” privileges on the system because you are listed in the sudoers file
as one user with all “root” access right, you can run the sudo command with the -s (shell)
option to become the super-user “root” on the server.
• To sudo as super-user “root” with shell access, use the following command:
[sysadmin@deep /]# sudo –s
Password:
To be able to use the above command, you should have all “root” access rights in the sudoers
configuration file. Please note that, in this example, you have to enter the super-user “root”
password and not the password of the user “sysadmin”.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-447-320.jpg)
![Sudo 1
CHAPTER 7
448
Running sudo for user with limited “root” access
If you are a user with limited “root” access rights on the server, you cannot use the previous
sudo command to sudo as super-user “root” on the server. Instead, you should prepend the
sudo command with the command you want to run as “root” user. This supposes that
commands you expect to run are listed as allowed commands to use inside the sudoers
configuration file.
• To sudo as super-user “root” with limited access, use the following command:
[mary@deep /]# sudo /etc/init.d/httpd restart
Password:
The above command will restart the httpd web server daemon on the system even if you are the
user called “mary” because the sudoers file allows you to do it as super-user “root”. Please
note that in this example you have to enter the password of user “mary” and not the password of
the super-user “root”.
Further documentation
For more details, there are some manual pages you can read:
$ man sudoers (5) - List of which users may execute what.
$ man sudo (8) - Execute a command as another user.
$ man visudo (8) - Used to edit the sudoers file.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-448-320.jpg)
![sXid 1
CHAPTER 8
453
Linux sXid
Abstract
SUID/SGID files can be a security hazard. To reduce the risks, we have previously removed the
's' bits from root-owned programs that won't require such privileges (See chapter related to
General System Security), but future and existing files may be set with these ‘s’ bits enabled
without you being notified.
sXid is an all in one suid/sgid monitoring program designed to be run by “cron” on a regular
basis. Basically, it tracks any changes in your s[ug]id files and folders. If there are any new ones,
ones that aren't set any more, or they have changed bits or other modes then it reports the
changes in an easy to read format via email or on the command line. sXid will automate the task
to find all SUID/SGID on your server and report them to you. Once installed you can forget it and
it will do the job for you.
These installation instructions assume
Commands are Unix-compatible.
The source path is /var/tmp (note that other paths are possible, as personal discretion).
Installations were tested on OpenNA Linux & Red Hat Linux 7.3.
All steps in the installation will happen using the super-user account “root”.
Whether kernel recompilation may be required: No
Latest sXid version number is 4.0.2
The procedures given in this chapter are likely to work on all Linux platforms, but we have only
tested it on OpenNA Linux and Red Hat Linux.
Packages
The following is based on information listed by sXid as of 2002/06/24. Please check
ftp://marcus.seva.net/pub/sxid/ regularly for the latest status. We chose to install from source
because it provides the facility to fine tune the installation.
Source code is available from:
sXid Homepage: ftp://marcus.seva.net/pub/sxid/
sXid FTP Site: 137.155.111.51
You must be sure to download: sxid_4.0.2.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-453-320.jpg)
![sXid 1
CHAPTER 8
454
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install sXid, and then
one afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > sXid1
• And the following one after you install the software:
[root@deep root]# find /* > sXid2
• Then use the following command to get a list of what changed:
[root@deep root]# diff sXid1 sXid2 > sXid-Installed
Compiling - Optimizing & Installing sXid
Below are the steps that you must make to configure, compile and optimize the sXid software
before installing it on your system. First off, we install the program as user 'root' so as to avoid
any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp sxid_version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf sxid_version.tar.gz
Step 2
Now move into the newly created sXid source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created sXid directory use the following command:
[root@deep tmp]# cd sxid-4.0.2/
• To configure and optimize sXid use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--sysconfdir=/etc
--mandir=/usr/share/man
WARNING: Pay special attention to the compile CFLAGS line above. We optimize sXid for an i686
CPU architecture with the parameter “-march=i686”. Please don’t forget to adjust this CFLAGS
line to reflect your own system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-454-320.jpg)
![sXid 1
CHAPTER 8
455
Step 3
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the sXid software:
[root@deep sXid-4.0.2]# cd
[root@deep root]# find /* > sXid1
[root@deep root]# cd /var/tmp/sxid-4.0.2/
[root@deep sxid-4.0.2]# make install
[root@deep sxid-4.0.2]# cd
[root@deep root]# find /* > sXid2
[root@deep root]# diff sXid1 sXid2 > sXid-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 4
Once the configuration, optimization, compilation, and installation of the sXid software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete sXid and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf sxid-version/
[root@deep tmp]# rm -f sxid_version_tar.gz
The rm command as used above will remove all the source files we have used to compile and
install sXid. It will also remove the sXid compressed archive from the /var/tmp directory.
Configuring sXid
After sXid has been built and installed successfully in your system, your next step is to configure
and customize its configuration files to fit your needs.
/etc/sxid.conf: (The sXid Configuration File)
/etc/cron.daily/sxid: (The sXid Cron File)
/etc/sxid.conf: The sXid Configuration File
The configuration file for sXid allows you to set options that modify the operation of the program.
It is well commented and very basic.
Step 1
We must change the default one to fit our requirements and operating system. The text in bold
are the parts of the configuration file that must be customized and adjusted to satisfy our needs.
• Edit the sxid.conf file (vi /etc/sxid.conf) and set your needs. Below is what we
recommend you.
# Configuration file for sXid
# Note that all directories must be absolute with no trailing /'s
# Where to begin our file search
SEARCH = "/"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-455-320.jpg)
![sXid 1
CHAPTER 8
456
# Which subdirectories to exclude from searching
EXCLUDE = "/proc /mnt /cdrom /floppy"
# Who to send reports to
EMAIL = "root"
# Always send reports, even when there are no changes?
ALWAYS_NOTIFY = "no"
# Where to keep interim logs. This will rotate 'x' number of
# times based on KEEP_LOGS below
LOG_FILE = "/var/log/sxid.log"
# How many logs to keep
KEEP_LOGS = "5"
# Rotate the logs even when there are no changes?
ALWAYS_ROTATE = "no"
# Directories where +s is forbidden (these are searched
# even if not explicitly in SEARCH), EXCLUDE rules apply
FORBIDDEN = "/home /tmp"
# Remove (-s) files found in forbidden directories?
ENFORCE = "yes"
# This implies ALWAYS_NOTIFY. It will send a full list of
# entries along with the changes
LISTALL = "no"
# Ignore entries for directories in these paths
# (this means that only files will be recorded, you
# can effectively ignore all directory entries by
# setting this to "/"). The default is /home since
# some systems have /home g+s.
IGNORE_DIRS = "/home"
# File that contains a list of (each on it's own line)
# other files that sxid should monitor. This is useful
# for files that aren't +s, but relate to system
# integrity (tcpd, inetd, apache...).
# EXTRA_LIST = "/etc/sxid.list"
# Mail program. This changes the default compiled in
# mailer for reports. You only need this if you have changed
# it's location and don't want to recompile sxid.
MAIL_PROG = "/bin/mail"
Step 2
Now, for security reasons, change the mode of this file to be 0400.
• This can be done with the following command:
[root@deep /]# chmod 400 /etc/sxid.conf](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-456-320.jpg)
![sXid 1
CHAPTER 8
457
/etc/cron.daily/sxid: The sXid Cron File
The sxid file is a small script executed automatically by the “crond” program of your server
each day to tracks any changes in your s[ug]id files and folders.
Step 1
If there are any new ones, ones that aren't set any more, or they have changed bits or other
modes then it reports the changes. If you intend to automate this task, follow the simple steps
below.
• Create the sxid script file (touch /etc/cron.daily/sxid) and add the following
lines:
#!/bin/sh
SXID_OPTS=
if [ -x /usr/bin/sxid ]; then
/usr/bin/sxid ${SXID_OPTS}
fi
Step2
Now, make this script executable and change its permissions to be 0510.
• This can be done with the following command:
[root@deep /]# chmod 510 /etc/cron.daily/sxid
sXid Administrative Tools
After your desired configuration options have been set and the program is running, we can play
with its utility. The sXid software is meant to run as a cronjob. It must run once a day, but busy
shell boxes may want to run it twice a day. You can also run this manually for spot-checking.
• To run sXid manually, use the command:
[root@deep /]# sxid -k
sXid Vers : 4.0.2
Check run : Thu Apr 25 19:35:36 2002
This host : deep.openna.com
Spotcheck : /root
Excluding : /proc /mnt /cdrom /floppy
Ignore Dirs: /home
Forbidden : /home /tmp
(enforcing removal of s[ug]id bits in forbidden paths)
No changes found
This checks for changes by recursing the current working directory. Log files will not be rotated
and no email sent. All output will go to stdout.
Further documentation
For more details, there are some manual pages you can read:
$ man sxid.conf (5) - Configuration settings for sxid.
$ man sxid (1) - Check for changes in s[ug]id files and directories.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-457-320.jpg)
![LogSentry 1
CHAPTER 9
462
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install LogSentry, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > LogSentry1
• And the following one after you install the software:
[root@deep root]# find /* > LogSentry2
• Then use the following command to get a list of what changed:
[root@deep root]# diff LogSentry1 LogSentry2 > LogSentry-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing LogSentry
Below are the steps that you must make to configure, compile and optimize the LogSentry
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp logsentry-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf logsentry-version.tar.gz
Step 2
In order to check that the version of LogSentry, which you are going to install, is an original and
unmodified one, use the command described below to check its MD5 hashes checksum.
• To verify the MD5 checksum of LogSentry, use the following command:
[root@deep tmp]# md5sum logsentry-1.1.1.tar.gz
This should yield an output similar to this:
e97c2f096e219e20310c1b80e9e1bc29 logsentry-1.1.1.tar.gz
Now check that this checksum is exactly the same as the one published on the LogSentry
website at the following URL: http://www.psionic.com/downloads/checksums.md5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-462-320.jpg)
![LogSentry 1
CHAPTER 9
463
Step 3
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to configure the program for our PATH environment variable under
Linux. Therefore, move into the newly created LogSentry source directory and perform the
following steps to configure and optimize the software for your system.
• To move into the newly created LogSentry directory use the following command:
[root@deep tmp]# cd logcheck-1.1.1/
Step 4
Here, we have to change default locations of different LogSentry configuration files on the
system. To archive these modifications, we must edit the logcheck.sh script file as follow.
• Edit the logcheck.sh file and change all of the targeted lines in the order shown below:
a) vi +47 systems/linux/logcheck.sh and change the line:
LOGTAIL=/usr/local/bin/logtail
To read:
LOGTAIL=/usr/bin/logtail
b) vi +55 systems/linux/logcheck.sh and change the line:
TMPDIR=/usr/local/etc/tmp
To read:
TMPDIR=/var/logsentry
c) vi +92 systems/linux/logcheck.sh and change the line:
HACKING_FILE=/usr/local/etc/logcheck.hacking
To read:
HACKING_FILE=/etc/logsentry/hacking
d) vi +101 systems/linux/logcheck.sh and change the line:
VIOLATIONS_FILE=/usr/local/etc/logcheck.violations
To read:
VIOLATIONS_FILE=/etc/logsentry/violations](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-463-320.jpg)
![LogSentry 1
CHAPTER 9
465
d) vi +30 Makefile and change the line:
INSTALLDIR_SH = /usr/local/etc
To read:
INSTALLDIR_SH = /usr/sbin
e) vi +30 Makefile and change the line:
TMPDIR = /usr/local/etc/tmp
To read:
TMPDIR = /var/logsentry
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the LogSentry software:
[root@deep logcheck-1.1.1]# cd
[root@deep root]# find /* > LogSentry1
[root@deep root]# cd /var/tmp/logcheck-1.1.1/
[root@deep logcheck-1.1.1]# mkdir –m0700 /etc/logsentry
[root@deep logcheck-1.1.1]# make linux
[root@deep logcheck-1.1.1]# strip /usr/bin/logtail
[root@deep logcheck-1.1.1]# cd /etc/logsentry/
[root@deep logsentry]# mv logcheck.hacking hacking
[root@deep logsentry]# mv logcheck.violations violations
[root@deep logsentry]# mv logcheck.violations.ignore violations.ignore
[root@deep logsentry]# mv logcheck.ignore ignore
[root@deep logsentry]# cd
[root@deep root]# find /* > LogSentry2
[root@deep root]# diff LogSentry1 LogSentry2 > LogSentry-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 7
Once the configuration, optimization, compilation, and installation of the LogSentry software
have been accomplished, we can free up some disk space by deleting the program tar archive
and the related source directory since they are no longer needed.
• To delete LogSentry and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf logcheck-version/
[root@deep tmp]# rm -f logsentry-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install LogSentry. It will also remove the LogSentry compressed archive from the /var/tmp
directory.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-465-320.jpg)
![LogSentry 1
CHAPTER 9
466
Configuring LogSentry
After LogSentry has been built and installed successfully in your system, your next step is to
check its configuration files to see if they fit your needs.
/etc/logsentry/hacking: (The LogSentry Hacking File)
/etc/logsentry/ignore: (The LogSentry Ignore File)
/etc/logsentry/violations: (The LogSentry Violation File)
/etc/logsentry/violations.ignore: (The LogSentry Violation Ignore File)
From the default install, there are no LogSentry configuration files to modify, the default entries
look fine and if you want to make some personal adjustment, all you have to do is to edit the
related LogSentry configuration files located under /etc/logsentry directory.
More information about the operation of each one is contained in the INSTALL file of LogSentry
under its uncompressed source directory.
Although the fact that there is no LogSentry configuration files to change, the last action to
make before using the program is to automate it.
Step 1
Create a file called logsentry under the /etc/cron.daily directory and add the following
lines to set LogSentry to run once per day.
• To create the logsentry file under /etc/cron.daily directory, type the following
lines in your terminal (as root):
cat <<EOF > /etc/cron.daily/logsentry
#!/bin/sh
# Hourly check Log files for security violations and unusual activity.
/usr/bin/logcheck.sh
EOF
Step 2
Now, make this script executable and change its permissions to be 0510.
• This can be done with the following commands:
[root@deep /]# chmod 510 /etc/cron.daily/logsentry
WARNING: Remember, in our configuration and installation, LogSentry does not report anything
via email if it has nothing useful to say.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-466-320.jpg)
![HostSentry 2
CHAPTER 0
470
Prerequisites
HostSentry requires that the listed software below be already installed on your system to be
able to compile successfully. If this is not the case, you must install it from your Linux CD-ROM or
source archive files. Please make sure you have this program installed on your machine before
you proceed with this chapter.
Python, which allows HostSentry to run, must already be installed on your system to
be able to compile and use the HostSentry software.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install HostSentry, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > HostSentry1
• And the following one after you install the software:
[root@deep root]# find /* > HostSentry2
• Then use the following command to get a list of what changed:
[root@deep root]# diff HostSentry1 HostSentry2 > HostSentry-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing HostSentry
Below are the steps that you must make to configure, compile and optimize the HostSentry
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp hostsentry-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf hostsentry-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-470-320.jpg)
![HostSentry 2
CHAPTER 0
471
Step 2
In order to check that the version of HostSentry, which you are going to install, is an original
and unmodified one, use the command described below to check its MD5 hashes checksum.
• To verify the MD5 checksum of HostSentry, use the following command:
[root@deep tmp]# md5sum hostsentry-0.02.tar.gz
This should yield an output similar to this:
3de0bbb7d456bb53683de56dfdf98362 hostsentry-0.02.tar.gz
Now check that this checksum is exactly the same as the one published on the HostSentry
website at the following URL: http://www.psionic.com/downloads/checksums.md5
Step 3
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to configure the program for our PATH environment variable under
Linux. Therefore, move into the newly created HostSentry source directory and perform the
following steps to configure and optimize the software for your system.
• To move into the newly created HostSentry directory use the following command:
[root@deep tmp]# cd hostsentry-0.02/
Step 4
First, we have to define directories where we want HostSentry to be installed on our system.
Editing the Makefile script file as follows does this:
• Edit the Makefile file and change all of the targeted lines in the order shown below:
a) vi +7 Makefile and change the line:
INSTALLDIR = /usr/local/abacus/hostsentry
To read:
INSTALLDIR = /etc/hostsentry
LIBDIR= /usr/lib/hostsentry
b) vi +21 Makefile and change the lines:
@echo "Installing HostSentry in: $(INSTALLDIR)"
install -d -g 0 -o root -m 0700 $(INSTALLDIR)
install -d -g 0 -o root -m 0700 $(INSTALLDIR)/modules
install -g 0 -o root -m 0700 host* $(INSTALLDIR)
install -g 0 -o root -m 0700 module* $(INSTALLDIR)/modules
To read:
install -d -m 0700 $(INSTALLDIR)
install -d -m 0700 $(LIBDIR)/modules
install -m 0400 host* $(LIBDIR)
install -m 0400 module* $(LIBDIR)/modules](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-471-320.jpg)
![HostSentry 2
CHAPTER 0
473
• Edit the moduleMultipleLogins.py file (vi +78 moduleMultipleLogins.py)
and change the line:
allowPath = config.parseToken('MODULE_PATH')
To read:
allowPath = '/etc/hostsentry/'
Step 6
Finally, we have to edit the hostsentry.py file and add a new line at the BEGINNING of the file
to set the environment variable of the python binary for the program to find and use it when it
runs.
• Edit the hostsentry.py file (vi hostsentry.py) and add the line:
#!/usr/bin/env python
Step 7
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the HostSentry software:
[root@deep hostsentry-0.02]# cd
[root@deep root]# find /* > HostSentry1
[root@deep root]# cd /var/tmp/hostsentry-0.02/
[root@deep hostsentry-0.02]# make install
[root@deep hostsentry-0.02]# mkdir -m0700 /var/hostsentry
[root@deep logsentry]# cd
[root@deep root]# find /* > HostSentry2
[root@deep root]# diff HostSentry1 HostSentry2 > HostSentry-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 8
Once the configuration, optimization, compilation, and installation of the HostSentry software
have been accomplished, we can free up some disk space by deleting the program tar archive
and the related source directory since they are no longer needed.
• To delete HostSentry and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf hostsentry-version/
[root@deep tmp]# rm -f hostsentry-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-473-320.jpg)
![HostSentry 2
CHAPTER 0
474
Configuring HostSentry
After HostSentry has been built and installed successfully in your system, your next step is to
configure and customize its configuration files.
/etc/hostsentry/hostsentry.conf: (The HostSentry Configuration File)
/etc/hostentry/hostsentry.ignore: (The HostSentry Ignore File)
/etc/hostentry/hostsentry.action: (The HostSentry Action File)
/etc/hostentry/hostsentry.modules: (The HostSentry Modules File)
/etc/hostentry/moduleForeignDomain.allow
/etc/hostentry/moduleMultipleLogins.allow
/etc/init.d/hostsentry: (The HostSentry Initialization File)
/etc/hostsentry/hostsentry.conf: The HostSentry Config File
The hostsentry.conf file is the main configuration file for HostSentry. It is in this file that
HostSentry gets all of its information about the way it should run on your system.
Step 1
By default, the hostsentry.conf file does not exist after installation and we have to create it.
• Create the hostsentry.conf file (touch /etc/hostsentry/hostsentry.conf)
and set your needs. Below is what we recommend you have in your file.
IGNORE_FILE = "/etc/hostsentry/hostsentry.ignore"
ACTION_FILE = "/etc/hostsentry/hostsentry.action"
MODULE_FILE = "/etc/hostsentry/hostsentry.modules"
MODULE_PATH = "/usr/lib/hostsentry/modules"
WTMP_FILE = "/var/log/wtmp"
DB_FILE = "/var/hostsentry/hostsentry.db"
DB_TTY_FILE = "/var/hostsentry/hostsentry.tty.db"
WTMP_FORMAT = "384/8:32/44:32/76:256"
Step2
Now, set the permissions of the hostsentry.conf file to be (0600/-rw-------) and owned
by the super-user ‘root’ for security reasons.
• To change the permissions and ownership of the hostsentry.conf file, use:
[root@deep /]# chmod 600 /etc/hostsentry/hostsentry.conf
[root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.conf](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-474-320.jpg)
![HostSentry 2
CHAPTER 0
475
/etc/hostsentry/hostsentry.ignore: The HostSentry Ignore File
The hostsentry.ignore file contains a list of users you want HostSentry to never process
or never take action against with the modules. This is useful for users such as "ftp" who show
up in wtmp but would cause a large number of false alarms because of the anonymous access. It
is important to note that each user must be placed one per line.
Step 1
By default, the hostsentry.ignore file doesn’t exist after installation and we have to create it.
• Create the hostsentry.ignore file (touch /etc/hostsentry/hostsentry.ignore)
and put in any user account you want to be ignored if it logs in to the system. Below is
what we recommend you put in the file. By default, we have no users defined.
# Place usernames in this file that you want to ignore (ftp, etc.)
Step2
Now, set the permissions of the hostsentry.ignore file to be (0600/-rw-------) and
owned by the super-user ‘root’ for security reasons.
• To change the permissions and ownership of the hostsentry.ignore file, use:
[root@deep /]# chmod 600 /etc/hostsentry/hostsentry.ignore
[root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.ignore
/etc/hostsentry/hostsentry.action: The HostSentry Action File
The hostsentry.action file is used to define actions we want HostSentry to take on the
specified module. In our example we inform it to logs, blocks the route connection, blocks the TCP
connection and disables the user access. It seems that this feature is not implemented but we
define and configure it for future version.
Step 1
By default, the hostsentry.action file doesn’t exist after installation, so we have to create it
manually.
• Create the hostsentry.action file (touch /etc/hostsentry/hostsentry.action)
and add in any action you want to be taken. Below is what we recommend you enter.
moduleFirstLogin=log,blockRoute,blockTCP,disable
Step2
Now, set the permissions of the hostsentry.action file to be (0600/-rw-------) and
owned by the super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the hostsentry.action file, use:
[root@deep /]# chmod 600 /etc/hostsentry/hostsentry.action
[root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.action](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-475-320.jpg)
![HostSentry 2
CHAPTER 0
476
/etc/hostsentry/hostsentry.modules: The HostSentry Modules File
The hostsentry.modules file tells HostSentry what modules to execute on login/logout and
in what order. If you don't want a particular module to run for whatever reason (false alarms, not
interested, etc.) then delete it here.
Step 1
By default, the hostsentry.modules file doesn’t exist after installation, so we have to create it.
• Create hostsentry.modules file (touch /etc/hostsentry/hostsentry.modules)
and add the following module lines to the file. Below is what we recommend.
moduleLoginLogout
moduleFirstLogin
moduleForeignDomain
moduleMultipleLogins
moduleRhostsCheck
moduleHistoryTruncated
moduleOddDirnames
Step2
Now, set the permissions of the hostsentry.modules file to be (0600/-rw-------) and
owned by the super-user ‘root’ for security reasons.
• To change the permission and ownership of the hostsentry.modules file, use:
[root@deep /]# chmod 600 /etc/hostsentry/hostsentry.modules
[root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.modules
/etc/hostsentry/moduleForeignDomain.allow
The moduleForeignDomain.allow file is used to list all domains from which we don’t want an
alert to be sent to the administrator when they log in to the system. Every domain listed in this file
will be processed as allowed domains. I recommend you only add your localhost to this file.
Step 1
By default, the moduleForeignDomain.allow file doesn’t exist after installation and we have
to create it.
• Create the moduleForeignDomain.allow file (touch
/etc/hostsentry/moduleForeignDomain.allow) and add the following line.
:0.0
Step2
Now, set the permissions of the moduleForeignDomain.allow file to be (0600/-rw-------)
and owned by the super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the moduleForeignDomain.allow
file, use the following commands:
[root@deep /]# chmod 600 /etc/hostsentry/moduleForeignDomain.allow
[root@deep /]# chown 0.0 /etc/hostsentry/moduleForeignDomain.allow](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-476-320.jpg)
![HostSentry 2
CHAPTER 0
477
/etc/hostsentry/moduleMultipleLogins.allow
The moduleMultipleLogins.allow file is used to list all hosts from which multiple loggings
are allowed. This mean that all hosts listed in this file will be allowed to make multiple connections
from different place without an alert to be send to the administrator. Again, I recommend you to
only add your localhost to this file as we do below.
Step 1
By default, the moduleMultipleLogins.allow file does not exist after installation; therefore
we have to create it.
• Create the moduleMultipleLogins.allow file (touch
/etc/hostsentry/moduleMultipleLogins.allow) and add the following line.
Below is what we recommend.
# Place hosts in here you want this module to disregard logins from.
localhost
Step2
Now, set the permissions of the moduleMultipleLogins.allow file to be (0600/-rw-------)
and owned by the super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the moduleMultipleLogins.allow
file, use the following commands:
[root@deep /]# chmod 600 /etc/hostsentry/moduleMultipleLogins.allow
[root@deep /]# chown 0.0 /etc/hostsentry/moduleMultipleLogins.allow
/etc/init.d/hostsentry: The HostSentry Initialization File
The /etc/init.d/hostsentry script file is responsible to automatically starting and stopping
the HostSentry server on your Linux system.
Please note that the following script is suitable for Linux operating systems that use SystemV. If
you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to
make it work for you.
Step 1
Create the hostsentry script file (touch /etc/init.d/hostsentry) and add the following
lines inside it:
#!/bin/bash
# This shell script takes care of starting and stopping HostSentry.
#
# chkconfig: 345 98 85
# description: HostSentry is a host based intrusion detection tool that
# performs Login Anomaly Detection (LAD). This tool allows
# administrators to spot strange login behavior and quickly
# respond to compromised accounts and unusual behavior.
#
# processname: hostsentry
# config: /etc/hostsentry/hostsentry.conf
# pidfile: /var/run/hostsentry.pid
# Source function library.
. /etc/init.d/functions](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-477-320.jpg)
![HostSentry 2
CHAPTER 0
478
# Source networking configuration.
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
RETVAL=0
prog="HostSentry"
start() {
if [ -f /var/run/hostsentry.pid ] ; then
pid=`cat /var/run/hostsentry.pid`
if [ "$pid" != "" ] ; then
echo $"HostSentry is already running"
exit 0
fi
fi
echo -n $"Starting $prog: "
cd /usr/lib/hostsentry
daemon python hostsentry.py
RETVAL=$?
echo
echo `ps aux | grep "python hostsentry.py" | cut --delimiter=" " -f 7`
>
/var/run/hostsentry.pid
[ $RETVAL -eq 0 ] && touch /var/lock/subsys/hostsentry
return $RETVAL
}
stop() {
echo -n $"Shutting down $prog: "
cd /usr/lib/hostsentry
killproc python hostsentry.py
RETVAL=$?
echo
[ $RETVAL -eq 0 ] && rm -f /var/lock/subsys/hostsentry && rm -f
/var/run
/hostsentry.pid
return $RETVAL
}
restart() {
stop
start
}
condrestart() {
if [ -f /var/lock/subsys/hostsentry ]; then
restart
fi
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;
restart)
restart](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-478-320.jpg)
![HostSentry 2
CHAPTER 0
479
;;
condrestart)
condrestart
;;
*)
echo $"Usage: $0 {start|stop|restart|condrestart}"
exit 1
;;
esac
Step 2
Once the /etc/init.d/hostsentry script file has been created, it is important to make it
executable, change its default permissions, create the necessary links and start it. Making this file
executable will allow the system to run it, changing its default permission is to allow only the root
user to change this file for security reason, and creation of the symbolic links will let your system
start the program automatically for you at each system boot.
• To make this script executable and to change its default permissions, use the commands:
[root@deep /]# chmod 700 /etc/init.d/hostsentry
[root@deep /]# chown 0.0 /etc/init.d/hostsentry
• To create the symbolic rc.d links for HostSentry, use the following commands:
[root@deep /]# chkconfig --add hostsentry
[root@deep /]# chkconfig --level 345 hostsentry on
• To start HostSentry software manually, use the following command:
[root@deep /]# /etc/init.d/hostsentry start
Starting HostSentry: [OK]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-479-320.jpg)
![PortSentry 2
CHAPTER 1
484
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install PortSentry, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > PortSentry1
• And the following one after you install the software:
[root@deep root]# find /* > PortSentry2
• Then use the following command to get a list of what changed:
[root@deep root]# diff PortSentry1 PortSentry2 > PortSentry-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing PortSentry
Below are the steps that you must make to configure, compile and optimize the PortSentry
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp portsentry-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf portsentry-version.tar.gz
Step 2
In order to check that the version of PortSentry, which you are going to install, is an original
and unmodified one, use the command described below to check its MD5 hashes checksum.
• To verify the MD5 checksum of PortSentry, use the following command:
[root@deep tmp]# md5sum portsentry-1.1.tar.gz
This should yield an output similar to this:
782839446b7eca554bb1880ef0882670 portsentry-1.1.tar.gz
Now check that this checksum is exactly the same as the one published on the PortSentry
website at the following URL: http://www.psionic.com/downloads/checksums.md5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-484-320.jpg)
![PortSentry 2
CHAPTER 1
485
Step 3
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to configure the program for our PATH environment variable under
Linux. Therefore, move into the newly created PortSentry source directory and perform the
following steps to configure and optimize the software for your system.
• To move into the newly created PortSentry directory use the following command:
[root@deep tmp]# cd portsentry-1.1/
Step 4
Here, we have to change default locations of different PortSentry configuration files on the
system and add the required optimization FLAGS for our CPU architecture. To make these
modifications, we must edit the Makefile script file as follows.
• Edit the Makefile file and change all of the targeted lines in the order shown below:
a) vi +29 Makefile and change the line:
CFLAGS = -O -Wall
To read:
CFLAGS = -O2 -march=i686 -funroll-loops -Wall
b) vi +40 Makefile and change the lines:
INSTALLDIR = /usr/local/psionic
CHILDDIR=/portsentry
To read:
INSTALLDIR = /usr/sbin
CONFIGDIR = /etc/portsentry
c) vi +71 Makefile and change the line:
@if [ ! -d $(INSTALLDIR) ]; then /bin/mkdir $(INSTALLDIR); fi
To read:
@if [ ! -d $(INSTALLDIR) ]; then /bin/mkdir -p $(INSTALLDIR); fi
d) vi +73 Makefile and change the lines:
@if [ "$(INSTALLDIR)" = "/usr/local/psionic" ]; then /bin/chmod 700 $(INSTALLDIR) ; fi
@echo "Creating portsentry directory $(INSTALLDIR)$(CHILDDIR)"
@if [ ! -d $(INSTALLDIR)$(CHILDDIR) ]; then /bin/mkdir $(INSTALLDIR)$(CHILDDIR); fi
To read:
@if [ "$(INSTALLDIR)" = "/usr/sbin" ]; then /bin/chmod 700 $(INSTALLDIR) ; fi
@echo "Creating portsentry directory $(CONFIGDIR)"
@if [ ! -d $(CONFIGDIR) ]; then /bin/mkdir -p $(CONFIGDIR); fi](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-485-320.jpg)
![PortSentry 2
CHAPTER 1
487
Step 5
The second file that we will modify is the portsentry_config.h header file. In this file, we will
change the default install location of the configuration file for PortSentry.
• Edit the portsentry_config.h file (vi +32 portsentry_config.h) and change
the following line:
#define CONFIG_FILE "/usr/local/psionic/portsentry/portsentry.conf"
To read:
#define CONFIG_FILE "/etc/portsentry/portsentry.conf"
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the PortSentry software:
[root@deep portsentry-1.1]# cd
[root@deep root]# find /* > PortSentry1
[root@deep root]# cd /var/tmp/portsentry-1.1/
[root@deep portsentry-1.1]# make linux
[root@deep portsentry-1.1]# make install
[root@deep portsentry-1.1]# strip /usr/sbin/portsentry
[root@deep portsentry-1.1]# mkdir -m0700 /var/portsentry
[root@deep portsentry-1.1]# cd
[root@deep root]# find /* > PortSentry2
[root@deep root]# diff PortSentry1 PortSentry2 > PortSentry-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 7
Once the configuration, optimization, compilation, and installation of the PortSentry software
have been accomplished, we can free up some disk space by deleting the program tar archive
and the related source directory since they are no longer needed.
• To delete PortSentry and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf portsentry-version/
[root@deep tmp]# rm -f portsentry-version.tar.gz
Configuring PortSentry
After PortSentry has been built and installed successfully on your system, your next step is to
configure and customize its configuration files to fit your needs.
/etc/portsentry/portsentry.conf: (The PortSentry Configuration File)
/etc/portsentry/portsentry.ignore: (The PortSentry Ignore File)
/etc/portsentry/portsentry.modes: (The PortSentry Modes File)
/etc/init.d/portsentry: (The PortSentry Initialization File)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-487-320.jpg)
![PortSentry 2
CHAPTER 1
492
Step2
Now, set the permissions of the portsentry.modes file to be (0600/-rw-------) and owned
by the super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the portsentry.modes file, use:
[root@deep /]# chmod 600 /etc/portsentry/portsentry.modes
[root@deep /]# chown 0.0 /etc/portsentry/portsentry.modes
/etc/init.d/portsentry: The PortSentry Initialization File
The /etc/init.d/portsentry script file is responsible to automatically starting and stopping
the PortSentry server on your Linux system.
Please note that the following script is suitable for Linux operating systems that use SystemV. If
you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to
make it work for you.
Step 1
Create the portsentry script file (touch /etc/init.d/portsentry) and add the following
lines inside it:
#!/bin/bash
# This shell script takes care of starting and stopping the Port Scan Detector.
#
# chkconfig: 345 98 05
# description: PortSentry Port Scan Detector is part of the Abacus Project
# suite of tools. The Abacus Project is an initiative to release
# low-maintenance, generic, and reliable host based intrusion
# detection software to the Internet community.
#
# processname: portsentry
# config: /etc/portsentry/portsentry.conf
# pidfile: /var/run/portsentry.pid
# Source function library.
. /etc/rc.d/init.d/functions
# Source networking configuration.
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
RETVAL=0
prog="PortSentry"
start() {
SENTRYDIR=/etc/portsentry
if [ -s $SENTRYDIR/portsentry.modes ] ; then
modes=`cut -d "#" -f 1 $SENTRYDIR/portsentry.modes`
else
modes="tcp udp"
fi
for i in $modes ; do
action "Starting $prog -$i: " /usr/sbin/portsentry -$i
RETVAL=$?](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-492-320.jpg)
![PortSentry 2
CHAPTER 1
493
done
echo
[ $RETVAL = 0 ] && touch /var/lock/subsys/portsentry
return $RETVAL
}
stop() {
echo -n "Shutting down $prog: "
killproc portsentry
RETVAL=$?
echo
[ $RETVAL = 0 ] && rm -f /var/lock/subsys/portsentry
return $RETVAL
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;
restart|reload)
stop
start
RETVAL=$?
;;
condrestart)
[ -f /var/lock/subsys/portsentry ] && restart || :
;;
status)
status portsentry
;;
*)
echo "Usage: portsentry {start|stop|restart|reload|condrestart|status}"
exit 1
esac
Step 2
Once the /etc/init.d/portsentry script file has been created, it is important to make it
executable, change its default permissions, create the necessary links and start it. Making this file
executable will allow the system to run it, changing its default permission is to allow only the root
user to change this file for security reasons, and creation of the symbolic links to start the
program automatically for you at each system boot.
• To make this script executable and to change its default permissions, use the command:
[root@deep /]# chmod 700 /etc/init.d/portsentry
[root@deep /]# chown 0.0 /etc/init.d/portsentry
• To create the symbolic rc.d links for PortSentry, use the following command:
[root@deep /]# chkconfig --add portsentry
[root@deep /]# chkconfig --level 345 portsentry on
• To start PortSentry software manually, use the following command:
[root@deep /]# /etc/init.d/portsentry start
Starting PortSentry: [OK]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-493-320.jpg)
![PortSentry 2
CHAPTER 1
494
Removing hosts that have been blocked by PortSentry
When PortSentry want to blocks attackers it uses the “route” command of Linux to do it.
Every host that has been blocked by PortSentry will appear under the ‘route” command.
Below, I show you how to remove hosts that have been blocked by PortSentry from your
system.
Step 1
We have to use the “route” command to list which hosts are presently blocked by the program.
The “route” command also lists other important information about your network routing but we
use it in this example to get the list of blocked hosts and to unlock them from the system.
• To list which hosts are presently blocked by PortSentry, use the command:
[root@deep /]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
www.hack.com – 255.255.255.255 !H 0 – 0 -
207.35.78.0 * 255.255.255.224 U 0 0 0 eth0
127.0.0.0 * 255.0.0.0 U 0 0 0 lo
default rt.openna.c 0.0.0.0 UG 0 0 0 eth0
In the above example, we can see that “www.hack.com” is listed into our routing table as a
domain that has been blocked by PortSentry because it tried to scan our system. The “-”
string inform us about the fact that this host is locked. Every host in the routing table with this
string “-” is marked as blocked by the system.
Step2
Now that we know about the host that has been blocked by PortSentry, we can decide to
remove it from the list of blocked hosts on our system.
• To remove the blocked host in question, use the following command:
[root@deep /]# route del –host www.hack.com reject
The above command will remove www.hack.com from the list of blocked hosts in the routing table
of our system. The option “del” in the “route” command is what makes it possible to remove the
host from the list. Your have to use the above command for any additional hosts that you want to
remove from the routing table.
Step 3
Finally, we have to edit the portsentry.history file and remove the line corresponding to
www.hack.com from the file. This is important for PortSentry to be able to add the site into the
list of blocked host in the event that the corresponding host tries to scan your system again.
• Edit the portsentry.history file and remove the line corresponding to the host:
[root@deep /]# vi /var/portsentry/portsentry.history
1020371099 - 05/02/2002 16:24:59 Host: 1.2.3.4/1.2.3.4 Port: 80 TCP Blocked](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-494-320.jpg)
![Snort 2
CHAPTER 2
498
Packages
The following is based on information listed by Snort as of 2002/07/08. Please check
http://www.snort.org/ regularly for the latest status. We chose to install from source because it
provides the facility to fine tune the installation.
Source code is available from:
Snort Homepage: http://www.snort.org/
You must be sure to download: snort-1.8.7.tar.gz
Prerequisites
Snort requires that the listed software below be already installed on your system to be able to
compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source
archive files. Please make sure you have this program installed on your machine before you
proceed with this chapter.
Libpcap, which is used extensively by Snort, must already be installed on your system.
Tcpdump, which allows some additional features with Snort.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install Snort, and then
one afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Snort1
• And the following one after you install the software:
[root@deep root]# find /* > Snort2
• Then use the following command to get a list of what changed:
[root@deep root]# diff Snort1 Snort2 > Snort-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-498-320.jpg)
![Snort 2
CHAPTER 2
499
Compiling - Optimizing & Installing Snort
Below are the steps that you must make to configure, compile and optimize the Snort software
before installing it on your system. First off, we install the program as user 'root' so as to avoid
any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp snort-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf snort-version.tar.gz
Step 2
In order to check that the version of Snort, which you are going to install, is an original and
unmodified one, use the command described below to check its MD5 hashes checksum.
• To verify the MD5 checksum of Snort, use the following command:
[root@deep tmp]# md5sum snort-1.8.7.tar.gz
This should yield an output similar to this:
29c81d0bc243edb21ba4ab33ee80457e snort-1.8.7.tar.gz
Now check that this checksum is exactly the same as the one published on the Snort website at
the following URL: http://www.snort.org/dl/snort-1.8.7.tar.gz.md5
Step 3
Snort needs a UID and GID to properly run on the system but this UID/GID cannot run as
super-user root; therefore we must create a special user with no shell privileges on the system for
running Snort daemon.
• To create this special Snort user on OpenNA Linux, use the following command:
[root@deep tmp]# groupadd -g 69 snort > /dev/null 2>&1 || :
[root@deep tmp]# useradd -c "Snort NIDS" -d /var/log/snort -g 69 -s
/bin/false -u 69 snort > /dev/null 2>&1 || :
• To create this special Snort user on Red Hat Linux, use the following command:
[root@deep tmp]# groupadd -g 69 snort > /dev/null 2>&1 || :
[root@deep tmp]# useradd -u 69 -g 69 -s /bin/false -M -r -d
/var/log/snort snort > /dev/null 2>&1 || :
The above command will create a null account, with no password, no valid shell, no files owned-
nothing but a UID and a GID for the program. Remember that Snort daemon does not need to
have a shell account on the server.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-499-320.jpg)
![Snort 2
CHAPTER 2
500
Step 4
Now, edit the shells file (vi /etc/shells) and add a non-existent shell name
“/bin/false”, which is the one we used in the passwd command above.
[root@deep tmp]# vi /etc/shells
/bin/bash2
/bin/bash
/bin/sh
/bin/false This is our added no-existent shell
Step 5
Next, move into the newly created Snort source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created Snort directory use the following command:
[root@deep tmp]# cd snort-1.8.7/
• To configure and optimize Snort use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--sysconfdir=/etc
--localstatedir=/var
--mandir=/usr/share/man
--with-openssl
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the Snort software:
[root@deep snort-1.8.7]# make
[root@deep snort-1.8.7]# cd
[root@deep root]# find /* > Snort1
[root@deep root]# cd /var/tmp/snort-1.8.7/
[root@deep snort-1.8.7]# make install
[root@deep snort-1.8.7]# mkdir -p /var/log/snort
[root@deep snort-1.8.7]# mkdir -p /etc/snort
[root@deep snort-1.8.7]# chown -R snort.snort /var/log/snort/
[root@deep snort-1.8.7]# install classification.config /etc/snort/
[root@deep snort-1.8.7]# install snort.conf *.rules /etc/snort/
[root@deep snort-1.8.7]# chmod 0644 /etc/snort/*
[root@deep snort-1.8.7]# strip /usr/bin/snort
[root@deep snort-1.8.7]# cd
[root@deep root]# find /* > Snort2
[root@deep root]# diff Snort1 Snort2 > Snort-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-500-320.jpg)
![Snort 2
CHAPTER 2
501
Step 7
Once the configuration, optimization, compilation, and installation of the Snort software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete Snort and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf snort-version/
[root@deep tmp]# rm -f snort-version.tar.gz
Configuring Snort
After Snort has been built and installed successfully in your system, your next step is to
configure and customize its configuration files to fit your needs.
/etc/snort/snort.conf: (The Snort Configuration File)
/etc/init.d/snort: (The Snort Initialization File)
/etc/snort/snort.conf: The Snort Config File
The snort.conf file is the main configuration file for Snort. It is in this file that Snort gets all
of its startup information and the way it should run on your system. You can edit it to specify your
network variables, preprocessors parameters, output plugging to use and Snort rules files.
The Snort configuration file is divided into four different sections. The first section is used to
define network variables, the second section is used to configure the preprocessor parameters
that Snort should use, the third section is used to configure output plugging to activate, and the
last section is used to enable specific Snort rule set. Below, we will explain each section and
how you should use them to configure Snort for your server.
• Edit the snort.conf file (vi /etc/snort/snort.conf) and set your needs. Below
is what we recommend you.
var HOME_NET $eth0_ADDRESS
var EXTERNAL_NET any
var SMTP $HOME_NET
var HTTP_SERVERS $HOME_NET
var SQL_SERVERS $HOME_NET
var DNS_SERVERS $HOME_NET
var RULE_PATH ./
preprocessor frag2
preprocessor stream4: detect_scans,detect_state_problems
preprocessor stream4_reassemble: both,ports all
preprocessor http_decode: 80
preprocessor rpc_decode: 111 32771
preprocessor bo
preprocessor telnet_decode
preprocessor portscan: $HOME_NET 4 3 portscan.log
preprocessor portscan-ignorehosts: 207.35.78.40 207.35.78.41
output alert_syslog: LOG_AUTH LOG_ALERT
include classification.config
This tells the snort.conf file to set itself up for this particular configuration with:](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-501-320.jpg)
![Snort 2
CHAPTER 2
506
/etc/init.d/snort: The Snort Initialization File
The /etc/init.d/snort script file is responsible to automatically starting and stopping the
Snort server on your Linux system. Please note that the following script is suitable for Linux
operating systems that use SystemV. If you Linux system use some other methods like BSD,
you’ll have to adjust the script bellow to make it work for you.
Step 1
Create the snort script file (touch /etc/init.d/snort) and add the following lines inside it:
#!/bin/bash
# This shell script takes care of starting and stopping the snort IDS daemon.
#
# chkconfig: 2345 40 60
# description: Snort is a lightweight network intrusion detection tool that
# currently detects more than 1100 host and network
# vulnerabilities, portscans, backdoors, and more.
# Source function library.
. /etc/init.d/functions
# Source networking configuration.
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
# Specify your network interface here
INTERFACE=eth0
RETVAL=0
prog="Snort"
start() {
echo -n $"Starting $prog: "
daemon /usr/bin/snort -A fast -u snort -g snort -b -s -z -d -D
-i $INTERFACE -c /etc/snort/snort.conf
RETVAL=$?
echo
[ $RETVAL = 0 ] && touch /var/lock/subsys/snort
return $RETVAL
}
stop() {
echo -n $"Shutting down $prog: "
killproc snort
RETVAL=$?
echo
[ $RETVAL = 0 ] && rm -f /var/lock/subsys/snort
return $RETVAL
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-506-320.jpg)
![Snort 2
CHAPTER 2
507
status)
status snort
;;
restart)
stop
start
;;
condrestart)
[ -f /var/lock/subsys/snort ] && restart
;;
*)
echo $"Usage: $prog {start|stop|status|restart|condrestart}"
exit 1
esac
exit $RETVAL
Step 2
Once the /etc/init.d/snort script file has been created, it is important to make it
executable, change its default permissions, create the necessary links and start it. Making this file
executable will allow the system to run it, changing its default permission to allow only the root
user to change this file for security reasons, and creation of the symbolic links will let the process
control initialization of Linux start the program automatically for you at each system boot.
• To make this script executable and to change its default permissions, use the command:
[root@deep /]# chmod 700 /etc/init.d/snort
[root@deep /]# chown 0.0 /etc/init.d/snort
• To create the symbolic rc.d links for Snort, use the following command:
[root@deep /]# chkconfig --add snort
[root@deep /]# chkconfig --level 2345 snort on
• To start Snort software manually, use the following command:
[root@deep /]# /etc/init.d/snort start
Starting Snort: [OK]
Running Snort in a chroot jail
This section applies only if you want to run Snort in chroot jail environment. To do it, we need to
create the required skeleton environment and copy necessary files into this chroot jail. Below are
the steps to follow if you want to run Snort with chroot jail support.
The main benefit of a chroot jail is that the jail will limit the portion of the file system the daemon
can see to the root directory of the jail. Additionally, since the jail only needs to support Snort,
the programs available in the jail can be extremely limited. Most importantly, there is no need for
setuid-root programs, which can be used to gain root access and break out of the jail.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-507-320.jpg)
![Snort 2
CHAPTER 2
508
Necessary steps to run Snort in a chroot jail:
What you're essentially doing is creating a skeleton root file system with just enough components
necessary (files, directories) to allow UNIX to do a chroot when Snort starts. The procedures to
run Snort in chroot jail are really easy to accomplish as follows.
Step 1
First, we have to create all the necessary chrooted environment subdirectories where we will
move Snort files and directories.
• Use the following command to create all the necessary chroot subdirectories.
[root@deep /]# mkdir -p /chroot/snort/etc/snort
[root@deep /]# mkdir -p /chroot/snort/var/log/snort
[root@deep /]# chown -R snort.snort /chroot/snort/var/log/snort
Step 2
Now, it is time to move the required Snort files to the related subdirectories in the chroot area for
Snort to work. We can copy these files to the chroot jail but it’s better to move them to avoid
unnecessary duplication of Snort files on the server.
• Use the following commands to move the require files into the chroot area.
[root@deep /]# mv /etc/snort/* /chroot/snort/etc/snort/
[root@deep /]# chmod 0644 /chroot/snort/etc/snort/*
Step 3
Once the Snort files have been moved to the chroot location, we can remove the old Snort
directories from the system since they are no longer required.
• This can be done with the following commands.
[root@deep /]# rm -rf /etc/snort/
[root@deep /]# rm -rf /var/log/snort/
Step 4
Next, we have to recreate a new snort initialization script file which starts Snort in the chroot
environment. Please note that the following script is suitable for Linux operating systems that use
SystemV. If you Linux system use some other method like BSD, you’ll have to adjust the script
below to make it work for you. The only difference with the previous Snort initialization script file
is that we use the “-t” option of Snort to specify the chroot location.
Edit the snort script file (vi /etc/init.d/snort) and add the following lines inside it:
#!/bin/bash
# This shell script takes care of starting and stopping the snort IDS daemon.
#
# chkconfig: 2345 40 60
# description: Snort is a lightweight network intrusion detection tool that
# currently detects more than 1100 host and network
# vulnerabilities, portscans, backdoors, and more.
# Source function library.
. /etc/init.d/functions
# Source networking configuration.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-508-320.jpg)
![Snort 2
CHAPTER 2
509
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
# Specify your network interface here
INTERFACE=eth0
RETVAL=0
prog="Snort"
start() {
echo -n $"Starting $prog: "
daemon /usr/bin/snort -A fast -u snort -g snort -b -s -z -d -D
-i $INTERFACE -c /etc/snort/snort.conf -t /chroot/snort/
RETVAL=$?
echo
[ $RETVAL = 0 ] && touch /var/lock/subsys/snort
return $RETVAL
}
stop() {
echo -n $"Shutting down $prog: "
killproc snort
RETVAL=$?
echo
[ $RETVAL = 0 ] && rm -f /var/lock/subsys/snort
return $RETVAL
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;
status)
status snort
;;
restart)
stop
start
;;
condrestart)
[ -f /var/lock/subsys/snort ] && restart
;;
*)
echo $"Usage: $prog {start|stop|status|restart|condrestart}"
exit 1
esac
exit $RETVAL](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-509-320.jpg)
![Snort 2
CHAPTER 2
510
Step 5
Finally, we must test the new chrooted jail configuration of our Snort program.
• Start the new chrooted jail Snort with the following command:
[root@deep /]# /etc/init.d/snort start
Starting Snort: [OK]
• If you don't get any errors, do a ps ax | grep snort and see if we're running:
[root@deep /]# ps ax | grep snort
16295 ? R 0:38 /usr/bin/snort -A fast -u snort -g snort -b -s -z -d
If so, lets check to make sure it's chrooted by picking its process number and doing ls -la
/proc/that_process_number/root/.
[root@deep /]# ls -la /proc/16295/root/
If you see something like:
total 4
drwxr-xr-x 4 root root 4096 May 7 19:10 ./
drwxr-xr-x 5 root root 4096 May 7 19:12 ../
drwxr-xr-x 3 root root 4096 May 7 19:12 etc/
drwxr-xr-x 3 root root 4096 May 7 19:12 var/
Congratulations! Your Snort in a chroot jail is working.
Further documentation
For more details, there is one manual page about Snort that you should read:
$ man snort (8) - Open source network intrusion detection system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-510-320.jpg)
![Tripwire 2
CHAPTER 3
514
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install Tripwire, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Tripwire1
• And the following one after you install the software:
[root@deep root]# find /* > Tripwire2
• Then use the following command to get a list of what changed:
[root@deep root]# diff Tripwire1 Tripwire2 > Tripwire-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing Tripwire
Below are the steps that you must make to configure, compile and optimize the Tripwire
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp tripwire-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf tripwire-version.tar.gz
Step 2
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to fix many bugs with Tripwire. Therefore, move into the newly
created Tripwire source directory and perform the following steps to configure and optimize the
software for your system.
• To move into the newly created Tripwire directory use the following command:
[root@deep tmp]# cd tripwire-2.3.1-2/](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-514-320.jpg)
![Tripwire 2
CHAPTER 3
515
Step 3
The first source file to modify is called “mailmessage.cpp”.
• Edit the mailmessage.cpp file (vi +244 src/tripwire/mailmessage.cpp) and
change:
const TCHAR* szFormat = _T("%a %d %b %Y %H:%M:%S %z");
To read:
const TCHAR* szFormat = _T("%a, %d %b %Y %H:%M:%S %z");
Step 4
The second file is called “platform.h” and we have to edit it and add a new line as follows.
• Edit the platform.h file (vi +294 src/core/platform.h) and change the line:
#define USES_FHS IS_LINUX
To read:
#define USE_FHS IS_LINUX
Step 5
The last file to modify is very important for Linux systems with GCC version 3; which should be the
default compiler for most Linux system now. The modifications are important to allow Tripwire
to compile with GCC v3. There is one problem, the modifications are too big to be listed in a book
and we have to retrieve it from the Internet as a patch file and patch our sources code.
The patch is available from the OpenNA website at the following URL:
http://www.openna.com/products/books/securing-optimizing-linux/3rdedition/index.htm
Please, download the patch and patch your Tripwire source codes as follow:
• To patch your Tripwire source codes, use the command:
[root@deep /]# cp tripwire-gcc3.patch /var/tmp/
[root@deep /]# cd /var/tmp/tripwire-2.3.1-2/
[root@deep tripwire-2.3.1-2]# patch -p1 < ../tripwire-gcc3.patch](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-515-320.jpg)
![Tripwire 2
CHAPTER 3
516
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the Tripwire software:
[root@deep tripwire-2.3.1-2]# cd src/
[root@deep src]# rm -rf STLport*
[root@deep src]# touch STLport_r STLport_d
[root@deep src]# export CXXFLAGS="-O2 -march=i686 -funroll-loops"
[root@deep src]# make release
[root@deep src]# cd
[root@deep root]# find /* > Tripwire1
[root@deep root]# cd /var/tmp/tripwire-2.3.1-2/bin/i686-pc-linux_r/
[root@deep i686-pc-linux_r]# install -m0500 siggen /usr/sbin/
[root@deep i686-pc-linux_r]# install -m0500 tripwire /usr/sbin/
[root@deep i686-pc-linux_r]# install -m0500 twadmin /usr/sbin/
[root@deep i686-pc-linux_r]# install -m0500 twprint /usr/sbin/
[root@deep i686-pc-linux_r]# cd ../../man/
[root@deep man]# install -m0440 man4/*.4 /usr/share/man/man4/
[root@deep man]# install -m0440 man5/*.5 /usr/share/man/man5/
[root@deep man]# install -m0440 man8/*.8 /usr/share/man/man8/
[root@deep man]# mkdir -m0700 /etc/tripwire
[root@deep man]# mkdir -p /var/lib/tripwire/report
[root@deep man]# chmod -R 0700 /var/lib/tripwire/
[root@deep man]# cd
[root@deep root]# find /* > Tripwire2
[root@deep root]# diff Tripwire1 Tripwire2 > Tripwire-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 7
Once the configuration, optimization, compilation, and installation of the Tripwire software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete Tripwire and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf tripwire-version/
[root@deep tmp]# rm -f tripwire-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install Tripwire. It will also remove the Tripwire compressed archive from the /var/tmp
directory.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-516-320.jpg)
![Tripwire 2
CHAPTER 3
517
Configuring Tripwire
After Tripwire has been built and installed successfully in your system, your next step is to
configure and customize its configuration files and policy file to fit your needs.
/etc/tripwire/twcfg.txt: (The Tripwire Configuration File)
/etc/tripwire/twpol.txt: (The Tripwire Policy File)
/etc/tripwire/twinstall.sh: (The Tripwire Cryptographic File)
/etc/cron.weekly/tripwire.cron: (The Tripwire Cron File)
/etc/tripwire/twcfg.txt: The Tripwire Configuration File
The twcfg.txt file is a small Tripwire configuration file used by the program the first time you
run it to get information about locations of different files and the way Tripwire runs and reports
on the integrity of the system. It stores system-specific information, such as the location of
Tripwire data files. In general, we use this file ONE time and REMOVE it from the server once
Tripwire is configured.
Step 1
By default, the twcfg.txt file do not exist after installation, we have to create it as follow.
• Create the twcfg.txt file (touch /etc/tripwire/twcfg.txt) and add the
following lines inside the file. Below is what we recommend you.
ROOT =/usr/sbin
POLFILE =/etc/tripwire/tw.pol
DBFILE =/var/lib/tripwire/$(HOSTNAME).twd
REPORTFILE =/var/lib/tripwire/report/$(HOSTNAME)-$(DATE).twr
SITEKEYFILE =/etc/tripwire/site.key
LOCALKEYFILE =/etc/tripwire/$(HOSTNAME)-local.key
EDITOR =/bin/vi
LATEPROMPTING =true
LOOSEDIRECTORYCHECKING =true
MAILNOVIOLATIONS =false
EMAILREPORTLEVEL =3
REPORTLEVEL =3
MAILMETHOD =SENDMAIL
SYSLOGREPORTING =true
MAILPROGRAM =/usr/sbin/sendmail -oi -t
Step2
Now, set the permissions of the twcfg.txt file to be (0640/-rw-r-----) and owned by the
super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the twcfg.txt file, use:
[root@deep /]# chmod 640 /etc/tripwire/twcfg.txt
[root@deep /]# chown 0.0 /etc/tripwire/twcfg.txt](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-517-320.jpg)
![Tripwire 2
CHAPTER 3
522
(
rulename = "/chroot directory",
severity = $(SIG_HI)
)
{
/chroot -> $(SEC_CRIT) (recurse = -1) ;
}
Step2
Now, set the permissions of the twpol.txt file to be (0640/-rw-r-----) and owned by the
super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the twpol.txt file, use:
[root@deep /]# chmod 640 /etc/tripwire/twpol.txt
[root@deep /]# chown 0.0 /etc/tripwire/twpol.txt
NOTE: Please, add to the above policy file all files, binaries, directories that you want the software
to monitor for you. Remove any files, binaries, directories that you don’t want the software to
monitor for you and don’t send emails to the mailing list if you receive error messages about the
fact that some files, binaries, directories don’t exist on your system. Instead, review your policy
file and make the changes related to the error that you received, because in most cases, this is
why you have this kind of errors messages. Finally, reads the twpolicy manual page for more
information on the parameters of this policy file.
/etc/tripwire/twinstall.sh: The Tripwire Cryptographic File
The twinstall.sh file is a script file used by Tripwire to configure, install and generate
cryptographic keys used by Tripwire during operation and verification.
Step 1
This script file asks you the passpharse that you want to run with Tripwire as well as other
operations related to the Tripwire database generation and policies. By default, the
twinstall.sh file does not exist after the installation; we have to create it as follows.
• Create the twinstall.sh file (touch /etc/tripwire/twinstall.sh) and add the
following lines inside the file.
#!/bin/sh
HOST_NAME='localhost'
if uname -n > /dev/null 2> /dev/null ; then
HOST_NAME=`uname -n`
fi
# Site Passphrase variable
TW_SITE_PASS=""
# Complete path to site key
SITE_KEY="/etc/tripwire/site.key"
# Local Passphrase variable
TW_LOCAL_PASS=""
# Complete path to local key](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-522-320.jpg)
![Tripwire 2
CHAPTER 3
523
LOCAL_KEY="/etc/tripwire/${HOST_NAME}-local.key"
# If clobber==true, overwrite files; if false, do not overwrite files.
CLOBBER="false"
# If prompt==true, ask for confirmation before continuing with install.
PROMPT="true"
# Name of twadmin executeable
TWADMIN="twadmin"
# Path to twadmin executeable
TWADMPATH=@sbindir@
# Path to configuration directory
CONF_PATH="/etc/tripwire"
# Name of clear text policy file
TXT_POL=$CONF_PATH/twpol.txt
# Name of clear text configuration file
TXT_CFG=$CONF_PATH/twcfg.txt
# Name of encrypted configuration file
CONFIG_FILE=$CONF_PATH/tw.cfg
# Path of the final Tripwire policy file (signed)
SIGNED_POL=`grep POLFILE $TXT_CFG | sed -e 's/^.*=(.*)/1/'`
if [ -z "$TW_SITE_PASS" ] || [ -z "$TW_LOCAL_PASS" ]; then
cat << END_OF_TEXT
----------------------------------------------
The Tripwire site and local passphrases are used to
sign a variety of files, such as the configuration,
policy, and database files.
Passphrases should be at least 8 characters in length
and contain both letters and numbers.
See the Tripwire manual for more information.
END_OF_TEXT
fi
echo
echo "----------------------------------------------"
echo "Creating key files..."
if [ "$CLOBBER" = "true" ] && [ "$PROMPT" = "false" ] && [ -f "$SITE_KEY" ] ;
then
rm -f "$SITE_KEY"
fi
if [ -f "$SITE_KEY" ] && [ "$CLOBBER" = "false" ] ; then
echo "The site key file "$SITE_KEY""
echo 'exists and will not be overwritten.'
else
cmdargs="--generate-keys --site-keyfile "$SITE_KEY""
if [ -n "$TW_SITE_PASS" ] ; then
cmdargs="$cmdargs --site-passphrase "$TW_SITE_PASS""
fi
eval ""$TWADMPATH/$TWADMIN" $cmdargs"
if [ $? -ne 0 ] ; then](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-523-320.jpg)
![Tripwire 2
CHAPTER 3
524
echo "Error: site key generation failed"
exit 1
else chmod 640 "$SITE_KEY"
fi
fi
if [ "$CLOBBER" = "true" ] && [ "$PROMPT" = "false" ] && [ -f "$LOCAL_KEY" ] ;
then
rm -f "$LOCAL_KEY"
fi
if [ -f "$LOCAL_KEY" ] && [ "$CLOBBER" = "false" ] ; then
echo "The site key file "$LOCAL_KEY""
echo 'exists and will not be overwritten.'
else
cmdargs="--generate-keys --local-keyfile "$LOCAL_KEY""
if [ -n "$TW_LOCAL_PASS" ] ; then
cmdargs="$cmdargs --local-passphrase "$TW_LOCAL_PASS""
fi
eval ""$TWADMPATH/$TWADMIN" $cmdargs"
if [ $? -ne 0 ] ; then
echo "Error: local key generation failed"
exit 1
else chmod 640 "$LOCAL_KEY"
fi
fi
echo
echo "----------------------------------------------"
echo "Signing configuration file..."
if [ "$CLOBBER" = "false" ] && [ -s "$CONFIG_FILE" ] ; then
backup="${CONFIG_FILE}.$$.bak"
echo "Backing up $CONFIG_FILE"
echo " to $backup"
`mv "$CONFIG_FILE" "$backup"`
if [ $? -ne 0 ] ; then
echo "Error: backup of configuration file failed."
exit 1
fi
fi
cmdargs="--create-cfgfile"
cmdargs="$cmdargs --cfgfile "$CONFIG_FILE""
cmdargs="$cmdargs --site-keyfile "$SITE_KEY""
if [ -n "$TW_SITE_PASS" ] ; then
cmdargs="$cmdargs --site-passphrase "$TW_SITE_PASS""
fi
eval ""$TWADMPATH/$TWADMIN" $cmdargs "$TXT_CFG""
if [ $? -ne 0 ] ; then
echo "Error: signing of configuration file failed."
exit 1
fi
# Set the rights properly
chmod 640 "$CONFIG_FILE"
cat << END_OF_TEXT
A clear-text version of the Tripwire configuration file
$TXT_CFG
has been preserved for your inspection. It is recommended](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-524-320.jpg)
![Tripwire 2
CHAPTER 3
525
that you delete this file manually after you have examined it.
END_OF_TEXT
echo
echo "----------------------------------------------"
echo "Signing policy file..."
if [ "$CLOBBER" = "false" ] && [ -s "$POLICY_FILE" ] ; then
backup="${POLICY_FILE}.$$.bak"
echo "Backing up $POLICY_FILE"
echo " to $backup"
mv "$POLICY_FILE" "$backup"
if [ $? -ne 0 ] ; then
echo "Error: backup of policy file failed."
exit 1
fi
fi
cmdargs="--create-polfile"
cmdargs="$cmdargs --cfgfile "$CONFIG_FILE""
cmdargs="$cmdargs --site-keyfile "$SITE_KEY""
if [ -n "$TW_SITE_PASS" ] ; then
cmdargs="$cmdargs --site-passphrase "$TW_SITE_PASS""
fi
eval ""$TWADMPATH/$TWADMIN" $cmdargs "$TXT_POL""
if [ $? -ne 0 ] ; then
echo "Error: signing of policy file failed."
exit 1
fi
# Set the proper rights on the newly signed policy file.
chmod 0640 "$SIGNED_POL"
cat << END_OF_TEXT
A clear-text version of the Tripwire policy file
$TXT_POL
has been preserved for your inspection. This implements
a minimal policy, intended only to test essential
Tripwire functionality. You should edit the policy file
to describe your system, and then use twadmin to generate
a new signed copy of the Tripwire policy.
END_OF_TEXT
Step 2
Now, set the permissions of the twinstall.sh file to be (0500/---x------) and owned by
the super-user ‘root’ for security reasons.
• This procedure can be accomplished with the following command:
[root@deep /]# chmod 500 /etc/tripwire/twinstall.sh
[root@deep /]# chown 0.0 /etc/tripwire/twinstall.sh
NOTE: The above script file can also be retrieved from the following URL:
http://www.openna.com/products/books/securing-optimizing-linux/3rdedition/index.htm](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-525-320.jpg)
![Tripwire 2
CHAPTER 3
526
/etc/cron.weekly/tripwire.cron: The Tripwire Cron File
The tripwire.cron file is a small script executed automatically by the crond program each
week to scan your hard disk for possible changes to files or directories and mail the results to the
system administrator.
Step 1
This script will automate the procedure of integrity checking for you. If you want to automate this
task, follow the simple steps below.
• Create the tripwire.cron script file (touch /etc/cron.weekly/tripwire.cron)
and add the following lines:
#!/bin/sh
HOST_NAME=`uname -n`
if [ ! -e /var/lib/tripwire/${HOST_NAME}.twd ] ; then
echo "**** Error: Tripwire database for ${HOST_NAME} not found. ****"
echo "**** Run "/etc/tripwire/twinstall.sh" and/or "tripwire --init". ****"
else
test -f /etc/tripwire/tw.cfg && /usr/sbin/tripwire --check
fi
Step 2
Now, set the permissions of the tripwire.cron file to be (0500/---x------) and owned by
the super-user ‘root’ for security reasons.
• This procedure can be accomplished with the following command:
[root@deep /]# chmod 500 /etc/cron.weekly/tripwire.cron
[root@deep /]# chown 0.0 /etc/cron.weekly/tripwire.cron
Running Tripwire for the first time
Once all the files are installed and configured on your server, it’s time to run Tripwire to
generate the cryptography key, passphrases and database files. These procedures should be
made the first time you install the software and ONLY the first time.
Step 1
Here we begin by running the twinstall.sh script file which will generate the cryptography
keys and will ask us to enter our passphrase (password) which is required each time we want to
update and accept Tripwire integrity reports.
• To run the twinstall.sh file use the following command:
[root@deep /]# /etc/tripwire/twinstall.sh
----------------------------------------------
The Tripwire site and local passphrases are used to
sign a variety of files, such as the configuration,
policy, and database files.
Passphrases should be at least 8 characters in length
and contain both letters and numbers.
See the Tripwire manual for more information.
----------------------------------------------](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-526-320.jpg)
![Tripwire 2
CHAPTER 3
527
Creating key files...
(When selecting a passphrase, keep in mind that good passphrases
typically have upper and lower case letters, digits and punctuation
marks, and are at least 8 characters in length.)
Enter the site keyfile passphrase: Your site keyfile passphrase
Verify the site keyfile passphrase: Your site keyfile passphrse again
Generating key (this may take several minutes)...Key generation complete.
(When selecting a passphrase, keep in mind that good passphrases
typically have upper and lower case letters, digits and punctuation
marks, and are at least 8 characters in length.)
Enter the local keyfile passphrase: Your local keyfile passphrase
Verify the local keyfile passphrase: Your local keyfile passphrase again
Generating key (this may take several minutes)...Key generation complete.
----------------------------------------------
Signing configuration file...
Please enter your site passphrase: Your site passphrase
Wrote configuration file: /etc/tripwire/tw.cfg
A clear-text version of the Tripwire configuration file
/etc/tripwire/twcfg.txt
has been preserved for your inspection. It is recommended
that you delete this file manually after you have examined it.
----------------------------------------------
Signing policy file...
Please enter your site passphrase: Your site passphrase
Wrote policy file: /etc/tripwire/tw.pol
A clear-text version of the Tripwire policy file
/etc/tripwire/twpol.txt
has been preserved for your inspection. This implements
a minimal policy, intended only to test essential
Tripwire functionality. You should edit the policy file
to describe your system, and then use twadmin to generate
a new signed copy of the Tripwire policy
Step 2
Once our passphrase keyfiles have been generated, it’s time to run Tripwire in its’ initialization
mode. The initialization mode will create the initial Tripwire database files based on what
information has been provided inside the twpol.txt file. Tripwire must have a database to
compare against, so we first create the file information database. This action will create a file
called “tw.db_[hostname]” in the directory you specified to hold your databases (where
[hostname] will be replaced with your machine hostname).
• To run the Tripwire in initialization mode, use the following command:
[root@deep /]# tripwire --init
Parsing policy file: /etc/tripwire/tw.pol
Generating the database...
*** Processing Unix File System ***
Please enter your local passphrase:
Wrote database file: /var/lib/tripwire/deep.twd
The database was successfully generated.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-527-320.jpg)
![Tripwire 2
CHAPTER 3
528
NOTE: Initialization of the database Tripwire uses should be done manually because the key
used to sign the database should be different for each system.
Step 3
Finally, if you have not received any kind of error message, then you can safety remove the
twcfg.txt and twpol.txt files from your system since they are no longer needed and it
would be a security risk to keep these files on your server.
• To remove the files from your system, use the following commands:
[root@deep /]# rm -f /etc/tripwire/twcfg.txt
[root@deep /]# rm -f /etc/tripwire/twpol.txt
NOTE: You have to remove the files from your server ONLY if you are sure that the initialization of
the databases has been completed without any errors. Otherwise you should keep these files and
regenerate a new database once all the errors have been fixed inside the twpol.txt file, since
in many cases errors come from twpol.txt file having some lines referring to files or directories
that do not exist in your system.
Securing Tripwire
It is highly recommended that the database (tw.db_[hostname]) file of Tripwire be moved
someplace (e.g. floppy) where it cannot be modified. This is important because data from
Tripwire is only as trustworthy as its database.
It is also recommended that you make a hardcopy printout of the database contents right away. In
the event that you become suspicious of the integrity of the database, you will be able to
manually compare information against this hardcopy.
Tripwire Administrative Tools
The commands listed below are some of the most used of this software, but many more exist.
Check the Tripwire manual pages for more details.
Running Tripwire in Interactive Checking Mode:
In “Interactive Checking Mode” feature, Tripwire verifies files or directories that have been
added, deleted, or changed from the original database and ask the user whether the database
entry should be updated. This mode is the most convenient way of keeping your database up-to-
date, but it requires that the user be "at the console". If you want to use this mode, then follow the
simple step below.
Once the file information database of Tripwire has been created, we can now run Tripwire in
“Interactive Checking Mode”. This mode will prompt the user for whether or not each changed
entry on the system should be updated to reflect the current state of the file.
• To run Tripwire in Interactive Checking Mode, use the following command:
[root@deep /]# tripwire --check --interactive
Parsing policy file: /etc/tripwire/tw.pol
*** Processing Unix File System ***
Performing integrity check...](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-528-320.jpg)
![ucspi-tcp 2
CHAPTER 4
536
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
installed files in the system in the event of an update in the future. To solve the problem, it is a
good idea to make a list of files on the system before you install ucsip-tcp, and one afterwards,
and then compares them using the diff utility to find out what files were placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > ucsip-tcp1
• And the following one after you install the software:
[root@deep root]# find /* > ucsip-tcp2
• Then use the following command to get a list of what changed:
[root@deep root]# diff ucsip-tcp1 ucsip-tcp2 > ucsip-tcp-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/85/Securing-optimizing-linux-the-hacking-solution-536-320.jpg)
![Preface
15
These installation instructions assume
You have a CD-ROM drive on your computer and the Official Red Hat Linux or OpenNA Linux
CD-ROM. Installations were tested on the Official Red Hat Linux version 7.3 and OpenNA Linux.
You should familiarize yourself with the hardware on which the operating system will be installed.
After examining the hardware, the rest of this document guides you, step-by-step, through the
installation process.
Obtaining the example configuration files
In a true server environment and especially when Graphical User Interface is not installed, we will
often use text files, scripts, shell, etc. Throughout this book we will see shell commands, script
files, configuration files and many other actions to execute on the terminal of the server. You can
enter them manually or use the compressed archive file that I made which contains all
configuration examples and paste them directly to your terminal. This seems to be useful in many
cases to save time.
The example configuration files in this book are available electronically via HTTP from this URL:
http://www.openna.com/products/books/securing-optimizing-linux/3rdedition/index.htm
• In either case, extract the files into your Linux server from the archive by typing:
[root@deep /]# cd /var/tmp
[root@deep tmp]# tar xzpf floppy-3.0.tgz
If you cannot get the examples from the Internet, please contact the author at this email address:
gmourani@openna.com
Problem with Securing & Optimizing Linux
When you encounter a problem in "Securing & Optimizing Linux" we want to hear about it. Your
reports are an important part in making the book more reliable, because even with the utmost
care we cannot guarantee that every part of the book will work on every platform under every
circumstance.
We cannot promise to fix every error right away. If the problem is obvious, critical, or affects a lot
of users, chances are that someone will look into it. It could also happen that we tell you to
update to a newer version to see if the problem persists there. Or we might decide that the
problem cannot be fixed until some major rewriting has been done. If you need help immediately,
consider obtaining a commercial support contract or try our Q&A archive from the mailing list for
an answer.
Below are some important links:
OpenNA web site: http://www.openna.com/
Mailing list: http://www.openna.com/support/mailing/
Support: http://www.openna.com/support/
RPM Download: http://www.openna.com/download/
Acknowledgments
I would like to thank all the OpenNA staff for their hard works and patience. A special gratitude
and many thanks to Colin Henry who made tremendous efforts to make this book grammatically
and orthographically sound in a professional manner. Adrian Pascalau for its time and help in the
open source community and all Linux users around the world who have participated by providing
good comments, ideas, recommendations and suggestions.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-15-2048.jpg)
![Introduction 0
CHAPTER 1
23
Libraries
Programs can be linked not only to object files (*.o) but also to libraries that are collections of
object files. There are two forms of linking to libraries: static, where the code goes in the
executable file, and dynamic, where the code is collected when the program starts to run.
Patches
It was common for executable files to be given corrections without recompiling them. Now this
practice has died out; in modern days, people change a small portion of the source code, putting
a change into a file called a “patch”. Where different versions of a program are required, small
changes to code can be released this way, saving the trouble of having two large distributions.
Errors in Compilation and Linking
Errors in compilation and linking are often due to typos, omissions, or misuse of the language.
You have to check that the right “includes file” is used for the functions you are calling.
Unreferenced symbols are the sign of an incomplete link step. Also check if the necessary
development libraries (GLIBC) or tools (GCC, DEV86, MAKE, etc) are installed on your system.
Debugging
Debugging is a large topic. It usually helps to have statements in the code that inform you of what
is happening. To avoid drowning in output you might sometimes get them to print out only the first
3 passes in a loop. Checking that variables have passed correctly between modules often helps.
Get familiar with your debugging tools.
Build & install software on your system
You will see in this book that we use many different compile commands to build and install
programs on the server. These commands are UNIX compatible and are used on all variants of
*NIX machines to compile and install software.
The procedures to compile and install software tarballs on your server are as follows:
1. First of all, you must download the tarball from your trusted software archive site. Usually
from the main site of the software you hope to install.
2. After downloading the tarball, change to the /var/tmp directory (note that other paths
are possible, at personal discretion) and untar the archive by typing the commands (as
root) as in the following example:
[root@deep /]# tar xzpf foo.tar.gz
The above command will extract all files from the example foo.tar.gz compressed archive and
will create a new directory with the name of the software from the path where you executed the
command.
The “x” option tells tar to extract all files from the archive.
The “z” option tells tar that the archive is compressed with gzip utility.
The “p” option maintains the original permissions the files had when the archive was created.
The “f” option tells tar that the very next argument is the file name.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-23-2048.jpg)
![32
Since this, is a relatively rare occurrence, you will save time if you try to use the standard diskette
images first, and then review the Errata only if you experience any problems completing the
installation. Below, we will show you two methods to create the installation Boot Disk, the first
method is to use an existing Microsoft Windows computer and the second using an existing Linux
computer.
Making a Diskette under MS-DOS:
Before you make the boot disk, insert the Official Red Hat Linux CD-ROM Disk 1 in your
computer that runs the Windows operating system. When the program asks for the filename,
enter boot.img for the boot disk. To make the floppies under MS-DOS, you need to use these
commands (assuming your CD-ROM is drive D: and contain the Official Red Hat Linux CD-ROM).
• Open the Command Prompt under Windows: Start | Programs | Command Prompt
C:> d:
D:> cd dosutils
D:dosutils> rawrite
Enter disk image source file name: ..imagesboot.img
Enter target diskette drive: a:
Please insert a formatted diskette into drive A: and press -ENTER- :
D:dosutils>exit
The rawrite.exe program asks for the filename of the disk image: Enter boot.img and insert
a blank floppy into drive A. It will then ask for a disk to write to: Enter a:, and when complete,
label the disk “Red Hat boot disk”, for example.
Making a Diskette under a Linux-Like OS:
To make a diskette under Linux or any other variant of Linux-Like operating system, you must
have permission to write to the device representing the floppy drive (known as /dev/fd0H1440
under Linux).
This permission is granted when you log in to the system as the super-user “root”. Once you
have logged as “root”, insert a blank formatted diskette into the diskette drive of your computer
without issuing a mount command on it. Now it’s time to mount the Red Hat Linux CD-ROM on
Linux and change to the directory containing the desired image file to create the boot disk.
• Insert a blank formatted diskette into the diskette drive
Insert the Red Hat Linux CD Part 1 into the CD-ROM drive
[root@deep /]# mount /dev/cdrom /mnt/cdrom
[root@deep /]# cd /mnt/cdrom/images/
[root@deep images]# dd if=boot.img of=/dev/fd0H1440 bs=1440k
1+0 records in
1+0 records out
[root@deep images]# cd /
[root@deep /]# umount /mnt/cdrom
Don’t forget to label the diskette “Red Hat boot disk”, for example.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-32-2048.jpg)
![Installation Issues 0
CHAPTER 2
39
Disk Partition (Manual Partitioning)
Now that we know exactly what partitions we need to create for our new Linux server, it is time to
choose the partitioning software we will use to make these partitions. With Red Hat Linux two
programs exist to assist you with this step:
• Manually partition with Disk druid
• Manually partition with fdisk [experts only]
Disk Druid is new software used by default in Red Hat Linux to partition your disk drive, this
program is easy to use, and allows you to use a graphical interface to create your partitions
tables.
fdisk was the first partitioning program available on Linux. It is more powerful then Disk
Druid and allows you to create your partition table in exactly the way you want it (if you want to
put your swap partition near the beginning of your drive, then you will need to use fdisk).
Unfortunately, it is also a little more complicated than Disk Druid and many Linux users prefer
to use Disk Druid for this reason.
Personally, I prefer to create the partitions with the fdisk program and I recommend you use
and be familiar with it, because if, in the future you want to add or change some file systems you
will need to use fdisk.
Partitioning with Disk Druid
This section applies only if you chose to use Disk Druid to partition your system. Disk Druid
is a program that partitions your hard drive for you. Choose “New” to add a new partition, “Edit”
to edit a partition, “Delete” to delete a partition and “Reset” to reset the partitions to the original
state. When you add a new partition, a new window appears on your screen and gives you
parameters to choose.
Mount Point: for where you want to mount your new partition in the filesystem.
Filesystem Type: Ext3 for Linux filesystem and Swap for Linux Swap Partition
Size (MB): for the size of your new partition in megabytes.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-39-2048.jpg)
![Installation Issues 0
CHAPTER 2
53
Boot Disk Creation
From this section of the installation, we have the possibility to create a boot disk for our newly
installed operating system. If you do not want to create a boot disk, you should check the “Skip
boot disk creation” checkbox before you click Next. Creating a boot disk must be made if you
decide to not install GRUB or LILO on the MBR (the Master Boot Record) or if you are not installing
GRUB or LILO at all.
How to use RPM Commands
This section contains an overview of using RPM for installing, uninstalling, upgrading, querying,
listing, and checking RPM packages on your Linux system. You must be familiar with these RPM
commands now because we’ll use them often in this book and especially later in this chapter for
software that must be uninstalled after installation of the server.
Install a RPM package:
Note that RPM packages have a file of names like foo-1.0-2.i386.rpm, which include the
package name (foo), version (1.0), release (2), and architecture (i386).
• To install a RPM package, use the command:
[root@deep /]# rpm -ivh foo-1.0-2.i386.rpm
foo ##################################################
Uninstall a RPM package:
Notice that we used the package name “foo”, not the name of the original package file “foo-
1.0-2.i386.rpm”.
• To uninstall a RPM package, use the command:
[root@deep /]# rpm -e foo](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-53-2048.jpg)
![54
Upgrade a RPM package:
With this command, RPM automatically uninstalls the old version of foo package and installs the
new one. Always use “rpm -Uvh” command to install packages, since it works fine even when
there are no previous versions of the package installed. This is the recommended method of
installing package on the system.
• To upgrade a RPM package, use the command:
[root@deep /]# rpm -Uvh foo-1.0-2.i386.rpm
foo ##################################################
Force the installation of a RPM package:
With this command, RPM will force the installation of the specified package even if some conflict
or other kind of problem exists. This command should be used with care and only if you know
what you do. In most case, RPM can correctly guest problem and refuse to install. To bypass RPM
warning, you can use the RPM command below.
• To force the installation of a RPM package, use the command:
[root@deep /]# rpm -Uvh --force foo-1.0-2.i386.rpm
foo ##################################################
Avoid RPM package dependency:
With this command, RPM will not take care of package dependency and will install the RPM
software on your system. Package dependency is an important concept in the RPM world.
Dependency is when some other packages depend of the RPM package you are trying to install.
By default, RPM check if all other RPM packages required for the RPM you try to install are present
before installing the RPM. If some required packages are not present, RPM will inform you. This is
made to avoid problem and be sure that the software you want to install will perfectly work. In
some special case, we don’t need to take care of dependency and can use the option below to
inform it to skip the dependency check when installing the software.
• To avoid RPM package dependency, use the command:
[root@deep /]# rpm -Uvh --nodeps foo-1.0-2.i386.rpm
foo ##################################################
Query a RPM package:
This command will print the package name, version, and release number of installed package
foo. Use this command to verify that a package is or is not installed on your system.
• To query a RPM package, use the command:
[root@deep /]# rpm -q foo
foo-2.3-8](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-54-2048.jpg)
![Installation Issues 0
CHAPTER 2
55
Display RPM package information:
This command displays package information; includes name, version, and description of the
installed program. Use this command to get information about the installed package.
• To display RPM package information, use the command:
[root@deep /]# rpm -qi foo
Name : foo Relocations: none
Version : 2.3 Vendor: OpenNA.com, Inc.
Release : 8 Build Date: Thu 24 Aug 2000 11:16:53 AM EDT
Install date: Mon 12 Feb 2001 01:17:24 AM EST Build Host: openna.com
Group : Applications/Archiving Source RPM: foo-2.3-8.src.rpm
Size : 271467 License: distributable
Packager : OpenNA.com, Inc. <http://www.openna.com/>
Summary : Here will appears summary of the package.
Description : Here will appears the description of the package.
Display RPM package information before installing the program:
This command displays package information; includes name, version, and description of the
program without the need to install the program first. Use this command to get information about
a package before you install it on your system.
• To display package information before installing the program, use the command:
[root@deep /]# rpm -qpi foo-2.3-8.i386.rpm
Name : foo Relocations: none
Version : 2.3 Vendor: OpenNA.com, Inc.
Release : 8 Build Date: Thu 24 Aug 2000 11:16:53 AM EDT
Install date: Mon 12 Feb 2001 01:17:24 AM EST Build Host: openna.com
Group : Applications/Archiving Source RPM: foo-2.3-8.src.rpm
Size : 271467 License: distributable
Packager : OpenNA.com, Inc. <http://www.openna.com/>
Summary : Here will appears summary of the package.
Description : Here will appears the description of the package.
List files in a installed RPM package:
This command will list all files in a installed RPM package. It works only when the package is
already installed on your system.
• To list files in a installed RPM package, use the command:
[root@deep /]# rpm -ql foo
/usr/bin/foo
/usr/bin/foo1
/usr/sbin/foo2
List files in RPM package that is not already installed:
This command will list all files in a RPM package that is not already installed on your system. It is
useful when you want to know which components are included in the package before installing it.
• To list files in RPM package that is not already installed, use the command:
[root@deep /]# rpm -qpl foo
/usr/lib/foo
/usr/bin/foo1
/usr/sbin/foo2](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-55-2048.jpg)
![56
Know which files is part of which RPM package:
This command will show you from which RPM package the file comes from. It works only when the
package is already installed on your system and it is very useful when you see some files into
Linux that you do not know about it and want to get more information about its RPM provenance.
• To know which files is part of which RPM package, use the command:
[root@deep /]# rpm -qf /etc/passwd
setup-2.3.4-1
Check RPM signature package:
This command checks the PGP signature of specified package to ensure its integrity and origin.
Always use this command first before installing new RPM package on your system. GnuPG or PGP
software must be already installed on your system before you can use this command. See the
chapter related to GnuPG installation and configuration for more information.
• To check a RPM signature package, use the command:
[root@deep /]# rpm --checksig foo
Examine the md5sum of RPM package:
The RPM md5sum is useful to verify that a package has not been corrupted or tampered with. You
can use it to be sure that the download of your new RPM package was not corrupted during
network transfer.
• To examine only the md5sum of the package, use the command:
[root@deep /]# rpm --checksig --nogpg foo
Starting and stopping daemon services
The init program of Linux (also known as process control initialization) is in charge of starting
all the normal and authorized processes that need to run at boot time on your system. These may
include the APACHE daemons, NETWORK daemons, and anything else that must be running
when your machine boots.
Each of these processes has a script file under the /etc/init.d directory written to accept an
argument, which can be start, stop, restart, etc. As you can imagine, those script are made
to simplify the administration of the server and the way we can start or stop services under Linux.
Of course, we can use the native way to start all required services under our server, but it is much
simple to have some kind of script files that should provide us some easy method to automate
and control the procedures. This is why init program and all initialization script files available
under the /etc/init.d directory exist.
Below are some examples showing you how to execute those scripts by hand.
For example:
• To start the httpd web server daemon manually under Linux, you’ll type:
[root@deep /]# /etc/init.d/httpd start
Starting httpd: [OK]
• To stop the httpd web server daemon manually under Linux, you’ll type:
[root@deep /]# /etc/init.d/httpd stop
Shutting down http: [OK]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-56-2048.jpg)
![Installation Issues 0
CHAPTER 2
57
• To restart the httpd web server daemon manually under Linux, you’ll type:
[root@deep /]# /etc/init.d/httpd restart
Shutting down http: [OK]
Starting httpd: [OK]
Check inside your /etc/init.d directory for services available and use the commands start |
stop | restart to work around. You will see along this book that we often use initialization
script file to administration and control the way we start, restart, stop, etc services under Linux.
Software that must be uninstalled after installation of the server
Red Hat Linux installs other programs on your system by default and doesn’t give you the choice
to uninstall them during the install setup. For this reason, you must uninstall the following
software after the complete installation of our Linux server.
Below is the list of programs and a short description of their purpose. We must uninstall them for
increased security and to make more space on our server. For more information and an
explanation of their capabilities and uses, please see your Red Hat manual or query the package
by making an “rpm -qi foo” command before uninstalling them.
The anacron package:
The anacron package is similar to the cron package but differ in the way that it does not
assume that the system is running continuously and it is a good command scheduler for system
which doesn’t run 24 hours a day. In server environment, the system should absolutely run 24/24
hours; therefore we simply don’t need to have this kind of package installed on a server.
• To remove the anacron package from your system, use the following commands:
[root@deep /]# /etc/init.d/anacron stop
[root@deep /]# rpm -e anacron
[root@deep /]# rm -rf /var/spool/anacron/
The apmd package:
The apmd package or Advanced Power Management Daemon utilities is used on notebook
computer. It can watch your notebook's battery and warn all users when the battery is low. As you
can imagine, there is no need to have it installed on a server machine.
• To remove the apmd package from your system, use the following commands:
[root@deep /]# /etc/init.d/apmd stop
[root@deep /]# rpm -e apmd
The at package:
The at package is a utility that will do time-oriented job control by scheduling a command to run
later. Unfortunately, it has had a rich history of problems and we can achieve the same
functionality with the more secure vixie-cron package. For this reason I recommend you to
uninstall it.
• To remove the at package from your system, use the following commands:
[root@deep /]# /etc/init.d/atd stop
[root@deep /]# rpm -e at](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-57-2048.jpg)
![58
The gpm package:
The gpm package provides mouse support to text-based Linux applications. It’s the software that
allows you to cut-and-paste operations using the mouse on your terminal. If most of your entire
administration of the server is made via remote connection, you can remove this package to save
some processes and memories. We can continue to use cut-and-paste operation via remote
connection to the server without problem. The gpm package is only useful if you stay at the
console terminal of your server to make administration tasks.
• To remove the gpm package from your system, use the following commands:
[root@deep /]# /etc/init.d/gpm stop
[root@deep /]# rpm -e gpm
The dhcpcd package:
The dhcpcd package contains the protocol, which allows systems to get their own network
configuration information from a DHCP server. If you are going to use DHCP on your network, it is
recommended to install the DHCP client included in the pump package, which provides a faster
and simpler DHCP client. For more information about DHCP, see its related chapter in this book.
• To remove the dhcpcd package from your system, use the following command:
[root@deep /]# rpm -e dhcpcd
The eject package:
The eject package contains an eject program that allows the user to eject removable media
(typically CD-ROMs, floppy disks, Iomega Jaz or Zip disks) using software. This is an unneeded
program to have installed in a server environment. You should keep it installed only if you’re
intended to run a tape backup on your system.
• To remove the eject package from your system, use the following command:
[root@deep /]# rpm -e eject
The hotplug package:
The hotplug package is a helper application for loading modules for USB devices. On a server
environment, USB devices are not used at all and are required only on Linux workstation.
• To remove the hotplug package from your system, use the following command:
[root@deep /]# rpm -e hotplug
The lokkit package:
The lokkit package is a Firewall configuration application for an average end user and it is not
designed to configure arbitrary firewalls since it is solely designed to handle typical dialup user
and cable modem set-ups. It is not the answer to a complex firewall configuration, and it is not the
equal of an expert firewall designer. Therefore remove it from your server and read the chapter
related to GIPTables in this book for more information about secure firewall configuration.
• To remove the lokkit package from your system, use the following command:
[root@deep /]# rpm -e lokkit](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-58-2048.jpg)
![Installation Issues 0
CHAPTER 2
59
The ipchains package:
The ipchains package is the old tool used with Linux kernel 2.2 for managing Linux kernel
packet filtering capabilities and to set up firewalling on the network. A new and more powerful tool
called “IPTables” exists and this is the one that we will use later in the book to set up our
firewall on Linux.
• To remove the ipchains package from your system, use the following command:
[root@deep /]# rpm -e ipchains
The ksymoops package:
The ksymoops package is a small tool used to report kernel oops and error message decoder.
This package is useful for developers that work on the Linux kernel and want to debug or for
users that want to report bugs with the kernel. The same result can be achieved with the dmesg
command of Linux. Therefore, you can remove this package from your secure server.
• To remove the ksymoops package from your system, use the following command:
[root@deep /]# rpm -e ksymoops
The kudzu package:
The kudzu package is a hardware-probing tool that runs at system boot time to determine what
hardware has been added or removed from the system. Again, in server environment, we don’t
upgrade, add or remove hardware every time. Therefore, we can safety remove this small
package from our server.
• To remove the kudzu package from your system, use the following command:
[root@deep /]# rpm -e kudzu
The mailcap package:
Metamail is a program that uses the mailcap file to determine how it should display non-text or
multimedia material. We don’t need to have it installed at all. Remove it.
• To remove the mailcap package from your system, use the following command:
[root@deep /]# rpm -e mailcap
The pciutils package:
The pciutils package contains various utilities for inspecting and setting devices connected to
the PCI bus. Keep it installed if you want, but I recommend removing it form the server.
• To remove the pciutils package from your system, use the following command:
[root@deep /]# rpm -e pciutils
The raidtools package:
The raidtools package includes the tools you need to set up and maintain a software RAID
device on a Linux system. You should keep this package only if you have configured your server
to run with RAID support. Otherwise, remove it.
• To remove the raidtools package from your system, use the following command:
[root@deep /]# rpm -e raidtools](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-59-2048.jpg)
![60
The redhat-logos package:
The redhat-logos package contains files of the Red Hat "Shadow Man" logo and the RPM logo.
• To remove the redhat-logos package from your system, use the following command:
[root@deep /]# rpm -e redhat-logos
The redhat-release package:
The redhat-release package contains the Red Hat Linux release files. Please note that if you
remove this package, the boot process of the system will generate error messages because it will
look for a file called “redhat-release” which will not be available anymore. To solve this
problem, we recreate the required file under the appropriated directory and add as content of this
file the word “Red Hat Linux”. Of course you can change it for whatever you want.
• To remove the redhat-release package from your system, use the command:
[root@deep /]# rpm -e redhat-release
[root@deep /]# echo Red Hat Linux > /etc/redhat-release
The setserial package:
The setserial package is a basic system utility for displaying or setting serial port information.
• To remove the setserial package from your system, use the command:
[root@deep /]# rpm -e setserial
The hdparm package:
The program hdparm is needed by IDE hard disks but not SCSI hard disks. If you have an IDE
disk on your system you must keep this program (hdparm), but if you don’t have an IDE hard
disk you can remove it safely from your system. hdparm is a small Linux tool used to optimize
your IDE hard drive. SCSI hard drives don’t need to be optimized since they are capable to run at
their full speed (80 Mps to 160 Mps) without modification.
• To remove the hdparm package from your system, use the following command:
[root@deep /]# rpm -e hdparm
The mkinitrd package:
The program mkinitrd is needed by SCSI or RAID hard disk but not IDE hard disks. If you
have a SCSI or RAID disk on your system you must keep this program (mkinitrd), but if you
don’t have a SCSI or RAID hard disk you can safely remove it from your system.
• To remove the mkinitrd package from your system, use the following command:
[root@deep /]# rpm -e --nodeps mkinitrd
The xconfig packages:
Use the programs kbdconfig, mouseconfig, timeconfig, netconfig, authconfig,
ntsysv, and setuptool in order to set your keyboard language and type, your mouse type,
your default time zone, your Ethernet devices, your NIS and shadow passwords, your numerous
symbolic links in /etc/rc.d directory, and text mode menu utility which allow you to access all
of these features.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-60-2048.jpg)
![Installation Issues 0
CHAPTER 2
61
After those configurations have been set during the installation stage of your Linux server it’s rare
that you would need to change them again. So, you can uninstall them, and if in the future you
need to change your keyboard, mouse, default time, etc again via test mode menu, all you have
to do is to install the program with the RPM from your original CD-ROM.
• To remove all the above programs from your system, use the following command:
[root@deep /]# rpm -e kbdconfig mouseconfig timeconfig netconfig
authconfig ntsysv setuptool
The newt package:
The newt package provides a development library for text mode user interfaces. It’s mainly used
by all the above config packages. Since all the config packages are removed from the server, we
can safety remove newt from our system. If you have decided to keep the above config
packages (kbdconfig, mouseconfig, etc), then you should keep newt, otherwise you should
remove it.
• To remove the newt package from your system, use the following command:
[root@deep /]# rpm -e newt
The lilo package:
The lilo package provides a boot loader for Linux. Remember that during our Linux installation,
we have chosen to go with GRUB instead of LILO as our boot loader for Linux. Therefore, you can
safety remove this package from your system since it’s really not used now.
• To remove the lilo package from your system, use the following command:
[root@deep /]# rpm -e lilo
[root@deep /]# rm -f /etc/lilo.conf.anaconda
The reiserfs-utils package:
The reiserfs-utils package contains a number of utilities for creating, checking, modifying,
and correcting any inconsistencies in ReiserFS file-systems. ReiserFS is another file-system
like Ext3 for Linux. In our configuration of Linux we use Ext3 as our file-system therefore we
don’t need to keep this package installed. Keep this utility package only if you intended to run
ReiserFS instead of Ext3 file-system on your Linux server.
• To remove the reiserfs-utils package from your system, use the command:
[root@deep /]# rpm -e reiserfs-utils
The quota package:
The program quota is a system administration tools for monitoring and limiting user/group disk
usage, per file system. This program must be installed only on servers where the need for
monitoring and restricting amount of disk space in user’s directories is require.
• To remove the quota package from your system, use the following command:
[root@deep /]# rpm -e quota](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-61-2048.jpg)
![62
The indexhtml package:
The indexhtml package contains the HTML page and graphics for a welcome page shown by
your browser under graphical interface installation. These HTML pages are information about Red
Hat software. You really don’t need this package under server installation and especially when
GUI is not available. Therefore, you can safety remove this package from your system.
• To remove the indexhtml package from your system, use the following command:
[root@deep /]# rpm -e indexhtml
The usbutils package:
The usbutils package provides Linux USB utilities for inspecting devices connected to a USB
bus on your system. In server environment, we really don’t use any USB devices and can safety
remove this package from our server installation. USB will usually be used only in Linux
workstation installation where you want to plug printer, camera and any other media of this type.
• To remove the usbutils package from your system, use the following command:
[root@deep /]# rpm -e usbutils
The hwdata package:
The hwdata package contains various hardware identification and configuration data mainly
used with USB and XFree86. Remember that XFree86 is related to graphical interface and since
we don’t use any GUI into our Linux server, we can remove this package from our system.
• To remove the hwdata package from your system, use the following command:
[root@deep /]# rpm -e hwdata
The parted package:
The parted package contains various utilities to create, destroy, resize, move and copy hard
disk partitions. It is useful when you need to play with your hard disk structure. In most case, we
partition and set hard disk organization at the installation of the operating system and don’t need
to play or change something once everything are installed and running properly. It’s rare to have
to use this package utility on a production server and this is the reason why I recommend you to
uninstall it. If you prefer to keep it installed for future possible usage, you are free to do it.
• To remove the parted package from your system, use the following command:
[root@deep /]# rpm -e parted
The hesiod package:
The hesiod package is another one that we can uninstall from our Linux server configuration
setup. It’s a system which uses existing DNS functionality to provide access to databases of
information that change infrequently. In most cases, we don’t need it and you should keep it
install only if you think that you will need it for some special configuration of your server.
• To remove the hesiod package from your system, use the following command:
[root@deep /]# rpm -e hesiod](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-62-2048.jpg)
![Installation Issues 0
CHAPTER 2
63
The mt-st package:
The mt-st package provides tools for controlling and managing tape drives on your system. You
should keep it installed only if you have and want to run a tape backup media on your system.
• To remove the mt-st package from your system, use the following command:
[root@deep /]# rpm -e mt-st
The man-pages package:
The man-pages package provides a large collection of additional manual pages (man pages)
from the Linux Documentation Project (LDP). By default many manual pages are installed with
the operating system and the man-pages package provides additional documentation for those
who want to read them on the system. In server environment, I really don’t see the need to have
additional manual pages installed since these manual pages can be read online from the Internet
or even on another computer running as a development or workstation.
• To remove the man-pages package from your system, use the following command:
[root@deep /]# rpm -e man-pages
The sendmail package:
Even if you don’t want to run your system as a full mail server, mailer software is always needed
for potential messages sent to the root user by different software services installed on your
machine.
Sendmail is a Mail Transport Agent (MTA) program that sends mail from one machine to another
and it’s the default mail server program installed on Red Hat Linux. Unfortunately, this software
has a long history of security problem and for this reason I highly recommend you to not use it on
your Linux server. You must uninstall Sendmail and see the part in this book that is related to
Mail Transfer Agent configuration and installation for some good alternative like Exim or Qmail.
• To remove the sendmail package from your system, use the following commands:
[root@deep /]# /etc/init.d/sendmail stop
[root@deep /]# rpm -e sendmail
The procmail package:
Procmail is a mail-processing program, which can be used by Sendmail for all local mail
delivery and filtering. This program is required only if you decide to install and use Sendmail on
your server and only if Sendmail is installed. Since we’ve decided to not go with Sendmail as
our MTA for security reason, we can uninstall procmail from our Linux server.
• To remove the procmail package from your system, use the following command:
[root@deep /]# rpm -e procmail
The openldap package:
The OpenLDAP software is a set of protocols for accessing directory services like phone book
style information and other kinds of information over the Internet. This useful program is not
suitable for everyone and depends of what you want to do with your system. If you want to give it
a try, see later in this book under the chapter related to databases for more information.
• To remove the OpenLDAP package from your system, use the following command:
[root@deep /]# rpm -e --nodeps openldap](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-63-2048.jpg)
![64
The cyrus-sasl packages:
The Cyrus SASL implementation is the Simple Authentication and Security Layer, a method for
adding authentication support to connection-based protocols. It is used in conjunction with
Cyrus, which is an electronic messaging program like Sendmail. Since Cyrus SASL is made to
be used with Sendmail that we have removed previously for security reason, we can safety
remove it.
• To remove the Cyrus SASL package from your system, use the following command:
[root@deep /]# rpm -e --nodeps cyrus-sasl cyrus-sasl-md5 cyrus-sasl-plain
The openssl package:
OpenSSL is an SSL encryption mechanism which ensures and provides safe and secure
transactions of data over networks. This piece of software is one of the most important tools for a
Linux server and it is highly recommended that it is installed. Unfortunately, the one that comes
with Red Hat Linux is not up to date and not optimized for our specific server. For this reason, we
will uninstall it now and see later in this book, under the chapters related to security software, how
to install, secure, optimize and use it.
• To remove the OpenSSL package from your system, use the following command:
[root@deep /]# rpm -e --nodeps openssl
[root@deep /]# rm -rf /usr/share/ssl/
The ash package:
The ash package is a smaller version of the bourne shell (sh). Since we already use sh, we can
uninstall this package from our system. If you use this program in your regular administration
task, then keep it installed on your server. In most case, we can remove it.
• To remove the ash package from your system, use the following command:
[root@deep /]# rpm -e ash
The tcsh package:
The tcsh package is an enhanced version of csh, another C shell. We already have bash as our
default shell program on Linux, and I don’t find any reason to keep another variant installed if we
don’t have any program or services that need it to run.
Most services under Linux can easily run with our default bash shell program and if you don’t
have any program that required tcsh to run, then I recommend you to uninstall it. If in the future,
you see that you need to have tcsh installed on your server for some specific program to run,
then all you have to do is to install it from your CD-ROM. In most cases, there is no program that
needs tsch to run, therefore you can remove it.
• To remove the tcsh package from your system, use the following command:
[root@deep /]# rpm -e tcsh](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-64-2048.jpg)
![Installation Issues 0
CHAPTER 2
65
The specspo package:
The specspo package contains the portable object catalogues used to internationalize Red Hat
packages. I don’t think that this kind of package is really required on a production server.
• To remove the specspo package from your system, use the following command:
[root@deep /]# rpm -e specspo
The krb5-libs package:
The krb5-libs package contains the shared libraries needed by Kerberos 5. Because we’re
not using Kerberos, we’ll need to uninstall this package. Kerberos is not secure as you can
think and can be cracked easily with some good knowledge of this program. Anyway it is yours to
decide if you really need it.
• To remove the krb5-libs package from your system, use the following command:
[root@deep /]# rpm -e krb5-libs
[root@deep /]# rm -rf /usr/kerberos/
The MAKEDEV package:
The MAKEDEV package contains the MAKEDEV program, which makes it easier to create and
maintain the files in the /dev directory. Program provided by this package is used for creating
device files in the /dev directory of your server. In general, we use it under development server
when we build new package for our Linux system. On production servers, it’s rare to use it.
Therefore we can remove it from our system without any problem.
• To remove the MAKEDEV package from your system, use the following command:
[root@deep /]# rpm -e MAKEDEV
Remove unnecessary documentation files
By default the majority of each RPM’s packages installed under Linux come with documentation
files related to the software. This documentation contains original files from the program tar
archive like README, FAQ, BUG, INSTALL, NEWS, PROJECTS and more.
Many of them can be easily retrieved from the website where the program has been downloaded
and it makes no sense for them to be kept on your system. I know that hard drives costs have
come down considerably recently, but why keep this kind of documentation on a secure server if
it unlikely they will not be read more than once. Anyway, have a look inside those files and decide
for yourself if you want to remove them or not.
• To remove all documentation files from your system, use the following commands:
[root@deep /]# cd /usr/share/doc/
[root@deep doc]# rm -rf *](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-65-2048.jpg)
![66
Remove unnecessary/empty files and directories
There are some files and directories we can remove manually from the file system of Linux to
make a clean install. These files and directories are not needed but still exist after our secure
installation of Linux and can be removed safely. Some are bugs from the Red Hat installation
script and others are created by default even if you don’t use them.
• To remove all unnecessary files and directories from your system, use the commands:
[root@deep /]# rm -f /etc/exports
[root@deep /]# rm -f /etc/printcap
[root@deep /]# rm -f /etc/ldap.conf
[root@deep /]# rm -f /etc/krb.conf
[root@deep /]# rm -f /etc/yp.conf
[root@deep /]# rm -f /etc/hosts.allow
[root@deep /]# rm -f /etc/hosts.deny
[root@deep /]# rm -f /etc/csh.login
[root@deep /]# rm -f /etc/csh.cshrc
[root@deep /]# rm -f /etc/fstab.REVOKE
[root@deep /]# rm -f /etc/pam_smb.conf
[root@deep /]# rm -rf /etc/xinetd.d/
[root@deep /]# rm -rf /etc/opt/
[root@deep /]# rm -rf /etc/X11/
[root@deep /]# rm -rf opt/
[root@deep /]# rm -rf /var/opt/
[root@deep /]# rm -rf /var/nis/
[root@deep /]# rm -rf /var/yp/
[root@deep /]# rm -rf /var/lib/games/
[root@deep /]# rm -rf /var/spool/lpd/
[root@deep /]# rm -rf /usr/lib/python1.5/
[root@deep /]# rm -rf /usr/lib/games/
[root@deep /]# rm -rf /usr/X11R6/
[root@deep /]# rm -rf /usr/etc/
[root@deep /]# rm -rf /usr/games/
[root@deep /]# rm -rf /usr/local/
[root@deep /]# rm -rf /usr/dict/
[root@deep /]# rm -f /usr/bin/X11
[root@deep /]# rm -f /usr/lib/X11
NOTE: If in the future you want to install a program which needs some of the files/directories we
have removed, then the program will automatically recreate the missing files or directories. Good!
Software that must be installed after installation of the server
There are certain programs required to be able to compile programs on your server, for this
reason you must install the following RPM packages. This part of the installation is very important
and requires that you install all the packages described below.
These are on your Red Hat Part 1 and Part 2 CD-ROMs under RedHat/RPMS directory and
represent the necessary base software needed by Linux to compile and install programs. Please
note that if you don’t want to compile software in your server or if you only use RPM’s packages to
update programs or if you use a dedicated server to develop, compile or create your own RPM’s
packages which will be installed later along your network on the servers, then you DON’T need to
install the packages described here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-66-2048.jpg)
![Installation Issues 0
CHAPTER 2
67
Step 1
First, we mount the CD-ROM drive and move to the RPMS subdirectory of the CD-ROM.
• To mount the CD-ROM drive and move to RPM directory, use the following commands:
[root@deep /]# mount /dev/cdrom /mnt/cdrom/
had: ATAPI 32X CD-ROM drive, 128kB Cache
mount: block device dev/cdrom is write-protected, mounting read-only
[root@deep /]# cd /mnt/cdrom/RedHat/RPMS/
These are the packages that we need to be able to compile and install programs on the Linux
system. Remember, this is the minimum number of packages that permits you to compile most of
the tarballs available for Linux. Other compiler packages exist on the Linux CD-ROM, so verify
with the README file that came with the tarballs program you want to install if you receive error
messages during compilation of the specific software.
The compiler packages:
Compiler packages contain programs and languages used to build software on the system.
Remember to uninstall the entire following compiler packages after successful installation of all
software required on your Linux server.
binutils-2.11.93.0.2-11.i386.rpm
bison-1.35-1.i386.rpm
byacc-1.9-19.i386.rpm
cdecl-2.5-22.i386.rpm
cpp-2.96-110.i386.rpm
cproto-4.6-9.i386.rpm
ctags-5.2.2-2.i386.rpm
dev86-0.15.5-1.i386.rpm
flex-2.5.4a-23.i386.rpm
gcc-2.96-110.i386.rpm
gcc-c++-2.96-110.i386.rpm
glibc-kernheaders-2.4-7.14.i386.rpm
m4-1.4.1-7.i386.rpm
make-3.79.1-8.i386.rpm
patch-2.5.4-12.i386.rpm
perl-5.6.1-34.99.6.i386.rpm
The development packages:
Development packages contain header and other files required during compilation of software. In
general, development packages are needed when we want to add some specific functionality to
the program that we want to compile. For example if I want to add PAM support to IMAP, I’ll need
pam-devel, which contains the required header files for IMAP to compile successfully.
As for compiler packages, all development packages must be uninstalled after successful
compilation of all the software that you need on your Linux server. Remember to uninstall them
since they are not needed for proper functionality of the server, but just to compile the programs.
aspell-devel-0.33.7.1-9.i386.rpm
db3-devel-3.3.11-6.i386.rpm
freetype-devel-2.0.9-2.i386.rpm
gdbm-devel-1.8.0-14.i386.rpm
gd-devel-1.8.4-4.i386.rpm
glibc-devel-2.2.5-34.i386.rpm
libjpeg-devel-6b-19.i386.rpm
libpng-devel-1.0.12-2.i386.rpm
libstdc++-devel-2.96-110.i386.rpm
ncurses-devel-5.2-26.i386.rpm
pam-devel-0.75-32.i386.rpm
pspell-devel-0.12.2-8.i386.rpm
zlib-devel-1.1.3-25.7.i386.rpm](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-67-2048.jpg)
![68
Dependencies packages:
Dependencies packages are other RPM packages needed by the RPM packages that we want to
install. This happens because some RPM’s are directly linked with others and depend on each
one to function properly. The following packages are required by the above RPM packages and
we will install them to satisfy dependencies. After proper compilation and installation of all needed
software on the Linux server, we can uninstall them safety (if not needed by special program that
we will install).
aspell-0.33.7.1-9.i386.rpm
freetype-2.0.9-2.i386.rpm
gd-1.8.4-4.i386.rpm
libjpeg-6b-19.i386.rpm
libpng-1.0.12-2.i386.rpm
libtool-libs-1.4.2-7.i386.rpm
pspell-0.12.2-8.i386.rpm
Step 2
It is better to install the software described above together if you don’t want to receive
dependencies error messages during the install. Some of the RPMs reside on CD-ROM Part 1
and other on CD-ROM Part2 of Red Hat. For easy installation, I recommend you to copy all of the
required packages (compilers and development) to your hard drive and install them from there.
• These procedures can be accomplished with the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rpm -Uvh *.rpm
Preparing... ########################################### [100%]
1:binutils ########################################### [ 2%]
2:bison ########################################### [ 5%]
3:byacc ########################################### [ 8%]
4:cdecl ########################################### [ 11%]
5:cpp ########################################### [ 13%]
6:cproto ########################################### [ 16%]
7:ctags ########################################### [ 19%]
8:db3-devel ########################################### [ 22%]
9:dev86 ########################################### [ 25%]
10:flex ########################################### [ 27%]
11:freetype ########################################### [ 30%]
12:freetype-devel ########################################### [ 33%]
13:gdbm-devel ########################################### [ 36%]
14:glibc-kernheaders ########################################### [ 38%]
15:glibc-devel ########################################### [ 41%]
16:gcc ########################################### [ 44%]
17:libjpeg ########################################### [ 47%]
18:libjpeg-devel ########################################### [ 50%]
19:libpng ########################################### [ 52%]
20:gd ########################################### [ 55%]
21:gd-devel ########################################### [ 58%]
22:libstdc++-devel ########################################### [ 61%]
23:gcc-c++ ########################################### [ 63%]
24:libtool-libs ########################################### [ 66%]
25:m4 ########################################### [ 69%]
26:make ########################################### [ 72%]
27:pam-devel ########################################### [ 75%]
28:patch ########################################### [ 77%]
29:perl ########################################### [ 80%]
30:ncurses-devel ########################################### [ 83%]
31:pspell ########################################### [ 86%]
32:aspell ########################################### [ 88%]
33:pspell-devel ########################################### [ 91%]
34:aspell-devel ########################################### [ 94%]
35:zlib-devel ########################################### [ 97%]
36:libpng-devel ########################################### [100%]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-68-2048.jpg)
![General Security 0
CHAPTER 3
75
Linux General Security
Abstract
A secure Linux server depends on how the administrator makes it. Once we have eliminated the
potential security risk by removing unneeded services, we can start to secure our existing
services and software on our server. Within a few hours of installing and configuring your system,
you can prevent many attacks before they occur. In this chapter we will discuss some of the more
general, basic techniques used to secure your system. The following is a list of features that can
be used to help prevent attacks from external and internal sources.
BIOS
It is recommended to disallow booting from floppy drives and set passwords on BIOS features.
You can check your BIOS manual or look at it thoroughly the next time you boot up your system
to find out how to do this. Disabling the ability to boot from floppy drives and being able to set a
password to access the BIOS features will improve the security of your system.
This will block unauthorized people from trying to boot your Linux system with a special boot disk
and will protect you from people trying to change BIOS features like allowing boot from floppy
drive or booting the server without prompt password. It is important to note that there is a
possibility to bypass this security measure if someone has a physical access to your server since
they can open the computer and unplug the BIOS battery. This will reset all features to their initial
values.
Unplug your server from the network
It is not wise to apply security changes in your newly installed Linux server if you are online. So it
is preferable to deactivate all network interfaces in the system before applying security changes.
• To stop specific network devices manually on your system, use the command:
[root@deep /]# ifdown eth0
• To start specific network devices manually on your system, use the command:
[root@deep /]# ifup eth0
• To shut down all network interfaces, use the following command:
[root@deep /]# /etc/init.d/network stop
Shutting down interface eth0 [OK]
Disabling Ipv4 packet forwarding [OK]
• To start all network interfaces, use the following command:
[root@deep /]# /etc/init.d/network start
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-75-2048.jpg)
![General Security 0
CHAPTER 3
79
The single-user login mode of Linux
This part applies for those who use LILO as their boot loader. Linux has a special command
(linux single) also known as ‘single-user mode’, which can be entered at the boot prompt
during startup of the system. The single-user mode is generally used for system maintenance.
You can boot Linux in single-user mode by typing at the LILO boot prompt the command:
LILO: linux single
This will place the system in Run level 1 where you'll be logged in as the super-user 'root', and
where you won't even have to type in a password!
Step 1
Requiring no password to boot into root under single-user mode is a bad idea! You can fix this by
editing the inittab file (vi /etc/inittab) and change the following line:
id:3:initdefault:
To read:
id:3:initdefault:
~~:S:wait:/sbin/sulogin
The addition of the above line will require entering the root password before continuing to boot
into single-user mode by making init (8) run the program sulogin (8) before dropping
the machine into a root shell for maintenance.
Step 2
Now, we have to restart the process control initialization of the server for the changes to take
effect.
• This can be done with the following command:
[root@deep /]# /sbin/init q
Disabling Ctrl-Alt-Delete keyboard shutdown command
Commenting out the line (with a “#”) listed below in your /etc/inittab file will disable the
possibility of using the Control-Alt-Delete command to shutdown your computer. This is
pretty important if you don't have the best physical security to the machine.
Step 1
• To do this, edit the inittab file (vi /etc/inittab) and change/comment the line:
ca::ctrlaltdel:/sbin/shutdown -t3 -r now
To read:
#ca::ctrlaltdel:/sbin/shutdown -t3 -r now](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-79-2048.jpg)
![80
Step 2
Now, we have to restart the process control initialization of the server for the changes to take
effect.
• This can be done with the following command:
[root@deep /]# /sbin/init q
Limiting the default number of started ttys on the server
Default installed Linux system comes with six virtual consoles in standard run levels. This means
that six mingetty processes will always by available at every time on the server. These virtual
consoles allow you to login with six different virtual consoles on the system.
Step 1
On secure server, we can limit the number to two virtual consoles and save some resources
which may be used for other work by the server when required.
• To do this, edit the inittab file (vi /etc/inittab) and remove/comment the lines:
1:2345:respawn:/sbin/mingetty tty1
2:2345:respawn:/sbin/mingetty tty2
3:2345:respawn:/sbin/mingetty tty3
4:2345:respawn:/sbin/mingetty tty4
5:2345:respawn:/sbin/mingetty tty5
6:2345:respawn:/sbin/mingetty tty6
To read:
1:2345:respawn:/sbin/mingetty tty1
2:2345:respawn:/sbin/mingetty tty2
#3:2345:respawn:/sbin/mingetty tty3
#4:2345:respawn:/sbin/mingetty tty4
#5:2345:respawn:/sbin/mingetty tty5
#6:2345:respawn:/sbin/mingetty tty6
Step 2
Now, we have to restart the process control initialization of the server for the changes to take
effect.
• This can be done with the following command:
[root@deep /]# /sbin/init q
The LILO and /etc/lilo.conf file
This part applies for those who use LILO as their boot loader. LILO is a boot loader that can be
used to manage the boot process, boot Linux kernel images from floppy disks, hard disks or can
even act as a "boot manager" for other operating systems.
LILO is very important in the Linux system and for this reason, we must protect it the best we
can. The most important configuration file of LILO is the lilo.conf file. It is with this file that we
can configure and improve the security of our LILO program and Linux system. Following are
three important options that will improve the security of our valuable LILO program.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-80-2048.jpg)
![General Security 0
CHAPTER 3
81
• Adding: timeout=00
This option controls how long (in seconds) LILO waits for user input before booting to the default
selection. One of the requirements of C2 security is that this interval be set to 0 unless the system
dual boots something else.
• Adding: restricted
This option asks for a password only, if parameters are specified on the command line (e.g.
linux single). The option “restricted” can only be used together with the “password”
option. Make sure you use this one on each additional image you may have.
• Adding: password=<password>
This option asks the user for a password when trying to load the image. Actually the effect of
using the password parameter in /etc/lilo.conf will protect the Linux image from booting.
This means, it doesn't matter if you load Linux in single mode or if you just do a normal boot. It
will always ask you for the password.
Now this can have a very bad effect, namely you are not able to reboot Linux remotely any more
since it won't come up until you type in the root password at the console. It is for this reason that
adding “restricted” with “password” is very important since the option "restricted" relaxes
the password protection and a password is required only if parameters are specified at the LILO
prompt, (e.g. single).
Passwords are always case-sensitive, also make sure the /etc/lilo.conf file is no longer
world readable, or any user will be able to read the password. Here is an example of our
protected LILO with the lilo.conf file.
Step 1
• Edit the lilo.conf file (vi /etc/lilo.conf) and add or change the three options
above as show:
boot=/dev/sda
map=/boot/map
install=/boot/boot.b
prompt remove this line if you don’t want to pass options at the LILO prompt.
timeout=00 change this line to 00 to disable the LILO prompt.
linear
message=/boot/message remove this line if you don’t want the welcome screen.
default=linux
restricted add this line to relaxes the password protection.
password=<password> add this line and put your password.
image=/boot/vmlinuz-2.4.2-2
label=linux
initrd=/boot/initrd-2.4.2-2.img
read-only
root=/dev/sda6
Step 2
Because the configuration file /etc/lilo.conf now contains unencrypted passwords, it should
only be readable for the super-user “root”.
• To make the /etc/lilo.conf file readable only by the super-user “root”, use the
following command:
[root@deep /]# chmod 600 /etc/lilo.conf (will be no longer world readable).](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-81-2048.jpg)
![82
Step 3
Now we must update our configuration file /etc/lilo.conf for the change to take effect.
• To update the /etc/lilo.conf file, use the following command:
[root@deep /]# /sbin/lilo -v
LILO version 21.4-4, copyright © 1992-1998 Wernerr Almesberger
‘lba32’ extentions copyright © 1999,2000 John Coffman
Reading boot sector from /dev/sda
had : ATAPI 32X CD-ROM drive, 128kB Cache
Merging with /boot/boot.b
Mapping message file /boot/message
Boot image : /boot/vmlinuz-2.2.16-22
Mapping RAM disk /boot/initrd-2.2.16-22.img
Added linux *
/boot/boot.0800 exists – no backup copy made.
Writing boot sector.
Step 4
One more security measure you can take to secure the lilo.conf file is to set it immutable,
using the chattr command.
• To set the file immutable simply, use the following command:
[root@deep /]# chattr +i /etc/lilo.conf
And this will prevent any changes (accidental or otherwise) to the lilo.conf file. If you wish to
modify the lilo.conf file you will need to unset the immutable flag:
• To unset the immutable flag, use the following command:
[root@deep /]# chattr -i /etc/lilo.conf
WARNING: When you use the password option, then LILO will always ask you for the password,
regardless if you pass options at the LILO prompt (e.g. single) or not EXCEPT when you set
the "restricted" option in /etc/lilo.conf.
The option "restricted" relaxes the password protection and a password is required only if
parameters are specified at the LILO prompt, (e.g. single).
If you didn't had this option set "restricted", Linux will always ask you for the password and
you will not be able to remotely reboot your system, therefore don’t forget to add the option
"restricted” with the option "password" into the /etc/lilo.conf file.
The GRUB and /boot/grub/grub.conf file
This part applies for those who use GRUB as their boot loader. GRUB is another boot loader like
LILO but with many more useful feature and power. One of its main advantages compared to
LILO is the fact that it provides a small shell interface to manage the operating system. Also, it
doesn’t need to be updated each time you recompile a new kernel on your server.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-82-2048.jpg)
![General Security 0
CHAPTER 3
83
GRUB is very important since it is the first software program that runs when the computer starts
and we have to secure it as much as possible to avoid any possible problem. In its default
installation it’s already well protected and below I explain how its configuration file is made. In
regard to LILO, GRUB is really easy to use and configure. Below is a default GRUB configuration
file and security I recommend you to apply. The text in bold are the parts of the configuration file
that must be customized and adjusted to satisfy our needs.
• Edit the grub.conf file (vi /boot/grub/grub.conf) and set your needs. Below is
what we recommend you:
default=0
timeout=0
splashimage=(hd0,0)/grub/splash.xpm.gz
password --md5 $1$oKr0ÝmFo$tPYwkkvQbtqo1erwHj5wb/
title Red Hat Linux (2.4.18-3)
root (hd0,0)
kernel /vmlinuz-2.4.18-3 ro root=/dev/sda5
initrd /initrd-2.4.18-3.img
This tells the grub.conf file to set itself up for this particular configuration with:
default=0
The option “default” is used to define the default entry of the configuration file. The number “0”
mean that the following parameters are the default entry for the configuration of GRUB. In a server
configuration where Linux is the only operating system to boot, the default entry of “0” will be the
only one to use and we don’t need to define any additional entry.
timeout=0
This option “timeout” is used to define the timeout, in sec seconds to wait, before automatically
booting the default entry. As for LILO, one of the requirements of C2 security is that this interval
be set to 0 unless the system dual boots something else. One of the disadvantages to set this
option to “0” is that you will no longer be able to have access at boot time to the shell interface of
the software but this is not really a problem since all we need from the GRUB software is to boot
our operating system.
splashimage=(hd0,0)/grub/splash.xpm.gz
This option “splashimage” is an option added by Red Hat to boot the system with a graphical
image. The value is the path of the compressed image to use when booting GRUB. It’s to you to
keep this parameter on your system or to remove it. If you want to remove it, just delete the
above line with the compressed image from your server.
password --md5 $1$bgGCL/$4yF3t0py.IjU0LU.q7YfB1
This option “password” is used to inform GRUB to ask for a password and disallows any
interactive control, until you press the key <p> and enter a correct password. The option --md5
tells GRUB that a password in MD5 format is required as a value. If it is omitted, GRUB assumes
the specified password is in clear text.
When we have installed the operating system, we have already configured GRUB with a
password protection. This password is what you see here. If you want to change it, you have to
use the “grub-md5-crypt” command to generate a new encrypt password it in MD5 format.
• This can be done with the following command:
[root@dev /]# grub-md5-crypt
Password:
$1$bgGCL/$4yF3t0py.IjU0LU.q7YfB1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-83-2048.jpg)
![84
Once the above command has been issued, you have to cut and paste the encrypted password
to your configuration file.
title Red Hat Linux (2.4.18-3)
This option “title” is used to define a name to the contents of the rest of the line. It is directly
related to the default boot entry. What you enter here is what you will see during boot time. This
option is useful when we have more than one OS to start on our computer and allow us to give
the name that we want to distinguish them. You are free to enter whatever name you like.
root (hd0,0)
This option “root” is one of the most important parameter with GRUB and without it nothing will
work. It is used to define the current root device to use for booting the operating system. Its
definition and configuration is a little bit special as you can see. Here is an explanation of its
meaning. The “hd0” parameter represents using the entire disk and the “hd0,0” represents using
the partition of the disk (or the boot sector of the partition when installing GRUB). Don’t be
confused here because “hd” is valid for IDE and SCSI drives. There is no difference; you always
use “hd” even on SCSI drive.
kernel /vmlinuz-2.4.18-3 ro root=/dev/sda5
This option “kernel” is used to load the primary boot image (our kernel). The parameter to this
option is simply the path where GRUB should find the kernel image from which we want it to boot.
The additional lines are to inform it that kernel image is located on the sda5 partition on our
server and that we want to load it as read only for security reason.
initrd /initrd-2.4.18-3.img
This option “initrd” is optional and will appear into your GRUB configuration file only if you run a
SCSI computer. For IDE computer, this option is not required and should not be defined inside
the configuration file of GRUB. The parameter simply informs GRUB software where our initial ram
disk image is located on the server. GRUB reads this initial ram disk and loads it during startup.
The /etc/services file
The port numbers on which certain "standard" services are offered are defined in the RFC 1700
"Assigned Numbers". The /etc/services file enables server and client programs to convert
service names to these numbers (ports). The list is kept on each host and it is stored in the file
/etc/services. Only the "root" user is allowed to make modifications to this file. It is rare to
edit the /etc/services file since it already contains the more common service names / port
numbers. To improve security, we can set the immutable flag on this file to prevent unauthorized
deletion or modification.
• To immunize the /etc/services file, use the following command:
[root@deep /]# chattr +i /etc/services](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-84-2048.jpg)
![General Security 0
CHAPTER 3
85
The /etc/securetty file
The /etc/securetty file allows you to specify which TTY and VC (virtual console) devices the
“root” user is allowed to login on. The /etc/securetty file is read by the login program
(usually /bin/login). Its format is a list of the tty and vc devices names allowed, and for all
others that are commented out or do not appear in this file, root login is disallowed.
Disable any tty and vc devices that you do not need by commenting them out (# at the
beginning of the line) or by removing them.
• Edit the securetty file (vi /etc/securetty) and comment out or remove the lines:
vc/1
#vc/2
#vc/3
#vc/4
#vc/5
#vc/6
#vc/7
#vc/8
#vc/9
#vc/10
#vc/11
tty1
#tty2
#tty3
#tty4
#tty5
#tty6
#tty7
#tty8
#tty9
#tty10
#tty11
Which means root is allowed to login on only tty1 and vc/1. This is my recommendation,
allowing “root” to log in on only one tty or vc device and use the su or sudo command to
switch to “root” if you need more devices to log in as “root”.
Special accounts
It is important to DISABLE ALL default vendor accounts that you don’t use on your system
(some accounts exist by default even if you have not installed the related services on your
server). This should be checked after each upgrade or new software installation. Linux provides
these accounts for various system activities, which you may not need if the services are not
installed on your server. If you do not need the accounts, remove them. The more accounts you
have, the easier it is to access your system.
We assume that you are using the shadow password suite on your Linux system. If you are not,
you should consider doing so, as it helps to tighten up security somewhat. This is already set if
you’ve followed our Linux installation procedure and selected, under the “Authentication
Configuration”, the option to “Enable Shadow Passwords” (see the chapter related to the
“Installation of your Linux Server” for more information).
• To delete user on your system, use the following command:
[root@deep /]# userdel username
• To delete group on your system, use the following command:
[root@deep /]# groupdel username](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-85-2048.jpg)
![86
Step 1
First we will remove all default vendor accounts into the /etc/passwd file that are unnecessary
for the operation of the secure server configuration that we use in this book.
• Type the following commands to delete all default users accounts listed below.
[root@deep /]# userdel adm
[root@deep /]# userdel lp
[root@deep /]# userdel shutdown
[root@deep /]# userdel halt
[root@deep /]# userdel news
[root@deep /]# userdel mailnull
[root@deep /]# userdel operator
[root@deep /]# userdel games
[root@deep /]# userdel gopher
[root@deep /]# userdel ftp
[root@deep /]# userdel vcsa
WARNING: By default, the userdel command will not delete a user’s home directory. If you want
the home directories of accounts to be deleted too, then add the -r option to the userdel
command. Finally, the -r option must be used only when you have added a new user to the
server and want to remove them. It doesn’t need to be used for the removal of the above default
user’s accounts since they do not have a home directory.
Once the above list of users has been deleted from your Linux system, the /etc/passwd file will
look like this:
root:x:0:0:root:/root:/bin/bash
bin:x:1:1:bin:/bin:/sbin/nologin
daemon:x:2:2:daemon:/sbin:/sbin/nologin
sync:x:5:0:sync:/sbin:/bin/sync
mail:x:8:12:mail:/var/spool/mail:/sbin/nologin
uucp:x:10:14:uucp:/var/spool/mail:/sbin/nologin
nobody:x:99:99:Nobody:/:/sbin/nologin
rpm:x:37:37::/var/lib/rpm:/bin/bash
Step 2
After that we have removed all the unnecessary default vendor accounts into the /etc/passwd
file from our system, we will remove all default vendor accounts into the /etc/group file.
• Type the following commands to delete all default users groups accounts listed below.
[root@deep /]# groupdel adm
[root@deep /]# groupdel lp
[root@deep /]# groupdel news
[root@deep /]# groupdel games
[root@deep /]# groupdel dip](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-86-2048.jpg)
![General Security 0
CHAPTER 3
87
Once the above list of group users has been deleted from your Linux system the /etc/group
file will look like this:
root:x:0:root
bin:x:1:root,bin,daemon
daemon:x:2:root,bin,daemon
sys:x:3:root,bin
tty:x:5:
disk:x:6:root
mem:x:8:
kmem:x:9:
wheel:x:10:root
mail:x:12:mail
uucp:x:14:uucp
man:x:15:
lock:x:54:
nobody:x:99:
users:x:100:
slocate:x:21:
floppy:x:19:
utmp:x:22:
rpm:x:37:
Step 3
Now, you can add all the necessary and allowed users into the system. Below I show you how
you should add new user into your Linux server. Adding a new user into your server mean that
you have to create the username and assign him/her a password.
• To add a new user on your system, use the following command:
[root@deep /]# useradd username
For example:
[root@deep /]# useradd admin
• To add or change password for user on your system, use the following command:
[root@deep /]# passwd username
For example:
[root@deep /]# passwd admin
The output should look something like this:
Changing password for user admin
New UNIX password:
Retype new UNIX password:
passwd: all authentication tokens updated successfully
Step 4
The immutable bit can be used to prevent accidentally deleting or overwriting a file that must be
protected. It also prevents someone from creating a symbolic link to this file, which has been the
source of attacks involving the deletion of /etc/passwd, /etc/shadow, /etc/group or
/etc/gshadow files.
• To set the immutable bit on the passwords and groups files, use the following commands:
[root@deep /]# chattr +i /etc/passwd
[root@deep /]# chattr +i /etc/shadow
[root@deep /]# chattr +i /etc/group](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-87-2048.jpg)
![88
[root@deep /]# chattr +i /etc/gshadow
WARNING: In the future, if you intend to add or delete users, passwords, user groups, or group
files, you must unset the immutable bit on all those files or you will not be able to make and
update your changes. Also if you intend to install an RPM program that will automatically add a
new user to the different immunized passwd and group files, then you will receive an error
message during the install if you have not unset the immutable bit from those files.
• To unset the immutable bit on the passwords and groups files, use the commands:
[root@deep /]# chattr -i /etc/passwd
[root@deep /]# chattr -i /etc/shadow
[root@deep /]# chattr -i /etc/group
[root@deep /]# chattr -i /etc/gshadow
Control mounting a file system
You can have more control on mounting file systems like /var/lib, /home or /tmp partitions
with some nifty options like noexec, nodev, and nosuid. This can be setup in the /etc/fstab
file. The fstab file contains descriptive information about the various file system mount options;
each line addresses one file system.
Information related to security options in the fstab text file are:
defaults Allow everything (quota, read-write, and suid) on this partition.
noquota Do not set users quotas on this partition.
nosuid Do not set SUID/SGID access on this partition.
nodev Do not set character or special devices access on this partition.
noexec Do not set execution of any binaries on this partition.
quota Allow users quotas on this partition.
ro Allow read-only on this partition.
rw Allow read-write on this partition.
suid Allow SUID/SGID access on this partition.
NOTE: For more information on options that you can set in this file (fstab) see the man pages
about mount (8).
Step 1
• Edit the fstab file (vi /etc/fstab) and change it depending of your needs.
For example change:
LABEL=/home /home ext3 defaults 1 2
LABEL=/tmp /tmp ext3 defaults 1 2
LABEL=/var/lib /var/lib ext3 defaults 1 2
To read:
LABEL=/home /home ext3 defaults,nosuid 1 2
LABEL=/tmp /tmp ext3 defaults,nosuid,noexec 1 2
LABEL=/var/lib /var/lib ext3 defaults,nodev 1 2](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-88-2048.jpg)
![General Security 0
CHAPTER 3
89
Meaning, <nosuid>, do not allow set-user-identifier or set-group-identifier bits to take effect,
<nodev>, do not interpret character or block special devices on this file system partition, and
<noexec>, do not allow execution of any binaries on the mounted file system.
Step 2
Once you have made the necessary adjustments to the /etc/fstab file, it is time to inform the
Linux system about the modifications.
• This can be accomplished with the following commands:
[root@deep /]# mount /var/lib -oremount
[root@deep /]# mount /home -oremount
[root@deep /]# mount /tmp -oremount
Each file system that has been modified must be remounted with the command show above. In
our example we have modified the /var/lib, /home, and /tmp file system and it is for this
reason that we remount these files systems with the above commands.
• You can verify if the modifications have been correctly applied to the Linux system with
the following command:
[root@deep /]# cat /proc/mounts
rootfs / rootfs rw 0 0
/dev/root / ext3 rw 0 0
/proc /proc proc rw 0 0
/dev/sda1 /boot ext3 rw 0 0
/dev/sda8 /chroot ext3 rw 0 0
none /dev/pts devpts rw 0 0
/dev/sda7 /home ext3 rw,nosuid 0 0
none /dev/shm tmpfs rw 0 0
/dev/sda11 /tmp ext3 rw,nosuid,noexec 0 0
/dev/sda6 /usr ext3 rw 0 0
/dev/sda9 /var ext3 rw 0 0
/dev/sda10 /var/lib ext3 rw,nodev 0 0
This command will show you all the files systems on your Linux server with parameters applied to
them.
Mounting the /usr directory of Linux as read-only
It is allowable to mount your /usr partition read-only since this is, by definition, static data. Of
course, anyone with root access can remount it as writable, but a generic attack script may not
know this. On many Linux variants this directory resides in its own partition and the default
parameter is to mount it as read-write. We can change this parameter to make it read-only for
better security. Please be sure that your /usr directory is on its own partition or the following
hack will not work for you.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-89-2048.jpg)
![90
Step 1
Mounting the /usr partition as read-only eliminates possible problems that someone may try to
change or modify vital files inside it. To mount the /usr file system of Linux as read-only, follow
the simple steps below.
• Edit the fstab file (vi /etc/fstab) and change the line:
LABEL=/usr /usr ext3 defaults 1 2
To read:
LABEL=/usr /usr ext3 defaults,ro 1 2
We add the “ro” option to this line to specify to mount this partition as read-only.
Step 2
Make the Linux system aware about the modification you have made to the /etc/fstab file.
• This can be accomplished with the following command:
[root@deep /]# mount /usr -oremount
• Then test your results with the following command:
[root@deep /]# cat /proc/mounts
rootfs / rootfs rw 0 0
/dev/root / ext3 rw 0 0
/proc /proc proc rw 0 0
/dev/sda1 /boot ext3 rw 0 0
/dev/sda8 /chroot ext3 rw 0 0
none /dev/pts devpts rw 0 0
/dev/sda7 /home ext3 rw,nosuid 0 0
none /dev/shm tmpfs rw 0 0
/dev/sda11 /tmp ext3 rw,nosuid,noexec 0 0
/dev/sda6 /usr ext3 ro 0 0
/dev/sda9 /var ext3 rw 0 0
/dev/sda10 /var/lib ext3 rw,nodev 0 0
If you see something like: /dev/sda6 /usr ext3 ro 0 0, congratulations!
WARNING: If in the future you want to install some RPM package or program from source code, it is
important to reset the modification you have made to the /usr directory to its initial state (read-
write) or you will not be able to install new software because the /usr partition is set as read-
only. All you have to do if you want to put the /usr partition to its original state is to edit the
/etc/fstab file again and remove the “ro” option then remount the /usr file system with the
“mount -oremount” command again.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-90-2048.jpg)
![General Security 0
CHAPTER 3
91
Tighten scripts under /etc/init.d
Fix the permissions of the script files that are responsible for starting and stopping all your normal
processes that need to run at boot time.
• To fix the permissions of those files, use the following command:
[root@deep /]# chmod 0700 /etc/init.d/*
Which means just the super-user “root” is allowed to Read, Write, and Execute scripts files on
this directory. I don’t think regular users need to know what’s inside those script files.
WARNING: If you install a new program or update a program that use the init system V script
located under /etc/init.d/ directory, don’t forget to change or verify the permission of this
script file again.
Tighten scripts under /etc/cron.daily/
As for the above hack, we can tighten the security of all script files that are responsible for
executing scheduled job on our server. Those files have a default permission mode of (0755-
rwxr-xr-x), which is too high for what they should accomplish.
• To fix the permissions of those files, use the following command:
[root@deep /]# chmod 0550 /etc/cron.daily/*
The same is true for other cron directories under the /etc directory of your system. If files exist
under the other cron directories, then use the above command to change their default permission
mode for better security.
WARNING: If you install a new program or update a program that provides and install a cron file on
your server, don’t forget to change or verify the permission of this script file again.
Bits from root-owned programs
A regular user will be able to run a program as root if it is set to SUID root. All programs and files
on your computer with the ’s’ bits appearing on its mode, have the SUID (-rwsr-xr-x) or SGID
(-r-xr-sr-x) bit enabled. Because these programs grant special privileges to the user who is
executing them, it is important to remove the 's' bits from root-owned programs that won't
absolutely require such privilege.
This can be accomplished by executing the command chmod a-s with the name(s) of the
SUID/SGID files as its arguments.
Such programs include, but aren't limited to:
Programs you never use.
Programs that you don't want any non-root users to run.
Programs you use occasionally, and don't mind having to su (1) to root to run.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-91-2048.jpg)
![92
Step 1
We've placed an asterisk (*) next to each program we personally might disable and consider
being not absolutely required for the duty work of the server. Remember that your system needs
some suid root programs to work properly, so be careful.
• To find all files with the ‘s’ bits from root-owned programs, use the command:
[root@deep]# find / -type f ( -perm -04000 -o -perm -02000 ) -exec ls -l {} ;
*-rwsr-xr-x 1 root root 34296 Mar 27 20:40 /usr/bin/chage
*-rwsr-xr-x 1 root root 36100 Mar 27 20:40 /usr/bin/gpasswd
-rwxr-sr-x 1 root slocate 25020 Jun 25 2001 /usr/bin/slocate
-r-s--x--x 1 root root 15104 Mar 13 20:44 /usr/bin/passwd
*-r-xr-sr-x 1 root tty 6920 Mar 14 15:24 /usr/bin/wall
*-rws--x--x 1 root root 12072 Apr 1 18:26 /usr/bin/chfn
*-rws--x--x 1 root root 11496 Apr 1 18:26 /usr/bin/chsh
*-rws--x--x 1 root root 4764 Apr 1 18:26 /usr/bin/newgrp
*-rwxr-sr-x 1 root tty 8584 Apr 1 18:26 /usr/bin/write
-rwsr-xr-x 1 root root 21080 Apr 15 00:49 /usr/bin/crontab
*-rwsr-xr-x 1 root root 32673 Apr 18 17:40 /usr/sbin/ping6
*-rwsr-xr-x 1 root root 13994 Apr 18 17:40 /usr/sbin/traceroute6
-rwxr-sr-x 1 root utmp 6604 Jun 24 2001 /usr/sbin/utempter
-rws--x--x 1 root root 22388 Apr 15 18:15 /usr/sbin/userhelper
*-rwsr-xr-x 1 root root 17461 Apr 19 12:35 /usr/sbin/usernetctl
*-rwsr-xr-x 1 root root 35192 Apr 18 17:40 /bin/ping
*-rwsr-xr-x 1 root root 60104 Apr 1 18:26 /bin/mount
*-rwsr-xr-x 1 root root 30664 Apr 1 18:26 /bin/umount
-rwsr-xr-x 1 root root 19116 Apr 8 12:02 /bin/su
-r-sr-xr-x 1 root root 120264 Apr 9 23:24 /sbin/pwdb_chkpwd
-r-sr-xr-x 1 root root 16992 Apr 9 23:24 /sbin/unix_chkpwd
*-rwxr-sr-x 1 root root 14657 Apr 19 12:35 /sbin/netreport
Step 2
• To disable the suid bits on selected programs above, use the following commands:
[root@deep /]# chmod a-s /usr/bin/chage
[root@deep /]# chmod a-s /usr/bin/gpasswd
[root@deep /]# chmod a-s /usr/bin/wall
[root@deep /]# chmod a-s /usr/bin/chfn
[root@deep /]# chmod a-s /usr/bin/chsh
[root@deep /]# chmod a-s /usr/bin/newgrp
[root@deep /]# chmod a-s /usr/bin/write
[root@deep /]# chmod a-s /usr/sbin/ping6
[root@deep /]# chmod a-s /usr/sbin/traceroute6
[root@deep /]# chmod a-s /usr/sbin/usernetctl
[root@deep /]# chmod a-s /bin/ping
[root@deep /]# chmod a-s /bin/mount
[root@deep /]# chmod a-s /bin/umount
[root@deep /]# chmod a-s /sbin/netreport](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-92-2048.jpg)
![General Security 0
CHAPTER 3
93
Don’t let internal machines tell the server what their MAC address is
To avoid the risk that a user could easily change a computers IP address and appear as
someone else to the firewall, you can force the ARP cache entries of Linux using the arp
command utility. A special option can be used with the arp utility to avoid letting INTERNAL
machines tell the server what their MAC (Media Access Control) address is and the IP address
associated with it.
Step1
ARP is a small utility, which manipulates the kernel’s ARP (Address Resolution Protocol) cache.
Through all possible options associated with this utility, the primary one is clearing an address
mapping entry and manually setting up one. In the hope to more secure our server from the
INTERNAL, we will manually set MAC address (sometimes called Hardware addresses) of all
known computers in our network statically by using static ARP entries.
• For each IP address of INTERNAL computers in your network, use the following
command to know the MAC address associate with the IP address:
[root@deep /]# ifconfig
eth0 Link encap:Ethernet HWaddr 00:50:DA:C6:D3:FF
inet addr:207.35.78.3 Bcast:207.35.78.32 Mask:255.255.255.224
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:1887318 errors:0 dropped:0 overruns:1 frame:0
TX packets:2709329 errors:0 dropped:0 overruns:0 carrier:1
collisions:18685 txqueuelen:100
Interrupt:10 Base address:0xb000
eth1 Link encap:Ethernet HWaddr 00:50:DA:C6:D3:09
inet addr:192.168.1.11 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:182937 errors:0 dropped:0 overruns:0 frame:0
TX packets:179612 errors:0 dropped:0 overruns:0 carrier:0
collisions:7434 txqueuelen:100
Interrupt:11 Base address:0xa800
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:3924 Metric:1
RX packets:7465 errors:0 dropped:0 overruns:0 frame:0
TX packets:7465 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
The MAC (Media Access Control) address will be the letters and numbers that come after
“HWaddr” (the Hardware Address). In the above example our MAC address are:
00:50:DA:C6:D3:FF for the interface eth0 and 00:50:DA:C6:D3:09 for the interface eth1.
Step 2
Once we know the MAC (Media Access Control) address associated with IP address, we can add
them manually to the ARP entries of the Linux server.
• To add manually MAC address to ARP entries, use the following commands:
[root@deep /]# arp -s 207.35.78.3 00:50:DA:C6:D3:FF
[root@deep /]# arp -s 192.168.1.11 00:50:DA:C6:D3:09](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-93-2048.jpg)
![94
The “-s” option means to manually create an ARP address mapping entry for host hostname with
hardware address set to hw_addr class. You can add you ARP commands to the
/etc/rc.local file if you want to keep your configuration if the system reboots.
Step 3
• To verify if the modifications have been added to the system, use the following command:
[root@deep /]# arp
Address Hwtype Hwaddress Flags Mask Iface
207.35.78.3 ether 00:20:78:13:86:92 CM eth1
192.168.1.11 ether 00:E0:18:90:1B:56 CM eth1
WARNING: If you receive error message like: SIOCSARP: Invalid argument, it is because the MAC
(Media Access Control) address you want to add is the one of your server. You must add only MAC
address of INTERNAL computers in your private network. This hack doesn’t apply to external
node on the Internet.
You can now be reassured that someone will not change the system's IP address of an
INTERNAL system and get through. If they do change the IP address, the server simply won't
talk to them. With the new iptables tool of Linux, which replace the old ipchains utility for
packet filter administration and firewall setup, MAC addresses can be filtered and configured in the
firewall rules too.
Unusual or hidden files
It is important to look everywhere on the system for unusual or hidden files (files that start with a
period and are normally not shown by the “ls” command), as these can be used to hide tools and
information (password cracking programs, password files from other systems, etc.). A common
technique on UNIX systems is to put a hidden directory or file in a user's account with an unusual
name, something like '...' or '.. ' (dot dot space) or '..^G' (dot dot control-G). The find
program can be used to look for hidden files.
• To look for hidden files, use the following commands:
[root@deep /]# find / -name ".. " -print -xdev
[root@deep /]# find / -name ".*" -print -xdev | cat –v
/etc/skel/.bash_logout
/etc/skel/.bash_profile
/etc/skel/.bashrc
/etc/.pwd.lock
/root/.bash_logout
/root/.Xresources
/root/.bash_profile
/root/.bashrc
/root/.cshrc
/root/.tcshrc
/root/.bash_history
/usr/lib/perl5/5.6.1/i386-linux/.packlist
/home/admin/.bash_logout
/home/admin/.bash_profile
/home/admin/.bashrc
/.autofsck](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-94-2048.jpg)
![General Security 0
CHAPTER 3
95
Finding Group and World Writable files and directories
Group and world writable files and directories, particularly system files (partitions), can be a
security hole if a cracker gains access to your system and modifies them. Additionally, world-
writable directories are dangerous, since they allow a cracker to add or delete files as he or she
wishes in these directories. In the normal course of operation, several files will be writable,
including some from the /dev, /var/mail directories, and all symbolic links on your system.
• To locate all group & world-writable files on your system, use the command:
[root@deep /]# find / -type f ( -perm -2 -o -perm -20 ) -exec ls -lg {} ;
-rw-rw-r-- 1 root utmp 107904 Jun 17 12:04 /var/log/wtmp
-rw-rw-r-- 1 root utmp 4608 Jun 17 12:04 /var/run/utmp
• To locate all group & world-writable directories on your system, use the command:
[root@deep /]# find / -type d ( -perm -2 -o -perm -20 ) -exec ls -ldg {} ;
drwxrwxr-x 12 root man 4096 Jun 17 06:50 /var/cache/man/X11R6
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat1
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat2
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat3
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat4
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat5
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat6
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat7
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat8
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat9
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/catn
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat1
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat2
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat3
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat4
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat5
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat6
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat7
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat8
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat9
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/catn
drwxrwxr-x 12 root man 4096 Jun 17 06:50 /var/cache/man/local
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat1
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat2
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat3
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat4
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat5
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat6
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat7
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat8
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat9
drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/catn
drwxrwxr-x 3 root lock 4096 Jun 17 06:49 /var/lock
drwxrwxr-x 2 root root 4096 Apr 19 12:35 /var/run/netreport
drwxrwxr-x 2 root 12 4096 Jun 17 12:30 /var/spool/mail
drwxrwxrwt 2 root root 4096 Jun 17 11:29 /var/tmp
drwxrwxrwt 2 root root 4096 Jun 17 06:52 /tmp
WARNING: A file and directory integrity checker like “Tripwire” software can be used regularly to
scan, manage and find modified group or world writable files and directories easily. See later in
this book for more information about Tripwire.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-95-2048.jpg)
![96
Unowned files
Don’t permit any unowned file on your server. Unowned files may also be an indication that an
intruder has accessed your system. If you find unowned file or directory on your system, verify its
integrity, and if all looks fine, give it an owner name. Some time you may uninstall a program and
get an unowned file or directory related to this software; in this case you can remove the file or
directory safely.
• To locate files on your system that do not have an owner, use the following command:
[root@deep /]# find / -nouser -o -nogroup
WARNING: It is important to note that files reported under /dev/ directory don’t count.
Finding .rhosts files
Finding all existing .rhosts files that could exist on your server should be a part of your regular
system administration duties, as these files should not be permitted on your system. Remember
that a cracker only needs one insecure account to potentially gain access to your entire network.
Step 1
If you are doing a new install of Linux (like we did) you should not have any .rhosts files on
your system. If the result returns nothing, then you are safe and your system contain no .rhosts
files in the /home directory at this time.
• You can locate all existing .rhosts files on your system with the following command:
[root@deep /]# find /home -name .rhosts
Step 2
You can also use a cron job to periodically check for, report the contents of, and delete
$HOME/.rhosts files. Also, users should be made aware that you regularly perform this type of
audit, as directed by your security policy.
• Create the rhosts.cron file (touch /etc/cron.daily/rhosts.cron) and add the
following lines inside the script file.
#!/bin/sh
/usr/bin/find /home -name .rhosts | (cat <<EOF
This is an automated report of possible existent ..rhosts files on
the server deep.openna.com, generated by the find utility command.
New detected ..rhosts. files under the ./home/. directory include:
EOF
cat
) | /bin/mail -s "Content of .rhosts file audit report" root
• Now make this script executable, verify the owner, and change the group to “root”.
[root@deep /]# chmod 550 /etc/cron.daily/rhosts.cron
[root@deep /]# chown 0.0 /etc/cron.daily/rhosts.cron
Each day mail will be sent to “root” with a subject:” Content of .rhosts file audit report” containing
potential new .rhosts files.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-96-2048.jpg)
![General Security 0
CHAPTER 3
97
Physical hard copies of all-important logs
One of the most important security considerations is the integrity of the different log files under
the /var/log directory on your server. If despite each of the security functions put in place on
our server, a cracker can gain access to it; our last defence is the log file system, so it is very
important to consider a method of being sure of the integrity of our log files.
If you have a printer installed on your server, or on a machine on your network, a good idea
would be to have actual physical hard copies of all-important logs. This can be easily
accomplished by using a continuous feed printer and having the syslog program sending all
logs you think are important out to /dev/lp0 (the printer device). Cracker can change the files,
programs, etc on your server, but can do nothing when you have a printer that prints a real paper
copy of all of your important logs.
As an example:
For logging of all telnet, mail, boot messages and ssh connections from your server to the
printer attached to THIS server, you would want to add the following line to the
/etc/syslog.conf file:
Step 1
• Edit the syslog.conf file (vi /etc/syslog.conf) and add at the end of this file the
following line:
authpriv.*;mail.*;local7.*;auth.*;daemon.info /dev/lp0
Step 2
• Now restart your syslog daemon for the change to take effect:
[root@deep /]# /etc/init.d/syslog restart
Shutting down kernel logger: [OK]
Shutting down system logger: [OK]
Starting system logger: [OK]
Starting kernel logger: [OK]
As an example:
For logging of all telnet, mail, boot messages and ssh connections from your server to the
printer attached to a REMOTE server in your local network, then you would want to add the
following line to /etc/syslog.conf file on the REMOTE server.
Step 1
• Edit the syslog.conf file (vi /etc/syslog.conf) on the REMOTE server (for
example: printer.openna.com) and add at the end of this file the following line:
authpriv.*;mail.*;local7.*;auth.*;daemon.info /dev/lp0
If you don’t have a printer in your network, you can also copy all the log files to another machine;
simply omit the above first step of adding /dev/lp0 to your syslog.conf file on remote and go
directly to the “-r” option second step on remote. Using the feature of copying all the log files to
another machine will give you the possibility to control all syslog messages on one host and will
tear down administration needs.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-97-2048.jpg)
![98
Step 2
Since the default configuration of the syslog daemon is to not receive any messages from the
network, we must enable on the REMOTE server the facility to receive messages from the
network. To enable the facility to receive messages from the network on the REMOTE server,
add the following option “-r” to your syslog daemon script file (only on the REMOTE host):
• Edit the syslog daemon (vi +24 /etc/rc.d/init.d/syslog) and change:
daemon syslogd -m 0
To read:
daemon syslogd -r -m 0
Step 3
• Restart your syslog daemon on the remote host for the change to take effect:
[root@mail /]# /etc/init.d/syslog restart
Shutting down kernel logger: [OK]
Shutting down system logger: [OK]
Starting system logger: [OK]
Starting kernel logger: [OK]
Step 4
• Edit the syslog.conf file (vi /etc/syslog.conf) on the LOCAL server, and add at
the end of this file the following line:
authpriv.*;mail.*;local7.*;auth.*;daemon.info @printer
Where (printer) represent the hostname of the REMOTE server. Now if anyone ever hacks
your machine and attempts to erase vital system logs, you still have a hard copy of everything. It
should then be fairly simple to trace where they came from and deal with it accordingly.
Step 5
• Restart your syslog daemon on the LOCAL server for the change to take effect:
[root@deep /]# /etc/init.d/syslog restart
Shutting down kernel logger: [OK]
Shutting down system logger: [OK]
Starting system logger: [OK]
Starting kernel logger: [OK]
WARNING: Never use your Gateway Server as a host to control all syslog messages; this is a
very bad idea. More options and strategies exist with the sysklogd program, see the man pages
about sysklogd (8), syslog(2), and syslog.conf(5) for more information.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-98-2048.jpg)
![General Security 0
CHAPTER 3
99
Getting some more security by removing manual pages
Here we have to think a little bit about manual pages installed on all Linux system. Manual pages
also known as man-pages are compressed files located under the /usr/share/man directory
on your system. These documentation files are very useful to get quick information on how
service, program, commands, and configuration files of specific software work. These files are
readable by the man program and depend of other installed software on Linux to work and display
the information.
On production servers where specific task are assigned and where we only run services to the
internal or external, does we really need to have these manual pages and related software
installed? Do we will connect to these production servers to read these manual pages? Does this
is really important to have them duplicated on all of our different servers? Personally, I don’t think
because we can have all of these useful documentation files available on our Linux workstation or
development server each time we need to consult them.
If you have made attention to what we have done previously to secure our server, you will
remember that most of all group and world-writable directories on our system comes from the
/var/cache directory which is owned by the man program associated with manual pages. By
removing manual pages and related software from our server, we can get some more security
and save some not negligible space which could help when we scan our server with integrity tool
like Tripwire. This also allow us to remove other software directly related to man program and
limit the number of installed component on our production server without scarifying in the
functionally of the server. If this is what you want to do, here are the steps to follow.
Step 1
First of all, we should remove the man software from our system. The man software is the
program we use to read manual pages. By removing this software we eliminate most of all group
and world-writable directories from our system.
• To remove the man software, use the following command:
[root@deep /]# rpm -e man
Step 2
Once the above software has been removed, we can continue with groff. Groff is a document
formatting system that takes standard text and formatting commands as input and produces
formatted output. This software is used by man to format man-pages.
• To remove the groff software, use the following command:
[root@deep /]# rpm -e groff
Step 3
Because we don’t use manual pages anymore on our production servers, we can remove all
man-pages that are already installed and available under the /usr/share/man directory.
• To remove all preinstalled man-pages from your server, use the following commands:
[root@deep /]# cd /usr/share/man/
[root@deep man]# rm -f man*/*.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-99-2048.jpg)
![100
Step 4
Finally, it is important to note that any future installation and upgrade of RPM packages on the
system should be made with the “--excludedocs” option. This RPM option allow us to install or
upgrade the RPM package without the need to install the documentation part that may comes with
the software. For example, if I want to install or upgrade the bind package, I will use the
following RPM command.
• To install or upgrade RPM without documentation, use the following command:
[root@deep /]# rpm –Uvh --exculdedocs bind-version.i386.rpm
System is compromised!
If you believe that your system has been compromised, contact CERT ® Coordination Center or
your representative in FIRST (Forum of Incident Response and Security Teams).
Internet Email: cert@cert.org
CERT Hotline: (+1) 412-268-7090
Facsimile: (+1) 412-268-6989
CERT/CC personnel answer 8:00 a.m. – 8:00 p.m. EST (GMT –5)/EDT (GMT –4)) on working
days; they are on call for emergencies during other hours and on weekends and holidays.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-100-2048.jpg)
![Pluggable Authentication Modules 0
CHAPTER 4
105
Disabling console program access
In a safe environment, where we are sure that console is secured because passwords for BIOS
and GRUB or LILO are set and all physical power and reset switches on the system are disabled,
it may be advantageous to entirely disable all console-equivalent access to programs like
poweroff, reboot, and halt for regular users on your server.
• To do this, run the following command:
[root@deep /]# rm -f /etc/security/console.apps/<servicename>
Where <servicename> is the name of the program to which you wish to disable console-
equivalent access.
• To disable console program access, use the following commands:
[root@deep /]# rm -f /etc/security/console.apps/halt
[root@deep /]# rm -f /etc/security/console.apps/poweroff
[root@deep /]# rm -f /etc/security/console.apps/reboot
This will disable console-equivalent access to programs halt, poweroff, and reboot.
Disabling all console access
The Linux-PAM library installed by default on your system allows the system administrator to
choose how applications authenticate users, such as for console access, program and file
access.
Step 1
In order to disable all these accesses for the users, you must comment out all lines that refer to
pam_console.so in the /etc/pam.d directory. This step is a continuation of the hack
“Disabling console program access”. The following script will do the trick automatically for you.
• As ‘root’ creates the disabling.sh script file (touch disabling.sh) and add the
following lines inside:
# !/bin/sh
cd /etc/pam.d
for i in * ; do
sed '/[^#].*pam_console.so/s/^/#/' < $i > foo && mv foo $i
done
Step 2
Now, we have to make the script executable and run it.
• This can be done with the following commands:
[root@deep /]# chmod 700 disabling.sh
[root@deep /]# ./disabling.sh
This will comment out all lines that refer to pam_console.so for all files located under
/etc/pam.d directory. Once the script has been executed, you can remove it from your system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-105-2048.jpg)
![108
Step 1
The default console.perms configuration file of Linux is secure enough for regular use of the
system where an Xwindow interface is considered to be installed but in a highly secure
environment where the Graphical User Interface (GUI) is not installed or where some special
devices like sound, jaz, etc have no reason to exist, we can tighten the console.perms
security file of Linux to be more secure by eliminating non-existent or unneeded privileged users
to have capabilities that they would not otherwise have.
• Edit the console.perms file (vi /etc/security/console.perms), and change the
default lines inside this file:
# file classes -- these are regular expressions
<console>=tty[0-9][0-9]* vc/[0-9][0-9]* :[0-9].[0-9] :[0-9]
<xconsole>=:[0-9].[0-9] :[0-9]
# device classes -- these are shell-style globs
<floppy>=/dev/fd[0-1]*
/dev/floppy/* /mnt/floppy*
<sound>=/dev/dsp* /dev/audio* /dev/midi*
/dev/mixer* /dev/sequencer
/dev/sound/* /dev/beep
<cdrom>=/dev/cdrom* /dev/cdroms/* /dev/cdwriter* /mnt/cdrom*
<pilot>=/dev/pilot
<jaz>=/mnt/jaz*
<zip>=/mnt/pocketzip* /mnt/zip*
<ls120>=/dev/ls120 /mnt/ls120*
<scanner>=/dev/scanner /dev/usb/scanner*
<rio500>=/dev/usb/rio500
<camera>=/mnt/camera* /dev/usb/dc2xx* /dev/usb/mdc800*
<memstick>=/mnt/memstick*
<flash>=/mnt/flash*
<diskonkey>=/mnt/diskonkey*
<rem_ide>=/mnt/microdrive*
<fb>=/dev/fb /dev/fb[0-9]*
/dev/fb/*
<kbd>=/dev/kbd
<joystick>=/dev/js[0-9]*
<v4l>=/dev/video* /dev/radio* /dev/winradio* /dev/vtx* /dev/vbi*
/dev/video/*
<gpm>=/dev/gpmctl
<dri>=/dev/nvidia* /dev/3dfx*
<mainboard>=/dev/apm_bios
# permission definitions
<console> 0660 <floppy> 0660 root.floppy
<console> 0600 <sound> 0600 root
<console> 0600 <cdrom> 0660 root.disk
<console> 0600 <pilot> 0660 root.uucp
<console> 0600 <jaz> 0660 root.disk
<console> 0600 <zip> 0660 root.disk
<console> 0600 <ls120> 0660 root.disk
<console> 0600 <scanner> 0600 root
<console> 0600 <camera> 0600 root
<console> 0600 <memstick> 0600 root
<console> 0600 <flash> 0600 root
<console> 0600 <diskonkey> 0660 root.disk
<console> 0600 <rem_ide> 0660 root.disk
<console> 0600 <fb> 0600 root
<console> 0600 <kbd> 0600 root
<console> 0600 <joystick> 0600 root](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-108-2048.jpg)
![Pluggable Authentication Modules 0
CHAPTER 4
109
<console> 0600 <v4l> 0600 root
<console> 0700 <gpm> 0700 root
<console> 0600 <mainboard> 0600 root
<console> 0600 <rio500> 0600 root
<xconsole> 0600 /dev/console 0600 root.root
<xconsole> 0600 <dri> 0600 root
To read :
# file classes -- these are regular expressions
<console>=tty[0-9][0-9]* vc/[0-9][0-9]* :[0-9].[0-9] :[0-9]
# device classes -- these are shell-style globs
<floppy>=/dev/fd[0-1]*
/dev/floppy/* /mnt/floppy*
<cdrom>=/dev/cdrom* /dev/cdroms/* /dev/cdwriter* /mnt/cdrom*
<pilot>=/dev/pilot
<fb>=/dev/fb /dev/fb[0-9]*
/dev/fb/*
<kbd>=/dev/kbd
<gpm>=/dev/gpmctl
<mainboard>=/dev/apm_bios
# permission definitions
<console> 0660 <floppy> 0660 root.floppy
<console> 0600 <cdrom> 0660 root.disk
<console> 0600 <pilot> 0660 root.uucp
<console> 0600 <fb> 0600 root
<console> 0600 <kbd> 0600 root
<console> 0700 <gpm> 0700 root
<console> 0600 <mainboard> 0600 root
Here we removed every privileged user related to the Graphical User Interface and others related
to sound, zip drive, jaz drive, scanner, joystick and video media at the physical console
on the server.
Putting limits on resource
The limits.conf file located under the /etc/security directory can be used to control and
limit resources for the users on your system. It is important to set resource limits on all your users
so they can't perform Denial of Service attacks (number of processes, amount of memory, etc) on
the server. These limits will have to be set up for the user when he or she logs in.
Step 1
For example, limits for all users on your system might look like this:
• Edit the limits.conf file (vi /etc/security/limits.conf) and change the lines:
#* soft core 0
#* hard rss 10000
#@student hard nproc 20
To read:
* hard core 0
* hard rss 5000
* hard nproc 35](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-109-2048.jpg)
![110
This says to prohibit the creation of core files “core 0”, restrict the number of processes to 35
“nproc 35”, and restrict memory usage to 5M “rss 5000” for everyone except the super user
“root”. All of the above only concerns users who have entered through the login prompt on your
system. With this kind of quota, you have more control on the processes, core files, and memory
usage that users may have on your system. The asterisk “*” mean: all users that logs in on the
server.
Putting an asterisk “*” to cover all users can pose problem with daemon users account like “www”
for a web server, “mysql” for a SQL database server, etc. If we put an asterisk, then, these users
will be affected by the restriction and limitation of processes or memory usage.
To solve the problem, we can choose an existing group name in our system and add every
regular user to this group. In this manner, the restrictions and limitations will apply to all users
who are members of this group name only. A special group account named “users” can be used
for this purpose. This is the recommended method on putting limit on user resources.
• Edit the limits.conf file (vi /etc/security/limits.conf) and change the lines:
#* soft core 0
#* hard rss 10000
#@student hard nproc 20
To read:
@users hard core 0
@users hard rss 5000
@users hard nproc 35
If you decide to use a group name like “@users” to control and limit resources for the users on
your system, then it is important to not forget to change the GUI (Group User ID) of these users
to be “100”. “100” is the numeric value of the user’s ID “users”.
• The command to create a new user with group name which is set by default to users is:
[root@deep /]# useradd -g100 admin
The “-g100” option represents the number of the user’s initial login group and in our case “100”
is the group account name “users”. The “admin” parameter is the user name we want to add to
the group name “users”.
WARNING: Use the same command above for all users on your system you want to be member of
the “users” group account. It is also preferable to set this parameter first before adding users to
the system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-110-2048.jpg)
![112
Blocking; su to root, by one and sundry
The su (Substitute User) command allows you to become other existing users on the system. For
example you can temporarily become ‘root’ and execute commands as the super-user ‘root’.
Step 1
If you don’t want anyone to su to root or want to restrict the su command to certain users then
uncomment the following line of your su configuration file in the /etc/pam.d directory. We
highly recommend that you limit the persons allowed to su to the root account.
• Edit the su file (vi /etc/pam.d/su) and uncomment the following line in the file:
auth required /lib/security/pam_wheel.so use_uid
After this line has been uncommented, the /etc/pam.d/su file should look like this:
#%PAM-1.0
auth sufficient /lib/security/pam_rootok.so
auth required /lib/security/pam_wheel.so use_uid
auth required /lib/security/pam_stack.so service=system-auth
account required /lib/security/pam_stack.so service=system-auth
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
session optional /lib/security/pam_xauth.so
Which means only those who are members of the “wheel” group can su to root; it also includes
logging. Note that the “wheel” group is a special account on your system that can be used for
this purpose. You cannot use any group name you want to make this hack. This hack combined
with specifying which TTY and VC devices super-user root is allowed to login on will improve
your security a lot on the system.
Step 2
Now that we have defined the “wheel” group in our /etc/pam.d/su file configuration, it is time
to add some users who will be allowed to su to super-user “root” account.
• If you want to make, for example, the user “admin” a member of the “wheel” group, and
thus be able to su to root, use the following command:
[root@deep /]# usermod -G10 admin
Which means “G” is a list of supplementary groups, where the user is also a member of. “10” is
the numeric value of the user’s ID “wheel”, and “admin” is the user we want to add to the
“wheel” group. Use the same command above for all users on your system you want to be able
to su to super-user “root” account.
NOTE: For Linux users, who use the Xwindow interface, it is important to note that if you can't su
in a GNOME terminal, it’s because you’ve used the wrong terminal. (So don't think that this advice
doesn't work simply because of a GNOME terminal problem!)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-112-2048.jpg)
![Pluggable Authentication Modules 0
CHAPTER 4
113
Facultative:
A special line exists in the su file /etc/pam.d/su which allows you to implicitly trust users in the
“wheel” group (for security reasons, I don’t recommend using this option). This mean that all
users who are members of the “wheel” group can su to root without the need to enter the
super-user “root” password.
• To allow users who are members of the “wheel” group to su to root account without the
need to enter the “root” password, edit the su file (vi /etc/pam.d/su) and
uncomment the following line in the file:
auth sufficient /lib/security/pam_wheel.so trust use_uid
After this line has been uncommented, the /etc/pam.d/su file should look like this:
#%PAM-1.0
auth sufficient /lib/security/pam_rootok.so
auth sufficient /lib/security/pam_wheel.so trust use_uid
auth required /lib/security/pam_stack.so service=system-auth
account required /lib/security/pam_stack.so service=system-auth
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
session optional /lib/security/pam_xauth.so
Using sudo instead of su for logging as super-user
There is a security tool called “sudo” that we discuss into this book. This security software allow
us to archive the same result as using the su command to get root privilege on the server but in
a more secure and informative way. With sudo installed in our server, we can get information
about who is connected as super-user root as well as many other useful features. Please see
the chapter related to this security program into this book for more information about sudo.
If you want to use sudo to allow and control which is allowed to logging as super-user root on
your server, then you no longer need to use the su command of Linux to archive this task and we
can remove the SUID bit on this command to completely disable su and use sudo.
This let us remove one more SUID bit on our secure server and have a more complete and
powerful security software to control access to super-user root. This is the method I highly
recommend you to use instead of the su command of Linux.
Step 1
To archive this result, we have to remove the SUID bit of the su command and install the sudo
security software as explained further down in this book. This also implies that we don’t need to
modify the above su configuration file on our system. To recap, all we need to do is to remove
the SUID bit on the su command, and install sudo in our server.
• To remove the SUID bit on the su binary, use the following command:
[root@deep /]# chmod a-s /bin/su](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-113-2048.jpg)
![General Optimization 0
CHAPTER 5
121
Some misunderstanding in the compiler flags options
At lot of discussions exist in the Linux community about the “-O” option and its level numbers.
Some Linux users try to convince that level number up to “-O3” like “-O9” will produce faster
program. The “-O9” flag doesn't do anything over “-O3”, if you don't believe me make a small file,
call it testO3.c and see:
Step 1
• Create the testO3.c file with the following command:
[root@deep tmp]# touch testO3.c
Step 2
• Run the GCC compiler with “-O3” flag through the testO3.c file with the command:
[root@deep tmp]# gcc -O3 -S -fverbose-asm testO3.c
Step 3
Look at testO3.s that it made, then run again with “-O9” and compare the output.
• Create the testO9.c file with the following command:
[root@deep tmp]# touch testO9.c
Step 4
• Run the GCC compiler again with “-O9” flag through the testO9.c file with command:
[root@deep tmp]# gcc -O9 -S -fverbose-asm testO9.c
Step 5
Now if you compare the output you will see no difference between the both files.
• To compare the output, use the following command:
[root@deep tmp]# diff testO3.s testO9.s > difference
WARNING: The “-O3” flag level number is the best and highest optimization flag you can use
during optimization of programs under Linux.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-121-2048.jpg)
![General Optimization 0
CHAPTER 5
122
The gcc specs file
The /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file of Red Hat Linux is a set of
defines that the gcc compiler uses internally to set various aspects of the compile environment.
All customizations that you put in this file will apply for the entire variable environment on your
system, so putting optimization flags in this file is a good choice.
To squeeze the maximum performance from your x86 programs, you can use full optimization
when compiling with the “-O3” flag. Many programs contain “-O2” in the Makefile. The “-O3”
level number is the highest level of optimization. It will increase the size of what it produces, but it
runs faster in most case. You can also use the “-march=cpu_type” switch to optimize the
program for the CPU listed to the best of GCC’s ability. However, the resulting code will only be run
able on the indicated CPU or higher.
Below are the optimization flags that we recommend you to put in your /usr/lib/gcc-
lib/i386-redhat-linux/2.96/specs file depending on your CPU architecture. The
optimization options apply only when we compile and install a new program in our server. These
optimizations don’t play any role in our Linux base system; it just tells our compiler to optimize the
new programs that we will install with the optimization flags we have specified in the
/usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file. Adding options listed below
depending of your CPU architecture to the gcc 2.96 specs file will save you having to change
every CFLAGS in future Makefiles.
Step 1
The first thing to do is to verify the compiler version installed on your Linux server.
• To verify the compiler version installed on your system, use the command:
[root@deep /]# gcc -v
Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs
gcc version 2.96 20000731 (Red Hat Linux 7.3 2.96-110)
Step 2
For CPU i686 or PentiumPro, Pentium II, Pentium III, and Athlon
Edit the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file, scroll down a ways...
You'll see a section like the following:
*cpp_cpu_default:
-D__tune_i386__
*cpp_cpu:
-Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__
%{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__
%{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__
%{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro -
D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__
%{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:-
D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:-
D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__
}%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:-
D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__
}%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}}
*cc1_cpu:
%{!mcpu*: %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium}
%{mpentiumpro:-mcpu=pentiumpro}}](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-122-2048.jpg)
![General Optimization 0
CHAPTER 5
126
Step3
Once our optimization flags have been applied to the gcc 2.96 specs file, it time to verify if the
modification work.
• To verify if the optimization work, use the following commands:
[root@deep tmp]# touch cpu.c
[root@deep tmp]# gcc cpu.c -S -fverbose-asm
[root@deep tmp]# less cpu.s
What you'll get is a file that contains depending of options you have chose, something like:
.file "ccnVPjeW.i"
.version "01.01"
# GNU C version 2.96 20000731 (Red Hat Linux 7.3 2.96-110) (i386-redhat-linux)
compiled by GNU C version 2.96 20000731 (Red Hat Linux 7.3 2.96-110).
# options passed: -O2 -march=i686 -funroll-loops -fverbose-asm
# options enabled: -fdefer-pop -foptimize-sibling-calls -fcse-follow-jumps
# -fcse-skip-blocks -fexpensive-optimizations -fthread-jumps
# -fstrength-reduce -funroll-loops -fpeephole -fforce-mem -ffunction-cse
# -finline -fkeep-static-consts -fcaller-saves -fpcc-struct-return -fgcse
# -frerun-cse-after-loop -frerun-loop-opt -fdelete-null-pointer-checks
# -fschedule-insns2 -fsched-interblock -fsched-spec -fbranch-count-reg
# -fnew-exceptions -fcommon -fverbose-asm -fgnu-linker -fregmove
# -foptimize-register-move -fargument-alias -fstrict-aliasing
# -fmerge-constants -fident -fpeephole2 -fmath-errno -m80387 -mhard-float
# -mno-soft-float -mieee-fp -mfp-ret-in-387 -march=i686
gcc2_compiled.:
.ident "GCC: (GNU) 2.96 20000731 (Red Hat Linux 7.3 2.96-110)"
WARNING: In our example we are optimized the specs file for a i686 CPU processor. It is important
to note that most of the “-f” options are automatically included when you use “-O2” and don't
need to be specified again. The changes that were shown were made so that a command like
"gcc" would really be the command "gcc -march=i686" without having to change every single
Makefile which can really be a pain.
Below is the explanation of the different optimization options we use:
• The “-march=cpu_type” optimization flag
The “-march=cpu_type” optimization option will set the default CPU to use for the
machine type when scheduling instructions.
• The “-funroll-loops” optimization flag
The “-funroll-loops” optimization option will perform the optimization of loop
unrolling and will do it only for loops whose number of iterations can be determined at
compile time or run time.
• The “-fomit-frame-pointer” optimization flag
The “-fomit-frame-pointer” optimization option, one of the most interesting, will
allow the program to not keep the frame pointer in a register for functions that don't need
one. This avoids the instructions to save, set up and restores frame pointers; it also
makes an extra register available in many functions and makes debugging impossible on
most machines.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-126-2048.jpg)
![General Optimization 0
CHAPTER 5
127
WARNING: All future optimizations that we will describe in this book refer by default to a Pentium
PRO/II/III and higher i686 CPU family. So you must adjust the compilation flags for your specific
CPU processor type in the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file and
during your compilation time.
Striping all binaries and libraries files
When compiler builds program it add many comments and other stuffs inside the resulting binary
and library code which are used to debug application on the system or to make the code more
readable by developers. To get the latest bit of optimization, we can remove all comments and
other unneeded stuffs since there are only used for debugging purpose. This will make the
software runs a little bit faster because it will have fewer codes to read when executing.
I don’t know if it’s a good idea to talk about this hack because it’s really dangerous to apply and
can make your system unstable or completely unworkable if you don’t take care of what you do.
The process of eliminating all unneeded comments and other unneeded stuffs from your binaries
and libraries files is made by the use of the strip command of Linux. This command should be
used with care and in the good manner or you will certainly have a bad surprise.
Bellow, I will explain you how to apply it on your system and on which files or where you should
use it. It is very important to know that it’s NOT all binaries and especially libraries files that need
to be striped by this method but ONLY some of them. If you apply this hack on your entire
system, then something will inevitably break, you have been warned.
Finally, you should use this hack on servers where you DON’T compile software. If you compile
software on the server where you want to apply this hack, then nothing will work and you will not
be able to compile any software on it. Use this hack on server which doesn’t have any compiler
packages installed to make compilation.
Step 1
The first step in our procedure will be to be sure that the strip command is available on our
server; this command comes from the “binutils” RPM package. Therefore, if it is not installed,
install it from your CD-ROM.
Step 2
Once the strip program is installed on our server, it’s time to strip the required files. With the
commands below, we strip all binaries program available under the /bin, /sbin, /usr/bin and
/usr/sbin directories of your server.
• To strip all binaries program, use the following commands:
[root@deep /]# strip /bin/*
[root@deep /]# strip /sbin/*
[root@deep /]# strip /usr/bin/*
[root@deep /]# strip /usr/sbin/*
NOTE: When issuing the above commands, you will receive some error messages like “File
format not recognized” on your terminal. This is normal because some of the binaries files
are symbolic link pointing to other binaries on your system and the strip command generate the
warning because it cannot strip symbolic links.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-127-2048.jpg)
![General Optimization 0
CHAPTER 5
128
Step 3
Now, it’s time to strip the libraries files. This is where the action can become dangerous if you
don’t take care or abuse the strip command. With the commands below, we strip all libraries
files available under the /lib and /usr/lib directory of the system.
• To strip all libraries files, use the following command:
[root@deep /]# strip -R .comment /usr/lib/*.so.*
[root@deep /]# strip -R .comment /lib/*.so.*
NOTE: Make attention to the above command, you can see here that I use the “-R” option with the
strip command. This option allows us to select a specific name to strip from the target libraries
files. With the “.comment” name, we inform the command to remove any lines inside the libraries
codes where this name appears. You can see that I don’t use the strip command without any
option as I do for the above step related to binaries program. This is very important and you
should never use the strip command without the above option to strip libraries files on your
system.
Tuning IDE Hard Disk Performance
Accessing a hard disk can be 50 to 100 times slower than reading data from RAM. File caches
using RAM can alleviate this. However, low memory conditions will reduce the amount of memory
available for the file-system cache slowing things down. File systems can also become heavily
fragmented, slowing down disk accesses. Heavy use of symbolic links on Unix systems can slow
down disk accesses too.
Default Linux installs are also notorious for setting hard disk default settings which are tuned for
compatibility and not for speed. Use the command hdparm to tune your Linux hard disk settings.
The hdparm is a tool, which can be used to tune and improve the performance of your IDE hard
disk. By default, any IDE drives you have in your Linux system are not optimized. Even if you
have an ULTRA DMA system you will not be able to take full advantage of its speed if you are not
using the hdparm tool to enable its features. This is because there is many different hard drive
makes and models and Linux cannot know every feature of each one.
Performance increases have been reported on massive disk I/O operations by setting the IDE
drivers to use DMA, 32-bit transfers and multiple sector modes. The kernel seems to use more
conservative settings unless told otherwise. The magic command to change the setting of your
drive is hdparm.
Before going into the optimization of your hard drive, it is important to verify that the hdparm
package is installed in your system. If you have followed every step during the installation of
Linux on your computer, then this package is not installed.
To verify if hdparm package is installed on your system, use the command:
[root@deep /]# rpm -q hdparm
package hdparm is not installed
If the hdparm package seems not to be installed, you’ll need to mount your CD-ROM drive
containing the Linux CD-ROM Part 1 and install it.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-128-2048.jpg)
![General Optimization 0
CHAPTER 5
129
• To mount the CD-ROM drive, use the following commands:
[root@deep /]# mount /dev/cdrom /mnt/cdrom/
had: ATAPI 32X CD-ROM drive, 128kB Cache
mount: block device dev/cdrom is write-protected, mounting read-only
• To install the hdparm package on your Linux system, use the following command:
[root@deep /]# cd /mnt/cdrom/RedHat/RPMS/
[root@deep RPMS]# rpm -Uvh hdparm-version.i386.rpm
hdparm ##################################################
• To unmount your CD-ROM drive, use the following command:
[root@deep RPMS]# cd /; umount /mnt/cdrom/
Once hdparm package is installed on the system, it is time to go into the optimization of your
hard drive. It is important to note that depending on your model and make, there will be some
parameters that will apply and other that don’t. It is to your responsibility to know and understand
your disk drive before applying any optimization parameters as described below.
Finally, and especially for UltraDMA systems, it is vital to verify under your BIOS settings if the
parameters related to DMA support on your computer are enabled or you will inevitably break your
hard disk. You have been warned.
Step 1
The first parameter applies to the majority of all modern drives and models in the market and
enables 32-bit I/O over PCI buses. This option is one of the most important and will usually
double the speed of your drive.
• To enable 32-bit I/O over the PCI buses, use the following command:
[root@deep /]# /sbin/hdparm -c3 /dev/hda (or hdb, hdc etc).
This will usually, depending on your IDE Disk Drive model, cut the timing buffered disk reads time
by two. The hdparm (8) manpage says that you may need to use “-c3” for many chipsets since it
works with nearly all 32-bit IDE chipsets. All (E)IDE drives still have only a 16-bit connection
over the ribbon cable from the interface card.
Step 2
The second parameter applies only on standard DMA disk and will activate the simple DMA feature
of the disk. This feature is for old disk drives with DMA capabilities.
• To enable DMA, use the following command:
[root@deep /]# /sbin/hdparm -d1 /dev/hda (or hdb, hdc etc).
This may depend on support for your motherboard chipset being compiled into your kernel. Also,
this command will enable DMA support for your hard drive only for interfaces which support DMA, it
will cut the timing buffered disk reads time and will improve the performance by two.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-129-2048.jpg)
![General Optimization 0
CHAPTER 5
130
Step 3
Multiword DMA mode 2, also known as ATA2 disk drive is the successor of the simple DMA drive. If
you have this kind of hard drive, then you must enable the parameter in your Linux system.
• To enable multiword DMA mode 2 transfers, use the following command:
[root@deep /]# /sbin/hdparm -d1 -X34 /dev/hda (or hdb, hdc etc).
This sets the IDE transfer mode for newer (E)IDE/ATA2 drives. (Check your hardware manual
to see if you have it).
Step 4
As for DMA mode 2, the UltraDMA mode 2 is an improvement of the DMA technology. If you have
this kind of drive in your system, then choose this mode.
• To enable UltraDMA mode 2 transfers, use the following command:
[root@deep /]# /sbin/hdparm -d1 -X66 /dev/hda (or hdb, hdc etc)
See your manual page about hdparm for more information. USE THIS OPTION WITH EXTREME
CAUTION!
Step 5
The UltraDMA mode 4 is one of the latest entries and one of the most popular at this time; it is
also known and referred as ATA/66. I guess that most of you have this kind of drive installed and
if it is the case then it is the one that you must choose for sure.
• To enable UltraDMA mode4 transfers, use the following command:
[root@deep /]# /sbin/hdparm -d1 -X12 -X68 /dev/hda (or hdb, hdc etc)
This will enable UltraDMA ATA/66 mode on your drive. See your manual page about hdparm
for more information. USE THIS OPTION WITH EXTREME CAUTION!
Step 6
Multiple sector mode (aka IDE Block Mode), is a feature of most modern IDE hard drives,
permitting the transfer of multiple sectors per I/O interrupt, rather than the usual one sector per
interrupt. When this feature is enabled, it typically reduces operating system overhead for disk I/O
by 30-50%. On many systems it also provides increased data throughput of anywhere from 5% to
50%.
• To set multiple sector mode I/O, use the following command:
[root@deep /]# /sbin/hdparm -mXX /dev/hda (or hdb, hdc etc)
Where “XX” represent the maximum setting supported by your drive. The “-i” flag can be used to
find the maximum setting supported by an installed drive: look for MaxMultSect in the output.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-130-2048.jpg)
![General Optimization 0
CHAPTER 5
131
• To find the maximum setting of your drive, use the following command:
[root@deep /]# /sbin/hdparm -i /dev/hda (or hdb, hdc etc)
/dev/hda:
Model=QUANTUM FIREBALLP LM15, FwRev=A35.0700, SerialNo=883012661990
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
BuffType=3(DualPortCache), BuffSize=1900kB, MaxMultSect=16, MultSect=16
DblWordIO=no, OldPIO=2, DMA=yes, OldDMA=2
CurCHS=16383/16/63, CurSects=-66060037, LBA=yes, LBAsects=29336832
tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
UDMA modes: mode0 mode1 mode2 mode3 *mode4
Step 7
The get/set sector count is used to improve performance in sequential reads of large files! The
default setting is 8 sectors (4KB) and we will double and change it for 16. USE THIS OPTION
WITH EXTREME CAUTION!
• To improve the get/set sector count for file system read-ahead, use the command:
[root@deep /]# /sbin/hdparm -a16 /dev/hda (or hdb, hdc etc)
Step 8
The get/set interrupt-unmask flag will greatly improve Linux's responsiveness and eliminates
"serial port overrun" errors. USE THIS OPTION WITH EXTREME CAUTION!
• To improve and get/set interrupt-unmask flag for the drive, use the command:
[root@deep /]# /sbin/hdparm -u1 /dev/hda (or hdb, hdc etc)
Step 9
The IDE drive's write-caching feature will improve the performance of the hard disk. USE THIS
OPTION WITH EXTREME CAUTION!
• To enable the IDE drive's write-caching feature, use the following command:
[root@deep /]# /sbin/hdparm -W1 /dev/hda (or hdb, hdc etc)
Step 10
These options will allow the drive to retain your settings over a soft reset (as done during the error
recovery sequence). It is important to note that not all drives support this feature.
• To enables the drive to retain your settings, use the command:
[root@deep /]# /sbin/hdparm -K1 -k1 /dev/hda (or hdb, hdc etc)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-131-2048.jpg)
![General Optimization 0
CHAPTER 5
132
Step 11
Once every tuning related to your specific drive have been set, you can test the results and see if
you want to keep them or not.
• You can test the results of your changes by running hdparm in performance test mode:
[root@deep /]# /sbin/hdparm -vtT /dev/hda (or hdb, hdc etc).
/dev/hda:
multcount = 16 (on)
I/O support = 3 (32-bit w/sync)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 1 (on)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 16 (on)
geometry = 1826/255/63, sectors = 29336832, start = 0
Timing buffer-cache reads: 128 MB in 0.85 seconds = 150.59 MB/sec
Timing buffered disk reads: 64 MB in 2.54 seconds = 25.20 MB/sec
Once you have a set of hdparm options, you can put the commands in your /etc/rc.local file
to run it every time you reboot the machine. When running from /etc/rc.local, you can add
the “-q” option for reducing screen clutter. In my case, I will put the following configuration in the
end of my rc.local file:
/sbin/hdparm -q -c3 -d1 -X12 -X68 -m16 -a16 -u1 -W1 -k1 -K1 /dev/had](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-132-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
136
A solution was found and this was the Modularized Kernel approach. A technique that allows
small pieces of compiled code to be inserted in or removed from the running kernel. In this way
the Kernel will only load and run drivers and features that your computer have and will forget
about the others. This practice is what all Linux vendors use to provide Linux kernels. They build
and link every driver and feature as a module (which keeps the binary kernel smaller) that can be
recognized and loaded if, and only if, they are needed by the kernel or the system.
Kernel developers provide the ability to build a Modularized kernel, through an option that asks
you during kernel configuration if you want to build the available drivers/features as a module.
This option appears at the beginning of the Kernel configuration in the following form "Enable
loadable module support (CONFIG_MODULES) [Y/n/?]". If you answer "Yes" here,
then the compiled Kernel will be a Modularized Kernel and all future questions appearing during
kernel configuration will give you the choice to compile the drivers/features into the Kernel code
as a module by answering "m" for module, "y" for yes includes the code, or "n" do not include the
code. Alternatively, if you answer "No" to the question "Enable loadable module support
(CONFIG_MODULES) [Y/n/?]", then the corresponding Kernel will be a Monolithic kernel and
all future questions appearing during kernel configuration will let you answer either "y" (yes,
include the driver/feature) or "n" (no, do not include the drivers/feature). This allows you to build a
Kernel where every driver/feature is compiled into it.
To recap, Modularized Kernels allow small pieces of compiled code, which reside under the
/lib/modules/2.4.x-x/ kernel directory to be inserted into or removed from the running
kernel and a Monolithic Kernel contains the drivers/features into its compiled code.
Some people will say that a loadable module is as good as hard-linked code. But what sort of
speed difference is seen when using loadable modules instead of hard-linked code? Well, here’s
an extract of the kernel mailing list archive:
The immediate response from some was "almost nothing," but further consideration has shown
this not to be true. There are, in fact, a number of costs associated with loadable modules. The
biggest, perhaps, relates to how loadable modules are placed in kernel memory. The code for a
module needs to live in a contiguous address space. The kernel sets up that address space with
a function called vmalloc, which allocates memory with virtual addresses. In other words, a
loadable module is in an address space that is visible to the kernel, but which is separate from
where the core kernel code goes. This difference is important. The core kernel address space is a
direct map of physical memory; it can be handled very efficiently in the processor's page table.
Indeed, on some processors, a single page table entry covers the entire kernel. Space obtained
from vmalloc, instead, uses one page table entry per memory page. A greater number of page
table entries mean more lookups, and more translation buffer misses.
One estimate is that the slowdown can be as much as 5%. Given this problem, why not load
modules into the regular kernel memory space? Module code requires a contiguous address
space. Since the standard kernel space is a direct map of physical memory, contiguous address
spaces must also be contiguous in physical memory. Once the system has been running for a
while, finding even two physically contiguous pages can be a challenge; finding enough to load a
large module can be almost impossible. Modules also seem to have endemic problems with race
conditions - it is possible, for example, for the kernel to attempt to access a newly-loaded module
before it is fully initialized. Modules can also, in some situations, be removed while still in use.
Such occurrences are obviously quite rare, but they can be catastrophic when they happen. The
race conditions can be fixed with enough work, but that may require changing some fundamental
kernel interfaces. In general, dealing with loadable modules is not an easy task; as one kernel
hacker told us in a private message: "Doing live surgery on the kernel is never going to be pretty."](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-136-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
138
Pristine source
If you don’t use the RPM package to install the kernel, it will be difficult for you to locate all the files
installed onto the system if you want to update your kernel in the future. To solve this problem, it’s
a good idea to make a list of files on the system before you install the kernel, and then one
afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the kernel:
[root@deep root]# find /* > Kernel1
• And the following one after you install the kernel:
[root@deep root]# find /* > Kernel2
• Then use this command to get a list of what changed:
[root@deep root]# diff Kernel1 Kernel2 > Kernel-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new kernel. In our example above, we use the /root directory of the system
to store all the generated file lists.
Making an emergency boot floppy
The first step before going into the configuration and compilation of our new kernel is to create an
emergency boot floppy in case something goes wrong during the build of your new Linux Kernel.
Here we create the boot floppy for a modularized kernel since the kernel that is presently installed
on our system should be a modularized kernel. We’ll see later that a method for creating the boot
disk for mololithic kernel exists and is different from what we use here. To create the emergency
boot floppy for a modularized kernel, follow these steps.
Step1
We have to find the present modularized kernel version used on our system. We need this
information to be able to create our emergency boot floppy disk.
• To know which kernel version is running on your system, use the command:
[root@dev /]# uname -a
Linux dev 2.4.18-3 #1 Thu Apr 18 07:37:53 EDT 2002 i686 unknown
From the above command, we know now that our kernel version is 2.4.18-3. Therefore we will
use this information in the next step to create the boot disk.
Step2
Once we know which kernel version we are currently running, we can use the command below to
create the boot disk.
• Put a floppy in your system and execute the following command as root:
[root@deep /]# mkbootdisk --device /dev/fd0H1440 2.4.18-3
Insert a disk in /dev/fd0. Any information on the disk will be lost.
Press <Enter> to continue or ^C to abort:](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-138-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
139
NOTE: In this example, the current kernel on our system is version 2.4.18-3 and this is why we
use “2.4.18-3” here. If you kernel version is different from what we use here, you just have to
change the example version number for the one that you have according to the result returned by
the “uname -a” command.
Following these guidelines, you will now have a boot floppy with a known working kernel in case
of problems with the upgrade. I recommend rebooting the system with the floppy to make sure
that the floppy works correctly before continuing.
Preparing the Kernel for the installation
We have to copy the new kernel tar archive to the appropriate location on our server /usr/src
and then we remove the old kernel from our system before installing a new one. Removing the
old kernel will not freeze your computer until you try to reboot it before installing the new one
because the Linux kernel resides in memory.
Step 1
We must copy the archive file of the kernel to the /usr/src directory and move to this directory.
• To copy the tar archive of the Linux kernel to the /usr/src directory, use the command:
[root@deep /]# cp linux-version.tar.gz /usr/src/
• To move to the /usr/src directory, use the following command:
[root@deep /]# cd /usr/src/
Step 2
Depending on how the Linux Kernel has been previously installed on your system, there are two
ways to uninstall it, these are shown below.
If you already have installed a Linux kernel with a tar archive before
These steps are required ONLY if you have previously installed a Linux kernel from a tar archive.
If it is a fresh, first install of the kernel, then uninstall the kernel-headers-version.
i386.rpm, kernel-version.i386.rpm packages that are on your system.
• Move to the /usr/src directory if you are not already in it with the following command:
[root@deep /]# cd /usr/src/
• Remove the Linux symbolic link with the following command:
[root@deep src]# rm -f linux
• Remove the Linux kernel headers directory with the following command:
[root@deep src]# rm -rf linux-2.4.x/
• Remove the Linux kernel with the following command:
[root@deep src]# rm -f /boot/vmlinuz-2.4.x
• Remove the Linux System.map file with the following command:
[root@deep src]# rm -f /boot/System.map-2.4.x
• Remove the Linux kernel modules directory (if available) with the following command:
[root@deep src]# rm -rf /lib/modules/2.4.x/](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-139-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
140
NOTE: Removing the old kernel modules is only required if you have installed a modularized
kernel version before. If the modules directory doesn’t exist under the /lib/modules directory,
it’s because your old kernel version is not a modularized kernel.
If the original kernel’s RPM packages are installed on your system
If the original kernel RPM packages are installed on your system, instead of the Linux kernel tar
archive, which they would be if you have just finished installing your new Linux system, or have
previously used an RPM package to upgrade your system, then use the following command to
uninstall the Linux kernel:
• Verify which kernel RPM packages are installed on your server with the command:
[root@deep src]# rpm -qa | grep kernel
kernel-2.4.18-3
• Also, verify if Kernel header package is installed on your server with the command:
[root@deep src]# rpm -q glibc-kernheaders
glibc-kernheaders-2.4-7.14
The above command shows us that kernel and glibc-kernheaders are the only kernel RPM
packages installed on our system. We uninstall them as show below.
• To uninstall the linux kernel RPM, use the following command:
[root@deep src]# rpm -e --nodeps kernel glibc-kernheaders
NOTE: If you receive an error message like: cannot remove /lib/modules/2.4.x directory,
directory not empty, then remove the directory manually with command like: rm –rf
/lib/modules/2.4.x/ form your system. This directory is related to the old kernel and it is not
required for the new kernel we want to install.
Step 3
Once we have uninstalled the old kernel and our new kernel tar archive has been copied to the
/usr/src directory, we must uncompress it and remove the tar archive (linux-version.
tar.gz) from the system to conserve disk space.
• To uncompress the kernel, use the following command:
[root@deep src]# tar xzpf linux-version.tar.gz
• To remove the kernel tar archive from the system, use the following command:
[root@deep src]# rm -f linux-version.tar.gz
WARNING: If kernel compilation is something new for you, then it is recommended to keep the
kernel tar archive (linux-version.tar.gz) until the end of the installation. This way, if you
make a mistake during compilation, you have the source available to try again.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-140-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
141
Applying the Grsecurity kernel patch
Grsecurity is a single patch file for the newest stable versions of Linux kernel that is an
attempt to greatly improve the security of a Linux system. It mainly accomplishes this by
physically patching the Linux kernel to make processes more restricted.
Many other projects like Grsecurity exist on the Internet. There are some well know like the
RSBAC (www.rsbac.org), LIDS (www.lids.org), SELinux (www.nsa.gov/selinux/), and OpenWall
(www.openwall.com/linux/), projects, all of which fulfills only one part of a complete security
system.
What makes Grsecurity so different and better than these other projects is mainly because
Grsecurity provides greatly needed additional security to Linux systems. In other words, it
covers all features of the other projects and adds additional security features that the other
projects do NOT cover. The types of added Grsecurity security are categorized as:
1. Complete Buffer Overflow Exploitation Protection.
2. File system Race Protection.
3. System Auditing.
4. Change-root Protections.
5. Portscan and OS Fingerprinting Protection.
6. Restricted Users.
7. Configurability Options.
8. Access Control.
Grsecurity patch may change from version to version, and some may contain various other
security features. It is important to use the Grsecurity patch that corresponds to the Linux
Kernel version that you compile on your server. If you compile kernel version 2.4.18, you have to
download Grsecurity patch for kernel version 2.4.18, etc.
WARNING: When applying the Grsecurity patch to the Kernel, a new “Grsecurity”
configuration section will be added at the end of your Linux Kernel configuration allowing you to
configure and enable the security features that you want.
Obtaining and Installing Grsecurity
To obtain the Grsecurity patch suitable for your Linux Kernel, simply follow the "download" link
on the Grsecurity home page, and download the file listed there. Grsecurity patch is
available directly via http://www.grsecurity.org/. Remember that Grsecurity is specific to one
kernel version, and as of this writing that version is 2.4.18.
• To apply the Grsecurity patch to the Linux kernel, use the commands:
[root@deep /]# cp grsecurity-1.9.4-2.4.18.patch /usr/src/
[root@deep /]# cd /usr/src/linux/
[root@deep linux]# patch -p1 < ../grsecurity-1.9.4-2.4.18.patch
[root@deep linux]# cd ../
[root@deep src]# rm -f grsecurity-1.9.4-2.4.18.patch
The step of patching your new kernel with Grsecurity patch is completed. Now follow the rest
of this Kernel installation guide to build the Linux kernel and reboot your system. Configuration
relating to Grsecurity will appears at the end of your Kernel configuration.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-141-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
143
• Edit the Makefile file (vi +20 /usr/src/linux/Makefile) and change the line:
HOSTCFLAGS = -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer
To read:
HOSTCFLAGS = -Wall -Wstrict-prototypes -O2 -march=i686 -funroll-loops -
fomit-frame- pointer
• Edit the Makefile file (vi +91 /usr/src/linux/Makefile) and change the line:
CFLAGS := $(CPPFLAGS) -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer
-fno-strict-aliasing
To read:
CFLAGS := $(CPPFLAGS) -Wall -Wstrict-prototypes -O2 -march=i686 -funroll-
loops -fomit-frame-pointer -fno-strict-aliasing
WARNING: In the last example, we optimize the code for an i686 CPU architecture, if you have a
different processor, you'll have to adjust the “-march=i686" options for your specific processor.
Never compile the Kernel with optimization code number superior to “-O2”, this do nothing more
and should produce an unstable kernel in some cases. Therefore use “-O2” at the maximum
optimization number but never something superior to it.
Cleaning up the Kernel
It is important to make sure that your /usr/include/asm, and /usr/include/linux
subdirectories are just symlinks to the kernel sources.
Step 1
The asm and linux subdirectories are soft links to the real include kernel source header
directories needed for our Linux architecture, for example /usr/src/linux/include/asm-
i386 for asm.
• To symlink the asm, and linux subdirectories to the kernel sources, type the following
commands on your terminal:
[root@deep src]# cd /usr/include/
[root@deep include]# rm -f asm linux
[root@deep include]# ln -s /usr/src/linux/include/asm-i386 asm
[root@deep include]# ln -s /usr/src/linux/include/linux linux
This is a very important part of the configuration: we remove the asm, and linux directories
under /usr/include then create new links that point to the same name directories under the
new kernel source version directory. The /usr/include directory contains important header
files needed by your kernel and programs to be able to compile on your system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-143-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
144
WARNING: If the previously installed kernel in your system was made by RPM packages, then the
asm and linux soft links will not exist since the uninstall of kernel-headers RPM package
removes them automatically for you. Don’t forget to create them.
Step 2
Make sure you have no stale .o files or dependencies lying around.
• To be sure that we have no stale .o files or dependencies lying around, type the
following commands on your terminal:
[root@deep include]# cd /usr/src/linux/
[root@deep linux]# make mrproper
WARNING: These two steps simply clean up anything that might have accidentally been left in the
source tree by the development team.
You should now have the source correctly installed. You can configure the kernel in one of three
ways. The first method is to use the make config command. It provides you with a text-based
interface for answering all the configuration options. You are prompted for all the options you
need to set up your kernel.
The second method is to use the make menuconfig command, which provides all the kernel
options in an easy-to-use menu. The third is to use the make xconfig command (only available
if the graphical interface of Linux is installed on the system), which provides a full graphical
interface to all the kernel options.
Step 3
For configuration in this guide, we will use the make config command because we have not
installed the XFree86 Window Interface on our Linux server or the necessary packages to use the
make menuconfig command.
• Type the following commands on your terminal to load the kernel configuration:
[root@deep /]# cd /usr/src/linux/ (if you are not already in this directory).
[root@deep linux]# make config
rm -f include/asm
( cd include ; ln -sf asm-i386 asm)
/bin/sh scripts/Configure arch/i386/config.in
#
# Using defaults found in arch/i386/defconfig
#](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-144-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
145
Configuring the Kernel
As soon as you enter make config at the prompt as described in the previous step, a list of
kernel configurable options will be displayed for you to choose to configure the kernel, you must
indicate what features and devices drivers you want to include on your system and select how to
include support for specific devices. Typically, for each configuration option, you have to respond
with one of the following choices:
[y] To compile into the kernel and always be loaded.
[m] To use a module for that feature and load that segment of code on demand.
[n] To skip and excludes the support for that specific device from the kernel.
WARNING: It is important to note that an [n] or [y] means the default choice. If a device does not
have a modular device driver or you have not compiled the Kernel as a Modularized Kernel, you
will not see the [m] option. Some time an [?] option will appear in the choices. This mean that
you can get more information about the feature when you type the ? + ENTER key. Choosing the
[?] help option opens another terminal describing the option.
A new Linux kernel is very specific to our computer hardware since we have to choose the right
drivers as well as features that we need to include and compile into the Kernel code. This implies
a good understanding and knowledge of your computer hardware. It is simply inconceivable to
build a Linux system if you don't know what hardware your computer has, especially if you spend
money to buy a computer and then take time to configure it. Therefore we assume that you know
all of your hardware and be aware that during the kernel configuration, you will be asked to
answer some important questions related to your specific computer hardware.
Be prepared to answer the following questions:
1. What type of processor do you have on your computer (i.e. Pentium III, AMD)?
2. How many processor do you have on your computer (i.e. 1, 2, 3)?
3. What kind of hard drive do you have on your computer (i.e. IDE, SCSI) ?
4. How many hard drives do you have on your computer? (i.e. want to make RAID)?
5. How much memories (RAM) do you have on your computer (i.e. 512 MB RAM)?
6. Do you have a network card? If so, who made it and what model is it?
7. Do you have a SCSI adapter? If so, who made it and what model is it?
8. Do you have a RAID system? If so, who made it and what model is it?
9. What type of mouse do you have (eg, PS/2, Microsoft, Logitech)?
10. If you have a serial mouse, what COM port is it connected to (eg, COM1)?
11. What is the make and model of your video card?
All of the above questions are very important and if you don't know the answers for all of them,
then it is recommended to get the information before going into Linux kernel configuration,
compilation and installation. If you have all of the information, then you can read the rest of this
guide.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-145-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
146
Monolithic kernel configuration
As we know now, they are two possible different configurations for the kernel. The first is called a
monolithic kernel the second is called a modularized kernel. Below we begin by
showing you the configuration of a monolithic kernel which is to compile the required code
and drivers directly into the kernel by answering the different kernel questions only by yes or no.
Don’t forget to only compile code that you need and use.
A new kernel is very specific to your computer hardware, in the monolithic kernel
configuration part below; we have the following hardware for our example. Of course you must
change them to fit whatever components you have in your system.
1 Pentium II 400 MHz (i686) processor
1 SCSI Motherboard
1 SCSI Hard Disk
1 SCSI Controler Adaptec AIC 7xxx
1 CD-ROM ATAPI IDE
1 Floppy Disk
2 Ethernet Cards Intel EtherExpressPro 10/100
1 Mouse PS/2
If you don’t want some of the options listed in the monolithic kernel configuration that I
enable by default, answer n (for no) instead of y (for yes) to the related questions. If you want
some other options that I disable, then answer y instead of n. Finally, the procedure of building a
new kernel is quite long, therefore I recommend you to take your time. Some coffees and
cigarettes will surely be welcome during these steps.
# Using defaults found in arch/i386/defconfig
#
*
* Code maturity level options
* Prompt for development and/or incomplete code/drivers (CONFIG_EXPERIMENTAL)
[N/y/?] Press Enter
This option relates to features or drivers that are currently considered to be in the alpha-test
phase and for a production system, it is highly recommended to answer by N to this question. Just
hit [Enter] to approve the default choice, which is NO (Do not prompt for development and/or
incomplete code/drivers).
*
* Loadable module support
* Enable loadable module support (CONFIG_MODULES) [Y/n/?] n
This option is very important since it asks us if we want to enable loadable module support into
the Kernel. Remember that our goal in this part of the guide is to build a Monolithic Linux Kernel
where all drivers and features are directly integrated and compiled into the Kernel code, therefore
our answer to this question must be No (Do not enable loadable module support). This means
that we have to enter n as the answer to this question to change the default value, which is Y for
Yes.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-146-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
147
*
* Processor type and features
* Processor family (386, 486, 586/K5/5x86/6x86/6x86MX, Pentium-Classic,
Pentium-MMX, Pentium-Pro/Celeron/Pentium-II, Pentium-III/Celeron(Coppermine),
Pentium-4, K6/K6-II/K6-III, Athlon/Duron/K7, Crusoe, Winchip-C6, Winchip-2,
Winchip-2A/Winchip-3, CyrixIII/C3) [Pentium-III/Celeron(Coppermine)] Pentium-4
This option asks you what type of processor you have on your computer and the default choice is
in brackets [Pentium-III/Celeron(Coppermine)]. Therefore, if you processor is not a
Pentium-III/Celerom(Coppermine) model, you have to enter your processor model here. The
available choices are listed in the question. In the above example, I changed the default value
[Pentium-III/Celeron(Coppermine)] and chose a Pentium-4 model.
Toshiba Laptop support (CONFIG_TOSHIBA) [N/y/?] Press Enter
This option asks you if you want to add a driver support for Toshiba Laptop. If you intend to run
this kernel on a Toshiba portable, then you have to answer Yes to this question and change the
default value which is No. In our example, we keep the default value of No by pressing the
[Enter] key.
Dell Inspiron 8000 support (CONFIG_I8K) [N/y/?] Press Enter
As for the previous option, this one asks you if you want to add a driver support for Dell Inspiron
8000 Laptop. If you intend to run this kernel on a Dell Inspiron 8000, then you have to answer
Yes to this question and change the default value which is No. In our example, we keep the
default value of No by pressing the [Enter] key again.
/dev/cpu/microcode - Intel IA32 CPU microcode support (CONFIG_MICROCODE)
[N/y/?] Press Enter
This option allows you to update the microcode on Intel processors in the IA32 family, e.g.
Pentium Pro, Pentium II, Pentium III, Pentium 4, Xeon etc. If you say Y here and change the
default value of N, then you'll need to say Y also to "/dev file system support" in the 'File
systems' section below. In most situations, this option is simply not needed and you can safety
keep the default setting of N by pressing the [Enter] key.
/dev/cpu/*/msr - Model-specific register support (CONFIG_X86_MSR) [N/y/?] y
This option allows you to enable a device that gives privileged processes access to the x86
Model-Specific Registers (MSRs). On multi-processor the MSR accesses are directed to a
specific CPU. It is a good idea to answer Y to this option to enable it, even if you only have one
processor on your system. Therefore answer Y to this question.
/dev/cpu/*/cpuid - CPU information support (CONFIG_X86_CPUID) [N/y/?] y
This option allows you to enable a device that gives processes access to the x86 CPUID
instructions to be executed on a specific processor. As for the previous option, it is a good idea to
change the default value of N to Y.
High Memory Support (off, 4GB, 64GB) [off] Press Enter
This option allows you to be able to use up to 64 Gigabytes of physical memory on x86 systems.
Usually, and it’s true for most of us, we have a system that will never run with more than 1
Gigabyte total physical RAM and if this is you case, then answer "off" here (default choice and
suitable for most users). In other part if you have a system that has between 1 and 4 Gigabytes
physical RAM, then answer "4GB" here. If more than 4 Gigabytes is used then answer "64GB"
here. In our example, we will configure the kernel to use a maximum of 1 Gigabyte total physical
RAM by pressing the [Enter] key to accept the default choice "off".](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-147-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
148
Math emulation (CONFIG_MATH_EMULATION) [N/y/?] Press Enter
This option allows Linux to emulate a math coprocessor (a feature used for floating point
operations). In general, only old 486SX and 386 processors do not have a a math coprocessor
built in. All modern Pentium I, II, III, IV and later, AMD, Cyrix, etc have a math coprocessor built in
and do not require this option to be turned on. If you have a very old 486SX or 386 processor in
your computer, then you will have to change the default value of N here to become Y, but in
general, everyone will keep the default value of N here since they have a math coprocessor built
in their processor chip.
MTRR (Memory Type Range Register) support (CONFIG_MTRR) [N/y/?] Press Enter
This option on Intel family processors (Pentium Pro, Pentium II and later) allows the Memory
Type Range Registers (MTRRs) to be used to control processor access to memory ranges. This
is most useful if you have XFree86 graphical interface installed on your computer or have more
than 1 processor on your system. Changing the default value of N here to become Y on a Linux
system where a graphical interface or multiple processors are present will increase the
performance of the system. If you don't use a graphical interface, or do not have more than one
processor installed on you system, you must keep the default value of N by pressing the
[Enter] key. Finally, it is important to note that this option is valid only if you have Intel family
processors (Pentium Pro, Pentium II and later) installed on your computer. It doesn't work with
AMD or Cyrix processors.
Symmetric multi-processing support (CONFIG_SMP) [Y/n/?] n
This option enables support for systems with more than one CPU. If you have a system with only
one CPU, like most personal computers, say N. If you have a system with more than one CPU,
say Y. In our example, we only have one processor installed on our computer and will change the
default value of Y to become N.
Local APIC support on uniprocessors (CONFIG_X86_UP_APIC) [N/y/?] (NEW) Press
Enter
This option appears only if you have answered N to the previous option and will let you enable
support for local APIC on system with single processor. Local APIC (Advanced Programmable
Interrupt Controller) is an integrated interrupt controller in the CPU that supports CPU-generated
self-interrupts (timer, performance counters), and the NMI watchdog, which detects hard lockups
on the server. On systems with single processor, you may have to answer Y here, but on systems
with multiple processors, you don't have to answer this question since the kernel will
automatically enable it. Usually we keep the default choice of N here since the feature is available
only on modern Pentium processors like the IV family. Pentium I, II, and III don’t support this
option, as well as AMD and Cyrix processors.
*
* General setup
* Networking support (CONFIG_NET) [Y/n/?] Press Enter
This option is very important and must be set to Y all the time, which is the default value. It allows
your Linux system to support and use networking. We use the default value of Y by pressing the
[Enter] key.
PCI support (CONFIG_PCI) [Y/n/?] Press Enter
This option allows us to enable or disable PCI support on Linux. In most cases, we have to say Y
here. PCI's are the white slots on your motherboard where you can add network cards, video
cards, etc. Since most of us use a PC and have this kind of slot available, it is important to say Y
here by pressing the [Enter] key to use the default choice.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-148-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
149
PCI access mode (BIOS, Direct, Any) [Any] Press Enter
On PCI systems, the BIOS can be used to detect the PCI devices and determine their
configuration. However, some old PCI motherboards have BIOS bugs and may crash if this is
done. Also, some embedded PCI-based systems don't have any BIOS at all. Linux can also try to
detect the PCI hardware directly without using the BIOS. With this option, you can specify how
Linux should detect the PCI devices. If you choose "BIOS", the BIOS will be used, if you choose
"Direct", the BIOS won't be used, and if you choose "Any", the kernel will try the direct access
method and falls back to the BIOS if that doesn't work. If unsure, go with the default, which is
"Any" by pressing the [Enter] key.
PCI device name database (CONFIG_PCI_NAMES) [Y/n/?] n
This option lets you enable a feature that allows the Kernel to contain a database of all known
PCI device names via different files under the /proc filesystem and make the information
comprehensible to the user or disable this feature and get device ID numbers instead of names.
Disabling this feature will save you about 80KB of kernel image size and will make the kernel
smaller in size, which is a good thing for increased performance. If you disable this feature, the
kernel will still run as usual, but will show you device ID numbers instead of names. Therefore, we
will change the default value of Y (get PCI device by names) to become N (get PCI device by ID
numbers) and will save 80KB of Kernel image size.
EISA support (CONFIG_EISA) [N/y/?] Press Enter
This option allows you to enable support for the Extended Industry Standard Architecture (EISA)
bus. EISA is now obsolete by the PCI bus and you may enable this option only if you use some
older ISA card on your system. Since most of us don't have these types of card, we can safety
keep the default value of N here by pressing the [Enter] key.
MCA support (CONFIG_MCA) [N/y/?] Press Enter
This option allows us to enable MCA support with the kernel. MCA (MicroChannel Architecture) is
found in some IBM PS/2 machines and laptops. It is a bus system similar to PCI or ISA. It is rare
that we have this kind of bus on our system and we can safety keep the default value of N (do not
support MCA on this kernel) by pressing the [Enter] key again.
Support for hot-pluggable devices (CONFIG_HOTPLUG) [Y/n/?] n
This option is often required only on laptop computer where you have a PCMCIA or PC-cards that
you can plug or unplug at any time. If the kernel that you compile is made to run on a standard
computer (not laptop, portable), then you have to say N here by changing the default option of Y
to become N.
System V IPC (CONFIG_SYSVIPC) [Y/n/?] Press Enter
This option allows us to enable IPC on Linux. IPC (Inter Process Communication) is a suite of
library functions and system calls which let processes (running programs) synchronize and
exchange information. It is generally considered to be a good thing, and some programs won't
run unless you say Y here. Therefore, we keep the default setting of Y by pressing the [Enter]
key.
BSD Process Accounting (CONFIG_BSD_PROCESS_ACCT) [N/y/?] Press Enter
This option allows us to enable a user level programs, which will be able to instruct the kernel (via
a special system call) to write process accounting information to a file. It is not vital or even really
necessary to enable this kind of option to get a working Linux kernel. Therefore, we keep the
default value of N here by pressing the [Enter] key.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-149-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
150
Sysctl support (CONFIG_SYSCTL) [Y/n/?] Press Enter
This option is very important and especially with Linux kernel 2.4.x generation. It is the feature
that allows us to dynamically change certain kernel parameters and variables on the fly through
the /proc filesystem with the use of the /etc/sysctl.conf file. We keep the default value of
Y by pressing the [Enter] key.
Kernel core (/proc/kcore) format (ELF, A.OUT) [ELF] Press Enter
This option allows a file under the /proc filesystem, which is called "kcore" to contain the
Kernel core image in two different formats, which are ELF or A.OUT. For newer Linux systems,
ELF (Executable and Linkable Format) is highly recommended and is the default option that we
choose by pressing the [Enter] key. A.OUT (Assembler.OUTput) was the old format method
used, but is now really obsolete.
Kernel support for a.out binaries (CONFIG_BINFMT_AOUT) [Y/n/?] n
This option is very important and allows us to provide support for A.OUT. A.OUT
(Assembler.OUTput) is a set of formats for libraries and executables used in the earliest versions
of UNIX and, as we said before, it is now really obsolete and was replaced with the ELF format.
To be sure none of our programs will use this older executable format, we will change the default
value of Y to become N.
Kernel support for ELF binaries (CONFIG_BINFMT_ELF) [Y/n/?] Press Enter
Here, it is important to answer Y to this question since this binary format is the one used now on
modern Linux systems. ELF (Executable and Linkable Format) is a format for libraries and
executables used across different architectures and operating systems. Many new executables
are distributed solely in ELF format and you definitely want to say Y to this option by pressing the
[Enter] key.
Kernel support for MISC binaries (CONFIG_BINFMT_MISC) [Y/n/?] Press Enter
This option enables wrapper-driven binary formats into the kernel. This feature is required by
programs that need an interpreter to run, like Java, Python or Emacs-Lisp. Since we're sure to
use one of these types of programs, it is safe to accept the default value of Y by pressing the
[Enter] key.
Power Management support (CONFIG_PM) [Y/n/?] n
This option allows Power Management Support for your computer. With this feature parts of your
computer are shut off or put into a power conserving "sleep" mode if they are not being used. In
general it is good to enable this option on portable computer (Laptop) only. Note that, even if you
say N here, Linux, on the x86 architecture, will issue the hlt instruction if nothing is to be done,
thereby sending the processor to sleep and saving power. Therefore our choice will be N here.
*
* Memory Technology Devices (MTD)
* Memory Technology Device (MTD) support (CONFIG_MTD) [N/y/?] Press Enter
This option enables MTD (Memory Technology Devices) on your computer. MTD are flash, RAM
and similar chips, often used for solid-state file systems on embedded devices. If you have some
of these devices in your system, then answer Y to the question. In most cases, the default choice
of N is recommended.
*
* Parallel port support
* Parallel port support (CONFIG_PARPORT) [N/y/?] Press Enter
If you want to use devices connected to your machine's parallel port like a printer, zip drive, etc,
then you need to say Y here otherwise the default value N is recommended.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-150-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
151
*
* Plug and Play configuration
* Plug and Play support (CONFIG_PNP) [Y/n/?] n
Plug and Play (PnP) is a standard for peripherals which allows those peripherals to be configured
by software. If you answer to this question by Y, then Linux will be able to configure your Plug and
Play devices. Under Linux, we really don't need to enable PNP support and our choice will be N
here.
*
* Block devices
* Normal PC floppy disk support (CONFIG_BLK_DEV_FD) [Y/n/?] Press Enter
This option allows us to use the floppy disk drive(s) in our PC under Linux. Since everyone has
and usually needs a floppy disk in their computer, the answer to this question will be Y. If you run
a Linux server in highly secure environment, you could answer to this question by N since we
never use floppy disk on this type of system.
XT hard disk support (CONFIG_BLK_DEV_XD) [N/y/?] Press Enter
This option allows us to enable the very old 8 bit hard disk controllers used in the IBM XT
computer and since it's pretty unlikely that you have one of these then you must answer to this
question by N. Therefore we simply press [Enter] because the default answer to this question
is N.
Compaq SMART2 support (CONFIG_BLK_CPQ_DA) [N/y/?] Press Enter
This option allows us to enable support for the Compaq Smart Array controllers. If you use these
kinds of boards in your system, then you should say Y here, otherwise the default value of N is
recommended.
Compaq Smart Array 5xxx support (CONFIG_BLK_CPQ_CISS_DA) [N/y/?] Press Enter
This option allows us to enable support for the Compaq Smart Array 5xxx controllers. If you use
these kinds of boards in your system, then you should say Y here, otherwise the default value of
N is recommended.
Mylex DAC960/DAC1100 PCI RAID Controller support (CONFIG_BLK_DEV_DAC960)
[N/y/?] Press Enter
This option enables support for the Mylex DAC960, AcceleRAID, and eXtremeRAID PCI RAID
controllers. If you use these kinds of boards in your system, then you should say Y here,
otherwise the default value of N is recommended.
Loopback device support (CONFIG_BLK_DEV_LOOP) [N/y/?] Press Enter
This option is a little different from other kernel options and if you enable it, you will be able to use
a regular file as a block device that let you have access to some special advanced features and
possibilities under Linux. The default value for this option will be ok for most of us and only
advanced users or user that know why they need these features will enable the "Loopback
device support" option. For SCSI systems with a modularized kernel, it is important to say Y
here since SCSI drivers will use this device.
Network block device support (CONFIG_BLK_DEV_NBD) [N/y/?] Press Enter
This option enables another advanced feature under Linux, the possibility for your system to be a
client for network block devices. The default value for this option would be ok for most of us and
only advanced users or user that know why they need this feature will enable it.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-151-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
152
RAM disk support (CONFIG_BLK_DEV_RAM) [N/y/?] Press Enter
This option will allow you to use a portion of your RAM memory as a block device, so that you can
make file systems on it, read and write to it and do all the other things that you can do with normal
block devices (such as hard drives). It is usually used to load and store a copy of a minimal root
file system off of a floppy into RAM during the initial install of Linux. Again, most normal users
won't need the RAM disk functionality and will answer N to this question. For SCSI systems with a
modularized kernel, it is important to say Y here since SCSI drivers will use this feature.
*
* Multi-device support (RAID and LVM)
* Multiple devices driver support (RAID and LVM) (CONFIG_MD) [N/y/?] Press
Enter
This option is required only for RAID and logical volume management (LVM). If you use them,
then change the default value of N to become Y.
*
* Networking options
* Packet socket (CONFIG_PACKET) [Y/n/?] Press Enter
This option allows you to enable applications, which communicate directly with network devices
without an intermediate network protocol implemented in the kernel like the tcpdump program. It
is a good idea to enable this feature for most of us.
Packet socket: mmapped IO (CONFIG_PACKET_MMAP) [N/y/?] y
This option allows packet protocol driver to use an IO (Input/Output) mechanism that results in
faster communication. Say Y here.
Kernel/User netlink socket (CONFIG_NETLINK) [N/y/?] y
This option allows us to enable two-way communication between the kernel and user processes.
Say Y here.
Routing messages (CONFIG_RTNETLINK) [N/y/?] (NEW) y
If we have said Y to the previous option, we must say Y here too or the previous option will not
work. Therefore our choice is Y.
Netlink device emulation (CONFIG_NETLINK_DEV) [N/y/?] (NEW) y
This option is a backward compatibility option, and we have to choose Y for now.
Network packet filtering (replaces ipchains) (CONFIG_NETFILTER) [N/y/?] y
This option enables support for a packet filter firewall on your system (Netfilter). It is very
important to answer to this question by Y if you want to support firewall and IPTables on your
computer. If you answer N to this question, then the firewalling features will not be available, even
if your have the IPTables software installed on your system.
Network packet filtering debugging (CONFIG_NETFILTER_DEBUG) [N/y/?] (NEW) y
This option turns on support for debugging the netfilter code. It is a good idea to enable it.
Socket Filtering (CONFIG_FILTER) [N/y/?] Press Enter
This option allows us to enable Linux Socket Filter, a feature needed by PPP packet filtering in
general. Therefore you only need to say Y here if you want to use PPP packet filtering on your
system. Since we use a network card to get a connection to the Internet and not PPP (modem
link), we keep the default value of N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-152-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
153
Unix domain sockets (CONFIG_UNIX) [Y/n/?] Press Enter
This option is very important and must always be set to Y. It allows us to include support for Unix
domain sockets; sockets are the standard Unix mechanism for establishing and accessing
network connections. It is vital to say Y to this option.
TCP/IP networking (CONFIG_INET) [Y/n/?] Press Enter
Another very important and vital option under Linux. This option allows us to enable support for
TCP/IP networking on the computer. TCP/IP are the protocols used on the Internet and on most
local Ethernets. It is highly recommended to say Y here even if you are not connected to the
Internet.
IP: multicasting (CONFIG_IP_MULTICAST) [Y/n/?] n
This option allows us to enable a code that addresses several networked computers at once. You
need it if you intend to participate in the MBONE, a high bandwidth network on top of the Internet
which carries audio and video broadcasts. For most people, it's safe to say N here.
IP: advanced router (CONFIG_IP_ADVANCED_ROUTER) [N/y/?] n
This option allows us to configure our Linux system to run mostly as a router. The answer to this
question won't directly affect the kernel: answering N will just cause the configuration to skip all
the questions about advanced routing. In many cases, we can safety keep the default value of N
here. Only users that want to run their Linux system primarily as a router will answer to this
question by Y to be presented a list of advanced routing features to enable or reject. If you want
to configure your Linux server as a Gateway server, you need to answer Y to this question and all
questions related to this option.
IP: kernel level autoconfiguration (CONFIG_IP_PNP) [N/y/?] Press Enter
This option must be set to Y only for diskless machines requiring network access to boot. For
most people, it's safe to say N here.
IP: tunneling (CONFIG_NET_IPIP) [N/y/?] Press Enter
This option will enable Tunneling on our system. Tunneling is a means of encapsulating data of
one protocol type within another protocol and sending it over a channel that understands the
encapsulating protocol. This is an advanced feature and only advanced users who know why they
need it must answer Y to this question.
IP: GRE tunnels over IP (CONFIG_NET_IPGRE) [N/y/?] Press Enter
This option will enable another kind of Tunneling feature on our system. This method is known as
GRE (Generic Routing Encapsulation). This is an advanced feature and only advanced users that
know why they need it must answer Y to this question.
IP: TCP Explicit Congestion Notification support (CONFIG_INET_ECN) [N/y/?]
Press Enter
This option enables Explicit Congestion Notification on your system. ECN allows routers to notify
clients about network congestion, resulting in fewer dropped packets and increased network
performance. This is a very good feature but, unfortunately, there are many broken firewalls on
the Internet, which refuse connections from ECN-enabled machines, and it may be a while before
these firewalls are fixed. Until then, to access a site behind such a firewall you will have to disable
this option by saying N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-153-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
154
IP: TCP syncookie support (disabled per default) (CONFIG_SYN_COOKIES) [N/y/?] y
This option is very important and every one must answer to this question by Y because normal
TCP/IP networking is open to an attack known as "SYN flooding". This denial-of-service attack
prevents legitimate remote users from being able to connect to your computer during an ongoing
attack and requires very little work from the attacker, who can operate from anywhere on the
Internet. SYN cookies provide protection against this type of attack. Therefore don't forget to
answer Y to this question.
*
* IP: Netfilter Configuration
All questions under the "IP: Netfilter Configuration" section of the Kernel configuration
are related to packet filter firewall support and features. We recommend you enable everything.
Below, we show you the answer for each question without any explanation on the features. If you
need to get more information about the features that you don't understand, you can simply type ?
[Enter] at the prompt to get help.
*
Connection tracking (required for masq/NAT) (CONFIG_IP_NF_CONNTRACK) [N/y/?]
(NEW) y
FTP protocol support (CONFIG_IP_NF_FTP) [N/y/?] (NEW) y
IRC protocol support (CONFIG_IP_NF_IRC) [N/y/?] (NEW) y
IP tables support (required for filtering/masq/NAT) (CONFIG_IP_NF_IPTABLES)
[N/y/?] (NEW) y
limit match support (CONFIG_IP_NF_MATCH_LIMIT) [N/y/?] (NEW) y
MAC address match support (CONFIG_IP_NF_MATCH_MAC) [N/y/?] (NEW) y
netfilter MARK match support (CONFIG_IP_NF_MATCH_MARK) [N/y/?] (NEW) y
Multiple port match support (CONFIG_IP_NF_MATCH_MULTIPORT) [N/y/?] (NEW) y
TOS match support (CONFIG_IP_NF_MATCH_TOS) [N/y/?] (NEW) y
LENGTH match support (CONFIG_IP_NF_MATCH_LENGTH) [N/y/?] (NEW) y
TTL match support (CONFIG_IP_NF_MATCH_TTL) [N/y/?] (NEW) y
tcpmss match support (CONFIG_IP_NF_MATCH_TCPMSS) [N/y/?] (NEW) y
Connection state match support (CONFIG_IP_NF_MATCH_STATE) [N/y/?] (NEW) y
Packet filtering (CONFIG_IP_NF_FILTER) [N/y/?] (NEW) y
REJECT target support (CONFIG_IP_NF_TARGET_REJECT) [N/y/?] (NEW) y
Full NAT (CONFIG_IP_NF_NAT) [N/y/?] (NEW) y
Packet mangling (CONFIG_IP_NF_MANGLE) [N/y/?] (NEW) y
TOS target support (CONFIG_IP_NF_TARGET_TOS) [N/y/?] (NEW) y
MARK target support (CONFIG_IP_NF_TARGET_MARK) [N/y/?] (NEW) y
LOG target support (CONFIG_IP_NF_TARGET_LOG) [N/y/?] (NEW) y
TCPMSS target support (CONFIG_IP_NF_TARGET_TCPMSS) [N/y/?] (NEW) y
WARNING: If you want to enable IPTables support into the kernel, the iptables program must
be installed first or you will receive error messages during kernel compilation. This is because
when IPTables support is enabled, the kernel will associate some part of the iptables
program with it configuration. Therefore don’t forget to install IPTables before configuring kernel
with IPTables support. Finally the same warning is true for quota support into the kernel.
*
*
* The IPX protocol (CONFIG_IPX) [N/y/?] Press Enter
This option allows us to enable Novell networking protocol support on Linux. You need it if you
want to access Novell NetWare file or print servers using Linux or if you want to configure your
Linux system to run as a Novell NetWare file or print server. In most cases, this is not required
and we can answer N to this question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-154-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
155
Appletalk protocol support (CONFIG_ATALK) [N/y/?] Press Enter
This option allows us to enable AppleTalk protocol support on Linux. You need it if you want to
access Apple computers using Linux. In most cases, this is not required and we can answer N to
this question.
DECnet Support (CONFIG_DECNET) [N/y/?] Press Enter
This option allows us to enable DECnet networking protocol support on Linux. The DECnet
networking protocol was used in many products made by Digital (now Compaq). In most cases,
this is not required and we can answer N to this question.
802.1d Ethernet Bridging (CONFIG_BRIDGE) [N/y/?] Press Enter
This option will allow your Linux system to act as an Ethernet bridge, which means that the
different Ethernet segments it is connected to will appear as one Ethernet to the participants.
Several such bridges can work together to create even larger networks of Ethernets using the
IEEE 802.1 spanning tree algorithm. In most cases, this is not required and we can answer N to
this question.
*
* QoS and/or fair queueing
* QoS and/or fair queueing (CONFIG_NET_SCHED) [N/y/?] Press Enter
This option allows us to enable QoS (Quality of Service) support on Linux. When the kernel has
several packets to send out over a network device, it has to decide which ones to send first,
which ones to delay, and which ones to drop. This is the job of the packet scheduler, and several
different algorithms for how to do this "fairly" have been proposed. If we answer N to this
question, the standard packet scheduler, which is a FIFO (first come, first served) will be used by
default. The standard packet scheduler is enough for most of us and if you are running a router
system or are an advanced user who wants to experiment in some new way with TCP/IP
networking, then you can say Y to this question and be able to choose from among several
alternative algorithms which can then be attached to different network devices. In most cases, we
say N to this question.
*
* Telephony Support
* Linux telephony support (CONFIG_PHONE) [N/y/?] Press Enter
This option allows us to use a regular phone for voice-over-IP applications. This also means that
you have to have a telephony card attached to your computer and you know what to do. Most
people will simply answer N to this question.
*
* ATA/IDE/MFM/RLL support
* ATA/IDE/MFM/RLL support (CONFIG_IDE) [Y/n/?] Press Enter
This option allows the kernel to manage low cost mass storage units such as ATA/(E)IDE and
ATAPI units. The most common cases are IDE hard drives and ATAPI CD-ROM drives and
since we all have one of these devices attached to our computer we can safety say Y to this
question. The only time that you can answer to this question by N is when you know that your
system is pure SCSI.
*
* IDE, ATA and ATAPI Block devices
* Enhanced IDE/MFM/RLL disk/cdrom/tape/floppy support (CONFIG_BLK_DEV_IDE)
[Y/n/?] Press Enter
This option allows us to enable the new enhanced driver with IDE/MFM/RLL
disk/cdrom/tape/floppy drives. If you have one or more IDE drives, it is required to answer Y to
this question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-155-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
156
Use old disk-only driver on primary interface (CONFIG_BLK_DEV_HD_IDE) [N/y/?]
Press Enter
This option allows us to enable the old hard disk driver to control IDE/MFM/RLL
disk/cdrom/tape/floppy drives. It is highly recommended not to enable it, since we've already used
the new enhanced driver option with IDE/MFM/RLL disk/cdrom/tape/floppy drives above. This
option may be useful for older systems.
Include IDE/ATA-2 DISK support (CONFIG_BLK_DEV_IDEDISK) [Y/n/?] Press Enter
This option allows us to enable another enhanced support for MFM/RLL/IDE hard disks. The only
time that you can answer N to this question is when you know that your system is pure SCSI.
Therefore, we'll enable support for this option by saying Y to the question.
Use multi-mode by default (CONFIG_IDEDISK_MULTI_MODE) [Y/n/?] n
This option allows us to fix possible error messages that can appear on IDE systems. This error
message may look like:
hda: set_multmode: status=0x51 { DriveReady SeekComplete Error }
hda: set_multmode: error=0x04 { DriveStatusError }
If you get this kind of error message on your system, then you have to say Y to this option. We
suppose that you have a good IDE disk drive and that this error will never appear for you, in this
case, we will change the default value of Y to become N.
Include IDE/ATAPI CDROM support (CONFIG_BLK_DEV_IDECD) [Y/n/?] n
This option allows us to instruct Linux to identify more than one CD-ROM drive along with other
IDE devices, as "hdb" or "hdc", or something similar at boot time. If you have only one CD-ROM
drive installed on your system, you can say N to this question even if it uses the ATAPI protocol
and save some KB into the kernel. You have to say Y to this option only if you handle more than
one CD-ROM drive in your computer. Since most people will usually have only one CD-ROM
installed into their computer, we will change the default value of Y to become N. If you have one
CD-ROM and another one CD-ROM for burning disk, then you have answer Y to this question.
Include IDE/ATAPI TAPE support (CONFIG_BLK_DEV_IDETAPE) [N/y/?] Press Enter
This option allows us to enable an IDE tape drive using the ATAPI protocol. If you have an IDE
tape drive installed on your system, you must answer Y to this question. Most people will say N
here.
Include IDE/ATAPI FLOPPY support (CONFIG_BLK_DEV_IDEFLOPPY) [N/y/?] Press Enter
This option allows us to enable an IDE floppy drive which uses the ATAPI protocol. If you have
this kind of floppy drive installed on your system, you must answer this question Y. Most people
will say N here.
SCSI emulation support (CONFIG_BLK_DEV_IDESCSI) [N/y/?] Press Enter
This option provides SCSI host adapter emulation for IDE ATAPI devices, and will allow us to
use a SCSI device driver instead of a native ATAPI driver. If you intend to install and use a CD-
RW drive on your computer, then you have to say Y here. Again, most people will say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-156-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
157
*
* IDE chipset support/bugfixes
* CMD640 chipset bugfix/support (CONFIG_BLK_DEV_CMD640) [Y/n/?] n
This option allows us to include code which tries to automatically detect and correct the CMD640
problems under Linux. The CMD-Technologies CMD640 IDE chip is used on many common 486
and Pentium motherboards, usually in combination with a "Neptune" or "SiS" chipset.
Unfortunately, it has a number of rather nasty design flaws that can cause severe data corruption
under many common conditions. To know if you need to enable this option for your system to
correct this bug, edit the /proc/cpuinfo file and see if the parameter "f00f_bug" is set to no
or yes. If the "f00f_bug" value is set to no, then you don't need to enable this option and can
say N to the question, otherwise you have to say Y here.
RZ1000 chipset bugfix/support (CONFIG_BLK_DEV_RZ1000) [Y/n/?] n
This option allows us to include code which tries to automatically detect and correct the RZ1000
problems under Linux. The PC-Technologies RZ1000 IDE chip is used on many common 486
and Pentium motherboards, usually along with the "Neptune" chipset. As for the CMD640 bug
above, it also has a number of rather nasty design flaws that can cause severe data corruption
under many common conditions. To know if you need to enable this option for your system to
correct this bug, edit the /proc/cpuinfo file and see if the parameter "coma_bug" is set to no
or yes. If the "coma_bug" value is set to no, then you don't need to enable this option and can
say N to the question, otherwise you have to say Y here.
Generic PCI IDE chipset support (CONFIG_BLK_DEV_IDEPCI) [Y/n/?] Press Enter
This option helps the IDE driver to automatically detect and configure all PCI-based IDE
interfaces in your system. If you have PCI systems which use IDE drive(s), then say Y to this
question. Most of us have PCI systems which use IDE and have answer Y this question.
Sharing PCI IDE interrupts support (CONFIG_IDEPCI_SHARE_IRQ) [Y/n/?] Press
Enter
This option allows us to enable support for sharing a single IRQ with other cards under ATA/IDE
chipsets. In general, everyone has answer Y this question.
Generic PCI bus-master DMA support (CONFIG_BLK_DEV_IDEDMA_PCI) [Y/n/?] Press
Enter
This option allows us to reduce CPU overhead with IDE drive(s) on PCI system capable of bus-
master DMA operation. If you have a modern IDE ATA/33/66/100 hard drive, then it is
recommended to answer this question Y.
Boot off-board chipsets first support (CONFIG_BLK_DEV_OFFBOARD) [N/y/?] Press
Enter
This option allows us to reverse the device scan order to improve the usability of some boot
managers such as lilo when booting from a drive on an off-board controller. In many cases, this
option is really not required and you can accept the default value of N here.
Use PCI DMA by default when available (CONFIG_IDEDMA_PCI_AUTO) [Y/n/?] Press
Enter
A very important option to enable on all modern IDE disk drives. This option allows us to use the
DMA feature of our disk drive under Linux to improve performance. Most people running a capable
DMA drive will answer to this question by Y.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-157-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
158
All of the following Kernel options are related to the special onboard chipsets that you may have
on your motherboard. Therefore, specific drivers are provided for each of them and you have to
choose from the list the one that matches you chipset. If you have an Intel onboard chipset, then
you can safety choose the default answer of N to all of the questions, since the kernel supports it
naturally.
Other chipset models must be selected from the list. In many cases, if your chipset is not listed,
this means that it is automatically supported by the Kernel. Note that two options have their
default answer set to Y (Intel PIIXn chipsets support (CONFIG_BLK_DEV_PIIX)
[Y/n/?] and PIIXn Tuning support (CONFIG_PIIX_TUNING) [Y/n/?]). If you have a
Pentium II or later processor, you must keep the default value of these two option to Y.
AEC62XX chipset support (CONFIG_BLK_DEV_AEC62XX) [N/y/?] Press Enter
ALI M15x3 chipset support (CONFIG_BLK_DEV_ALI15X3) [N/y/?] Press Enter
CMD64X chipset support (CONFIG_BLK_DEV_CMD64X) [N/y/?] Press Enter
CY82C693 chipset support (CONFIG_BLK_DEV_CY82C693) [N/y/?] Press Enter
Cyrix CS5530 MediaGX chipset support (CONFIG_BLK_DEV_CS5530) [N/y/?] Press
Enter
HPT34X chipset support (CONFIG_BLK_DEV_HPT34X) [N/y/?] Press Enter
HPT366 chipset support (CONFIG_BLK_DEV_HPT366) [N/y/?] Press Enter
Intel PIIXn chipsets support (CONFIG_BLK_DEV_PIIX) [Y/n/?] Press Enter
PIIXn Tuning support (CONFIG_PIIX_TUNING) [Y/n/?] Press Enter
NS87415 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_NS87415) [N/y/?] Press
Enter
PROMISE PDC202{46|62|65|67|68} support (CONFIG_BLK_DEV_PDC202XX) [N/y/?] Press
Enter
ServerWorks OSB4/CSB5 chipsets support (CONFIG_BLK_DEV_SVWKS) [N/y/?] Press
Enter
SiS5513 chipset support (CONFIG_BLK_DEV_SIS5513) [N/y/?] Press Enter
SLC90E66 chipset support (CONFIG_BLK_DEV_SLC90E66) [N/y/?] Press Enter
Tekram TRM290 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_TRM290) [N/y/?]
Press Enter
VIA82CXXX chipset support (CONFIG_BLK_DEV_VIA82CXXX) [N/y/?] Press Enter
Other IDE chipset support (CONFIG_IDE_CHIPSETS) [N/y/?] Press Enter
IGNORE word93 Validation BITS (CONFIG_IDEDMA_IVB) [N/y/?] Press Enter
*
* SCSI support
* SCSI support (CONFIG_SCSI) [Y/n/?] Press Enter
This option allows us to enable SCSI hard disks, SCSI tape drives, SCSI CD-ROM's or any other
SCSI devices under Linux. If you have a SCSI like system, you need to answer Y to this question.
If you don't have any SCSI devices on your system, you can safety answer N to the question. For
users that have a SCSI system, it is very important for you to know the name of your SCSI host
adapter (the card inside your computer that "speaks" the SCSI protocol, also called SCSI
controller), because you will be asked for it if you enable this option. Once again, if you don't have
a SCSI system, simply answer N to this question and skip this section of the Linux Kernel
configuration.
*
* SCSI support type (disk, tape, CD-ROM)
* SCSI disk support (CONFIG_BLK_DEV_SD) [Y/n/?] Press Enter
This option allows us to enable support for a SCSI hard disk under Linux. If you have enabled the
SCSI support feature above because you have a SCSI hard drive on your system, then it's here
that you have to specify it by answering Y to the question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-158-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
159
Maximum number of SCSI disks that can be loaded as modules
(CONFIG_SD_EXTRA_DEVS) [40] Press Enter
This option allows us to control the amount of additional space allocated in tables for drivers that
are loaded as modules after the kernel is booted. In the event that the SCSI core itself was
loaded as a module, this value is the number of additional disks that can be loaded after the first
host driver is loaded. Since we're compiling a Monolithic Kernel where no modules are available,
this option doesn't concern us and we can safety press the [Enter] key to accept the default
value.
SCSI tape support (CONFIG_CHR_DEV_ST) [N/y/?] Press Enter
This option allows us to enable support for a SCSI tape drive under Linux. If you have enabled
the SCSI support feature above because you have a SCSI tape drive on your system, then it's
here that you have to specify it by answering Y to the question. For most SCSI users, the answer
is N (no, we don't have a SCSI tape drive on this computer).
SCSI OnStream SC-x0 tape support (CONFIG_CHR_DEV_OSST) [N/y/?] Press Enter
This option allows us to enable support for the OnStream SC-x0 SCSI tape drives under Linux. If
you have this kind of SCSI tape drive installed on your computer, then you have to answer to this
question by Y. Most SCSI users will simply say N to this question.
SCSI CD-ROM support (CONFIG_BLK_DEV_SR) [N/y/?] Press Enter
This option allows us to enable support for a SCSI CD-ROM under Linux. If you have enabled the
SCSI support feature above because you have a SCSI CD-ROM on your system, then it's here
that you have to specify it by answering Y to the question. For most SCSI users, the answer is N
(no, we don't have a SCSI CD-ROM on this computer).
SCSI generic support (CONFIG_CHR_DEV_SG) [N/y/?] Press Enter
This option allows us to enable support to use SCSI scanners, synthesizers or CD-writers or just
about anything having "SCSI" in its name other than hard disks, CD-ROMs or tapes. If you have
one of these SCSI items installed on your computer, then you have to say Y here as well as for
the "SCSI disk support" option above to enable the driver.
VERY IMPORTANT NOTE: For users having IDE CD-writers, you have to say Y to this question too,
even if your CD-writers are not SCSI CD-writers. Most SCSI users will simply say N to his
question.
*
* Some SCSI devices (e.g. CD jukebox) support multiple LUNs
* Enable extra checks in new queueing code (CONFIG_SCSI_DEBUG_QUEUES) [Y/n/?]
Press Enter
This option turns on a lot of additional consistency checking for the new queuing code on SCSI
devices. It is a good idea to enable it by saying Y to the question.
Probe all LUNs on each SCSI device (CONFIG_SCSI_MULTI_LUN) [Y/n/?] n
This option force the SCSI driver to probe for multiple LUN's (Logical Unit Number) on your
system and will certainly affect the performance of the system. Most SCSI users will simply
disable this option by saying N to this question to improve performance. If you enable this option,
then we assume that you know what you're doing.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-159-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
160
Verbose SCSI error reporting (kernel size +=12K) (CONFIG_SCSI_CONSTANTS)
[Y/n/?] n
This option allows any error messages regarding your SCSI hardware to be more
understandable, this enlarges your kernel by about 12 KB. If performance is important to you, we
highly recommend you to disable this option by answering the question N.
SCSI logging facility (CONFIG_SCSI_LOGGING) [N/y/?] Press Enter
This option allows us to turns on a logging facility that can be used to debug a number of SCSI
related problems. Again, if performance is important to you, we highly recommend you to disable
this option by keeping the default value of N here.
*
* SCSI low-level drivers
Below you will be presented a list of available SCSI controllers to choose from, simply select the
SCSI controller that is installed on your system and disable all the others. As an example, we will
pretend that we have an Adaptec AIC7080 controller and will enable it further down. We chose
an Adaptec AIC7080 model for our example; don't forget to change our choice if you have
another kind of SCSI controller installed on your system.
*
3ware Hardware ATA-RAID support (CONFIG_BLK_DEV_3W_XXXX_RAID) [N/y/?] Press
Enter
7000FASST SCSI support (CONFIG_SCSI_7000FASST) [N/y/?] Press Enter
ACARD SCSI support (CONFIG_SCSI_ACARD) [N/y/?] Press Enter
Adaptec AHA152X/2825 support (CONFIG_SCSI_AHA152X) [N/y/?] Press Enter
Adaptec AHA1542 support (CONFIG_SCSI_AHA1542) [N/y/?] Press Enter
Adaptec AHA1740 support (CONFIG_SCSI_AHA1740) [N/y/?] Press Enter
Adaptec AIC7xxx support (CONFIG_SCSI_AIC7XXX) [N/y/?] y
Maximum number of TCQ commands per device (CONFIG_AIC7XXX_CMDS_PER_DEVICE)
[253] (NEW) Press Enter
Initial bus reset delay in milli-seconds (CONFIG_AIC7XXX_RESET_DELAY_MS)
[15000] (NEW) Press Enter
Build Adapter Firmware with Kernel Build (CONFIG_AIC7XXX_BUILD_FIRMWARE)
[N/y/?] (NEW) Press Enter
Adaptec I2O RAID support (CONFIG_SCSI_DPT_I2O) [N/y/?] Press Enter
AdvanSys SCSI support (CONFIG_SCSI_ADVANSYS) [N/y/?] Press Enter
Always IN2000 SCSI support (CONFIG_SCSI_IN2000) [N/y/?] Press Enter
AM53/79C974 PCI SCSI support (CONFIG_SCSI_AM53C974) [N/y/?] Press Enter
AMI MegaRAID support (CONFIG_SCSI_MEGARAID) [N/y/?] Press Enter
BusLogic SCSI support (CONFIG_SCSI_BUSLOGIC) [N/y/?] Press Enter
Compaq Fibre Channel 64-bit/66Mhz HBA support (CONFIG_SCSI_CPQFCTS) [N/y/?]
Press Enter
DMX3191D SCSI support (CONFIG_SCSI_DMX3191D) [N/y/?] Press Enter
DTC3180/3280 SCSI support (CONFIG_SCSI_DTC3280) [N/y/?] Press Enter
EATA ISA/EISA/PCI (DPT and generic EATA/DMA-compliant boards) support
(CONFIG_SCSI_EATA) [N/y/?] Press Enter
EATA-DMA [Obsolete] (DPT, NEC, AT&T, SNI, AST, Olivetti, Alphatronix) support
(CONFIG_SCSI_EATA_DMA) [N/y/?] Press Enter
EATA-PIO (old DPT PM2001, PM2012A) support (CONFIG_SCSI_EATA_PIO) [N/y/?] Press
Enter
Future Domain 16xx SCSI/AHA-2920A support (CONFIG_SCSI_FUTURE_DOMAIN) [N/y/?]
Press Enter
Intel/ICP (former GDT SCSI Disk Array) RAID Controller support
(CONFIG_SCSI_GDTH) [N/y/?] Press Enter
Generic NCR5380/53c400 SCSI support (CONFIG_SCSI_GENERIC_NCR5380) [N/y/?] Press
Enter
IBM ServeRAID support (CONFIG_SCSI_IPS) [N/y/?] Press Enter
Initio 9100U(W) support (CONFIG_SCSI_INITIO) [N/y/?] Press Enter
Initio INI-A100U2W support (CONFIG_SCSI_INIA100) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-160-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
161
NCR53c406a SCSI support (CONFIG_SCSI_NCR53C406A) [N/y/?] Press Enter
NCR53c7,8xx SCSI support (CONFIG_SCSI_NCR53C7xx) [N/y/?] Press Enter
SYM53C8XX Version 2 SCSI support (CONFIG_SCSI_SYM53C8XX_2) [N/y/?] Press Enter
NCR53C8XX SCSI support (CONFIG_SCSI_NCR53C8XX) [N/y/?] Press Enter
SYM53C8XX SCSI support (CONFIG_SCSI_SYM53C8XX) [Y/n/?] n
PAS16 SCSI support (CONFIG_SCSI_PAS16) [N/y/?] Press Enter
PCI2000 support (CONFIG_SCSI_PCI2000) [N/y/?] Press Enter
PCI2220i support (CONFIG_SCSI_PCI2220I) [N/y/?] Press Enter
PSI240i support (CONFIG_SCSI_PSI240I) [N/y/?] Press Enter
Qlogic FAS SCSI support (CONFIG_SCSI_QLOGIC_FAS) [N/y/?] Press Enter
Qlogic ISP SCSI support (CONFIG_SCSI_QLOGIC_ISP) [N/y/?] Press Enter
Qlogic ISP FC SCSI support (CONFIG_SCSI_QLOGIC_FC) [N/y/?] Press Enter
Qlogic QLA 1280 SCSI support (CONFIG_SCSI_QLOGIC_1280) [N/y/?] Press Enter
Seagate ST-02 and Future Domain TMC-8xx SCSI support (CONFIG_SCSI_SEAGATE)
[N/y/?] Press Enter
Simple 53c710 SCSI support (Compaq, NCR machines) (CONFIG_SCSI_SIM710) [N/y/?]
Press Enter
Symbios 53c416 SCSI support (CONFIG_SCSI_SYM53C416) [N/y/?] Press Enter
Tekram DC390(T) and Am53/79C974 SCSI support (CONFIG_SCSI_DC390T) [N/y/?] Press
Enter
Trantor T128/T128F/T228 SCSI support (CONFIG_SCSI_T128) [N/y/?] Press Enter
UltraStor 14F/34F support (CONFIG_SCSI_U14_34F) [N/y/?] Press Enter
UltraStor SCSI support (CONFIG_SCSI_ULTRASTOR) [N/y/?] Press Enter
*
* Fusion MPT device support
* Fusion MPT (base + ScsiHost) drivers (CONFIG_FUSION) [N/y/?] Press Enter
*
* I2O device support
* I2O support (CONFIG_I2O) [N/y/?] Press Enter
This option allows us to enable support for Intelligent Input/Output (I2O) architecture. In order for
this to work, you need to have an I2O interface adapter card in your computer. If you have this
kind of I2O interface adapter card installed on your system, then you can say Y to the question
and you will get a choice of interface adapter drivers and OSM's. Most users simply say N here.
*
* Network device support
* Network device support (CONFIG_NETDEVICES) [Y/n/?] Press Enter
This option is one of the most important and allows us to enable support and feature for network
cards under Linux. Therefore, we have to answer Y to this question.
*
* ARCnet devices
* ARCnet support (CONFIG_ARCNET) [N/y/?] Press Enter
This option allows us to enable ARCnet chipset support under Linux. If you have a network card
of this type installed on your system, then say Y here, otherwise and for most users, you have to
keep the default value of N here.
Dummy net driver support (CONFIG_DUMMY) [Y/n/?] n
This option is only useful for PPP dial up modem users on Linux. If you don't use your system to
make PPP connections, then you can safety answer N to this question. Only users who have a
modem and want to use it to establish a connection via PPP or SLIP need to say Y here. Since,
in our example, we use a network card to make a network connection, we will answer the
question N.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-161-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
162
Bonding driver support (CONFIG_BONDING) [N/y/?] Press Enter
This option allows us to 'bond' multiple Ethernet Channels together. This is called 'Etherchannel'
by Cisco, 'Trunking' by Sun, and 'Bonding' in Linux. It is a technique to merge Ethernet segments
together for doubling the speed of a connection. In most cases, we can safety choose the default
choice of N here. You must have two Ethernet cards installed on your system and on the remote
computer where you want to use this technique to be able to use it. Also, this is an advanced
feature and only experienced Linux users may need it. Therefore, we will answer to the question
by N.
EQL (serial line load balancing) support (CONFIG_EQUALIZER) [N/y/?] Press Enter
This option allows us to enable the same feature as the previous option, but this time for two
modems and two telephone lines. Therefore, we will simply say N to this question.
Universal TUN/TAP device driver support (CONFIG_TUN) [N/y/?] Press Enter
This option allows us to enable TUN/TAP support under Linux. TUN/TAP provides packet
reception and transmission for user space programs. It can be viewed as a simple Point-to-Point
or Ethernet device, which instead of receiving packets from a physical media, receives them from
user space program and instead of sending packets via physical media, writes them to the user
space program. It is rare that we need this feature, if you need it then simply say Y here. Most
users will say N here.
*
* Ethernet (10 or 100Mbit)
* Ethernet (10 or 100Mbit) (CONFIG_NET_ETHERNET) [Y/n/?] Press Enter
This option is very important and allows us to enable support for Ethernet Network Interface
Cards (NIC's) under Linux. Now, everyone has a NIC in their computer and if you want to be able
to use your network card, then you have to say Y here. Note that the answer to this question
won't directly affect the kernel: saying N will just cause the configuration to skip all the questions
about Ethernet network cards. We must say Y here to be able to select the network card that we
have in our computer from the list of supported network cards.
It is very important to know the name of the network card(s) installed in your system because you
will be asked for it. As an example we will pretend that we have an "EtherExpressPro/100"
network card in our computer and we will enable support for it. This is an example and don't
forget to change our default choice if you have another kind of network card installed in your
system. In general, we say Y for the network card that we have and N for all other network cards.
Sun Happy Meal 10/100baseT support (CONFIG_HAPPYMEAL) [N/y/?] Press Enter
Sun GEM support (CONFIG_SUNGEM) [N/y/?] Press Enter
3COM cards (CONFIG_NET_VENDOR_3COM) [N/y/?] Press Enter
AMD LANCE and PCnet (AT1500 and NE2100) support (CONFIG_LANCE) [N/y/?] Press
Enter
Western Digital/SMC cards (CONFIG_NET_VENDOR_SMC) [N/y/?] Press Enter
Racal-Interlan (Micom) NI cards (CONFIG_NET_VENDOR_RACAL) [N/y/?] Press Enter
DEPCA, DE10x, DE200, DE201, DE202, DE422 support (CONFIG_DEPCA) [N/y/?] Press
Enter
HP 10/100VG PCLAN (ISA, EISA, PCI) support (CONFIG_HP100) [N/y/?] Press Enter
Other ISA cards (CONFIG_NET_ISA) [N/y/?] Press Enter
EISA, VLB, PCI and on board controllers (CONFIG_NET_PCI) [Y/n/?] Press Enter
AMD PCnet32 PCI support (CONFIG_PCNET32) [N/y/?] Press Enter
Apricot Xen-II on board Ethernet (CONFIG_APRICOT) [N/y/?] Press Enter
CS89x0 support (CONFIG_CS89x0) [N/y/?] Press Enter
DECchip Tulip (dc21x4x) PCI support (CONFIG_TULIP) [N/y/?] Press Enter
Generic DECchip & DIGITAL EtherWORKS PCI/EISA (CONFIG_DE4X5) [N/y/?] Press
Enter
Digi Intl. RightSwitch SE-X support (CONFIG_DGRS) [N/y/?] Press Enter
Davicom DM910x/DM980x support (CONFIG_DM9102) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-162-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
163
EtherExpressPro/100 support (CONFIG_EEPRO100) [Y/n/?] Press Enter
Myson MTD-8xx PCI Ethernet support (CONFIG_FEALNX) [N/y/?] Press Enter
National Semiconductor DP8381x series PCI Ethernet support (CONFIG_NATSEMI)
[N/y/?] Press Enter
PCI NE2000 and clones support (see help) (CONFIG_NE2K_PCI) [N/y/?] Press Enter
RealTek RTL-8139 PCI Fast Ethernet Adapter support (CONFIG_8139TOO) [N/y/?]
Press Enter
SiS 900/7016 PCI Fast Ethernet Adapter support (CONFIG_SIS900) [N/y/?] Press
Enter
SMC EtherPower II (CONFIG_EPIC100) [N/y/?] Press Enter
Sundance Alta support (CONFIG_SUNDANCE) [N/y/?] Press Enter
TI ThunderLAN support (CONFIG_TLAN) [N/y/?] Press Enter
VIA Rhine support (CONFIG_VIA_RHINE) [N/y/?] Press Enter
Winbond W89c840 Ethernet support (CONFIG_WINBOND_840) [N/y/?] Press Enter
Pocket and portable adapters (CONFIG_NET_POCKET) [N/y/?] Press Enter
*
* Ethernet (1000 Mbit)
* Alteon AceNIC/3Com 3C985/NetGear GA620 Gigabit support (CONFIG_ACENIC)
[N/y/?] Press Enter
D-Link DL2000-based Gigabit Ethernet support (CONFIG_DL2K) [N/y/?] Press Enter
National Semiconduct DP83820 support (CONFIG_NS83820) [N/y/?] Press Enter
Packet Engines Hamachi GNIC-II support (CONFIG_HAMACHI) [N/y/?] Press Enter
SysKonnect SK-98xx support (CONFIG_SK98LIN) [N/y/?] Press Enter
FDDI driver support (CONFIG_FDDI) [N/y/?] Press Enter
This option allows us to enable FDDI card support under Linux. FDDI (Fiber Distributed Data
Interface) is a high speed local area network design to runs over copper or fiber. If you are
connected to such a network and want a driver for the FDDI card in your computer, say Y here.
Most users will simply say N here.
PPP (point-to-point protocol) support (CONFIG_PPP) [N/y/?] Press Enter
This option allows us to enable PPP support under Linux. PPP (Point to Point Protocol) is the
protocol used by modern modems to establish a remote connection with your ISP. If you have a
modem card installed on your system to make a remote connection with your ISP, then you need
to answer Y to this question. If you don't use PPP to connect on the Internet, then you can safety
say N here. In our example, we assume that you use another method, like a network interface, to
connect to the Internet and say N here.
SLIP (serial line) support (CONFIG_SLIP) [N/y/?] Press Enter
This option allows us to enable SLIP or CSLIP support under Linux. These protocols are really
old now and have been replaced by the PPP protocol (see the above option). If for any reason
you still use them, then say Y here. Most users will answer this question N.
*
* Wireless LAN (non-hamradio)
* Wireless LAN (non-hamradio) (CONFIG_NET_RADIO) [N/y/?] Press Enter
This option allows us to enable support for wireless LANs and everything having to do with radio,
but not with amateur radio or FM broadcasting. If you need such support on your system, then
say Y here, also if you need Wireless Extensions with wireless PCMCIA (PC-) cards on Linux, you
need to say Y here too. Most users will simply say N here.
*
* Token Ring devices
* Token Ring driver support (CONFIG_TR) [N/y/?] Press Enter
This option allows us to enable support for Token Ring under Linux. Token Ring is IBM's way of
communication on a local network; the rest of the world uses Ethernet. If you need Token Ring
support on your computer, then say Y here. Most people will select the default choice of N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-163-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
164
Fibre Channel driver support (CONFIG_NET_FC) [N/y/?] Press Enter
This option allows us to enable Fibre Channel support under Linux. Fibre Channel is a high speed
serial protocol mainly used to connect large storage devices to the computer. If you have a Fibre
channel adapter card in your computer, then you can say Y here. Most users will simply say N
here.
*
* Wan interfaces
* Wan interfaces support (CONFIG_WAN) [N/y/?] Press Enter
This option allows us to enable support for Wan interfaces under Linux. Wide Area Networks
(WANs), such as X.25, frame relay and leased lines, are used to interconnect Local Area
Networks (LANs) over vast distances. If you have these kinds of cards installed on your system,
then you can answer Y to the question. Most users will say N here.
*
* Amateur Radio support
* Amateur Radio support (CONFIG_HAMRADIO) [N/y/?] Press Enter
This option allows us to enable Amateur Radio support under Linux. If you want to connect your
Linux box to an amateur radio, answer Y here. Note that the answer to this question won't directly
affect the kernel: saying N will just cause the configuration to skip all the questions about amateur
radio. Most people will say N here.
*
* IrDA (infrared) support
* IrDA subsystem support (CONFIG_IRDA) [N/y/?] Press Enter
This option allows us to enable support for the IrDA (TM) protocols under Linux. IrDA (Infrared
Data Associations) is a support for wireless infrared communication. Laptops or computers that
use infrared and PDA's users will say Y here. Most users will say N here.
*
* ISDN subsystem
* ISDN support (CONFIG_ISDN) [N/y/?] Press Enter
This option allows us to enable ISDN support under Linux. ISDN (Integrated Services Digital
Networks) is a special type of fully digital telephone service; it's mostly used to connect to your
Internet service provider (with SLIP or PPP). If you have this type of card installed on your
computer (popular in Europe), then you have to say Y here otherwise, you will certainly keep the
default value of N here.
*
* Old CD-ROM drivers (not SCSI, not IDE)
* Support non-SCSI/IDE/ATAPI CDROM drives (CONFIG_CD_NO_IDESCSI) [N/y/?] Press
Enter
If you have a CD-ROM drive that is neither SCSI nor IDE/ATAPI, say Y here, otherwise N. Note
that the answer to this question doesn't directly affect the kernel: saying N will just cause the
configuration to skip all the questions about these CD-ROM drives. Most users will say N here.
*
* Input core support
* Input core support (CONFIG_INPUT) [N/y/?] Press Enter
This option allows us to enable any of the USB HID (Human Interface Device) options in the USB
support section which require Input core support. If you intended to use USB on Linux, say Y here
otherwise, you can safety say N here. It is rare that we have to use USB on server systems.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-164-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
165
*
* Character devices
* Virtual terminal (CONFIG_VT) [Y/n/?] Press Enter
This option is very important and allows us to enable support for terminal devices with display and
keyboard devices. On Linux, you need at least one virtual terminal device in order to make use of
your keyboard and monitor. Therefore, we have to say Y here.
Support for console on virtual terminal (CONFIG_VT_CONSOLE) [Y/n/?] Press Enter
This option allows us to enable support for consoles on virtual terminals. Most users will simply
keep the default value of Y here.
Standard/generic (8250/16550 and compatible UARTs) serial support
(CONFIG_SERIAL) [Y/n/?] n
This option allows us to use serial mice, modems and similar devices that connect to the standard
serial ports under Linux. If you use your Linux system as a workstation with graphical interface
installed, then you must answer Y to the question. If you use your Linux system for dedicated
Ethernet WWW/FTP servers, then you can say N here. In our example, we assume that your
Linux system is dedicated and configured to run as a server only and answer N to the question.
Non-standard serial port support (CONFIG_SERIAL_NONSTANDARD) [N/y/?] Press
Enter
This option allows us to enable support for any non-standard serial boards under Linux. These
are usually used for systems that need many serial ports, because they serve many terminals or
dial-in connections. If this is not true in your case, then you can safety say N here. Most people
will simply say N here.
Unix98 PTY support (CONFIG_UNIX98_PTYS) [Y/n/?] Press Enter
This option is important in modern Linux system and allows us to enable pseudo terminal (PTY)
support under Linux. Everyone must say Y here. Information about pseudo terminal (PTY) can be
found in the Kernel documentation.
Maximum number of Unix98 PTYs in use (0-2048) (CONFIG_UNIX98_PTY_COUNT) [256]
128
This option allows us to set the maximum number of Unix98 PTY's that can be used at any one
time. It is important to note that each additional set of 256 PTY's occupy approximately 8 KB of
kernel memory on 32-bit architectures and for security as well as performance reasons, we must
keep the number as low as possible. Therefore, we will change the default value of 256 to 128.
*
* I2C support
* I2C support (CONFIG_I2C) [N/y/?] Press Enter
This option allows us to enable I2C and SMBus support under Linux. IC2 is a slow serial bus
protocol used in many micro controller applications and developed by Philips. If you need this
feature, then say Y to the question otherwise say N here. Most people will say N here.
Bus Mouse Support (CONFIG_BUSMOUSE) [N/y/?] Press Enter
This option allows us to enable bus mouse support under Linux. All modern computer now use a
serial mouse. If you still continue to use a bus mouse on your system, then you have to say Y
here. Laptop users also need to say Y here. Most users will simply say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-165-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
166
Mouse Support (not serial and bus mice) (CONFIG_MOUSE) [Y/n/?] Press Enter
This option allows us to enable support for no serial or bus mice support under Linux. This is for
machines with a mouse, which is neither a serial, nor a bus mouse. Examples are PS/2 mice. If
you have a PS/2 mouse, then say Y here, otherwise say N here. Laptop and workstation users
also need to say Y here. If you use your Linux system for dedicated Ethernet WWW/FTP servers,
then you can say N here and save some space in your Kernel code.
PS/2 mouse (aka "auxiliary device") support (CONFIG_PSMOUSE) [Y/n/?] Press
Enter
This option enables support for a PS/2 mouse on your system. The PS/2 mouse connects to a
special mouse port that looks much like the keyboard port (small circular connector with 6 pins). If
you have this kind of mouse (most modern computers and laptops have and use it), then say Y
here otherwise say N.
C&T 82C710 mouse port support (as on TI Travelmate) (CONFIG_82C710_MOUSE)
[N/y/?] Press Enter
This option allows us to enable support for a certain kind of PS/2 mouse used on the TI
Travelmate. If you have this kind of mouse installed on your system, then say Y here. Most users
will say N to this question.
PC110 digitizer pad support (CONFIG_PC110_PAD) [N/y/?] Press Enter
This drives the digitizer pad on the IBM PC110 palmtop. It can turn the digitizer pad into PS/2
mouse emulation with tap gestures or into an absolute pad. Most users will answer this question
N.
QIC-02 tape support (CONFIG_QIC02_TAPE) [N/y/?] Press Enter
This option allows us to enable QIC-02 tape support under Linux. QIC-02 is a non-SCSI tape
drive and if you use it, then says Y to the question. Most users will say N here.
*
* Watchdog Cards
* Watchdog Timer Support (CONFIG_WATCHDOG) [N/y/?] Press Enter
This option enables Watchdog Timer support under Linux. For details about watchdog, please
read Documentation/watchdog.txt in the kernel source. Most users will say N here.
Intel i8x0 Random Number Generator support (CONFIG_INTEL_RNG) [N/y/?] Press
Enter
This option allows us to enable support for a driver that provides kernel-side support for the
Random Number Generator hardware found on Intel i8xx-based motherboards. If you have this
generator hardware installed on your computer, then you can say Y to the question. Most people
will say N here.
/dev/nvram support (CONFIG_NVRAM) [N/y/?] Press Enter
This option allows us to get read and write access to the 50 bytes of non-volatile memory in the
real time clock (RTC). This memory is conventionally called "CMOS RAM" on PCs and may be
used to view settings there, or to change them (with some utility). Most users will say N to this
question.
Enhanced Real Time Clock Support (CONFIG_RTC) [N/y/?] Press Enter
This option allows us to get access to the real time clock (or hardware clock) built into the
computer. If you run Linux on a multiprocessor machine and said Y to "Symmetric Multi
Processing" above, you should say Y here to read and set the RTC in an SMP compatible
fashion.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-166-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
167
Double Talk PC internal speech card support (CONFIG_DTLK) [N/y/?] Press Enter
This option allows us to enable support for the DoubleTalk PC under Linux. If you have a speech
card installed on your computer, then answer to this question by Y. Most users will say N here.
Siemens R3964 line discipline (CONFIG_R3964) [N/y/?] Press Enter
This option allows synchronous communication with devices using the Siemens R3964 packet
protocol. Unless you are dealing with special hardware like PLC's, you are unlikely to need this.
Most users will simply say N here.
Applicom intelligent fieldbus card support (CONFIG_APPLICOM) [N/y/?] Press
Enter
This option provides the kernel-side support for the intelligent fieldbus cards made by Applicom
International. Unless you are dealing with such a card, you are unlikely to need this. Most users
will simply say N here.
*
* Ftape, the floppy tape device driver
* Ftape (QIC-80/Travan) support (CONFIG_FTAPE) [N/y/?] Press Enter
This option allows us to enable support for some well know tape drives under Linux. If you enable
this option, then you'll be asked to select your make and model from the available list. If you don't
use tape drive on your computer, then you can say N to this option.
/dev/agpgart (AGP Support) (CONFIG_AGP) [Y/n/?] n
This option is only pertinent if you use XFree86 on your computer and want to use the GLX or
DRI features for better performance. Enable this option only if you use graphical interface on your
computer. If you system is a server, then you really don't need this option. In our example, we are
configuring the kernel for a server purpose, therefore, we will say N here.
Direct Rendering Manager (XFree86 4.1.0 and higher DRI support) (CONFIG_DRM)
[Y/n/?] n
This option is directly related to the use of XFree86 and graphical interface on your computer. It
allows us to enable Kernel-level support for the Direct Rendering Infrastructure (DRI) for better
performance of the system. If your computer doesn't have a graphical interface installed, and is
configured as a server, then you don't need to enable this option. If you say Y here because you
have and use a graphical interface, then you need to select the module that's right for your
graphics card from the available list. In our example, we are configuring the kernel for a server
purpose, therefore, we will say N here.
ACP Modem (Mwave) support (CONFIG_MWAVE) [N/y/?] Press Enter
This option allows us to enable ACP modem (Mwave) support under Linux. ACP is a WinModem
composed of a kernel driver and a user level application. Together these components support
direct attachment to public switched telephone networks (PSTNs) and support selected countries.
Most user will simply say N here.
*
* Multimedia devices
* Video For Linux (CONFIG_VIDEO_DEV) [N/y/?] Press Enter
This option allows us to enable support for audio/video capture and overlay devices and FM radio
cards under Linux. If you use graphical interface on your system, you need to say Y here. If your
system is configured as a server and you don't have graphical interface installed on it, then you
can safety say N here. In our example, we are configuring the kernel for a server purpose,
therefore, we will say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-167-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
168
*
* File systems
* Quota support (CONFIG_QUOTA) [N/y/?] y
This option is important to allow us to set per user/group limits for disk usage (also called disk
quotas) under Linux. It is sometimes required on server environments where such a purpose is
required. If you use Linux as a workstation with graphical interface, it is not useful to enable it.
Only enable this option on server environment when required.
Kernel automounter support (CONFIG_AUTOFS_FS) [N/y/?] Press Enter
This option allows us to automatically mount remote file systems on demand. A newer version of
the automounter with more features is now available; therefore we really don't need to enable this
option. Say N here.
Kernel automounter version 4 support (also supports v3) (CONFIG_AUTOFS4_FS)
[Y/n/?] n
This option allows us to enable support for the new automounter tool. If you are not a part of a
fairly large, distributed network or don't have a laptop which needs to dynamically reconfigure to
the local network, you probably do not need an automounter, and can say N here.
Ext3 journalling file system support (EXPERIMENTAL) (CONFIG_EXT3_FS) [N/y/?] y
This option is very important and allows us to enable support for the new EXT3 journaling file
system support under Linux. EXT3 is the journaling version of the Second extended file system
(often called ext3), the de facto standard Linux file system (method to organize files on a storage
device) for hard disks. Most users will say Y here to be able to get advantage of the new EXT3
journaling file system.
JBD (ext3) debugging support (CONFIG_JBD_DEBUG) [N/y/?] y
This option allows us to enable debugging output while the system is running EXT3 under Linux,
in order to help track down any problems you are having. It is a good idea to enable it, therefore
answer Y to the question.
DOS FAT fs support (CONFIG_FAT_FS) [N/y/?] Press Enter
If you want to use one of the FAT-based file systems (the MS-DOS, VFAT (Windows 95) and
UMSDOS (used to run Linux on top of an ordinary DOS partition) file systems, then you must say Y
here. In many case, we really don't need this kind of feature and can safety say N here.
Recommended choice is N and especially for a linux server system.
Compressed ROM file system support (CONFIG_CRAMFS) [N/y/?] Press Enter
This option allows us to enable CramFs support under Linux. CramFs (Compressed ROM File
System) is designed to be a simple, small, and compressed file system for ROM based embedded
systems. Most users will simply say N here.
Virtual memory file system support (former shm fs) (CONFIG_TMPFS) [Y/n/?] Press
Enter
This option allows us to enable support for Tmpfs under Linux. Tmpfs is a file system which
keeps all files in virtual memory. It is a good idea to enable it on your computer.
Simple RAM-based file system support (CONFIG_RAMFS) [N/y/?] Press Enter
This option allows us to enable support for Ramfs under Linux. Ramfs is a file system which
keeps all files in RAM. It allows read and write access. Most users will say N here.
ISO 9660 CDROM file system support (CONFIG_ISO9660_FS) [Y/n/?] Press Enter
This option is important and allows Linux to access and read your CD-ROM. Since everyone has
and uses a CD-ROM, it is vital to say Y to this question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-168-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
169
Microsoft Joliet CDROM extensions (CONFIG_JOLIET) [N/y/?] Press Enter
This option allows us to be able to read Joliet CD-ROM's under Linux. Joliet is a Microsoft
extension for the ISO 9660 CD-ROM file system, which allows for long filenames in unicode
format. If you think that you'll need to access some files written in this format under Linux with
your CD-ROM, then you need to say Y here. For workstations, you may need to say Y here but
for Linux servers, you have to say N here.
Transparent decompression extension (CONFIG_ZISOFS) [N/y/?] Press Enter
This is a Linux-specific extension to RockRidge which lets you store data in compressed form on
a CD-ROM and have it transparently decompressed when the CD-ROM is accessed. Again, for
workstations you may need it but for servers, you don't need it. Say N.
Minix fs support (CONFIG_MINIX_FS) [N/y/?] Press Enter
This option allows us to enable Minix fs support with Linux. The minix file system (method to
organize files on a hard disk partition or a floppy disk) was the original file system for Linux, but
has been superseded by the second extended file system (ext2fs). Simply say N here.
FreeVxFS file system support (VERITAS VxFS(TM) compatible) (CONFIG_VXFS_FS)
[N/y/?] Press Enter
FreeVxFS is a file system driver that support the VERITAS VxFS(TM) file system format.
VERITAS VxFS(TM) is the standard file system of SCO UnixWare (and possibly others) and
optionally available for Sunsoft Solaris, HP-UX and many other operating systems. If you want to
support such file system on your computer, then say Y here otherwise say N.
NTFS file system support (read only) (CONFIG_NTFS_FS) [N/y/?] Press Enter
NTFS is the file system of Microsoft Windows NT. Say Y if you want to get read access to files on
NTFS partitions of your hard drives otherwise say N here. Most users will say N.
OS/2 HPFS file system support (CONFIG_HPFS_FS) [N/y/?] Press Enter
OS/2 is IBM's operating system for PC's, the same as Warp, and HPFS is the file system used for
organizing files on OS/2 hard disk partitions. Say Y if you want to be able to read files from and
write files to an OS/2 HPFS partition on your hard drive. Most users will say N here.
/proc file system support (CONFIG_PROC_FS) [Y/n/?] Press Enter
This is a virtual file system providing information about the status of the system. Several
programs depend on this, so everyone should say Y here.
/dev/pts file system for Unix98 PTYs (CONFIG_DEVPTS_FS) [Y/n/?] Press Enter
This option allows us to get a virtual file system which can be mounted on /dev/pts with
"mount -t devpts". This, together with the pseudo terminal master multiplexer /dev/ptmx, is
used for pseudo terminal support as described in The Open Group's Unix98 standard. Again,
everyone should say Y here.
ROM file system support (CONFIG_ROMFS_FS) [N/y/?] Press Enter
This is a very small read-only file system mainly intended for initial ram disks of installation disks,
but it could be used for other read-only media as well. Most people will simply say N to this
question. If you want to run SCSI system on modularized kernel, you should say Y here.
Second extended fs support (CONFIG_EXT2_FS) [Y/n/?] Press Enter
This is the de facto standard Linux file system (method to organize files on a storage device) for
hard disks and you must say Y to this question.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-169-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
170
System V/Xenix/V7/Coherent file system support (CONFIG_SYSV_FS) [N/y/?] Press
Enter
SCO, Xenix and Coherent are commercial Unix systems for Intel machines, and Version 7 was
used on the DEC PDP-11. Saying Y here would allow you to read from their floppies and hard
disk partitions. Most people will say N to this question.
UDF file system support (read only) (CONFIG_UDF_FS) [N/y/?] Press Enter
This is the new file system used on some CD-ROMs and DVD's. Say Y if you intend to mount
DVD discs or CD-RW's written in packet mode, or if written to by other UDF utilities, such as
DirectCD. Only enable this option if you have some such need.
UFS file system support (read only) (CONFIG_UFS_FS) [N/y/?] Press Enter
BSD and derivative versions of Unix (such as SunOS, FreeBSD, NetBSD, OpenBSD and
NeXTstep) use a file system called UFS. Some System V Unixes can create and mount hard disk
partitions and diskettes using this file system as well. Saying Y here will allow you to read from
these partitions. Most users will say N here.
*
* Network File Systems
* Coda file system support (advanced network fs) (CONFIG_CODA_FS) [N/y/?] Press
Enter
Coda is an advanced network file system, similar to NFS in that it enables you to mount file
systems of a remote server and access them with regular Unix commands as if they were sitting
on your hard disk. Enable this option only if you need it otherwise disable it.
NFS file system support (CONFIG_NFS_FS) [Y/n/?] n
If you are connected to some other (usually local) Unix computer (using SLIP, PLIP, PPP or
Ethernet) and want to mount files residing on that computer (the NFS server) using the Network
File Sharing protocol, say Y here.
NFS server support (CONFIG_NFSD) [Y/n/?] n
If you want your Linux box to act as an NFS *server*, so that other computers on your local
network which support NFS can access certain directories on your box transparently, you have
two options: you can use the self-contained user space program nfsd, in which case you should
say N here, or you can say Y and use the kernel based NFS server. The advantage of the kernel
based solution is that it is faster. Finally, if you don't want to support NFS server at all, simply say
N here.
SMB file system support (to mount Windows shares etc.) (CONFIG_SMB_FS) [N/y/?]
Press Enter
SMB (Server Message Block) is the protocol Windows for Workgroups (WfW), Windows 95/98,
Windows NT, 2000, XP and OS/2 Lan Manager use to share files and printers over local
networks. Saying Y here allows you to mount their file systems (often called "shares" in this
context) and access them just like any other Unix directory. Enable this option only if you need it.
In most cases the answer to this question will be N even if you install Samba on your system.
NCP file system support (to mount NetWare volumes) (CONFIG_NCP_FS) [N/y/?]
Press Enter
NCP (NetWare Core Protocol) is a protocol that runs over IPX and is used by Novell NetWare
clients to talk to file servers. It is to IPX what NFS is to TCP/IP, if that helps. Saying Y here allows
you to mount NetWare file server volumes and to access them just like any other Unix directory.
Enable this option only if you need it. In most cases the answer to this question will be N. You do
not have to say Y here if you want your Linux box to act as a file *server* for Novell NetWare
clients.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-170-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
171
*
* Partition Types
* Advanced partition selection (CONFIG_PARTITION_ADVANCED) [N/y/?] Press Enter
This option allows us to enable the use of hard disks under Linux which were partitioned under an
operating system running on a different architecture than the Linux system. Note that the answer
to this question won't directly affect the kernel: saying N will just cause the configuration to skip all
the questions about foreign partitioning schemes.
*
* Console drivers
* VGA text console (CONFIG_VGA_CONSOLE) [Y/n/?] Press Enter
This option allows us to use Linux in text mode through a display that complies with the generic
VGA standard. Virtually everyone wants that. Everyone should say Y here.
Video mode selection support (CONFIG_VIDEO_SELECT) [N/y/?] Press Enter
This option allows us to enable support for text mode selection on kernel startup. In general we
don't need to enable this option. Say N here and read the file Documentation/svga.txt for
more information about the Video mode selection support if you are curious.
*
* Sound
* Sound card support (CONFIG_SOUND) [Y/n/?] n
This option allows us to enable sound support under Linux. If you have a sound card in your
computer, then say Y here and select from the available list of sound card the one that you have.
If you run Linux as a workstation, you may need to say Y here, if you run Linux as a server, you
really don't need to enable this option and can safety say N here.
*
* USB support
* Support for USB (CONFIG_USB) [Y/n/?] n
This option allows us to enable USB support under Linux. If your computer has a USB port and
you want to use USB devices, then you have to say Y here. For servers you really don't need to
say Y here and can safety say N.
*
* Kernel hacking
* Kernel debugging (CONFIG_DEBUG_KERNEL) [N/y/?] Press Enter
You have to say Y here only if you are developing drivers or trying to debug and identify kernel
problems. Most users will simply say N here.
*
* Grsecurity
*
Grsecurity (CONFIG_GRKERNSEC) [N/y/?] y
This option allows us to enable Grsecurity support under Linux. If you say Y to this option, you
will be able to configure many Grsecurity features that will enhance the security of your Linux
system in many ways. This option is available ONLY if you have patched your Linux kernel with
the Grsecurity patch as discussed previously in this chapter. For best security of your Linux
server say Y here.
Security level (Low, Medium, High, Customized) [Customized] Press Enter
This Grsecurity option allows us to choose between three predefined Grsecurity
configurations or to customize the Grsecurity configuration as we want. For better control of
what the patch does and what we may or may not need, we chose to have full control about what
should or shouldn’t be enable with Grsecurity by pressing the [Enter] key to accept the
default choice of [Customized], which let us see all available security features and chose only
the one we need for our server and kernel security setup.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-171-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
172
*
* Buffer Overflow Protection
*
Openwall non-executable stack (CONFIG_GRKERNSEC_STACK) [N/y/?] y
This Grsecurity option allows us to enable the non-executable stack protection on the system.
If you say Y here, your system will not allow execution of code on the stack, making buffer
overflow exploitation more difficult. It’s a good idea to say Y here.
Gcc trampoline support (CONFIG_GRKERNSEC_STACK_GCC) [N/y/?] Press Enter
This Grsecurity option allows us to support trampoline code along with the above stack
protection. Trampolining is an action to use the ability of the stack to contain executable code,
which can improve program efficiency in a few cases. Since few programs and some version of
GCC use and need 'trampolining', it is preferable to NOT enable this option to avoid break of some
program on the system. Say N here by pressing the [Enter] key.
Read-only kernel memory (CONFIG_GRKERNSEC_KMEM) [N/y/?] y
This Grsecurity option allows us to enable the read-only kernel memory protection on the
system. If you say Y here, root will not be able to modify the contents of kernel memory. It’s a
good idea to say Y here. If you are building a Monolithic Kernel, then the ability of an attacker to
insert foreign code into a running kernel is completely removed. Yes, another good idea to build a
Monolithic Kernel instead of a Modularized Kernel. Say Y here.
*
* Access Control Lists
*
Grsecurity ACL system (CONFIG_GRKERNSEC_ACL) [N/y/?] y
This Grsecurity option allows us to enable the Access Control List system (ACL) for
Grsecurity. It’s a good idea to say Y here. ACL allows us to better control what program, files,
etc on the system are allowed to do. We use it to apply a security policy that will work for the
entire system. You can install and run Grsecurity without ACL but it is recommended for
optimum security to enable this feature and use it. Once properly implemented, it will become
impossible for a cracker to access and damage your Linux server. Personally, with Grsecurity
ACL, I don’t know how someone could break into a Linux system. Say Y here.
ACL Debugging Messages (CONFIG_GR_DEBUG) [N/y/?] y
This Grsecurity option allows the Grsecurity ACL system to print debugging messages as
an aid to finding problems in your ACL sets. It’s really a good idea to say Y here since it can
become very difficult to debug problems with ACL if this option is not enable. Say Y here.
Extra ACL Debugging Messages (CONFIG_GR_SUPERDEBUG) [N/y/?] Press Enter
This Grsecurity option allows us to enable additional debugging messages that can help in
finding problems in ACL sets or to gain a better understanding of the internal workings of the ACL
system. In many cases, it is not necessary to enable this additional debugging messages feature
for ACL and we will say N here.
Denied capability logging (CONFIG_GRKERNSEC_ACL_CAPLOG) [N/y/?] y
This Grsecurity option allows us to enable the denied capability logging support protection with
ACL. If you say Y here, logs will be produced when a root-owned process does not have a
needed capability raised in his set. This can help to debug the ACL of Grsecurity again. It’s a
good idea to say Y here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-172-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
173
Path to gradm (CONFIG_GRADM_PATH) [/sbin/gradm] Press Enter
This Grsecurity option allows us to specify the path of the gradm binary installed on the
system. gradm is the binary program used to manage Grsecurity ACL on the server. You have
to download and install it to be able to use ACL on your server. Please read the next chapter in
this book to get information about gradm and how to setup Grsecurity ACL on your system.
The default path for gradm as shown above is correct and we press the [Enter] key to accept
the default value for the path.
Maximum tries before password lockout (CONFIG_GR_MAXTRIES) [3] 2
Once the gradm program is installed on your server to control and manage Grsecurity ACL,
you will have to create a password for gradm to work. This option allows us to specify the number
of times a user can attempt to authorize them selves with the Grsecurity ACL system. Here we
change the default value of 3 to become 2, meaning that users are allowed to authorize
themselves with the Grsecurity ACL system twice.
Time to wait after max password tries, in seconds (CONFIG_GR_TIMEOUT) [30]
Press Enter
This option specifies the time the user must wait after attempting to authorize to the ACL system
with the maximum number of invalid passwords. Just press [Enter] to accept the default entry.
*
* Filesystem Protections
*
Proc restrictions (CONFIG_GRKERNSEC_PROC) [N/y/?] y
This Grsecurity option allows us to enable the proc restrictions protection on the system. If you
say Y here, the permissions of the /proc file system will be altered to enhance system security
and privacy. It’s a very good idea to say Y here.
Restrict to user only (CONFIG_GRKERNSEC_PROC_USER) [N/y/?] y
This Grsecurity option allows us to enable restrict to user only protection on the system. If you
say Y here, non-root users will only be able to view their own processes, restricts them from
viewing network-related information, viewing kernel symbol and module information. It’s a very
good idea to say Y here.
Additional restrictions (CONFIG_GRKERNSEC_PROC_ADD) [N/y/?] y
This Grsecurity option allows us to enable additional restrictions protection on the system. If
you say Y here, additional restrictions will be placed on the /proc file system that keep normal
users from viewing cpu and device information. Again, it’s a good idea to say Y here.
Linking restrictions (CONFIG_GRKERNSEC_LINK) [N/y/?] y
This Grsecurity option allows us to enable the linking restrictions protection on the system. If
you say Y here, /tmp race exploits will be prevented, since users will no longer be able to follow
symlinks owned by other users in world-writeable +t directories, unless the owner of the symlink
is the owner of the directory. Users will also not be able to hard link to files they do not own. It’s
really a good idea to say Y here.
FIFO restrictions (CONFIG_GRKERNSEC_FIFO) [N/y/?] y
This Grsecurity option allows us to enable FIFO restrictions protection on the system. If you
say Y here, users will not be able to write to FIFOs they don't own in world-writeable +t
directories, unless the owner of the FIFO is the same owner of the directory it's held in. It’s a
good idea to say Y here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-173-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
174
Secure file descriptors (CONFIG_GRKERNSEC_FD) [N/y/?] y
This Grsecurity option allows us to enable secure file descriptors protection on the system. If
you say Y here, binaries will be protected from data spoofing attacks. It’s a very good idea to say
Y here.
Chroot jail restrictions (CONFIG_GRKERNSEC_CHROOT) [N/y/?] y
This Grsecurity option allows us to enable chroot jail restrictions protection on the system. If
you say Y here, you will be able to choose several options that will make breaking out of a
chrooted jail much more difficult. It’s a very good idea to say Y here.
Restricted signals (CONFIG_GRKERNSEC_CHROOT_SIG) [N/y/?] y
This Grsecurity option allows us to enable the restricted signals protection on the system. If
you say Y here, processes inside a chroot will not be able to send signals outside of the chroot.
It’s a good idea to say Y here.
Deny mounts (CONFIG_GRKERNSEC_CHROOT_MOUNT) [N/y/?] y
This Grsecurity option allows us to enable deny mounts protection on the system. If you say Y
here, processes inside a chroot will not be able to mount or remount file systems. It’s a good idea
to say Y here.
Deny double-chroots (CONFIG_GRKERNSEC_CHROOT_DOUBLE) [N/y/?] y
This Grsecurity option allows us to enable the deny double-chroot protection on the system. If
you say Y here, processes inside a chroot will not be able to chroot again. This is a widely used
method of breaking out of a chroot jail and should not be allowed. It’s a good idea to say Y here.
Enforce chdir("/") on all chroots (CONFIG_GRKERNSEC_CHROOT_CHDIR) [N/y/?] y
This Grsecurity option allows us to enable the enforce chdir("/") on all chroots protection
on the system. If you say Y here, the current working directory of all newly-chrooted applications
will be set to the root directory of the chroot. It’s a good idea to say Y here.
Deny (f)chmod +s (CONFIG_GRKERNSEC_CHROOT_CHMOD) [N/y/?] y
This Grsecurity option allows us to enable the deny (f)chmod +s protection on the system.
If you say Y here, processes inside a chroot will not be able to chmod or fchmod files to make
them have suid or sgid bits. It’s a really good idea to say Y here.
Deny mknod (CONFIG_GRKERNSEC_CHROOT_MKNOD) [N/y/?] y
This Grsecurity option allows us to enable the deny mknod protection on the system. If you
say Y here, processes inside a chroot will not be allowed to mknod (create device on the system).
It’s a good idea to say Y here, unless you run into software incompatibilities.
Deny ptraces (CONFIG_GRKERNSEC_CHROOT_PTRACE) [N/y/?] y
This Grsecurity option allows us to enable the deny ptraces protection on the system. If you
say Y here, processes inside a chroot will not be able to ptrace other processes. It’s a good idea
to say Y here.
Restrict priority changes (CONFIG_GRKERNSEC_CHROOT_NICE) [N/y/?] y
This Grsecurity option allows us to enable the restrict priority changes protection on the
system. If you say Y here, processes inside a chroot will not be able to raise the priority of
processes in the chroot, or alter the priority of processes outside the chroot. It’s a good idea to
say Y here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-174-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
175
Capability restrictions within chroot (CONFIG_GRKERNSEC_CHROOT_CAPS) [N/y/?]
Press Enter
This Grsecurity option allows us to enable the capability restrictions within chroot protection on
the system. If you say Y here, the capabilities on all root processes within a chroot jail will be
lowered to stop module insertion, raw i/o, system and net admin tasks, rebooting the system,
modifying immutable files, and changing the system time. This option can break some
applications on the server and we disable it by answering to the question by N.
Secure keymap loading (CONFIG_GRKERNSEC_KBMAP) [N/y/?] y
This Grsecurity option allows us to enable the secure keymap loading protection on the
system. If you say Y here, KDSKBENT and KDSKBSENT ioctl calls being called by unprivileged
users will be denied. This means that unprivileged users with access to the console will NOT be
able to modify keyboard bindings. It’s a good idea to say Y here.
*
* Kernel Auditing
*
Single group for auditing (CONFIG_GRKERNSEC_AUDIT_GROUP) [N/y/?] Press Enter
This Grsecurity option allows us to enable single group for auditing protection on the system. If
you say Y here, the exec, chdir, (un)mount, and ipc logging features of Grsecurity will
only operate on a group you specify. By default Grsecurity produces a large amount of logs
from the entire system on a production server; we don’t really need the entire auditing feature
provided by Grsecurity even on specified group. Therefore we simply say N to this question
and enable later in this Grsecurity kernel configuration the auditing log that we really need for
production servers.
Exec logging (CONFIG_GRKERNSEC_EXECLOG) [N/y/?] Press Enter
This Grsecurity option allows us to enable the exec logging protection on the system. If you
say Y here, execution of any program on the server will be logged to syslog. This option when
enabled will produce a LOT of logs, especially on an active system. Therefore, we don’t
recommend you to enable this security option. If you are those people that like to spend all their
time reading log files, you could enable this option but be aware that it will take you a day to read
all the logs generated by this option on a production server. Be reasonable, and say N here.
Log execs within chroot (CONFIG_GRKERNSEC_CHROOT_EXECLOG) [N/y/?] y
This Grsecurity option allows us to enable the log execs within chroot protection on the
system. If you say Y here, all executions of any program inside a chroot jail environment will be
logged to syslog. Contrary to the previous option that logs execution of any program on the
system, this option ONLY logs the execution of program inside a chroot jail. In general, services
running in chroot jail environment are limited, meaning that your log file will not become too BIG
to read; therefore we can say Y here to enable this security option on the server.
Chdir logging (CONFIG_GRKERNSEC_AUDIT_CHDIR) [N/y/?] Press Enter
This Grsecurity option allows us to enable the chdir logging protection on the system. If you
say Y here, all chdir() calls will be logged. Chdir() calls is when you navigate via your console
on your server by using the ‘cd’ command. Imagine how many times you use the ‘cd’ command
on your server when you performing some administration tasks. I recommend you to NOT enable
this option if you want to avoid some BIG log files to read again. Say N here.
(Un)Mount logging (CONFIG_GRKERNSEC_AUDIT_MOUNT) [N/y/?] Press Enter
This Grsecurity option allows us to enable the (Un)Mount logging protection on the system. If
you say Y here, all mounts and unmounts calls will be logged to syslog. This means that each
time you mount or unmount drive on your server; the action will be logged to syslog. You can
enable this security option if you like, but personally, I prefer to disable it. Say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-175-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
176
IPC logging (CONFIG_GRKERNSEC_AUDIT_IPC) [N/y/?] y
This Grsecurity option allows us to enable the IPC logging protection on the system. If you say
Y here, creation and removal of message queues, semaphores, and shared memory will be
logged. It’s a good idea to say Y here.
Ptrace logging (CONFIG_GRKERNSEC_AUDIT_PTRACE) [N/y/?] Press Enter
This Grsecurity option allows us to enable the ptrace logging protection on the system. If you
say Y here, all successful ptraces will be logged. Ptraces are special operations performed
when programs like strace or gdb are run. In general we never install programs like strace
and gdb on a production server. These programs are only required on development server to
debug software. If you don’t have these kinds of programs installed on your server, you can
safety say N here, otherwise say Y to this security option.
Signal logging (CONFIG_GRKERNSEC_SIGNAL) [N/y/?] y
This Grsecurity option allows us to enable the signal logging protection on the system. If you
say Y here, certain important signals will be logged, such as SIGSEGV that will inform you of
when an error in a program occurred, which in some cases could mean a possible exploit
attempt. It’s a good idea to say Y here.
Fork failure logging (CONFIG_GRKERNSEC_FORKFAIL) [N/y/?] y
This Grsecurity option allows us to enable the fork failure logging protection on the system. If
you say Y here, all failed fork() attempts will be logged. This could suggest a fork bomb, or
someone attempting to overstep their process limit. It’s a good idea to say Y here.
Set*id logging (CONFIG_GRKERNSEC_SUID) [N/y/?] Press Enter
This Grsecurity option allows us to enable the set*id logging protection on the system. If you
say Y here, all set*id() calls will be logged. Enabling this security option could produce a lot of
logs on an active system that run some services that use set*id() calls to operate. Mailman,
Exim, Sendmail are known to software that uses many set*id() calls. Also, we already have
other security programs doing the same job like sXid, therefore we don’t really need to enable
this option. It is your’s to decide if you really need it or not. Personally, I disable this option by
saying N here and use sXid to achieve the same result.
Log set*ids to root (CONFIG_GRKERNSEC_SUID_ROOT) [N/y/?] y
This Grsecurity option allows us to enable the log set*ids to root protection on the system. If
you say Y here, only set*id() calls where a user is changing to the GID or UID of the root user
will be logged. Such information could be useful when detecting a possible intrusion attempt. This
option will produce smaller logs than logging all calls; therefore it’s a good idea to say Y here.
Time change logging (CONFIG_GRKERNSEC_TIME) [N/y/?] y
This Grsecurity option allows us to enable the time change logging protection on the system. If
you say Y here, any changes of the system clock will be logged. It’s a good idea to say Y here.
*
* Executable Protections
*
Exec process limiting (CONFIG_GRKERNSEC_EXECVE) [N/y/?] y
This Grsecurity option allows us to enable the exec process limiting protection on the system.
If you say Y here, users with a resource limit on processes will have the value checked during
execve() calls (execution of program). It’s really a good idea to say Y here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-176-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
177
Dmesg(8) restriction (CONFIG_GRKERNSEC_DMESG) [N/y/?] y
This Grsecurity option allows us to enable the dmesg restriction protection on the system. If
you say Y here, non-root users will not be able to use dmesg(8) to view up to the last 4kb of
messages in the kernel's log buffer. Again, it’s really a good idea to say Y here.
Randomized PIDs (CONFIG_GRKERNSEC_RANDPID) [N/y/?] y
This Grsecurity option allows us to enable the randomized PIDs protection on the system. If
you say Y here, all PIDs created on the system will be pseudo-randomly generated. This is
extremely effective to disallow an attacker from guessing pids of daemons, etc. It’s a good idea to
say Y here.
Altered default IPC permissions (CONFIG_GRKERNSEC_IPC) [N/y/?] Press Enter
This Grsecurity option allows us to enable the altered default IPC permissions protection on
the system. If you say Y here, the default permissions for IPC objects will be set based on the file
system umask of the user creating the object. This is a good security feature but unfortunately, it
is known to break software like Apache. Therefore we say N here.
Limit uid/gid changes to root (CONFIG_GRKERNSEC_TTYROOT) [N/y/?] y
This Grsecurity option allows us to enable the limit UID/GID changes to root protection on the
system. If you say Y here, you will be able choose from three options that will allow you to restrict
access to the root account by console type. Therefore we say Y here.
Deny physical consoles (tty) (CONFIG_GRKERNSEC_TTYROOT_PHYS) [N/y/?] Press
Enter
This Grsecurity option allows us to enable the deny physical consoles (tty) protection on the
system. If you say Y here, access to root from physical consoles will be denied. This is only
recommended for rare cases where you will never need to be physically at the machine. For most
of us, this is not the case and we have to say N here.
Deny serial consoles (ttyS) (CONFIG_GRKERNSEC_TTYROOT_SERIAL) [N/y/?] y
This Grsecurity option allows us to enable deny serial consoles (ttyS) protection on the
system. If you say Y here, access to root from serial consoles will be denied. Most people can say
Y here, since most don't use serial devices for their console access.
Deny pseudo consoles (pty) (CONFIG_GRKERNSEC_TTYROOT_PSEUDO) [N/y/?] Press
Enter
This Grsecurity option allows us to enable the deny pseudo consoles (pty) protection on the
system. If you say Y here, access to root from pseudo consoles will be denied. Pseudo consoles
include consoles from telnet, ssh, or any other kind of interactive shell initiated from the
network. In general, most of us use at least SSH to make a remote connection. Therefore we
have to say N here, if we want to continue to use SSH for secure remote connection.
Fork-bomb protection (CONFIG_GRKERNSEC_FORKBOMB) [N/y/?] y
This Grsecurity option allows us to enable fork-bomb protection on the system. If you say Y
here, you will be able to configure a group to add to users on your system that you want to be
unable to fork-bomb the system. It’s a good idea to say Y here and chose in the next security
option the GID to run this protection with.
GID for restricted users (CONFIG_GRKERNSEC_FORKBOMB_GID) [1006] Press Enter
This Grsecurity option allows us to enable the GID for restricted users’ protection on the
system. Here we have to enter a GID number as the value, the default value is 1006 and we can
keep it by pressing the [Enter] key. It is important to note that this GID should be added to any
user for which the feature should be activated. See the next chapter of this book for more
information about the procedures to follow. At this time you only need to accept the default value.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-177-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
178
Forks allowed per second (CONFIG_GRKERNSEC_FORKBOMB_SEC) [40] Press Enter
This Grsecurity option allows us to enable the forks allowed per second protection on the
system. This option is a continuation of the above forks options, here we have to enter the
maximum number of forks allowed per second, and the default setting should be fine for most
users.
Maximum processes allowed (CONFIG_GRKERNSEC_FORKBOMB_MAX) [20] 35
This Grsecurity option allows us to enable the maximum processes allowed on the system.
Here we have to enter the maximum number of processes users in the fork-bomb protected
group can run. We change the default value of 20 to become 35. 35 is the number you have set
into the /etc/security/limit.conf file. Please see what is set into your limit.conf file
and report it here. In general, 35 is a good value to go with.
Trusted path execution (CONFIG_GRKERNSEC_TPE) [N/y/?] y
This Grsecurity option allows us to enable trusted path execution protection on the system. If
you say Y here, you will be able to choose a GID to add to the supplementary groups of users
you want to mark as "untrusted." These users will not be able to execute any files that are not in
root-owned directories writeable only by root. It’s a good idea to say Y here.
Glibc protection (CONFIG_GRKERNSEC_TPE_GLIBC) [N/y/?] y
This Grsecurity option allows us to enable the glibc protection on the system. If you say Y
here, all non-root users executing any files while glibc specific environment variables such as
LD_PRELOAD are set, will have their environment cleared of these variables, since they could be
used to evade the trusted path execution protection. It also protects against evasion through
executing the dynamic linker to run a rogue binary. It is highly recommended you say Y here.
Partially restrict non-root users (CONFIG_GRKERNSEC_TPE_ALL) [N/y/?] y
This Grsecurity option allows us to enable the partially restrict non-root users protection on the
system. If you say Y here, All non-root users other than the ones in the group specified in the
main TPE option will only be allowed to execute files in directories they own that are not group or
world-writeable, or in directories owned by root and writeable only by root. It’s a good idea to say
Y here.
GID for untrusted users: (CONFIG_GRKERNSEC_TPE_GID) [1005] Press Enter
This Grsecurity option allows us to enable the GID for untrusted user’s protection on the
system. Here we have to enter a GID number as the value, the default value is 1005 and we can
keep it by pressing the [Enter] key. It is important to note that this GID should be added to any
user for which the feature should be activated. See the next chapter of this book for more
information about the procedures to follow. At this time you only need to accept the default value.
Restricted ptrace (CONFIG_GRKERNSEC_PTRACE) [N/y/?] y
This Grsecurity option allows us to enable the restricted ptrace protection on the system. If
you say Y here, no one but root will be able to ptrace processes. Tracing syscalls inside the
kernel will also be disabled. It’s a good idea to say Y here.
Allow ptrace for group (CONFIG_GRKERNSEC_PTRACE_GROUP) [N/y/?] Press Enter
This Grsecurity option allows us to enable the allow ptrace for group protection on the system.
If you say Y here, you will be able to choose a GID of users will be able to ptrace. Remember
that ptraces are special operations performed when programs like strace or gdb are run for
debugging purpose. Since these kind of program should in general be run on development server
and by a root user only, we can safety say N here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-178-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
179
*
* Network Protections
*
Randomized IP IDs (CONFIG_GRKERNSEC_RANDID) [N/y/?] y
This Grsecurity option allows us to enable the allow randomized IP IDs protection on the
system. If you say Y here, the entire ID field on all outgoing packets will be randomized. This
hinders OS fingerprinters and keeps your machine from being used as a bounce for an
untraceable portscan. It’s a good idea to say Y here.
Randomized TCP source ports (CONFIG_GRKERNSEC_RANDSRC) [N/y/?] y
This Grsecurity option allows us to enable the randomized TCP source ports protection on the
system. If you say Y here, situations where a source port is generated on the fly for the TCP
protocol (ie. with connect() ) will be altered so that the source port is generated at random,
instead of a simple incrementing algorithm. It’s a good idea to say Y here.
Randomized RPC XIDs (CONFIG_GRKERNSEC_RANDRPC) [N/y/?] y
This Grsecurity option allows us to enable the randomized RPC XIDs protection on the
system. If you say Y here, the method of determining XIDs for RPC requests will be randomized,
instead of using linux's default behavior of simply incrementing the XID. This allows us to have a
more secure RPC connection on the system. It’s a good idea to say Y here.
Altered Ping IDs (CONFIG_GRKERNSEC_RANDPING) [N/y/?] y
This Grsecurity option allows us to enable the altered Ping IDs protection on the system. If you
say Y here, the way Linux handles echo replies will be changed so that the reply uses an ID equal
to the ID of the echo request. This will help in confusing OS detection. It’s a good idea to say Y
here.
Randomized TTL (CONFIG_GRKERNSEC_RANDTTL) [N/y/?] y
This Grsecurity option allows us to enable the randomized TTL protection on the system. If
you say Y here, your TTL (time to live) for packets will be set at random, with a base of the
sysctl ttl default, to further confuse OS detection. It’s a good idea to say Y here.
Socket restrictions (CONFIG_GRKERNSEC_SOCKET) [N/y/?] y
This Grsecurity option allows us to enable the socket restrictions protection on the system. If
you say Y here, you will be able to choose from three options related to socket protection on the
server. From these three available security options, you’ll have to choose the one that best fits
your requirements. Therefore, it’s a good idea to say Y here since we have to choose one of the
three available security options for our needs.
Deny any sockets to group (CONFIG_GRKERNSEC_SOCKET_ALL) [N/y/?] y
This Grsecurity option allows us to enable the socket restrictions protection on the system. If
you say Y here, you will be able to choose a GID of whose users will be unable to connect to
other hosts from your machine or run server applications from your machine. This is the security
option that we’ll choose. Say Y here.
GID to deny all sockets for: (CONFIG_GRKERNSEC_SOCKET_ALL_GID) [1004] Press
Enter
This Grsecurity option allows us to enable the GID to deny all sockets protection on the
system. Here we have to enter a GID number as the value, the default value is 1004 and we can
keep it by pressing the [Enter] key. It is important to note that this GID should be added to any
user for which the feature should be activated. See the next chapter of this book for more
information about the procedures to follow. At this time you only need to accept the default value.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-179-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
180
Deny client sockets to group (CONFIG_GRKERNSEC_SOCKET_CLIENT) [N/y/?] Press
Enter
This Grsecurity option allows us to enable the deny client sockets to group protection on the
system. If you say Y here, you will be able to choose a GID of whose users will be unable to
connect to other hosts from your machine, but will be able to run servers. We have already
chosen the above option that enables socket protection on both ways (users cannot connect to
other hosts or run server applications from our machine), therefore we don’t need to say Y here.
Say N here.
Deny server sockets to group (CONFIG_GRKERNSEC_SOCKET_SERVER) [N/y/?] Press
Enter
This Grsecurity option allows us to enable the deny server sockets to group protection on the
system. If you say Y here, you will be able to choose a GID of whose users will be unable to run
server applications from your machine. As for the above option, we already have chosen the first
option that enable socket protection on both way (users cannot connect to other hosts or run
server applications from our machine), therefore we don’t need to say Y here. Say N here.
*
* Sysctl support
*
Sysctl support (CONFIG_GRKERNSEC_SYSCTL) [N/y/?] Press Enter
This Grsecurity option allows us to enable Grsecurity sysctl support protection on the
system. If you say Y here, you will be able to change the options that Grsecurity runs with at
bootup, without having to recompile your kernel. You can echo values to files in
/proc/sys/kernel/grsecurity to enable (1) or disable (0) various features. Enabling this
option will reduce the effectiveness of the added security of the Grsecurity patch, therefore we
say N here.
*
* Miscellaneous Features
*
Seconds in between log messages(minimum) (CONFIG_GRKERNSEC_FLOODTIME) [30]
Press Enter
This Grsecurity option allows us to enable the seconds in between log messages protection
on the system. This option allows you to enforce the number of seconds between Grsecurity
log messages. The default should be suitable for most people. Just press the [Enter] key here
to accept the default value.
BSD-style coredumps (CONFIG_GRKERNSEC_COREDUMP) [N/y/?] y
This Grsecurity option allows us to enable the BSD-style coredumps protection on the
system. If you say Y here, Linux will use a style similar to BSD for coredumps, core.processname.
Not a security feature, just a useful one. Say Y here.
*** End of Linux kernel configuration.
*** Check the top-level Makefile for additional configuration.
*** Next, you must run 'make dep'.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-180-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
181
Modularized kernel configuration
Building kernel with modules (Modularized kernel) has some advantages. It allows easy
portability between different Linux systems, since you can choose and build different parts of the
kernel as a module and load that segment of code on demand. Below we show you the
configuration of Modularized kernel, which is to compile some needed codes and drivers as
a module into the kernel by answering the different questions using y, n or m. As for the previous
Monolithic kernel configuration, don’t forget to only compile code that you need and use.
A new kernel is very specific to your computer hardware, in the Modularized kernel
configuration part below; we assume the following hardware for our example. Of course you must
change them to fit your system components.
1 Pentium-III 667 MHz (i686) processor
1 Motherboard Asus P3V4X Pro 133Mhz EIDE
1 Hard Disk Ultra ATA/100 EIDE
1 Chipset Apollo Pro133A
1 CD-ROM ATAPI IDE
1 Floppy Disk
2 Ethernet Cards 3COM 3c597 PCI 10/100
1 Mouse PS/2
If you don’t want some options listed in the Modularized kernel configuration that I enable by
default, answer n (for no) instead of y (for yes) or m (for modularized if possible) to the related
questions. If you want some other options that I disable, then answer y or m instead of n.
rm -f include/asm
( cd include ; ln -sf asm-i386 asm)
/bin/sh scripts/Configure arch/i386/config.in
#
# Using defaults found in arch/i386/defconfig
#
*
* Code maturity level options
*
Prompt for development and/or incomplete code/drivers (CONFIG_EXPERIMENTAL) [N/y/?] Press
Enter
*
* Loadable module support
*
Enable loadable module support (CONFIG_MODULES) [Y/n/?] Press Enter
Set version information on all module symbols (CONFIG_MODVERSIONS) [Y/n/?] n
Kernel module loader (CONFIG_KMOD) [Y/n/?] Press Enter
*
* Processor type and features
*
Processor family (386, 486, 586/K5/5x86/6x86/6x86MX, Pentium-Classic, Pentium-MMX,
Pentium-Pro/Celeron/Pentium-II, Pentium-III/Celeron(Coppermine), Pentium-4, K6/K6-II/K6-
III, Athlon/Duron/K7, Elan, Crusoe, Winchip-C6, Winchip-2, Winchip-2A/Winchip-3,
CyrixIII/C3) [Pentium-III/Celeron(Coppermine)] Press Enter
Toshiba Laptop support (CONFIG_TOSHIBA) [N/y/m/?] Press Enter
Dell laptop support (CONFIG_I8K) [N/y/m/?] Press Enter
/dev/cpu/microcode - Intel IA32 CPU microcode support (CONFIG_MICROCODE) [N/y/m/?] m
/dev/cpu/*/msr - Model-specific register support (CONFIG_X86_MSR) [N/y/m/?] m
/dev/cpu/*/cpuid - CPU information support (CONFIG_X86_CPUID) [N/y/m/?] m
High Memory Support (off, 4GB, 64GB) [off] Press Enter
Math emulation (CONFIG_MATH_EMULATION) [N/y/?] Press Enter
MTRR (Memory Type Range Register) support (CONFIG_MTRR) [N/y/?] Press Enter
Symmetric multi-processing support (CONFIG_SMP) [Y/n/?] n
Local APIC support on uniprocessors (CONFIG_X86_UP_APIC) [N/y/?] (NEW) y
IO-APIC support on uniprocessors (CONFIG_X86_UP_IOAPIC) [N/y/?] (NEW) y
*
* General setup](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-181-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
182
*
Networking support (CONFIG_NET) [Y/n/?] Press Enter
PCI support (CONFIG_PCI) [Y/n/?] Press Enter
PCI access mode (BIOS, Direct, Any) [Any] Press Enter
PCI device name database (CONFIG_PCI_NAMES) [Y/n/?] n
EISA support (CONFIG_EISA) [N/y/?] Press Enter
MCA support (CONFIG_MCA) [N/y/?] Press Enter
Support for hot-pluggable devices (CONFIG_HOTPLUG) [Y/n/?] n
System V IPC (CONFIG_SYSVIPC) [Y/n/?] Press Enter
BSD Process Accounting (CONFIG_BSD_PROCESS_ACCT) [N/y/?] Press Enter
Sysctl support (CONFIG_SYSCTL) [Y/n/?] Press Enter
Kernel core (/proc/kcore) format (ELF, A.OUT) [ELF] Press Enter
Kernel support for a.out binaries (CONFIG_BINFMT_AOUT) [Y/m/n/?] m
Kernel support for ELF binaries (CONFIG_BINFMT_ELF) [Y/m/n/?] Press Enter
Kernel support for MISC binaries (CONFIG_BINFMT_MISC) [Y/m/n/?] m
Power Management support (CONFIG_PM) [Y/n/?] n
*
* Memory Technology Devices (MTD)
*
Memory Technology Device (MTD) support (CONFIG_MTD) [N/y/m/?] Press Enter
*
* Parallel port support
*
Parallel port support (CONFIG_PARPORT) [N/y/m/?] Press Enter
*
* Plug and Play configuration
*
Plug and Play support (CONFIG_PNP) [Y/m/n/?] n
*
* Block devices
*
Normal PC floppy disk support (CONFIG_BLK_DEV_FD) [Y/m/n/?] Press Enter
XT hard disk support (CONFIG_BLK_DEV_XD) [N/y/m/?] Press Enter
Compaq SMART2 support (CONFIG_BLK_CPQ_DA) [N/y/m/?] Press Enter
Compaq Smart Array 5xxx support (CONFIG_BLK_CPQ_CISS_DA) [N/y/m/?] Press Enter
Mylex DAC960/DAC1100 PCI RAID Controller support (CONFIG_BLK_DEV_DAC960) [N/y/m/?] Press
Enter
Loopback device support (CONFIG_BLK_DEV_LOOP) [N/y/m/?] Press Enter
Network block device support (CONFIG_BLK_DEV_NBD) [N/y/m/?] Press Enter
RAM disk support (CONFIG_BLK_DEV_RAM) [N/y/m/?] Press Enter
*
* Multi-device support (RAID and LVM)
*
Multiple devices driver support (RAID and LVM) (CONFIG_MD) [N/y/?] Press Enter
*
* Networking options
*
Packet socket (CONFIG_PACKET) [Y/m/n/?] Press Enter
Packet socket: mmapped IO (CONFIG_PACKET_MMAP) [N/y/?] y
Netlink device emulation (CONFIG_NETLINK_DEV) [N/y/m/?] (NEW) m
Network packet filtering (replaces ipchains) (CONFIG_NETFILTER) [N/y/?] y
Network packet filtering debugging (CONFIG_NETFILTER_DEBUG) [N/y/?] (NEW) y
Socket Filtering (CONFIG_FILTER) [N/y/?] Press Enter
Unix domain sockets (CONFIG_UNIX) [Y/m/n/?] Press Enter
TCP/IP networking (CONFIG_INET) [Y/n/?] Press Enter
IP: multicasting (CONFIG_IP_MULTICAST) [Y/n/?] n
IP: advanced router (CONFIG_IP_ADVANCED_ROUTER) [N/y/?] Press Enter
IP: kernel level autoconfiguration (CONFIG_IP_PNP) [N/y/?] Press Enter
IP: tunneling (CONFIG_NET_IPIP) [N/y/m/?] Press Enter
IP: GRE tunnels over IP (CONFIG_NET_IPGRE) [N/y/m/?] Press Enter
IP: TCP Explicit Congestion Notification support (CONFIG_INET_ECN) [N/y/?] Press Enter
IP: TCP syncookie support (disabled default) (CONFIG_SYN_COOKIES) [N/y/?] y
*
* IP: Netfilter Configuration
*
Connection tracking (required for masq/NAT) (CONFIG_IP_NF_CONNTRACK) [N/y/m/?] (NEW) m
FTP protocol support (CONFIG_IP_NF_FTP) [N/m/?] (NEW) m
IRC protocol support (CONFIG_IP_NF_IRC) [N/m/?] (NEW) m
IP tables support (required for filtering/masq/NAT) (CONFIG_IP_NF_IPTABLES) [N/y/m/?]
(NEW) m
limit match support (CONFIG_IP_NF_MATCH_LIMIT) [N/m/?] (NEW) m](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-182-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
183
MAC address match support (CONFIG_IP_NF_MATCH_MAC) [N/m/?] (NEW) m
netfilter MARK match support (CONFIG_IP_NF_MATCH_MARK) [N/m/?] (NEW) m
Multiple port match support (CONFIG_IP_NF_MATCH_MULTIPORT) [N/m/?] (NEW) m
TOS match support (CONFIG_IP_NF_MATCH_TOS) [N/m/?] (NEW) m
AH/ESP match support (CONFIG_IP_NF_MATCH_AH_ESP) [N/m/?] (NEW) m
LENGTH match support (CONFIG_IP_NF_MATCH_LENGTH) [N/m/?] (NEW) m
TTL match support (CONFIG_IP_NF_MATCH_TTL) [N/m/?] (NEW) m
tcpmss match support (CONFIG_IP_NF_MATCH_TCPMSS) [N/m/?] (NEW) m
Connection state match support (CONFIG_IP_NF_MATCH_STATE) [N/m/?] (NEW) m
Packet filtering (CONFIG_IP_NF_FILTER) [N/m/?] (NEW) m
REJECT target support (CONFIG_IP_NF_TARGET_REJECT) [N/m/?] (NEW) m
Full NAT (CONFIG_IP_NF_NAT) [N/m/?] (NEW) m
MASQUERADE target support (CONFIG_IP_NF_TARGET_MASQUERADE) [N/m/?] (NEW) m
REDIRECT target support (CONFIG_IP_NF_TARGET_REDIRECT) [N/m/?] (NEW) m
Packet mangling (CONFIG_IP_NF_MANGLE) [N/m/?] (NEW) m
TOS target support (CONFIG_IP_NF_TARGET_TOS) [N/m/?] (NEW) m
MARK target support (CONFIG_IP_NF_TARGET_MARK) [N/m/?] (NEW) m
LOG target support (CONFIG_IP_NF_TARGET_LOG) [N/m/?] (NEW) m
TCPMSS target support (CONFIG_IP_NF_TARGET_TCPMSS) [N/m/?] (NEW) m
ipchains (2.2-style) support (CONFIG_IP_NF_COMPAT_IPCHAINS) [N/y/m/?] (NEW) Press Enter
ipfwadm (2.0-style) support (CONFIG_IP_NF_COMPAT_IPFWADM) [N/y/m/?] (NEW) Press Enter
*
*
*
The IPX protocol (CONFIG_IPX) [N/y/m/?] Press Enter
Appletalk protocol support (CONFIG_ATALK) [N/y/m/?] Press Enter
DECnet Support (CONFIG_DECNET) [N/y/m/?] Press Enter
802.1d Ethernet Bridging (CONFIG_BRIDGE) [N/y/m/?] Press Enter
*
* QoS and/or fair queueing
*
QoS and/or fair queueing (CONFIG_NET_SCHED) [N/y/?] Press Enter
*
* Telephony Support
*
Linux telephony support (CONFIG_PHONE) [N/y/m/?] Press Enter
*
* ATA/IDE/MFM/RLL support
*
ATA/IDE/MFM/RLL support (CONFIG_IDE) [Y/m/n/?] Press Enter
*
* IDE, ATA and ATAPI Block devices
*
Enhanced IDE/MFM/RLL disk/cdrom/tape/floppy support (CONFIG_BLK_DEV_IDE) [Y/m/n/?] Press
Enter
*
* Please see Documentation/ide.txt for help/info on IDE drives
*
Use old disk-only driver on primary interface (CONFIG_BLK_DEV_HD_IDE) [N/y/?] Press
Enter
Include IDE/ATA-2 DISK support (CONFIG_BLK_DEV_IDEDISK) [Y/m/n/?] Press Enter
Use multi-mode by default (CONFIG_IDEDISK_MULTI_MODE) [Y/n/?] n
Include IDE/ATAPI CDROM support (CONFIG_BLK_DEV_IDECD) [Y/m/n/?] Press Enter
Include IDE/ATAPI TAPE support (CONFIG_BLK_DEV_IDETAPE) [N/y/m/?] Press Enter
Include IDE/ATAPI FLOPPY support (CONFIG_BLK_DEV_IDEFLOPPY) [N/y/m/?] Press Enter
SCSI emulation support (CONFIG_BLK_DEV_IDESCSI) [N/y/m/?] Press Enter
*
* IDE chipset support/bugfixes
*
CMD640 chipset bugfix/support (CONFIG_BLK_DEV_CMD640) [Y/n/?] n
RZ1000 chipset bugfix/support (CONFIG_BLK_DEV_RZ1000) [Y/n/?] n
Generic PCI IDE chipset support (CONFIG_BLK_DEV_IDEPCI) [Y/n/?] Press Enter
Sharing PCI IDE interrupts support (CONFIG_IDEPCI_SHARE_IRQ) [Y/n/?] Press Enter
Generic PCI bus-master DMA support (CONFIG_BLK_DEV_IDEDMA_PCI) [Y/n/?] Press Enter
Boot off-board chipsets first support (CONFIG_BLK_DEV_OFFBOARD) [N/y/?] Press Enter
Use PCI DMA by default when available (CONFIG_IDEDMA_PCI_AUTO) [Y/n/?] Press Enter
AEC62XX chipset support (CONFIG_BLK_DEV_AEC62XX) [N/y/?] Press Enter
ALI M15x3 chipset support (CONFIG_BLK_DEV_ALI15X3) [N/y/?] Press Enter
AMD Viper support (CONFIG_BLK_DEV_AMD74XX) [N/y/?] Press Enter
CMD64X chipset support (CONFIG_BLK_DEV_CMD64X) [N/y/?] Press Enter
CY82C693 chipset support (CONFIG_BLK_DEV_CY82C693) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-183-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
184
Cyrix CS5530 MediaGX chipset support (CONFIG_BLK_DEV_CS5530) [N/y/?] Press Enter
HPT34X chipset support (CONFIG_BLK_DEV_HPT34X) [N/y/?] Press Enter
HPT366 chipset support (CONFIG_BLK_DEV_HPT366) [N/y/?] Press Enter
Intel PIIXn chipsets support (CONFIG_BLK_DEV_PIIX) [Y/n/?] Press Enter
PIIXn Tuning support (CONFIG_PIIX_TUNING) [Y/n/?] Press Enter
NS87415 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_NS87415) [N/y/?] Press Enter
PROMISE PDC202{46|62|65|67|68} support (CONFIG_BLK_DEV_PDC202XX) [N/y/?] Press Enter
ServerWorks OSB4/CSB5 chipsets support (CONFIG_BLK_DEV_SVWKS) [N/y/?] Press Enter
SiS5513 chipset support (CONFIG_BLK_DEV_SIS5513) [N/y/?] Press Enter
SLC90E66 chipset support (CONFIG_BLK_DEV_SLC90E66) [N/y/?] Press Enter
Tekram TRM290 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_TRM290) [N/y/?] Press
Enter
VIA82CXXX chipset support (CONFIG_BLK_DEV_VIA82CXXX) [N/y/?] y
Other IDE chipset support (CONFIG_IDE_CHIPSETS) [N/y/?] Press Enter
IGNORE word93 Validation BITS (CONFIG_IDEDMA_IVB) [N/y/?] Press Enter
*
* SCSI support
*
SCSI support (CONFIG_SCSI) [Y/m/n/?] n
*
* Fusion MPT device support
*
*
* I2O device support
*
I2O support (CONFIG_I2O) [N/y/m/?] Press Enter
*
* Network device support
*
Network device support (CONFIG_NETDEVICES) [Y/n/?] Press Enter
*
* ARCnet devices
*
ARCnet support (CONFIG_ARCNET) [N/y/m/?] Press Enter
Dummy net driver support (CONFIG_DUMMY) [M/n/y/?] Press Enter
Bonding driver support (CONFIG_BONDING) [N/y/m/?] Press Enter
EQL (serial line load balancing) support (CONFIG_EQUALIZER) [N/y/m/?] Press Enter
Universal TUN/TAP device driver support (CONFIG_TUN) [N/y/m/?] Press Enter
*
* Ethernet (10 or 100Mbit)
*
Ethernet (10 or 100Mbit) (CONFIG_NET_ETHERNET) [Y/n/?] Press Enter
Sun Happy Meal 10/100baseT support (CONFIG_HAPPYMEAL) [N/y/m/?] Press Enter
Sun GEM support (CONFIG_SUNGEM) [N/y/m/?] Press Enter
3COM cards (CONFIG_NET_VENDOR_3COM) [N/y/?] y
3c501 "EtherLink" support (CONFIG_EL1) [N/y/m/?] (NEW) Press Enter
3c503 "EtherLink II" support (CONFIG_EL2) [N/y/m/?] (NEW) Press Enter
3c505 "EtherLink Plus" support (CONFIG_ELPLUS) [N/y/m/?] (NEW) Press Enter
3c509/3c529 (MCA)/3c579 "EtherLink III" support (CONFIG_EL3) [N/y/m/?] (NEW) Press
Enter
3c515 ISA "Fast EtherLink" (CONFIG_3C515) [N/y/m/?] (NEW) Press Enter
3c590/3c900 series (592/595/597) "Vortex/Boomerang" support (CONFIG_VORTEX) [N/y/m/?]
(NEW) y
AMD LANCE and PCnet (AT1500 and NE2100) support (CONFIG_LANCE) [N/y/m/?] Press Enter
Western Digital/SMC cards (CONFIG_NET_VENDOR_SMC) [N/y/?] Press Enter
Racal-Interlan (Micom) NI cards (CONFIG_NET_VENDOR_RACAL) [N/y/?] Press Enter
DEPCA, DE10x, DE200, DE201, DE202, DE422 support (CONFIG_DEPCA) [N/y/m/?] Press Enter
HP 10/100VG PCLAN (ISA, EISA, PCI) support (CONFIG_HP100) [N/y/m/?] Press Enter
Other ISA cards (CONFIG_NET_ISA) [N/y/?] Press Enter
EISA, VLB, PCI and on board controllers (CONFIG_NET_PCI) [Y/n/?] n
Pocket and portable adapters (CONFIG_NET_POCKET) [N/y/?] Press Enter
*
* Ethernet (1000 Mbit)
*
Alteon AceNIC/3Com 3C985/NetGear GA620 Gigabit support (CONFIG_ACENIC) [N/y/m/?] Press
Enter
D-Link DL2000-based Gigabit Ethernet support (CONFIG_DL2K) [N/y/m/?] Press Enter
National Semiconduct DP83820 support (CONFIG_NS83820) [N/y/m/?] Press Enter
Packet Engines Hamachi GNIC-II support (CONFIG_HAMACHI) [N/y/m/?] Press Enter
SysKonnect SK-98xx support (CONFIG_SK98LIN) [N/y/m/?] Press Enter
FDDI driver support (CONFIG_FDDI) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-184-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
185
PPP (point-to-point protocol) support (CONFIG_PPP) [N/y/m/?] Press Enter
SLIP (serial line) support (CONFIG_SLIP) [N/y/m/?] Press Enter
*
* Wireless LAN (non-hamradio)
*
Wireless LAN (non-hamradio) (CONFIG_NET_RADIO) [N/y/?] Press Enter
*
* Token Ring devices
*
Token Ring driver support (CONFIG_TR) [N/y/?] Press Enter
Fibre Channel driver support (CONFIG_NET_FC) [N/y/?] Press Enter
*
* Wan interfaces
*
Wan interfaces support (CONFIG_WAN) [N/y/?] Press Enter
*
* Amateur Radio support
*
Amateur Radio support (CONFIG_HAMRADIO) [N/y/?] Press Enter
*
* IrDA (infrared) support
*
IrDA subsystem support (CONFIG_IRDA) [N/y/m/?] Press Enter
*
* ISDN subsystem
*
ISDN support (CONFIG_ISDN) [N/y/m/?] Press Enter
*
* Old CD-ROM drivers (not SCSI, not IDE)
*
Support non-SCSI/IDE/ATAPI CDROM drives (CONFIG_CD_NO_IDESCSI) [N/y/?] Press Enter
*
* Input core support
*
Input core support (CONFIG_INPUT) [N/y/m/?] Press Enter
*
* Character devices
*
Virtual terminal (CONFIG_VT) [Y/n/?] Press Enter
Support for console on virtual terminal (CONFIG_VT_CONSOLE) [Y/n/?] Press Enter
Standard/generic (8250/16550 and compatible UARTs) serial support (CONFIG_SERIAL)
[Y/m/n/?] Press Enter
Support for console on serial port (CONFIG_SERIAL_CONSOLE) [N/y/?] Press Enter
Extended dumb serial driver options (CONFIG_SERIAL_EXTENDED) [N/y/?] Press Enter
Non-standard serial port support (CONFIG_SERIAL_NONSTANDARD) [N/y/?] Press Enter
Unix98 PTY support (CONFIG_UNIX98_PTYS) [Y/n/?] Press Enter
Maximum number of Unix98 PTYs in use (0-2048) (CONFIG_UNIX98_PTY_COUNT) [256] 128
*
* I2C support
*
I2C support (CONFIG_I2C) [N/y/m/?] Press Enter
*
* Mice
*
Bus Mouse Support (CONFIG_BUSMOUSE) [N/y/m/?] Press Enter
Mouse Support (not serial and bus mice) (CONFIG_MOUSE) [Y/m/n/?] n
*
* Joysticks
*
*
* Input core support is needed for gameports
*
*
* Input core support is needed for joysticks
*
QIC-02 tape support (CONFIG_QIC02_TAPE) [N/y/m/?] Press Enter
*
* Watchdog Cards
*
Watchdog Timer Support (CONFIG_WATCHDOG) [N/y/?] Press Enter
Intel i8x0 Random Number Generator support (CONFIG_INTEL_RNG) [N/y/m/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-185-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
186
/dev/nvram support (CONFIG_NVRAM) [N/y/m/?] Press Enter
Enhanced Real Time Clock Support (CONFIG_RTC) [N/y/m/?] Press Enter
Double Talk PC internal speech card support (CONFIG_DTLK) [N/y/m/?] Press Enter
Siemens R3964 line discipline (CONFIG_R3964) [N/y/m/?] Press Enter
Applicom intelligent fieldbus card support (CONFIG_APPLICOM) [N/y/m/?] Press Enter
*
* Ftape, the floppy tape device driver
*
Ftape (QIC-80/Travan) support (CONFIG_FTAPE) [N/y/m/?] Press Enter
/dev/agpgart (AGP Support) (CONFIG_AGP) [Y/m/n/?] n
Direct Rendering Manager (XFree86 DRI support) (CONFIG_DRM) [Y/n/?] n
ACP Modem (Mwave) support (CONFIG_MWAVE) [N/y/m/?] Press Enter
*
* Multimedia devices
*
Video For Linux (CONFIG_VIDEO_DEV) [N/y/m/?] Press Enter
*
* File systems
*
Quota support (CONFIG_QUOTA) [N/y/?] y
Kernel automounter support (CONFIG_AUTOFS_FS) [N/y/m/?] Press Enter
Kernel automounter version 4 support (also supports v3) (CONFIG_AUTOFS4_FS) [Y/m/n/?] n
Reiserfs support (CONFIG_REISERFS_FS) [N/y/m/?] Press Enter
Ext3 journalling file system support (EXPERIMENTAL) (CONFIG_EXT3_FS) [N/y/m/?] y
JBD (ext3) debugging support (CONFIG_JBD_DEBUG) [N/y/?] y
DOS FAT fs support (CONFIG_FAT_FS) [N/y/m/?] m
MSDOS fs support (CONFIG_MSDOS_FS) [N/y/m/?] m
VFAT (Windows-95) fs support (CONFIG_VFAT_FS) [N/y/m/?] m
Compressed ROM file system support (CONFIG_CRAMFS) [N/y/m/?] Press Enter
Virtual memory file system support (former shm fs) (CONFIG_TMPFS) [Y/n/?] Press Enter
Simple RAM-based file system support (CONFIG_RAMFS) [N/y/m/?] Press Enter
ISO 9660 CDROM file system support (CONFIG_ISO9660_FS) [Y/m/n/?] m
Microsoft Joliet CDROM extensions (CONFIG_JOLIET) [N/y/?] y
Transparent decompression extension (CONFIG_ZISOFS) [N/y/?] Press Enter
Minix fs support (CONFIG_MINIX_FS) [N/y/m/?] Press Enter
FreeVxFS file system support (VERITAS VxFS(TM) compatible) (CONFIG_VXFS_FS) [N/y/m/?]
Press Enter
NTFS file system support (read only) (CONFIG_NTFS_FS) [N/y/m/?] Press Enter
OS/2 HPFS file system support (CONFIG_HPFS_FS) [N/y/m/?] Press Enter
/proc file system support (CONFIG_PROC_FS) [Y/n/?] Press Enter
/dev/pts file system for Unix98 PTYs (CONFIG_DEVPTS_FS) [Y/n/?] Press Enter
ROM file system support (CONFIG_ROMFS_FS) [N/y/m/?] Press Enter
Second extended fs support (CONFIG_EXT2_FS) [Y/m/n/?] Press Enter
System V/Xenix/V7/Coherent file system support (CONFIG_SYSV_FS) [N/y/m/?] Press Enter
UDF file system support (read only) (CONFIG_UDF_FS) [N/y/m/?] Press Enter
UFS file system support (read only) (CONFIG_UFS_FS) [N/y/m/?] Press Enter
*
* Network File Systems
*
Coda file system support (advanced network fs) (CONFIG_CODA_FS) [N/y/m/?] Press Enter
NFS file system support (CONFIG_NFS_FS) [Y/m/n/?] n
NFS server support (CONFIG_NFSD) [Y/m/n/?] n
SMB file system support (to mount Windows shares etc.) (CONFIG_SMB_FS) [N/y/m/?] Press
Enter
NCP file system support (to mount NetWare volumes) (CONFIG_NCP_FS) [N/y/m/?] Press Enter
*
* Partition Types
*
Advanced partition selection (CONFIG_PARTITION_ADVANCED) [N/y/?] Press Enter
*
* Native Language Support
*
Default NLS Option (CONFIG_NLS_DEFAULT) [iso8859-1] (NEW) Press Enter
Codepage 437 (United States, Canada) (CONFIG_NLS_CODEPAGE_437) [N/y/m/?] (NEW) Press
Enter
Codepage 737 (Greek) (CONFIG_NLS_CODEPAGE_737) [N/y/m/?] (NEW) Press Enter
Codepage 775 (Baltic Rim) (CONFIG_NLS_CODEPAGE_775) [N/y/m/?] (NEW) Press Enter
Codepage 850 (Europe) (CONFIG_NLS_CODEPAGE_850) [N/y/m/?] (NEW) Press Enter
Codepage 852 (Central/Eastern Europe) (CONFIG_NLS_CODEPAGE_852) [N/y/m/?] (NEW) Press
Enter
Codepage 855 (Cyrillic) (CONFIG_NLS_CODEPAGE_855) [N/y/m/?] (NEW) Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-186-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
187
Codepage 857 (Turkish) (CONFIG_NLS_CODEPAGE_857) [N/y/m/?] (NEW) Press Enter
Codepage 860 (Portuguese) (CONFIG_NLS_CODEPAGE_860) [N/y/m/?] (NEW) Press Enter
Codepage 861 (Icelandic) (CONFIG_NLS_CODEPAGE_861) [N/y/m/?] (NEW) Press Enter
Codepage 862 (Hebrew) (CONFIG_NLS_CODEPAGE_862) [N/y/m/?] (NEW) Press Enter
Codepage 863 (Canadian French) (CONFIG_NLS_CODEPAGE_863) [N/y/m/?] (NEW) Press Enter
Codepage 864 (Arabic) (CONFIG_NLS_CODEPAGE_864) [N/y/m/?] (NEW) Press Enter
Codepage 865 (Norwegian, Danish) (CONFIG_NLS_CODEPAGE_865) [N/y/m/?] (NEW) Press Enter
Codepage 866 (Cyrillic/Russian) (CONFIG_NLS_CODEPAGE_866) [N/y/m/?] (NEW) Press Enter
Codepage 869 (Greek) (CONFIG_NLS_CODEPAGE_869) [N/y/m/?] (NEW) Press Enter
Simplified Chinese charset (CP936, GB2312) (CONFIG_NLS_CODEPAGE_936) [N/y/m/?] (NEW)
Press Enter
Traditional Chinese charset (Big5) (CONFIG_NLS_CODEPAGE_950) [N/y/m/?] (NEW) Press Enter
Japanese charsets (Shift-JIS, EUC-JP) (CONFIG_NLS_CODEPAGE_932) [N/y/m/?] (NEW) Press
Enter
Korean charset (CP949, EUC-KR) (CONFIG_NLS_CODEPAGE_949) [N/y/m/?] (NEW) Press Enter
Thai charset (CP874, TIS-620) (CONFIG_NLS_CODEPAGE_874) [N/y/m/?] (NEW) Press Enter
Hebrew charsets (ISO-8859-8, CP1255) (CONFIG_NLS_ISO8859_8) [N/y/m/?] (NEW) Press Enter
Windows CP1250 (Slavic/Central European Languages) (CONFIG_NLS_CODEPAGE_1250) [N/y/m/?]
(NEW) Press Enter
Windows CP1251 (Bulgarian, Belarusian) (CONFIG_NLS_CODEPAGE_1251) [N/y/m/?] (NEW) Press
Enter
NLS ISO 8859-1 (Latin 1; Western European Languages) (CONFIG_NLS_ISO8859_1) [N/y/m/?]
(NEW) Press Enter
NLS ISO 8859-2 (Latin 2; Slavic/Central European Languages) (CONFIG_NLS_ISO8859_2)
[N/y/m/?] (NEW) Press Enter
NLS ISO 8859-3 (Latin 3; Esperanto, Galician, Maltese, Turkish) (CONFIG_NLS_ISO8859_3)
[N/y/m/?] (NEW) Press Enter
NLS ISO 8859-4 (Latin 4; old Baltic charset) (CONFIG_NLS_ISO8859_4) [N/y/m/?] (NEW)
Press Enter
NLS ISO 8859-5 (Cyrillic) (CONFIG_NLS_ISO8859_5) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-6 (Arabic) (CONFIG_NLS_ISO8859_6) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-7 (Modern Greek) (CONFIG_NLS_ISO8859_7) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-9 (Latin 5; Turkish) (CONFIG_NLS_ISO8859_9) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-13 (Latin 7; Baltic) (CONFIG_NLS_ISO8859_13) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-14 (Latin 8; Celtic) (CONFIG_NLS_ISO8859_14) [N/y/m/?] (NEW) Press Enter
NLS ISO 8859-15 (Latin 9; Western European Languages with Euro) (CONFIG_NLS_ISO8859_15)
[N/y/m/?] (NEW) Press Enter
NLS KOI8-R (Russian) (CONFIG_NLS_KOI8_R) [N/y/m/?] (NEW) Press Enter
NLS KOI8-U/RU (Ukrainian, Belarusian) (CONFIG_NLS_KOI8_U) [N/y/m/?] (NEW) Press Enter
NLS UTF8 (CONFIG_NLS_UTF8) [N/y/m/?] (NEW) Press Enter
*
* Console drivers
*
VGA text console (CONFIG_VGA_CONSOLE) [Y/n/?] Press Enter
Video mode selection support (CONFIG_VIDEO_SELECT) [N/y/?] Press Enter
*
* Sound
*
Sound card support (CONFIG_SOUND) [Y/m/n/?] n
*
* USB support
*
Support for USB (CONFIG_USB) [Y/m/n/?] n
*
* USB Controllers
*
*
* USB Device Class drivers
*
*
* SCSI support is needed for USB Storage
*
*
* USB Human Interface Devices (HID)
*
*
* Input core support is needed for USB HID
*
*
* USB Imaging devices
*](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-187-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
188
*
* USB Multimedia devices
*
*
* Video4Linux support is needed for USB Multimedia device support
*
*
* USB Network adaptors
*
*
* USB port drivers
*
*
* USB Serial Converter support
*
*
* USB Miscellaneous drivers
*
*
* Kernel hacking
*
Kernel debugging (CONFIG_DEBUG_KERNEL) [N/y/?] Press Enter
*
* Grsecurity
*
Grsecurity (CONFIG_GRKERNSEC) [N/y/?] y
Security level (Low, Medium, High, Customized) [Customized] Press Enter
*
* Buffer Overflow Protection
*
Openwall non-executable stack (CONFIG_GRKERNSEC_STACK) [N/y/?] y
Gcc trampoline support (CONFIG_GRKERNSEC_STACK_GCC) [N/y/?] Press Enter
Read-only kernel memory (CONFIG_GRKERNSEC_KMEM) [N/y/?] y
*
* Access Control Lists
*
Grsecurity ACL system (CONFIG_GRKERNSEC_ACL) [N/y/?] y
ACL Debugging Messages (CONFIG_GR_DEBUG) [N/y/?] y
Extra ACL Debugging Messages (CONFIG_GR_SUPERDEBUG) [N/y/?] Press Enter
Denied capability logging (CONFIG_GRKERNSEC_ACL_CAPLOG) [N/y/?] y
Path to gradm (CONFIG_GRADM_PATH) [/sbin/gradm] Press Enter
Maximum tries before password lockout (CONFIG_GR_MAXTRIES) [3] 2
Time to wait after max password tries, in seconds (CONFIG_GR_TIMEOUT) [30] Press Enter
*
* Filesystem Protections
*
Proc restrictions (CONFIG_GRKERNSEC_PROC) [N/y/?] y
Restrict to user only (CONFIG_GRKERNSEC_PROC_USER) [N/y/?] y
Additional restrictions (CONFIG_GRKERNSEC_PROC_ADD) [N/y/?] y
Linking restrictions (CONFIG_GRKERNSEC_LINK) [N/y/?] y
FIFO restrictions (CONFIG_GRKERNSEC_FIFO) [N/y/?] y
Secure file descriptors (CONFIG_GRKERNSEC_FD) [N/y/?] y
Chroot jail restrictions (CONFIG_GRKERNSEC_CHROOT) [N/y/?] y
Restricted signals (CONFIG_GRKERNSEC_CHROOT_SIG) [N/y/?] y
Deny mounts (CONFIG_GRKERNSEC_CHROOT_MOUNT) [N/y/?] y
Deny double-chroots (CONFIG_GRKERNSEC_CHROOT_DOUBLE) [N/y/?] y
Enforce chdir("/") on all chroots (CONFIG_GRKERNSEC_CHROOT_CHDIR) [N/y/?] y
Deny (f)chmod +s (CONFIG_GRKERNSEC_CHROOT_CHMOD) [N/y/?] y
Deny mknod (CONFIG_GRKERNSEC_CHROOT_MKNOD) [N/y/?] y
Deny ptraces (CONFIG_GRKERNSEC_CHROOT_PTRACE) [N/y/?] y
Restrict priority changes (CONFIG_GRKERNSEC_CHROOT_NICE) [N/y/?] y
Capability restrictions within chroot (CONFIG_GRKERNSEC_CHROOT_CAPS) [N/y/?] Press Enter
Secure keymap loading (CONFIG_GRKERNSEC_KBMAP) [N/y/?] y
*
* Kernel Auditing
*
Single group for auditing (CONFIG_GRKERNSEC_AUDIT_GROUP) [N/y/?] Press Enter
Exec logging (CONFIG_GRKERNSEC_EXECLOG) [N/y/?] Press Enter
Log execs within chroot (CONFIG_GRKERNSEC_CHROOT_EXECLOG) [N/y/?] y
Chdir logging (CONFIG_GRKERNSEC_AUDIT_CHDIR) [N/y/?] Press Enter](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-188-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
189
(Un)Mount logging (CONFIG_GRKERNSEC_AUDIT_MOUNT) [N/y/?] Press Enter
IPC logging (CONFIG_GRKERNSEC_AUDIT_IPC) [N/y/?] y
Ptrace logging (CONFIG_GRKERNSEC_AUDIT_PTRACE) [N/y/?] Press Enter
Signal logging (CONFIG_GRKERNSEC_SIGNAL) [N/y/?] y
Fork failure logging (CONFIG_GRKERNSEC_FORKFAIL) [N/y/?] y
Set*id logging (CONFIG_GRKERNSEC_SUID) [N/y/?] Press Enter
Log set*ids to root (CONFIG_GRKERNSEC_SUID_ROOT) [N/y/?] y
Time change logging (CONFIG_GRKERNSEC_TIME) [N/y/?] y
*
* Executable Protections
*
Exec process limiting (CONFIG_GRKERNSEC_EXECVE) [N/y/?] y
Dmesg(8) restriction (CONFIG_GRKERNSEC_DMESG) [N/y/?] y
Randomized PIDs (CONFIG_GRKERNSEC_RANDPID) [N/y/?] y
Altered default IPC permissions (CONFIG_GRKERNSEC_IPC) [N/y/?] Press Enter
Limit uid/gid changes to root (CONFIG_GRKERNSEC_TTYROOT) [N/y/?] y
Deny physical consoles (tty) (CONFIG_GRKERNSEC_TTYROOT_PHYS) [N/y/?] Press Enter
Deny serial consoles (ttyS) (CONFIG_GRKERNSEC_TTYROOT_SERIAL) [N/y/?] y
Deny pseudo consoles (pty) (CONFIG_GRKERNSEC_TTYROOT_PSEUDO) [N/y/?] Press Enter
Fork-bomb protection (CONFIG_GRKERNSEC_FORKBOMB) [N/y/?] y
GID for restricted users (CONFIG_GRKERNSEC_FORKBOMB_GID) [1006] Press Enter
Forks allowed per second (CONFIG_GRKERNSEC_FORKBOMB_SEC) [40] Press Enter
Maximum processes allowed (CONFIG_GRKERNSEC_FORKBOMB_MAX) [20] 35
Trusted path execution (CONFIG_GRKERNSEC_TPE) [N/y/?] y
Glibc protection (CONFIG_GRKERNSEC_TPE_GLIBC) [N/y/?] y
Partially restrict non-root users (CONFIG_GRKERNSEC_TPE_ALL) [N/y/?] y
GID for untrusted users: (CONFIG_GRKERNSEC_TPE_GID) [1005] Press Enter
Restricted ptrace (CONFIG_GRKERNSEC_PTRACE) [N/y/?] y
Allow ptrace for group (CONFIG_GRKERNSEC_PTRACE_GROUP) [N/y/?] Press Enter
*
* Network Protections
*
Randomized IP IDs (CONFIG_GRKERNSEC_RANDID) [N/y/?] y
Randomized TCP source ports (CONFIG_GRKERNSEC_RANDSRC) [N/y/?] y
Randomized RPC XIDs (CONFIG_GRKERNSEC_RANDRPC) [N/y/?] y
Altered Ping IDs (CONFIG_GRKERNSEC_RANDPING) [N/y/?] y
Randomized TTL (CONFIG_GRKERNSEC_RANDTTL) [N/y/?] y
Socket restrictions (CONFIG_GRKERNSEC_SOCKET) [N/y/?] y
Deny any sockets to group (CONFIG_GRKERNSEC_SOCKET_ALL) [N/y/?] y
GID to deny all sockets for: (CONFIG_GRKERNSEC_SOCKET_ALL_GID) [1004] Press Enter
Deny client sockets to group (CONFIG_GRKERNSEC_SOCKET_CLIENT) [N/y/?] Press Enter
Deny server sockets to group (CONFIG_GRKERNSEC_SOCKET_SERVER) [N/y/?] Press Enter
*
* Sysctl support
*
Sysctl support (CONFIG_GRKERNSEC_SYSCTL) [N/y/?] Press Enter
*
* Miscellaneous Features
*
Seconds in between log messages(minimum) (CONFIG_GRKERNSEC_FLOODTIME) [30] Press Enter
BSD-style coredumps (CONFIG_GRKERNSEC_COREDUMP) [N/y/?] y
*** End of Linux kernel configuration.
*** Check the top-level Makefile for additional configuration.
*** Next, you must run 'make dep'.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-189-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
190
Compiling the Kernel
This section applies to both Monolithic and Modularized kernels. Once the kernel configuration
has been completed, return to the /usr/src/linux directory (if you are not already in it) and
compile the new kernel. You do so by using the following command:
• To compile the Kernel, use the following command:
[root@deep linux]# make dep; make clean; make bzImage
This line contains three commands in one. The first one, make dep, actually takes your
configuration and builds the corresponding dependency tree. This process determines what gets
compiled and what doesn’t. The next step, make clean, erases all previous traces of a
compilation so as to avoid any mistakes in which the wrong version of a feature gets tied into the
kernel. Finally, make bzImage does the full compilation of the kernel.
After the process is complete, the kernel is compressed and ready to be installed on your system.
Before we can install the new kernel, we must know if we need to compile the corresponding
modules. This is required ONLY if you said yes to “Enable loadable module support
(CONFIG_MODULES)” and have compiled some options in the kernel configuration above as a
module (See Modularized kernel configuration). In this case, you must execute the following
commands:
• To compile the corresponding modules for your kernel, use the following commands:
[root@deep linux]# make modules
[root@deep linux]# make modules_install
WARNING: The make modules and make modules_install commands are required ONLY if
you say yes to “Enable loadable module support (CONFIG_MODULES)” in your kernel
configurations (See Modularized kernel configuration) because you want to build a
modularized kernel.
Installing the Kernel
This section applies to both the Monolithic and Modularized kernel. Ok, the kernel has been
configured, compiled and is now ready to be installed your system. Below are the steps required
to install all the necessary kernel components in your system.
Step 1
Copy the file /usr/src/linux/arch/i386/boot/bzImage from the kernel source tree to the
/boot directory, and give it an appropriate new name.
• To copy the bzImage file to the /boot directory, use the following commands:
[root@deep /]# cd /usr/src/linux/ (if you are not already in it)
[root@deep linux]# cp arch/i386/boot/bzImage /boot/vmlinuz-2.4.18](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-190-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
191
Step 2
A new System.map file is generated when you compile a kernel, and is a list of all the addresses
in that kernel and their corresponding symbols. Every time that you create a new kernel, such a
file System.map is created and saved in /usr/src/linux. It's a text file, which is read by a
few programs to do address <-> symbol translation, and which you need if you ever get an Oops.
Certain commands, like klog, ps, and lsof, use the System.map file to get the name of kernel
symbols. Without it some commands like lsof will complain that they can't find a System.map
file to match the currently booted kernel.
Copy the file /usr/src/linux/System.map from the kernel source tree to the /boot
directory, and give it an appropriate new name.
• To copy the System.map file to the /boot directory, use the following commands:
[root@deep /]# cd /usr/src/linux/ (if you are not already in it)
[root@deep linux]# cp System.map /boot/System.map-2.4.18
Step 3
Move into the /boot directory and rebuild the links vmlinuz and System.map.
• To rebuild the vmlinuz and System.map files, use the following commands:
[root@deep linux]# cd /boot/
[root@deep /boot]# ln -fs vmlinuz-2.4.18 vmlinuz
[root@deep /boot]# ln -fs System.map-2.4.18 System.map
We must rebuild the links of vmlinuz and System.map to point them to the new installed kernel
version. Without the new links LILO or GRUB program will look, by default, for the old version of
your Linux kernel.
Step 4
Remove obsolete and unnecessary files under the /boot directory to increase disk space:
• To remove obsolete and unnecessary files under the /boot directory, use commands:
[root@deep /]# cd /boot/ (if you are not already in it)
[root@deep /boot]# rm -f module-info
[root@deep /boot]# rm -f initrd-2.4.x.img
The module-info is a link, which points to the old modules directory of your original kernel.
Since we have installed a brand new kernel, we don’t need to keep this broken link.
The initrd-2.4.x.img is a file that contains an initial RAM disk image that serves as a
system before the disk is available. This file is only available and is installed by the Linux initial
setup installation if your system has a SCSI adapter present and only if your system has a SCSI
adapter. If we use and have a SCSI system, the required driver now will have been incorporated
into our new Linux kernel since we have build it by answering Y to the question related to our
SCSI model during the configuration of the kernel, so we can safely remove this file (initrd-
2.4.x.img).](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-191-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
192
Step 5
Create a new Linux kernel directory that will handle all the header files related to Linux kernel for
future compilation of other programs on your system.
Recall, we had created two symlinks under the /usr/include directory that pointed to the Linux
kernel header files to be able to compile it without receiving errors and also be able to compile
future programs. The /usr/include directory is where all the header files for your Linux system
are kept for reference and dependencies when you compile and install new programs.
The asm, and linux links are used when programs need to know some functions which are
compile-time specific to the kernel installed on your system. Programs call other headers as well
in the /usr/include directory when they must know specific information, dependencies, etc of
your system.
• To create a new Linux kernel directory to handle all header files, use the commands:
[root@deep /]# mkdir -p /usr/src/linux-2.4.18/include
[root@deep /]# cd /usr/src/linux/
[root@deep linux]# cp -r include/asm-i386 ../linux-2.4.18/include/
[root@deep linux]# cp -r include/linux ../linux-2.4.18/include/
[root@deep linux]# cd ../
[root@deep src]# rm -rf /usr/src/linux
[root@deep src]# cd /usr/src/ (to be sure that we are into the src directory)
[root@deep src]# ln -s /usr/src/linux-2.4.18 linux
First we create a new directory named “linux-2.4.18” based on the version of the kernel we
have installed for easy interpretation, then we copy directories asm-i386, and linux from
/usr/src/linux/include to our new location /usr/src/linux-2.4.18/include. After
we remove the entire source directory where we compiled the new kernel, we create a new
symbolic link named “linux” under /usr/src that points to our new /usr/src/linux-
2.4.18 directory. With these steps, future compiled programs will know where to look for
headers related to the kernel on your server.
NOTE: This step will allow us to gain space on our hard drive and will reduce the security risks.
The Linux kernel source directory handles a lot files and is about 94M in size when
uncompressed. With the procedure described above, our Linux kernel directory began
approximately 4M in size so we save 90MB for the same functionalities.
Verifying or upgrading your boot loader
Once the new kernel image has been installed on your server, we have to inform our boot loader
about it. This is done inside the configuration file of your boot loader software. Below I show you
how to do it depending if you use GRUB or LILO as your boot loader.
LILO:
This step applies only if you use LILO as your boot loader on the system. If you use GRUB as
your boot loader instead of LILO (highly recommended), then you can skip this section and go
directly to the next one.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-192-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
193
Step 1
You need to edit the lilo.conf file to make your new kernel one of the boot time options:
• Edit the lilo.conf file (vi /etc/lilo.conf) and make the appropriate change on
the line that reads “image=/boot/vmlinuz-x.x.x”.
[root@deep /]# vi /etc/lilo.conf
boot=/dev/sda
map=/boot/map
install=/boot/boot.b
timeout=00
default=linux
restricted
password=somepasswd
image=/boot/vmlinuz
label=linux
read-only
root=/dev/sda6
Step 2
Once the necessary modifications have been made in the /etc/lilo.conf file as shown
above, we update our lilo.conf file for the change to take effect.
• This can be done with the following command:
[root@deep /]# /sbin/lilo
Added linux *
GRUB:
This step applies only if you use GRUB as your boot loader on the system. In most case, GRUB
does not need to be updated when you install a new kernel on your computer but here we have to
verify inside our grub.conf file if all default setting are still available and configured because
when we uninstall Red Hat kernel RPM package, the software automatically remove some needed
parameters from the GRUB configuration file.
Step 1
Edit your GRUB configuration file and be sure that everything is correct and look like the following.
Your setting should differ from the example below.
• Edit the grub.conf file (vi /etc/grub.conf) and check your setting.
[root@deep /]# vi /etc/grub.conf
default 0
timeout 00
title Red Hat Linux
kernel (hd0,0)/vmlinuz ro root=/dev/sda5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-193-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
195
Rebooting your system to load the new kernel
Whether you have installed a new Monolithic Kernel where codes and drivers are compiled into
the kernel and are always loaded or a Modularized Kernel where some segment of codes are
compiled into the kernel as a module and loaded on demand, it is time to Reboot your system
and test your results.
• To reboot your Linux system, use the following command:
[root@deep /]# reboot
When the system is rebooted and you are logged in, verify the new version of your kernel with the
following command:
• To verify the version of your new kernel, use the following command:
[root@deep /]# uname -a
Linux dev 2.4.18-grsec-1.9.4 #1 Wed Jun 19 15:14:55 EDT 2002 i686 unknown
Congratulations!
Delete programs, edit files pertaining to modules
This section applies only if you chose to install a Monolithic Kernel on your system. By default
when you install Linux for the first time (like we did), the kernel is built as a Modularized Kernel.
This means that each device or function we need exists as a module and is controlled by the
Kernel Daemon program named kmod. kmod automatically loads some modules and functions
into memory as they are needed, and unloads them when they’re no longer being used.
kmod and other module management programs included in the modutils RPM package use the
modules.conf file located in the /etc directory to know for example which Ethernet card you
have, if your Ethernet card requires special configuration and so on. If we don’t use any modules
in our newly compiled kernel because we have compiled the kernel as a Monolithic Kernel and
ONLY in this case, we can remove the modules.conf file and completely uninstall the
modutils RPM package.
• To remove the modules.conf file, use the following command:
[root@deep /]# rm -f /etc/modules.conf
• To uninstall the modutils package, use the following command:
[root@deep /]# rpm -e --nodeps modutils
WARNING: Once again, the above is required only if you said no to “Enable loadable module
support (CONFIG_MODULES)” in your kernel configuration because you have decided to build
a Monolithic Kernel.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-195-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
196
Making a new rescue floppy for Modularized Kernel
This section applies only if you chose to install a Modularized Kernel on your system.
Immediately after you successfully start your system and log in as root, you should create a new
emergency boot floppy disk. The procedure to achieve it is the same as shown at the beginning
of this chapter related to Linux Kernel.
Please go back to the beginning of this chapter and follow the procedures to recreate a new
emergency boot floppy disk suitable for the new install Linux kernel on your system. Don’t forget
to test the boot disk to be sure that it works.
The mkbootdisk program runs only on Modularized Kernel. So you can’t use it on a Monolithic
Kernel; instead create an emergency boot floppy disk for Monolithic kernel as shown below.
Making a emergency boot floppy disk for Monolithic Kernel
This section applies only if you chose to install a Monolithic Kernel in your system. Because it is
possible to create a rescue floppy only on modularized kernel, we must find another way to boot
our Linux system for a monolithic kernel if the Linux kernel on the hard disk is damaged.
This is possible with a Linux emergency boot floppy disk. You should create it immediately after
you successfully start your system and log in as root. To create the emergency boot floppy, follow
these steps:
Step 1
Insert a floppy disk and format it with the following command:
[root@deep /]# fdformat /dev/fd0H1440
Double-sided, 80 tracks, 18 sec/track. Total capacity 1440 kB.
Formatting ... done
Verifying ... done
Step 2
Copy the actual file “vmlinuz” from the /boot directory to the floppy disk:
[root@deep /]# cp /boot/vmlinuz /dev/fd0H1440
cp: overwrite `/dev/fd0H1440'? y
NOTE: The vmlinuz file is a symbolic link that points to the real Linux kernel.
Step 3
Determine the kernel’s root device with the following command:
[root@deep /]# rdev
/dev/sda6 /
NOTE: The kernel’s root device is the disk partition where the root file system is located. In this
example, the root device is /dev/sda6; the device name should be different on your system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-196-2048.jpg)
![Kernel Security & Optimization 0
CHAPTER 6
197
Step 4
Set the kernel’s root device with the following command:
[root@deep /]# rdev /dev/fd0H1440 /dev/sda6
NOTE: To set the kernel’s root device, use the device reported by the “rdev” command utility in
the previous step.
Step 5
Mark the root device as read-only with the following command:
[root@deep /]# rdev -R /dev/fd0H1440 1
NOTE: This causes Linux to initially mount the root file system as read-only. By setting the root
device as read-only, you avoid several warnings and error messages.
Step 6
Now put the boot floppy in the drive A: and reboot your system with the following command to be
sure that your new boot disk is working:
[root@something /]# reboot
Following these guidelines, you will now have a boot floppy with a known working kernel in case
of problems with the upgrade. I recommend rebooting the system with the floppy to make sure
that the floppy works correctly.
Step 7
Because the mkbootdisk and dosfstools program are required only when you have a
Modularized kernel installed in your Linux system, we can remove the unneeded mkbootdisk
and dosfstools packages from the system when we have a Monolithic kernel installed on our
server.
• To uninstall the mkbootdisk and dosfstools utility, use the following command:
[root@deep /]# rpm –e mkbootdisk dosfstools](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-197-2048.jpg)
![Process file system management 0
CHAPTER 7
202
What is sysctl?
sysctl is an interface that allows you to make changes to a running Linux system. It serves two
functions: to read and to modify system settings.
• To view all readable variables, use the following command:
[root@deep /]# sysctl -a
• To read a particular variable, for example, fs.file-max, use the following command:
[root@deep /]# sysctl fs.file-max
fs.file-max = 8192
• To set a particular variable for fs.file-max, use the following command:
[root@deep /]# sysctl -w fs.file-max=5536
fs.file-max = 16384
Settings of sysctl variables are usually strings, numbers, or booleans (a boolean being 1 for
yes or a 0 for no). If you set and change variable manually with the sysctl command as show
above, your changes will not resists on the next reboot of the system. For this reason, we will use
and show you further down in this chapter how to make your changes permanent even on
possible reboot of the server by using the /etc/sysctl.conf file.
/proc/sys/vm: The virtual memory subsystem of Linux
All parameters described in this chapter reside under the /proc/sys/vm directory of the server
and can be used to tune the operation of the virtual memory (VM) subsystem of the Linux kernel.
Be very careful when attempting this. You can optimize your system, but you can also cause it to
crash. Since every system is different, you'll probably want some control over this piece of the
system.
Finally, these are advanced setting and if you don’t understand them, then don’t try to play in this
area or try to use all the examples below in your system. Remember that all systems are different
and require different settings and customizations. The majority of the following hacks will work
fine on a server with >= at 512MB of RAM or a minimum of 256MB of RAM. Below this amount of
memory, nothing is guaranteed and the default setting will just be fine for you.
Next I’ll show you parameters that can be optimized. All suggestions I make in this section are
valid for all kinds of servers. The only difference depends on the amount of RAM your machine
has and this is where the settings will change.
The above figure shows a snapshot of /proc/sys/vm directory on an OpenNA Linux & Red Hat
Linux system running kernel version 2.4. Please note that this picture may look different on your
system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-202-2048.jpg)
![Process file system management 0
CHAPTER 7
204
The default kernel setup for the bdflush parameters is:
"40 64 64 256 500 3000 60 0 0"
The default setup for the bdflush parameters under OpenNA Linux is:
"60 64 64 256 500 3000 80 0 0"
The default setup for the bdflush parameters under Red Hat Linux is:
"30 64 64 256 500 3000 60 0 0"
Step 1
To change the values of bdflush, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following line:
# Improve file system performance
vm.bdflush = 60 64 64 256 500 3000 80 0 0
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command in your terminal screen:
[root@deep /]# sysctl -w vm.bdflush="60 64 64 256 500 3000 80 0 0"
The kswapd parameter:
The kswapd file is related to the kernel swapout daemon. This file /proc/sys/vm/kswapd
frees memory on the system when it gets fragmented or full. Its task is to keep the memory
management system operating efficiently. Since every system is different, you'll probably want
some control over this piece of the system.
There are three parameters to tune in this file and two of them (tries_base and
swap_cluster) have the largest influence on system performance. The kswapd file can be
used to tune the operation of the virtual memory (VM) subsystem of the Linux kernel.
Parameter 1 (tries_base):
The kswapd parameter 1 specifies the maximum number of pages kswapd tries to free in one
round. Usually this number will be divided by 4 or 8, so it isn't as big as it looks. Increase this
number to cause swap to be released faster, and increase overall swap throughput. The default
value is 512 pages. We keep the default value here.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-204-2048.jpg)
![Process file system management 0
CHAPTER 7
205
Parameter 2 (tries_min):
The kswapd parameter 2 specifies the minimum number of pages kswapd tries to free a least
each time it is called. Basically it's just there to make sure that kswapd frees some pages even
when it's being called with minimum priority. The default value is 32 pages. We keep the default
value here.
Parameter 3 (swap_cluster):
The kswapd parameter 3 specifies the number of pages kswapd writes in one iteration. You
want this large to increase performance so that kswapd does its I/O in large chunks and the disk
doesn't have to seek often, but you don't want it to be too large since that would flood the request
queue. The default value is 8 pages. We improve the default value here.
The default kernel setup for the kswapd parameters is:
"512 32 8"
The default setup for the kswapd parameters under OpenNA Linux is:
"512 32 32"
The default setup for the kswapd parameters under Red Hat Linux is:
"512 32 8"
Step 1
To change the values of kswapd, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Increase swap bandwidth system performance
vm.kswapd = 512 32 32
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w vm.kswapd=”512 32 32”](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-205-2048.jpg)
![Process file system management 0
CHAPTER 7
206
The overcommit_memory parameter:
The overcommit_memory parameter is simply a flag that enables memory overcommitment.
Memory overcommitment is a procedure to check that a process has enough memory to allocate
a new virtual mapping. When this flag is 0, the kernel checks before each malloc() to see if
there's enough memory left. If the flag is 1, the system pretends there's always enough memory
and don't make the check on the system. This feature can be very useful ONLY on big servers
with a lot of pysical memories available (>= 2GB) because there are a lot of programs that
malloc() huge amounts of memory "just-in-case" and don't use much of it.
The default kernel setup for the overcommit_memory parameter is:
"0"
The default setup for the overcommit_memory parameter under OpenNA Linux is:
"0"
The default setup for the overcommit_memory parameter under Red Hat Linux is:
"0"
Step 1
To change the value of overcommit_memory, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enables/Disables memory overcommitment
vm.overcommit_memory = 0
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
WARNING: Only change the default value of 0 to become 1 on systems with more than 2GB of
RAM. Recall that on small systems the value must be set to 0 (overcommit_memory=0).
There is another way to update the entry without restarting the network by using the following
command into your terminal screen:
[root@deep /]# sysctl -w overcommit_memory=0](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-206-2048.jpg)
![Process file system management 0
CHAPTER 7
207
The page-cluster parameter:
The Linux virtual memory subsystem avoids excessive disk seeks by reading multiple pages on a
page fault. The number of pages it reads is highly dependent on the amount of memory in your
machine. The number of pages the kernel reads in at once is equal to 2 ^ page-cluster. Values
above 2 ^ 5 don't make much sense for swap because we only cluster swap data in 32-page
groups. The page-cluster parameter is used to tune the operation of the virtual memory (VM)
subsystem of the Linux kernel.
The default kernel setup for the kswapd parameter is:
"3"
The default setup for the kswapd parameter under OpenNA Linux is:
"5"
The default setup for the kswapd parameter under Red Hat Linux is:
"4"
Step 1
To change the value of page-cluster, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Increase number of pages kernel reads in at once
vm.page-cluster = 5
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w vm.page-cluster=5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-207-2048.jpg)
![Process file system management 0
CHAPTER 7
208
The pagetable_cache parameter:
The kernel keeps a number of page tables in a per-processor cache (this helps a lot on SMP
systems). The cache size for each processor will be between the low and the high value. On SMP
systems it is used so that the system can do fast pagetable allocations without having to acquire
the kernel memory lock.
For large systems, the settings are probably OK. For normal systems they won't hurt a bit. For
small systems (<16MB RAM) and on a low-memory, single CPU system it might be
advantageous to set both values to 0 so you don't waste the memory.
The default kernel setup for the kswapd parameters is:
"25 50"
The default setup for the kswapd parameters under OpenNA Linux is:
"25 50"
The default setup for the kswapd parameters under Red Hat Linux is:
"25 50"
Step 1
To change the values of pagetable_cache, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Improve number of page tables keeps in a per-processor cache
vm.pagetable_cache = 25 50
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
WARNING: Only change these values on systems with multiple processors (SMP) or on small
systems (single processor) with less than 16MB of RAM. Recall that on small systems the both
values must be set to 0 (vm.pagetable_cache = 0 0).
There is another way to update the entry without restarting the network by using the following
command into your terminal screen:
[root@deep /]# sysctl -w vm.pagetable_cache=”25 50”](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-208-2048.jpg)
![Process file system management 0
CHAPTER 7
209
/proc/sys/fs: The file system data of Linux
All parameters described later in this chapter reside under the /proc/sys/fs directory of the
server and can be used to tune and monitor miscellaneous things in the operation of the Linux
kernel. Be very careful when attempting this. You can optimize your system, but you can also
cause it to crash. Since every system is different, you'll probably want some control over these
pieces of the system.
Finally, these are advanced settings and if you don’t understand them, then don’t play in this area
or try to use all the examples below in your system. Remember that all systems are different and
required different setting and customization.
Below I show you only parameters that can be optimized for the system. All suggestions I
enumerate in this section are valid for every kind of servers. The only difference depends on the
amount of MB of RAM your machines have and this is where settings will change.
The above figure shows a snapshot of /proc/sys/fs directory on a OpenNA Linux & Red Hat
Linux system running kernel version 2.4. Please note that this picture may look different on your
system.
The file-max & file-nr parameters:
The file-max and file-nr files work together on Linux, we use the file-max parameter to
sets the maximum number of file-handles that the Linux kernel will allocate and the file-nr file
to get information about the number of allocated file handles, the number of used file handles and
the maximum number of file handles presently on the system. A large-scale production server
may easily require many thousands of file-handles, depending on the kind and number of
services running concurrently on the server. On busy servers where many services are running
concurrently, we generally tune this file (file-max) to improve the number of open files available
on the system.
It is important to note that you need to increase the limit of open files available on your server
ONLY when you get lots of error messages about running out of file handles. If you don’t receive
this kind of error message, you really DON’T need to increase the default value.
• To know the number of allocated file handles, the number of used file handles and the
maximum number of file handles on your system, use the following command:
[root@deep /]# cat /proc/sys/fs/file-nr
405 137 8192](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-209-2048.jpg)
![Process file system management 0
CHAPTER 7
210
The first value (405) in our result is the number of allocated file handles, the second value (137) is
the number of used file handles, and the last value (8192) is the maximum number of file handles.
When the allocated file handles (405) come close to the maximum (8192), but the number of
actually used ones (137) is far less, you've encountered a peak in your usage of file handles and
you don't need to increase the maximum. The default kernel setup is suitable for most of us.
The default kernel setup for the file-max parameter is:
"8192"
The default setup for the file-max parameter under OpenNA Linux is:
"8192"
The default setup for the file-max parameter under Red Hat Linux is:
"8192"
Step 1
To adjust the value of file-max, type the following command on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Increase limit of file-handles
fs.file-max = 8192
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w fs.file-max=8192](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-210-2048.jpg)
![Process file system management 0
CHAPTER 7
212
Prevent your system responding to ping request:
Preventing your system for responding to ping requests can make a big improvement in your
network security since no one can ping your server and receive an answer. The TCP/IP
protocol suite has a number of weaknesses that allows an attacker to leverage techniques in the
form of covert channels to surreptitiously pass data in otherwise benign packets.
Step 1
Preventing your server from responding to ping requests can help to minimize this problem. Not
responding to pings would at least keep most "crackers" out because they would never know it's
there. ICMP blocking can hurt the performance of long-duration TCP connections, and this is due
to the fact that MTU discovery relies on ICMP packets to work.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable ignoring ping request
net.ipv4.icmp_echo_ignore_all = 1
When this key is on (1), the computer will ignore all ICMP packets. It is not recommended to turn
on this key, except in special situations when you receive an ICMP packet based attack. In the
above parameter, we enable this option.
Step 2
Once the configuration has been set, you must restart your network for the change to take effect.
The command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.icmp_echo_ignore_all=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-212-2048.jpg)
![Process file system management 0
CHAPTER 7
213
Refuse responding to broadcasts request:
As for the ping request, it’s also important to disable broadcast requests. When a packet is
sent to an IP broadcast address (i.e. 192.168.1.255) from a machine on the local network,
that packet is delivered to all machines on that network. Then all the machines on a network
respond to this ICMP echo request and the result can be severe network congestion or outages
(Denial-of-Service attacks). See the RFC 2644 for more information.
Step 1
To disable broadcast requests, type the following command on your terminal.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable ignoring broadcasts request
net.ipv4.icmp_echo_ignore_broadcasts = 1
When this key is on (1), the server will never answer to "ping" if its destination address is
multicast or broadcast. It is good to turn on (1) this key to avoid your server becoming an
involuntary partner of a DoS attack. In the above parameter, we enable this option.
Step 2
Once the configuration has been set, you must restart your network for the change to take effect.
The command to restart the network is the following:
• To restart all networks devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.icmp_echo_ignore_broadcasts=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-213-2048.jpg)
![Process file system management 0
CHAPTER 7
214
Routing Protocols:
Routing and routing protocols can create several problems. IP source routing, where an IP packet
contains details of the path to its intended destination, is dangerous because according to RFC
1122 the destination host must respond along the same path. If an attacker was able to send a
source routed packet into your network, then he would be able to intercept the replies and fool
your host into thinking it is communicating with a trusted host.
Step 1
I strongly recommend that you disable IP source routing on all network interfaces on the system
to protect your server from this hole. If IP source routing is set to off (0), the server will not accept
source-routed frames. Remember that Source-routed frames have an embedded, explicit
description of the route between source and destination. Normally, IP packet routing is based
solely on destination's address.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable IP source routing
net.ipv4.conf.all.accept_source_route = 0
net.ipv4.conf.default.accept_source_route = 0
Source-routed packets are a powerful concept but were never used, and can bring security
problems because they allow a non-blind, REMOTE spoof. It is a very good idea to turn off (0)
these keys. In the above parameters, we disable these options.
Step 2
Once configurations have been set, you must restart your network for the change to take effect.
The command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.conf.all.accept_source_route=1
[root@deep /]# sysctl -w net.ipv4.conf.default.accept_source_route=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-214-2048.jpg)
![Process file system management 0
CHAPTER 7
215
Enable TCP SYN Cookie Protection:
A "SYN Attack" is a Denial of Service (DoS) attack that consumes all the resources on your
machine, forcing you to reboot. Denials of Service attacks (attacks which incapacitate a server
due to high traffic volume or ones those tie-up system resources enough that the server cannot
respond to a legitimate connection request from a remote system) are easily achievable from
internal resources or external connections via extranets and Internet. Enabling TCP SYN Cookie
Protection will help to eliminate the problem. Below is a simple explanation of the concept.
Every TCP/IP connection begins with a 3-way handshake:
1. Client sends a packet (packet 1) to server with SYN bit on, and waits;
2. Server sends a confirmation packet (packet 2) to client, and waits;
3. Client sends a third packet (packet 3) that consolidates the connection.
Once the 3-way handshake is done, the server keeps data of packet 1 in a queue to compare it
with packet 3 and establish the connection. This queue is limited in size and a quite high timeout.
The SYN-flood attack exploits this fact and sends a lot of type-1 packets with random IP source
addresses; the phase 3 answers never arrive; and once the queue is full, the server cannot
receive more connections, be they legitimate or forged.
The SYN cookie "trick" is to embed a code in the header of phase 2 packets, so server DOES
NOT NEED TO KEEP any information about the client. If the phase 3 packet arrives someday,
the server will calculate the port and client initial sequence number based solely on that packet -
and will be able to establish the connection.
Step 1
Since this embedded codification reduces randomness of the server initial sequence number, and
thus can increase the "chance" of IP spoof family attacks, SYN cookies are used only in
emergency situations, that is, when the half-open connections queue is full.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable TCP SYN Cookie Protection
net.ipv4.tcp_syncookies = 1
If this key is on (1), the kernel will send "SYN cookies" ONLY and ONLY when the half-open
connections queue becomes full. This will mitigate the effects of SYN-flood DoS attacks. In the
above parameter, we enable this option.
Step 2
Once the configuration has been set, you must restart your network for the change to take effect.
The command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
WARNING: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.tcp_syncookies=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-215-2048.jpg)
![Process file system management 0
CHAPTER 7
216
Disable ICMP Redirect Acceptance:
When hosts use a non-optimal or defunct route to a particular destination, an ICMP redirect
packet is used by routers to inform the hosts what the correct route should be. If an attacker is
able to forge ICMP redirect packets, he or she can alter the routing tables on the host and
possibly subvert the security of the host by causing traffic to flow via a path you didn't intend.
Step 1
A legitimate ICMP REDIRECT packet is a message from a router that says "router X is better
than me to reach network Y". Therefore, in complex networks, it will be highly recommended to
keep these keys activated. On simple networks, it’s strongly recommended to disable ICMP
Redirect Acceptance into all available interfaces on the server to protect it from this hole.
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable ICMP Redirect Acceptance
net.ipv4.conf.all.accept_redirects = 0
net.ipv4.conf.default.accept_redirects = 0
When these keys are off (0), the kernel does not honor ICMP_REDIRECT packets, thus avoiding
a whole family of attacks based on forging of this type of packet. In the above parameters, we
disable these options.
Step 2
Once the configurations have been set, you must restart your network for the change to take
effect. The command to restart the network is the following:
• To restart all networks devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.conf.all.accept_redirects=0
[root@deep /]# sysctl -w net.ipv4.conf.default.accept_redirects=0](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-216-2048.jpg)
![Process file system management 0
CHAPTER 7
217
Enable bad error message Protection:
This option will alert you about all bad error messages on your network.
Step 1
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following line:
# Enable/Disable bad error message Protection
net.ipv4.icmp_ignore_bogus_error_responses = 1
Step 2
Once configuration has been set, you must restart your network for the change to take effect. The
command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.icmp_ignore_bogus_error_responses=1
Enable IP spoofing protection:
The spoofing protection prevents your network from being the source of spoofed (i.e. forged)
communications that are often used in DoS Attacks.
Step 1
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable IP spoofing protection
net.ipv4.conf.all.rp_filter = 2
net.ipv4.conf.default.rp_filter = 2
These keys control IP Spoof detection and can have the following values:
0=absolutely no checking (the default)
1=simple checking (only obvious spoofs)
2=strong checking (positive source verification)
The recommended standard is level 2, which can bring "problems in complex (non loop free)
networks". In the above parameters, we enable the “strong checking” option.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-217-2048.jpg)
![Process file system management 0
CHAPTER 7
218
Step 2
Once the configurations have been made, you must restart your network for the change to take
effect. The command to restart the network is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
NOTE: This parameter will prevent spoofing attacks against your internal networks but your
external addresses can still be spoofed.
There is another way to update the entry without restarting the network by using the following
command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.conf.all.rp_filter=2
[root@deep /]# sysctl -w net.ipv4.conf.default.rp_filter=2
Enable Log Spoofed, Source Routed and Redirect Packets:
This change will log all Spoofed Packets, Source Routed Packets, and Redirect Packets to your
log files.
Step 1
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable/Disable Log Spoofed, Source Routed, Redirect Packets
net.ipv4.conf.all.log_martians = 1
net.ipv4.conf.default.log_martians = 1
When this key is on (1), the kernel will log any "impossible" packets, where the IP source address
spoofing is obvious. Example: packet with source address equal to 127.0.0.1 coming from an
Ethernet interface. In the above parameter, we enable this option.
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.conf.all.log_martians=1
[root@deep /]# sysctl -w net.ipv4.conf.default.log_martians=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-218-2048.jpg)
![Process file system management 0
CHAPTER 7
219
Other possible optimization of the system
All information described next relates to tuning we can make on the system. Be very careful when
attempting this. You can optimize your system, but you can also cause it to crash.
The shared memory limit parameters:
Shared memory is used for inter-process communication, and to store resources that are shared
between multiple processes such as cached data and code. If insufficient shared memory is
allocated any attempt to access resources that use shared memory such as database
connections or executable code will perform poorly.
Under UNIX world, shared memory is referred to as "System V IPC". Almost all modern operating
systems provide these features, but not all of them have them turned on or sufficiently sized by
default, especially systems with BSD heritage. With Linux the default shared memory limit (both
SHMMAX and SHMALL) is 32 MB in 2.4 kernels, but fortunately can be changed into the /proc file
system.
On system with small MB of RAM (>= 128MB), the default setting of 32MB for shared memory on
the system is enough and should not be changed. On system with lot of RAM, we can readjust
the default setting to better fit our machine and server performance.
Below, I show you an example, to allow 128MB of shared memory on the system. The new value
you enter for the shared memory should be four time less than what your total MB of RAM is. For
example if you have 512MB of RAM installed on your computer, then you can set the default
shared memory to 128MB as we do here. If you have less than what I use in this example, you
have to adjust the value to fit your needs.
Step 1
To change the default values of shared memory, type the following commands on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Improve shared memory size
kernel.shmall = 134217728
kernel.shmmax = 134217728
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following commands into your terminal screen:
[root@deep /]# sysctl -w kernel.shmall="134217728"
[root@deep /]# sysctl -w kernel.shmmax="134217728"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-219-2048.jpg)
![Process file system management 0
CHAPTER 7
220
Tuning the default and maximum window size parameters:
The following hack allow us to raise network limits on Linux by requesting that the kernel provide
a larger send buffer for the server's connections to its clients. This is possible by tuning the
"default send window" and "maximum send window" parameters.
Default parameter under Linux is 64kb, this is correct for regular use of the OS but if you run your
system as a server, it is recommended to change the default parameter to a sufficiently-high
value like 2000kb. 64kb is equal to 65536 (64 * 1024 = 65536), to set the values to 2000kb, we
should enter new values of 2048000 (2000 * 1024 = 2048000).
Step 1
To change the default values of default and maximum window size, type the following
commands on your terminal:
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Improve default and maximum window size
net.core.wmem_max = 2048000
net.core.wmem_default = 2048000
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
NOTE: There is another way to update the entry without restarting the network by using the
following commands into your terminal screen:
[root@deep /]# sysctl -w net.core.wmem_max="2048000"
[root@deep /]# sysctl -w net.core.wmem_default="2048000"
The ulimit parameter:
The ulimit command provides control over the resources available to the shell and to
processes started by it, on systems that allow such control. We can use it to change the default
resource set by the system to users or super user “root” but actually, we use it for changing
resources for the super user “root” only because resources for normal users are controlled and
managed through the /etc/security/limit.conf file.
It is not all default resources available through the ulimit parameter that need to be changed
but only those which can improve the performance of the server in a high load environment. Most
default values for the super user “root” are acceptable and should be kept unchanged. Linux itself
has a “Maximum Processes” per user limit. This feature allows us to control the number of
processes an existing user or the “root” super user on the server may be authorized to have. To
increase performance on highly loaded servers, we can safely set the limit of processes for the
super-user “root” to be unlimited. This is what we will do in the following steps.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-220-2048.jpg)
![Process file system management 0
CHAPTER 7
221
One question remains, how can we change the default resources for a specific user on the
system? Each new user has a hidden file called “.bashrc” available in their home directory. It is
into this file that we can change the default value of resources for the specific user. In the
example below, we do it for the super user “root” but the procedure is the same for any users on
the system. Just edit their corresponding “.bashrc” file and make the change.
Step 1
• Edit the .bashrc file for super user “root” (vi /root/.bashrc) and add the line:
ulimit -u unlimited
NOTE: You must exit and re-login from your terminal for the change to take effect.
Step 2
To verify that you are ready to go, make sure that when you type as root the command ulimit
-a on your terminal, it shows "unlimited" next to max user processes.
[root@deep /]# ulimit -a
core file size (blocks, -c) 0
data seg size (kbytes, -d) unlimited
file size (blocks, -f) unlimited
max locked memory (kbytes, -l) unlimited
max memory size (kbytes, -m) unlimited
open files (-n) 1024
pipe size (512 bytes, -p) 8
stack size (kbytes, -s) 8192
cpu time (seconds, -t) unlimited
max user processes (-u) unlimited
virtual memory (kbytes, -v) unlimited
NOTE: You may also do ulimit -u unlimited at the command prompt instead of adding it to
the /root/.bashrc file but the value will not survive to a reboot.
The atime attribute:
Linux records information about when files were created and last modified as well as when it was
last accessed. There is a cost associated with recording the last access time. The ext2 file
system of Linux has an attribute that allows the super-user to mark individual files such that their
last access time is not recorded. This may lead to significant performance improvements on often
accessed, frequently changing files such as the contents of News Server, Web Server, Proxy
Server, Database Server among other directories.
• To set the attribute to a file, use:
[root@deep /]# chattr +A filename For a specific file
For a whole directory tree, do something like:
[root@deep /root]# chattr -R +A /var/spool For a News and Mail Server directory
[root@deep /root]# chattr -R +A /home/httpd/html For a Web Server directory
[root@deep /root]# chattr -R +A /var/lib/mysql For a SQL Database directory](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-221-2048.jpg)
![Process file system management 0
CHAPTER 7
222
The noatime attribute:
Linux has a special mount option for file systems called noatime that can be added to each line
that addresses one file system in the /etc/fstab file. If a file system has been mounted with
this option, reading accesses to the file system will no longer result in an update to the atime
information associated with the file like we have explained previously. The importance of the
noatime setting is that it eliminates the need by the system to make writes to the file system for
files, which are simply being read. Since writes can be somewhat expensive, this can result in
measurable performance gains. Note that the write time information to a file will continue to be
updated anytime the file is written to. In our example below, we will set the noatime option to our
/chroot file system.
Step 1
• Edit the fstab file (vi /etc/fstab) and add in the line that refers to the /chroot file
system, the noatime option after the defaults option as show below:
LABEL=/chroot /chroot ext3 defaults,noatime 1 2
Step 2
Once you have made the necessary adjustments to the /etc/fstab file, it is time to inform the
system about the modification.
• This can be accomplished with the following commands:
[root@deep /]# mount /chroot -oremount
Each file system that has been modified must be remounted with the command shown above. In
our example we have modified the /chroot file system and it is for this reason that we remount
this file system with the above command.
Step 3
• You can verify if the modifications have been correctly applied to the Linux system with
the following command:
[root@deep /]# cat /proc/mounts
/dev/root / ext3 rw 0 0
/proc /proc proc rw 0 0
/dev/sda1 /boot ext3 rw 0 0
/dev/sda10 /cache ext3 rw,nodev 0 0
/dev/sda9 /chroot ext3 rw,noatime 0 0
/dev/sda8 /home ext3 rw,nosuid 0 0
/dev/sda13 /tmp ext3 rw,noexec,nosuid 0 0
/dev/sda7 /usr ext3 rw 0 0
/dev/sda11 /var ext3 rw 0 0
/dev/sda12 /var/lib ext3 rw 0 0
none /dev/pts devpts rw 0 0
This command will show you all file systems on your server with parameters applied to them. If
you see something like:
/dev/sda9 /chroot ext3 rw,noatime 0 0
Congratulations!](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-222-2048.jpg)
![TCP/IP Network Management 0
CHAPTER 8
227
Linux TCP/IP
Abstract
This chapter has been inserted here because it is preferable not to be connected to the network if
the parts "Installation-Related Reference" and "Security and Optimization-Related Reference" of
the book have not been completed. It is not wise to apply new security configurations to your
system if you are online. Also, don’t forget that the firewall, which represents 50% of networking
security, is still not configured on the Linux server. Finally it is very important and I say VERY
IMPORTANT that you check all configuration files related to Linux networking to be sure that
everything is configured correctly. Please follow all recommendations and steps in this chapter
before continuing reading this book. This will allow us to be sure that if something goes wrong in
the other chapters, it will be not related to your networking configurations.
• To stop specific network device manually on your system, use the following command:
[root@deep /]# ifdown eth0
• To start specific network device manually on your system, use the following command:
[root@deep /]# ifup eth0
• To stop all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network stop
Shutting down interface eth0 [OK]
Disabling IPv4 packet forwarding [OK]
• To start all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network start
Enabling IPv4 packet forwarding [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Until now, we have not played with the networking capabilities of Linux. Linux is one of the best
operating systems in the world for networking features. Most Internet sites around the world
already know this, and have used it for some time. Understanding your network hardware and all
the files related to it is very important if you want to have a full control of what happens on your
server. Good knowledge of primary networking commands is vital. Network management covers
a wide variety of topics. In general, it includes gathering statistical data and monitoring the status
of parts of your network, and taking action as necessary to deal with failures and other changes.
The most primitive technique for network monitoring is periodic "pinging" of critical hosts. More
sophisticated network monitoring requires the ability to get specific status and statistical
information from a range of devices on the network. These should include various sorts of
datagram counts, as well as counts of errors of different kinds. For these reasons, in this chapter
we will try to answer fundamental questions about networking devices, files related to network
functionality, and essential networking commands.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-227-2048.jpg)
![TCP/IP Network Management 0
CHAPTER 8
235
The following values may be used:
NETWORKING=answer, where answer is yes or no (Configure networking or not configure networking).
HOSTNAME=hostname, where hostname is the hostname of your server.
GATEWAY=gwip, where gwip is the IP address of the remote network gateway (if available).
GATEWAYDEV=gwdev, where gwdev is the device name (eth#) you use to access the remote gateway.
The /etc/sysctl.conf file:
With the new version of Red Hat Linux, all kernel parameters available under the /proc/sys/
subdirectory of Linux can be configured at runtime. You can use the new /etc/sysctl.conf
file to modify and set kernel parameters at runtime. The sysctl.conf file is read and loaded
each time the system reboots or when you restart your network. All settings are now stored in the
/etc/sysctl.conf file. All modifications to /proc/sys should be made through
/etc/sysctl.conf, because they are better for control, and are executed before rc.local or
any other "users" scripts.
Below, we’ll focus only on the kernel option for IPv4 forwarding support. See later in this chapter
the TCP/IP security parameters related to the sysctl.conf file.
To enable IPv4 forwarding on your Linux system, use the following command:
Step 1
• Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following line:
# Enable packet forwarding (required only for Gateway, VPN, Proxy, PPP)
net.ipv4.ip_forward = 1
Step 2
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
WARNING: You must enable packet forwarding only on a machine that serves as a Gateway
Server, VPN Server, Proxy Server or with PPP connection. Forwarding allows packets that are
destined for another network interface (if you have another one) to pass through the network.
There is another way to update the entry without restarting the network by using the following
command into your terminal screen:
[root@deep /]# sysctl -w net.ipv4.ip_forward=1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-235-2048.jpg)
![TCP/IP Network Management 0
CHAPTER 8
236
The /etc/hosts file:
As your machine gets started, it will need to know the mapping of some hostnames to IP
addresses before DNS can be referenced. This mapping is kept in the /etc/hosts file. In the
absence of a name server, any network program on your system consults this file to determine
the IP address that corresponds to a host name.
Following is a sample /etc/hosts file:
IP Address Hostname Alias
127.0.0.1 localhost.localdomain localhost
208.164.186.1 deep.openna.com deep
208.164.186.2 mail.openna.com mail
208.164.186.3 web.openna.com web
The leftmost column is the IP address to be resolved. The next column is that host’s name. Any
subsequent columns are the aliases for that host. In the second line, for example, the IP address
208.164.186.1 if for the host deep.openna.com. Another name for deep.openna.com is
deep.
WARNING: Some people have reporting that a badly formed line in the /etc/hosts file may result
to a "Segmentation fault (core dumped)" with the syslogd daemon, therefore I recommend you
to double check your entry under this file and be sure that its respond to the example as shown
above. The “Alias” part of the line is important if you want to be able to use the FQDN (Fully
Qualified Domain Name) of the system reported by the hostname -f command.
After you are finished adding and configuring your networking files, don’t forget to restart your
network for the changes to take effect.
• To restart your network, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
WARNING: Time out problems for telnet or ftp connection are often caused by the server trying
to resolve the client IP address to a DNS name. Either DNS isn’t configured properly on your
server or the client machines aren’t known to the DNS server. If you intend to run telnet or ftp
services on your server, and aren’t using DNS, don’t forget to add the client machine name and
IP in your /etc/hosts file on the server or you can expect to wait several minutes for the DNS
lookup to time out, before you get a login prompt.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-236-2048.jpg)
![TCP/IP Network Management 0
CHAPTER 8
237
Testing TCP/IP Networking
Once we have applied TCP/IP security and optimization parameters to our server and checked
or configured all files related to network functionality, we can run some tests to verify that
everything works as expected. It is important to note that at this stage every test must be
successful and not have any errors. It is to your responsibility to know and understand networking
architecture and basic TCP/IP protocols before testing any parts of your networking configuration
and topology.
Step 1
To begin, we can use the ifconfig utility to display all the network interfaces on the server.
• To display all the interfaces you have on your server, use the command:
[root@deep /]# ifconfig
The output should look something like this:
eth0 Link encap:Ethernet HWaddr 00:E0:18:90:1B:56
inet addr:208.164.186.2 Bcast:208.164.186.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:1295 errors:0 dropped:0 overruns:0 frame:0
TX packets:1163 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
Interrupt:11 Base address:0xa800
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:3924 Metric:1
RX packets:139 errors:0 dropped:0 overruns:0 frame:0
TX packets:139 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
NOTE: If the ifconfig tool is invoked without any parameters, it displays all interfaces you
configured. An option of “-a” shows the inactive one as well.
Step 2
If all network interfaces on the server look as you expect, then it is time to verify that you can
reach your hosts. Choose a host from your internal network, for instance 192.168.1.1
• To verify that you can reach your internal hosts, use the command:
[root@deep /]# ping 192.168.1.1
The output should look something like this:
PING 192.168.1.1 (192.168.1.1) from 192.168.1.1 : 56 data bytes
64 bytes from 192.168.1.1: icmp_seq=0 ttl=128 time=1.0 ms
64 bytes from 192.168.1.1: icmp_seq=1 ttl=128 time=1.0 ms
64 bytes from 192.168.1.1: icmp_seq=2 ttl=128 time=1.0 ms
64 bytes from 192.168.1.1: icmp_seq=3 ttl=128 time=1.0 ms
--- 192.168.1.1 ping statistics ---
4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 1.0/1.0/1.0 ms](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-237-2048.jpg)
![TCP/IP Network Management 0
CHAPTER 8
238
WARNING: Do not try to ping a host in which you have applied the previous TCP/IP security
settings to prevent your system to respond to ping request. Instead try to ping another host
without this feature enable. Also if you don’t receive an answer from the internal host you try to
ping, verify if your hubs, routers, network cards, and network topology are correct.
If you are able to ping your internal host, congratulations! Now we must ping an external
network, for instance 216.148.218.195
• To verify that you can reach the external network, use the command:
[root@deep /]# ping 216.148.218.195
The output should look something like this:
PING 216.148.218.195 (216.148.218.195) from 216.148.218.195 :56 data byte
64 bytes from 216.148.218.195: icmp_seq=0 ttl=128 time=1.0 ms
64 bytes from 216.148.218.195: icmp_seq=1 ttl=128 time=1.0 ms
64 bytes from 216.148.218.195: icmp_seq=2 ttl=128 time=1.0 ms
64 bytes from 216.148.218.195: icmp_seq=3 ttl=128 time=1.0 ms
--- 216.148.218.195 ping statistics ---
4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 1.0/1.0/1.0 ms
Step 3
You should now display the routing information with the command route to see if the hosts have
the correct routing entries.
• To display the routing information, use the command:
[root@deep /]# route -n
The output should look something like this:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
208.164.186.2 0.0.0.0 255.255.255.255UH 0 0 0 eth0
208.164.186.0 208.164.186.2 255.255.255.0 UG 0 0 0 eth0
208.164.186.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-238-2048.jpg)
![Step 4
Another useful option is “netstat -vat”, which shows all active and listen TCP connections.
• To shows all active and listen TCP connections, use the command:
[root@deep /]# netstat -vat
The output may look something similar to this example depending if the related services are
running. Be aware that your results will almost certainly vary from the ones shown below:
Active Internet connections (servers and established)
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 0 deep.openna.co:domain *:* LISTEN
tcp 0 0 localhost:domain *:* LISTEN
tcp 0 0 deep.openna.com:ssh gate.openna.com:1682ESTABLISHED
tcp 0 0 *:webcache *:* LISTEN
tcp 0 0 deep.openar:netbios-ssn *:* LISTEN
tcp 0 0 localhost:netbios-ssn *:* LISTEN
tcp 0 0 localhost:1032 localhost:1033 ESTABLISHED
tcp 0 0 localhost:1033 localhost:1032 ESTABLISHED
tcp 0 0 localhost:1030 localhost:1031 ESTABLISHED
tcp 0 0 localhost:1031 localhost:1030 ESTABLISHED
tcp 0 0 localhost:1028 localhost:1029 ESTABLISHED
tcp 0 0 localhost:1029 localhost:1028 ESTABLISHED
tcp 0 0 localhost:1026 localhost:1027 ESTABLISHED
tcp 0 0 localhost:1027 localhost:1026 ESTABLISHED
tcp 0 0 localhost:1024 localhost:1025 ESTABLISHED
tcp 0 0 localhost:1025 localhost:1024 ESTABLISHED
tcp 0 0 deep.openna.com:www *:* LISTEN
tcp 0 0 deep.openna.com:https *:* LISTEN
tcp 0 0 *:389 *:* LISTEN
tcp 0 0 *:ssh *:* LISTEN
Step 5
Sometimes machines on your network will discard your IP packets and finding the offending
Gateway responsilbe can be difficult. Fortunately the tracepath utility attempts to trace the
route an IP packet would follow to some Internet host. Choose an Internet host, for instance
64.81.28.146
• To print the route packets take to network host, use the command:
[root@deep /]# tracepath 64.81.28.146
The output should look something like this:
1?: [LOCALHOST] pmtu 1500
1?: 207.35.78.1
2?: 10.70.1.1
3?: 206.47.228.178
4?: 206.108.97.149
5?: 206.108.103.214
6?: 206.108.103.228
7?: 208.51.134.9
8?: 208.48.234.189
9?: 206.132.41.78 asymm 10
10?: 204.246.213.226 asymm 13
11?: 206.253.192.217 asymm 13
12?: 206.253.195.218 asymm 14
13: 64.81.28.146 asymm 15 139ms reached
Resume: pmtu 1500 hops 13 back 15](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-239-2048.jpg)
![TCP/IP Network Management 0
CHAPTER 8
240
Step 6
Finally, we will use the hostname command of Linux to show if our systems host name is
correct.
• To display and print the current host name of your server, use the command:
[root@deep /]# hostname
deep
The hostname command without any options will print the current host name of our system, in
this example “deep”.
Now, it’s important to verify if the Fully Qualified Domain Name (FQDN) of our server is reported
correctly.
• To display and print the FQDN of your server, use the command:
[root@deep /]# hostname -f
deep.openna.com
The last checkup
If you can answer, “Yes” to each of the questions below, then your network is working and you
can continue .
Parameters inside ifcfg-ethN files are corrects
The /etc/resolv.conf file contain your primary and secondary Domain Name Server
All parameters included in the /etc/host.conf file are corrects
All parameters included in the /etc/sysconfig/network file are corrects
The /etc/hosts file contain the mapping of your hostnames to IP addresses
All network interfaces on the server have the right parameter
You can reach the internal and external hosts
Your hosts have the correct routing entry
The status of the interfaces has been checked and looks fine
You are able to print the route packets take to network host](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-240-2048.jpg)
![Firewall Basic Concept 0
CHAPTER 9
244
The Well Known Ports are assigned by the IANA and on most systems can only be used by
system (or root) processes or by programs executed by privileged users (our daemons running in
the background). Ports are used by the TCP protocol [RFC793] to name the ends of logical
connections, which carry long-term conversations. For the purpose of providing services to
unknown callers, a service contact port is defined. The contact port is sometimes called the "well-
known port". Wherever possible, the same port assignments are also used by UDP protocol
[RFC768]. For many years Well Known Ports were in the range 0-255. Recently, the range for
the Well Known Ports has been expanded from 0 to 1023 to respond to the exponential growth of
the Internet.
The Registered Ports are also listed by the IANA and, on most systems, can be used by ordinary
user processes or programs executed by ordinary users. The IANA registers the use of these
ports as a convenience to the community. Again, wherever possible, these same port
assignments are used with UDP [RFC768]. The Registered Ports are in the range 1024-49151.
Finally, the Dynamic and/or Private Ports are those ranging from 49152 through to 65535.
1
2
3
[1] The Well Known Ports represent 2% of all available ports [0-1023].
[2] The Registered Ports represent 73% of all available ports [1024-49151].
[3] The Dynamic and/or Private Ports represent 25% of all available ports [49152-65535].
Ports Numbers Graphical Representation](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-244-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
262
Pristine source
As we don’t use the RPM package to install the program, it would be difficult for us to locate all the
files installed on the system if in the future we want to upgrade. To solve this problem, it is a good
idea to make a list of files on the system before you install GIPTables, and then one afterwards,
we can then compare them using the diff utility to find out what files were installed and where
they were placed.
• Simply run the following command before installing the software:
[root@deep root]# find /* > GIPTables1
• And the following one after you install the software:
[root@deep root]# find /* > GIPTables2
• Then use the following command to get a list of what changed:
[root@deep root]# diff GIPTables1 GIPTables2 > GIPTables-Installed
With this procedure, if any future upgrade appears, all you have to do is to read the generated list
of what files were added or changed by the program and remove them manually from your
system before installing the new software. In the above example, we use the /root directory of
the system to store the generated list of files.
Compiling - Optimizing & Installing GIPTables
To install the GIPTables software on your system, just download the latest version of the
software from http://www.giptables.org/ site, and then as user ‘root’ expand the archive under
your /var/tmp directory.
• To accomplish this use the following commands:
[root@deep /]# cp giptables-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf giptables-version.tar.gz
Next, move into the newly created GIPTables source directory and perform the following steps
to install the software for your system.
• To move into the newly created GIPTables source directory use the command:
[root@deep tmp]# cd giptables-1.1/
• To install GIPTables enter the following command:
[root@deep giptables-1.1]# ./install.sh
The “install.sh” script file will simply install any GIPTables components on your system to
the right location.
Once the installation of GIPTables has been completed, we can free up some disk space by
deleting both the program tar archive and the related source directory since they are no longer
needed.
• To delete GIPTables and its related source directory, use the commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf giptables-version/
[root@deep tmp]# rm -f giptables-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-262-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
264
Step1
First of all, choose one of the default configuration files that can be found in the
/lib/giptables/conf directory. Find one that mostly meets your needs, make a backup copy
of it and open it with any kind of text editor to configure it. In our example, we will configure our
GIPTables firewall software for a Gateway/Proxy Server with two network interfaces since it is
the most complicated configuration we are likely to encounter.
• These procedures can be accomplished with the following commands:
[root@deep /]# cd /lib/giptables/conf/
[root@deep conf]# cp giptables.conf.gateway giptables.conf.mybox
[root@deep conf]# ln -sf /lib/giptables/conf/giptables.conf.mybox
/etc/giptables.conf
In the above steps, we make a copy of our original “giptables.conf.gateway” file and create
a symbolic link pointing to the copy.
NOTE: It is a good idea not to directly modify an example configuration file, because if it gets
damage, then you have to install the entire package again in order to get it back.
Step2
Now our giptables.conf file that is a symbolic link pointing to the original configuration file for
a Gateway set-up exists. It is time to edit it and provide or change some minimal values to make it
work for our system.
In the GIPTables configuration file below, we’ll ONLY explain how to configure and set
parameters that are the same for all types of GIPTables firewall configuration. Parts that differ
are associated with different available GIPTables modules that must be loaded by the firewall
configuration file to enable different services. All available modules with GIPTables firewall are
explained later in this document.
• Edit the giptables.conf file (vi /etc/giptables.conf) and configure it to fit in
with your system and networking setup. Text in bold is what you should change to make
the firewall work with your server:
The Debug Definition:
The first configurable option that is the same for all types of GIPTables firewall configuration is
“DEBUG” and by default it is set to “off”. This option is useful only for debugging purposes.
# ----------------------------------------------------------------------------
# DEBUG
#
DEBUG="off"
If you set this option to "on", the firewall will display all IPTables rules relating to the
GIPTables configuration file that you use to the screen, nothing will go to the kernel. The
displayed set of rules will be commented so that you will not end up with lots of rules on the
screen that you do not understand. This way you can see only what the firewall is generating, and
also you will be able to better understand which rule is for what.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-264-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
265
When this option is set to "off" (the default setting), the firewall will send all generated rules to
the kernel, nothing will be displayed on your screen and the firewall will run on your system.
Therefore is you want to run GIPTables on your system, you must be sure that this option is set
to ‘off’.
NOTE: When the “DEBUG” option is set to “on”, it is possible to redirect the output of the firewall
rules to a file, and use this file as learning example of how to set up IPTables rules for different
kind of services. This is possible with the following command:
[root@deep tmp]# /etc/init.d/giptables start > output-result.txt
The Monolithic Kernel Definition:
The second configurable option that is the same for all types of GIPTables configuration is the
“MONOLITIC_KERNEL” option and it is set to “no” by default. This option exists because the Linux
kernel can be compiled so that it contains all the required IPTables driver code directly in it.
This way we have a MONOLITHIC KERNEL and we would need to answer by “yes” to the option.
# ----------------------------------------------------------------------------
# Some definitions for easy maintenance
# Edit these to suit your system
#
MONOLITIC_KERNEL="no"
If you set this option to ‘yes’, then GIPTables will be informed that all native IPTables
modules are directly compiled into the kernel. It is important to say ‘yes’ here only if you have a
Monolithic Linux Kernel installed on your computer otherwise say ‘no’. Then the firewall will look
for and load all the IPTables modules that are required, depending on your configuration file.
NOTE: If you compile your kernel as Monolithic, you should know what IPTables modules you
need to compile directly into the kernel, since the firewall will not try to load them. If you missed
some modules, you will inevitably get errors, or the firewall might not work as expected. The best
solution for a Monolithic Kernel set-up is to compile all native iptables modules into the kernel.
Also, don’t forget to set MONOLITIC_KERNEL="yes" in the firewall configuration file.
The External Network Interface Definition:
The next configurable option that is the same for all type of GIPTables firewall configuration is
one of the most important settings in the configuration file.
# Interface 0: This is our external network interface
# It is directly connected to Internet
INTERFACE0="eth0"
INTERFACE0_IPADDR="x.x.x.x"
ANY_IPADDR="0/0"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-265-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
273
# Refuse incoming packets claiming to be from the IP addresses of our
interfaces
REFUSE_SPOOFING_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_IN_REFUSE_SPOOFING[0]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[0]="no"
NETWORK1_IN_REFUSE_SPOOFING[0]="yes"
REFUSE_SPOOFING_IPADDR[1]=$INTERFACE1_IPADDR
INTERFACE0_IN_REFUSE_SPOOFING[1]="no"
INTERFACE1_IN_REFUSE_SPOOFING[1]="yes"
NETWORK1_IN_REFUSE_SPOOFING[1]="no"
B) We log and drop all incoming packets claiming to be from the broadcast source address
range. We accept broadcast source packets only in one situation: when we have a DHCP
Server, and this, because a DHCP Client will request its IP address by sending out and
DHCP discovery packet that has source IP address "0.0.0.0" and destination IP
address "255.255.255.255". In this situation, the Gateway Server is also a DHCP
Server, so we will accept by default those broadcast source packets only on the internal
interface.
# Refuse incoming packets claiming to be from broadcast-src address range
REFUSE_SPOOFING_IPADDR[2]="0.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[2]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[2]="no"
NETWORK1_IN_REFUSE_SPOOFING[2]="yes"
C) We log and drop all incoming packets claiming to be from the reserved loopback IP
address range. This is so obvious. We should never have incoming packets with source
IP address from the loopback address range. We can refuse them safely on all our
interfaces.
# Refuse incoming packets claiming to be from reserved loopback address range
REFUSE_SPOOFING_IPADDR[3]="127.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[3]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[3]="yes"
NETWORK1_IN_REFUSE_SPOOFING[3]="yes"
E) We log and drop all incoming packets claiming to be from the well-known private
networks: A, B, C. We can safely refuse all packets claiming to be from those private
networks on all of our interfaces, and internal network.
# Refuse incoming packets claiming to be from class A private network
REFUSE_SPOOFING_IPADDR[4]="10.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[4]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[4]="yes"
NETWORK1_IN_REFUSE_SPOOFING[4]="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-273-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
274
# Refuse incoming packets claiming to be from class B private network
REFUSE_SPOOFING_IPADDR[5]="172.16.0.0/12"
INTERFACE0_IN_REFUSE_SPOOFING[5]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[5]="yes"
NETWORK1_IN_REFUSE_SPOOFING[5]="yes"
# Refuse incoming packets claiming to be from class C private network
REFUSE_SPOOFING_IPADDR[6]="192.168.0.0/16"
INTERFACE0_IN_REFUSE_SPOOFING[6]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[6]="no"
NETWORK1_IN_REFUSE_SPOOFING[6]="yes"
WARNING: There is only one exception in which case we do not refuse incoming packets on our
internal interface claiming to be from our internal private network. This appears only for a
Gateway Server when your internal network is from class C. You should not refuse incoming
packets on internal interface from your internal network.
F) We log and drop all incoming packets claiming to be from class D, E, and unallocated IP
addresses. These are classes that are not currently used or that are unallocated. There is
no reason for an incoming packet to have a source IP address from one of those classes.
# Refuse incoming packets claiming to be from class D, E, and unallocated
REFUSE_SPOOFING_IPADDR[7]="224.0.0.0/3"
INTERFACE0_IN_REFUSE_SPOOFING[7]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[7]="yes"
NETWORK1_IN_REFUSE_SPOOFING[7]="yes"
The above Spoofing and bad address protection assume that you have two network interfaces
installed on your system. This configuration is suitable for a Gateway Server. If you only have one
network interface on your server, then you can ignore, comment out or remove those options that
refer to internal interface and network (Interface1 and Network1).
If REFUSE_SPOOFING="no" then the entire spoofing protection section is ignored.
The above configuration closes our discussion about parameters that are the same for all types of
GIPTables firewall configurations. Once you have configured all of the customized values in this
part of the GIPTables configuration file, suitable for your type of system, you are ready to start
the software.
/etc/rc.d/rc.giptables.blocked: The GIPTables Blocked File
Sometimes you’ll know an address that you would like to block from having any access at all to
your server. Instead of entering the entire iptables line per IP address for those jerks on the
internet, you can write them into the rc.giptables.blocked file, that will take the IP
addresses, strip out any comments and run the resulting list through an iptables routine.
The net effect is the /etc/giptables.conf file increases no more than needed, especially
when one might have a large number of IP addresses to deny.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-274-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
276
• To start GIPTables on your system, use the following command:
[root@deep /]# /etc/init.d/giptables start
The GIPTables Firewall Module Files
Once you have chosen the right GIPTables configuration file suitable for the type of server that
you want to protect and all the parameters, which are the same for all type of GIPTables firewall
have been configured, GIPTables should have enough information to run properly on your
system. Really, your work is done and your firewall is well protected against attacks.
Everyone has a different set-up for his firewall design and sometimes we need to implement a
new service and then open and control the port associated with this service on our server.
GIPTables allows us to add, modify, delete, and customize any existing or expected services in
a simple manner through its modules feature.
With GIPTables, each service like DNS, FTP, HTTPD, etc have their own modules. Those
modules are loaded only when defined in the giptables.conf file, so that if there are no
options related to FTP for example, the FTP module will not be loaded. You can specify on which
interface or network the module will work, and what kind of requests (incoming or outgoing) can
go thought that interface or network.
All GIPTables modules are located under the /lib/giptables/modules directory and it’s in
these module files that we handle all rules relating to the specific service. When we configure,
customize and enable service parameters in the giptables.conf file, the parameter in
question get its information about IPTables rules that must be used through the modules files
available under the /lib/giptables/modules directory. If the parameter of the specific
service that you want to enable is not defined into the GIPTables configuration file, then this
service will not load its IPTables rules from its modules file and will not run with your
GIPTables Firewall software.
If you look in the /lib/giptables/modules directory, you’ll find the following modules for the
services that can be enabled with GIPTables Firewall.
giptables-ANY The ANY module, which refer to ANY services
giptables-AUTH The AUTH module, which refer to AUTH services
giptables-DHCP The DHCP module, which refer to DHCP services
giptables-DNS The DNS module, which refer to DNS services
giptables-FINGER The FINGER module, which refer to FINGER services
giptables-FTP The FTP module, which refer to FTP services
giptables-HTTP The HTTP module, which refer to HTTP services
giptables-HTTPS The HTTPS module, which refer to HTTPS services
giptables-ICMP The ICMP module, which refer to ICMP services
giptables-IMAP The IMAP module, which refer to IMAP services
giptables-IMAPS The IMAPS module, which refer to IMAPS services
giptables-LDAP The LDAP module, which refer to LDAP services
giptables-LDAPS The LDAPS module, which refer to LDAPS services
giptables-MYSQL The MYSQL module, which refer to MYSQL services
giptables-NETBIOS The NetBIOS module, which refer to NetBIOS services
giptables-NNTP The NNTP module, which refer to NNTP services
giptables-NNTPS The NNTPS module, which refer to NNTPS services
giptables-NTP The NTP module, which refer to NTP services
giptables-ORACLE The ORACLE module, which refer to ORACLE services
giptables-POP3 The POP3 module, which refer to POP3 services
giptables-POP3S The POP3S module, which refer to POP3S services
giptables-POSTGRES The POSTGRES module, which refer to POSTGRES services](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-276-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
278
Enabling outgoing client requests
In the example below, we define and enable MySQL outgoing client requests for a Gateway
Server. The difference about parameters with other type of servers is that we need to define
additional network interface (INTERFACE1) and network (NETWORK1) for a Gateway Server set-
up. All text in bold should be configured to define and enable the MySQL service.
# ----------------------------------------------------------------------------
# MYSQL outgoing client request
#
# Interface 0 MYSQL outgoing client request
INTERFACE0_MYSQL_CLIENT="yes"
INTERFACE0_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_OUT_DST_IPADDR[0]=$ANY_IPADDR
In the above example, we first enable MySQL outgoing client request on the external interface
(INTERFACE0_MYSQL_CLIENT="yes").
Next, we instruct the system that the parameters apply to interface 0 (INTERFACE0) for the
MySQL service (MYSQL) for outgoing requests (OUT) with the source IP address (SRC_IPADDR)
coming from our external interface IP address ($INTERFACE0_IPADDR). Which means, packets
having our external interface IP address, as a source IP address will be able to go out and/or start
a new connection.
Finally, we inform the system that the parameters also apply to interface 0 (INTERFACE0) for the
MySQL service (MYSQL) for outgoing requests (OUT) with the destination IP address
(DST_IPADDR) going to anywhere ($ANY_IPADDR). And this means, packets having our external
interface, as the destination IP address will be able to go out and/or start a new connection.
Using the connection tracking capability of IPTables, the related MySQL incoming packets are
automatically allowed back in by the firewall. In this case, our machine can be a MySQL client that
is allowed to access any MySQL server on the Internet.
If we want to restrict access to only one external MySQL server, the parameters should be
configured like in the example below:
# Interface 0 MYSQL outgoing client request
INTERFACE0_MYSQL_CLIENT="yes"
INTERFACE0_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_OUT_DST_IPADDR[0]="x.x.x.x"
In this case, "x.x.x.x" is the IP address of the external MySQL server that we want to access.
For a second MySQL server, another set of parameters should be added, like in the example
below:](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-278-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
279
# Interface 0 MYSQL outgoing client request
INTERFACE0_MYSQL_CLIENT="yes"
INTERFACE0_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_OUT_DST_IPADDR[0]="x.x.x.x"
INTERFACE0_MYSQL_OUT_SRC_IPADDR[1]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_OUT_DST_IPADDR[1]="y.y.y.y"
"x.x.x.x" is the IP address of the first external MySQL server that we want to access and
"y.y.y.y" is the IP address of the second external MySQL server that we want to access. Please
note that the index of parameters has been increased, so that the first set of parameters have the
index 0, and the second set of parameters have the index 1.
NOTE: This rule is the same for all GIPTables Firewall parameters that have an index. If you
would like to add a second set of parameters, just copy/paste them, make the required changes
and do not forget to increase the index.
On a Gateway Server or machines with two networks interfaces, we need to define the following
additional parameters for the firewall to recognize the other network interface and the private
network behind it.
# Interface 1 MYSQL outgoing client request
INTERFACE1_MYSQL_CLIENT="yes"
INTERFACE1_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_MYSQL_OUT_DST_IPADDR[0]=$NETWORK1
In the above example, we enable MySQL outgoing client request on the internal interface
(INTERFACE1_MYSQL_CLIENT="yes").
We instruct the system that the parameters apply to internal interface 1 (INTERFACE1) for the
MySQL service (MYSQL) to outgoing requests (OUT) with source IP address (SRC_IPADDR)
coming from our internal interface IP address ($INTERFACE1_IPADDR). Therefore, any packets
having our internal interface IP address, as source IP address will be able to go out and/or start a
new connection.
Next, we inform the system that the parameters also apply to internal interface 1 (INTERFACE1)
for the MySQL service (MYSQL) for outgoing requests (OUT) with a destination IP address
(DST_IPADDR) going from our internal subnet IP address range ($NETWORK1). Therefore, any
packets from our internal subnet will be able to go out and/or start new connections.
Using the connection tracking capability of IPTables, the related MySQL incoming packets are
automatically allowed back in by the firewall. In this case, our machine can be a MySQL client that
is allowed to access any MySQL server from our internal subnet.
# Network 1 MYSQL forwarded outgoing client request
NETWORK1_MYSQL_CLIENT="yes"
NETWORK1_MYSQL_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_MYSQL_OUT_DST_IPADDR[0]=$ANY_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-279-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
280
Here, we enable MySQL outgoing client requests on our internal subnet
(NETWORK1_MYSQL_CLIENT="yes").
We instruct the system that the parameters apply to our internal subnet (NETWORK1) for the
MySQL service (MYSQL) for outgoing requests (OUT) with the source IP address (SRC_IPADDR)
coming from our internal subnet IP address range ($NETWORK1).
In the second line, we inform the system that the parameters also apply to our internal subnet
(NETWORK1) for the MySQL service (MYSQL) to outgoing requests (OUT) with destination IP
address (DST_IPADDR) going to anywhere ($ANY_IPADDR).
Using the connection tracking capability of IPTables, the related MySQL incoming packets are
automatically allowed back in by the firewall. In this case, our machines from our internal subnet
are the MySQL clients and are allowed to access any MySQL server on the Internet.
NOTE: The requests are automatically SNATed (MASQUERADEd) by the GIPTables Firewall, so
that the MySQL server from the Internet thinks that talks with our Gateway server.
In general, you should only replace MYSQL with the name of the service that you want to define
for the parameters to work for other type of services. In our example, we use MYSQL; it is to you
to change it for the service of your choice.
Enabling incoming client requests
As we can see, all of the above parameters apply only to outgoing client requests for the MySQL
service on a Gateway Server. Now for incoming server requests, we should add the related lines
to the configuration file to allow them in. In the example below, we define and enable MySQL
incoming client requests for a Gateway Server. The difference with parameters for other types of
servers is that here we need to define an additional network interface (INTERFACE1) and a
network (NETWORK1) for a Gateway Server set-up.
# ----------------------------------------------------------------------------
# MYSQL incoming client request
#
# Interface 0 MYSQL incoming client request
INTERFACE0_MYSQL_SERVER="yes"
INTERFACE0_MYSQL_IN_SRC_IPADDR[0]=$ANY_IPADDR
INTERFACE0_MYSQL_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
In the above example, we first enable incoming client request for MySQL on the external interface
(INTERFACE0_MYSQL_SERVER="yes").
Next, we instruct the system that the parameters apply to external interface 0 (INTERFACE0) for
the MySQL service (MYSQL) for incoming requests (IN) with the source IP address (SRC_IPADDR)
coming from anywhere ($ANY_IPADDR). This mean that we permit the firewall to receive packets
coming from anywhere on our external interface to start a new connection.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-280-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
281
Finally, we inform the system that parameters also apply to external interface 0 (INTERFACE0)
for MySQL service (MYSQL) on incoming requests (IN) with destination IP address (DST_IPADDR)
coming from our external IP address ($INTERFACE0_IPADDR). In other terms, incoming packets
having our external interface, as destination IP address will be able to come in and/or start a new
connection.
Using the connection tracking capability of IPTables, the related MySQL outgoing packets are
automatically allowed back out by the firewall. In this case, our machine is a MySQL server that is
allowed to receive requests from any MySQL client from the Internet.
If we want to allow access to only one external client machine on the MySQL server, the
parameters should be configured like in the example below:
# Interface 0 MYSQL incoming client request
INTERFACE0_MYSQL_SERVER="yes"
INTERFACE0_MYSQL_IN_SRC_IPADDR[0]="x.x.x.x"
INTERFACE0_MYSQL_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
In this case, "x.x.x.x" is the IP address of the external client machine that is allowed to access
our MySQL server. For a second external client machine allowed, another set of parameters
should be added, like in the example below:
# Interface 0 MYSQL incoming client request
INTERFACE0_MYSQL_SERVER="yes"
INTERFACE0_MYSQL_IN_SRC_IPADDR[0]="x.x.x.x"
INTERFACE0_MYSQL_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_MYSQL_IN_SRC_IPADDR[1]="y.y.y.y"
INTERFACE0_MYSQL_IN_DST_IPADDR[1]=$INTERFACE0_IPADDR
"x.x.x.x" is the IP address of the first external client machine that is allowed to access our
MySQL server and "y.y.y.y" is the IP address of the second external client machine that is
allowed to access our MySQL server. Please note that the index of parameters has been
increased, so that the first set of parameters have the index 0, and the second set of parameters
have the index 1.
NOTE: This rule is the same for all GIPTables Firewall parameters that have an index. If you
would like to add a second set of parameters, just copy/paste them, make the required changes
and do not forget to increase the index.
Don’t forget that we need to add all of the lines below for a Gateway Server set-up for the firewall
to recognize the second network interface and our internal subnet. The definitions and
explanations are the same as for outgoing client requests explained earlier.
# Interface 1 MYSQL incoming client request
INTERFACE1_MYSQL_SERVER="yes"
INTERFACE1_MYSQL_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_MYSQL_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-281-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
282
In the above example, we enable MySQL incoming client request on the internal interface
(INTERFACE1_MYSQL_SERVER="yes").
Next, we instruct the firewall that on the internal interface (INTERFACE1), all MySQL (MYSQL)
incoming packets (IN) with source IP address (SRC_IPADDR) from our internal subnet IP address
range ($NETWORK1) and with destination IP address (DST_IPADDR) coming from our internal
interface IP address ($INTERFACE1_IPADDR) will be allowed to come in and/or start a new
connection.
In other terms, any incoming MySQL packets with source IP address from our internal subnet IP
address range and with our internal interface IP address as destination IP address will be allowed
to come in and/or start a new connection.
Using the connection tracking capability of IPTables, the related MySQL outgoing packets are
automatically allowed back out by the firewall. In this case, our machine is a MySQL server that is
allowed to receive requests from any MySQL client from our internal subnet.
There might be a situation when we would like to access the MySQL server from our internal
subnet using the external interface IP address ($INTERFACE0_IPADDR) as destination IP
address (DST_IPADDR). This is the case when we connect to the MySQL server using its host
name instead of the IP address. Our DNS server might resolve the MySQL server's IP address as
the external interface IP address. In this case, the parameters should be configured like in the
example below:
# Interface 1 MYSQL incoming client request
INTERFACE1_MYSQL_SERVER="yes"
INTERFACE1_MYSQL_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_MYSQL_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_MYSQL_IN_SRC_IPADDR[1]=$NETWORK1
INTERFACE1_MYSQL_IN_DST_IPADDR[1]=$INTERFACE0_IPADDR
As you can see, we have copy/paste the first set of parameters, then changes the destination IP
address (DST_IPADDR) to our external interface IP address ($INTERFACE0_IPADDR) and also
increase the index number.
# Network 1 MYSQL forwarded incoming server request
NETWORK1_MYSQL_SERVER="yes"
NETWORK1_MYSQL_IN_CLI_IPADDR[0]=$ANY_IPADDR
NETWORK1_MYSQL_IN_SRV_IPADDR[0]="192.168.1.1"
In the above example, we enable MySQL incoming client request on our internal subnet
(NETWORK1_MYSQL_SERVER="yes").
Next, we instruct the firewall that in our internal subnet (NETWORK1), all MySQL (MYSQL) incoming
packets (IN) with source IP address (SRC_IPADDR) of any IP address ($ANY_IPADDR) and with
destination IP address (DST_IPADDR) "192.168.1.1" will be allowed to come in and/or start a
new connection.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-282-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
283
In other terms, any incoming MySQL packets with any IP address as source IP address and with
192.168.1.1 as destination IP address will be allowed to come in and/or start a new
connection.
Using the connection tracking capability of IPTables, the related MySQL outgoing packets are
automatically allowed back out by the firewall. In this case, our machine from our internal subnet
that has the IP address 192.168.1.1 is the MySQL server and it is allowed to receive requests
from any MySQL client from the Internet.
NOTE: The MySQL client from the Internet thinks that it talks to our Gateway server, so the actual
destination IP address of the packet is our external interface IP address
($INTERFACE1_IPADDR), but the packet is automatically DNATed to 192.168.1.1.
Pay special attention to the above parameters. We noted that IP address “192.168.1.1” is used
as the value for the incoming client requests with the forwarding feature. This is important, if your
internal workstation IP address is different, you will have to adjust the setting to fit your own IP
address for each of the forwarding definitions.
Step3
Now that our parameters for MySQL service have been correctly entered in the GIPTables
configuration file, we need to restart our GIPTables firewall for the changes to take effect.
• To restart GIPTables on your system, use the following command:
[root@deep /]# /etc/init.d/giptables restart
Well, now we have a better idea about what these cryptic definitions do and how to change them
to fit our needs depending of the type of firewall that we need for our server. Human error is
inevitable and if we entered all the additional parameters into GIPTables by hand, we could in
inadvertently make some errors. To avoid this risk, GIPTables provides through it’s
“giptables.conf.README” file all the possible definitions for available services that can be
used with it.
Therefore, if you need to add some additional services, which do not exist by default in the
giptables.conf file, you can refer to this file to get the parameters to make your service run
with GIPTables Firewall. All you’ll need to do is to cut and paste the required lines into your
GIPTables configuration file and set up each parameter by answering “yes” or “no” to the
questions.
Running the type of GIPTables firewall that you need
All servers should be configured to block the unused ports, even if they are not a firewall
server. This is required for increased security. Imagine that someone gains access to your main
firewall server: if your other servers are not configured to block unused ports, this can result a
serious network security risk. The same is true for local connections; unauthorized employees
can gain access to your other servers from inside the network.
As you should know now, before running GIPTables in your system, you must create a symbolic
link under the /etc directory that points to the GIPTables configuration file suitable for your
system. Once this configuration file exists under your /etc directory, all you have to do is to edit
it and set-up your networking configuration to make it work for you. This is true with all of server
types except for a Gateway Server which differs as explained below.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-283-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
285
Step1
The IP masquerading code will only work if IP forwarding is enabled on your system. This feature
is by default disabled and you can enable it with the following command:
• To enable IPv4 forwarding on your Linux system, use the following command:
Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines:
# Enable packet forwarding (required only for Gateway, VPN, Proxy, PPP)
net.ipv4.ip_forward = 1
You must restart your network for the change to take effect. The command to restart the network
is:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
Step 2
Create the symbolic link to the giptables.conf file that points to the right GIPTables
configuration file suitable for our setup of a Gateway Server.
• This can be accomplished with the following command:
[root@deep /]# ln -s /lib/giptables/conf/giptables.conf.gateway
/etc/giptables.conf
Step3
Once the symbolic link is created, we will edit it to suit our requirements. The text in bold are the
parts of the configuration that must be modified to satisfy your needs.
This is the configuration script file for a Gateway/Proxy Server, it will:
1 Log and limit the amount of incoming dropped packets.
2 Implement the Syn-flood protection.
3 Make sure that all NEW incoming tcp connections are SYN packets.
4 Protect from Spoofing and bad addresses.
5 Allow DNS client requests on internal & external interfaces.
6 Allow FTP client requests on internal & external interfaces.
7 Allow SSH client requests on internal & external interfaces.
8 Allow SSH server requests on the external interface.
9 Allow SMTP client requests on internal & external interfaces.
10 Allow POP3 client requests on the internal interface.
11 Allow POP3S client requests on the internal interface.
12 Allow HTTP client requests on internal & external interfaces.
13 Allow HTTPS client requests on internal & external interfaces.
14 Allow SQUID client requests on internal & external interfaces.
15 Allow SQUID server requests on the external interface.
16 Allow NETBIOS client requests on the internal interface.
17 Allow NETBIOS server requests on the internal interface.
18 Allow TRACEROUTE client requests on internal & external interfaces.
19 Allow TRACEROUTE server requests on internal & external interfaces.
20 Allow ICMP client requests on internal & external interfaces.
21 Allow DHCP client requests on the internal interface.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-285-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
286
If you don’t want some of the services listed in the firewall rules files for the Gateway/Proxy
Server, disable them by saying “no” to the questions. If you want some other services that are not
enabled, simply say, “yes” to the questions. If the service does not exist, add it to your
configuration based on the available examples from the giptables.conf.README file.
• To edit the giptales.conf file, use the following command:
[root@deep /]# vi /etc/giptables.conf
# ----------------------------------------------------------------------------
# GIPTables Firewall v1.1 http://www.giptables.org
# Copyright (C) 2002 Adrian Pascalau <apascalau@openna.com>
# GATEWAY main configuration file
#
# ----------------------------------------------------------------------------
# This file is part of GIPTables Firewall
#
# GIPTables Firewall is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
# ----------------------------------------------------------------------------
# DEBUG
#
DEBUG="off"
# ----------------------------------------------------------------------------
# Some definitions for easy maintenance
# Edit these to suit your system
#
MONOLITIC_KERNEL="no"
# Interface 0: This is our external network interface
# It is directly connected to Internet
INTERFACE0="eth0"
INTERFACE0_IPADDR="x.x.x.x"
ANY_IPADDR="0/0"
# Interface 1: This is our internal network interface
# It is directly connected to our internal Network 1
INTERFACE1="eth1"
INTERFACE1_IPADDR="192.168.1.254"
NETWORK1="192.168.1.0/24"
# Do you need Network Address Translation for your internal network?
NETWORK1_NAT="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-286-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
289
# ----------------------------------------------------------------------------
# Spoofing and bad addresses
#
REFUSE_SPOOFING="yes"
# Refuse incoming packets claiming to be from the ip addresses of our
interfaces
REFUSE_SPOOFING_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_IN_REFUSE_SPOOFING[0]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[0]="no"
NETWORK1_IN_REFUSE_SPOOFING[0]="yes"
REFUSE_SPOOFING_IPADDR[1]=$INTERFACE1_IPADDR
INTERFACE0_IN_REFUSE_SPOOFING[1]="no"
INTERFACE1_IN_REFUSE_SPOOFING[1]="yes"
NETWORK1_IN_REFUSE_SPOOFING[1]="no"
# Refuse incoming packets claiming to be from broadcast-src address range
REFUSE_SPOOFING_IPADDR[2]="0.0.0.0/8"
# If you provide DHCP services on one of your interfaces, do not forget to
# set the following option related to that interface to "no"
INTERFACE0_IN_REFUSE_SPOOFING[2]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[2]="no"
NETWORK1_IN_REFUSE_SPOOFING[2]="yes"
# Refuse incoming packets claiming to be from reserved loopback address range
REFUSE_SPOOFING_IPADDR[3]="127.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[3]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[3]="yes"
NETWORK1_IN_REFUSE_SPOOFING[3]="yes"
# Refuse incoming packets claiming to be from class A private network
REFUSE_SPOOFING_IPADDR[4]="10.0.0.0/8"
INTERFACE0_IN_REFUSE_SPOOFING[4]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[4]="yes"
NETWORK1_IN_REFUSE_SPOOFING[4]="yes"
# Refuse incoming packets claiming to be from class B private network
REFUSE_SPOOFING_IPADDR[5]="172.16.0.0/12"
INTERFACE0_IN_REFUSE_SPOOFING[5]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[5]="yes"
NETWORK1_IN_REFUSE_SPOOFING[5]="yes"
# Refuse incoming packets claiming to be from class C private network
REFUSE_SPOOFING_IPADDR[6]="192.168.0.0/16"
INTERFACE0_IN_REFUSE_SPOOFING[6]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[6]="no"
NETWORK1_IN_REFUSE_SPOOFING[6]="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-289-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
290
# Refuse incoming packets claiming to be from class D, E, and unallocated
REFUSE_SPOOFING_IPADDR[7]="224.0.0.0/3"
INTERFACE0_IN_REFUSE_SPOOFING[7]="yes"
INTERFACE1_IN_REFUSE_SPOOFING[7]="yes"
NETWORK1_IN_REFUSE_SPOOFING[7]="yes"
# ****************************************************************************
# *
# A N Y *
# *
# ****************************************************************************
ACCEPT_ANY="no"
# ****************************************************************************
# *
# D N S *
# *
# ****************************************************************************
ACCEPT_DNS="yes"
# ----------------------------------------------------------------------------
# DNS outgoing client request
#
# Interface 0 DNS outgoing client request
INTERFACE0_DNS_CLIENT="yes"
INTERFACE0_DNS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_DNS_OUT_DST_IPADDR[0]=$ISP_PRIMARY_DNS_SERVER
INTERFACE0_DNS_OUT_UDP_REQUEST[0]="yes"
INTERFACE0_DNS_OUT_TCP_REQUEST[0]="yes"
INTERFACE0_DNS_OUT_SPORT53_REQUEST[0]="no"
INTERFACE0_DNS_OUT_SRC_IPADDR[1]=$INTERFACE0_IPADDR
INTERFACE0_DNS_OUT_DST_IPADDR[1]=$ISP_SECONDARY_DNS_SERVER
INTERFACE0_DNS_OUT_UDP_REQUEST[1]="yes"
INTERFACE0_DNS_OUT_TCP_REQUEST[1]="yes"
INTERFACE0_DNS_OUT_SPORT53_REQUEST[1]="no"
# Network 1 DNS forwarded outgoing client request
NETWORK1_DNS_CLIENT="yes"
NETWORK1_DNS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_DNS_OUT_DST_IPADDR[0]=$ISP_PRIMARY_DNS_SERVER
NETWORK1_DNS_OUT_UDP_REQUEST[0]="yes"
NETWORK1_DNS_OUT_TCP_REQUEST[0]="yes"
NETWORK1_DNS_OUT_SPORT53_REQUEST[0]="no"
# ****************************************************************************
# *
# F T P *
# *
# ****************************************************************************
ACCEPT_FTP="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-290-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
291
# ----------------------------------------------------------------------------
# FTP outgoing client request
#
# Interface 0 FTP outgoing client request
INTERFACE0_FTP_CLIENT="yes"
INTERFACE0_FTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_FTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
INTERFACE0_FTP_OUT_PASIVE[0]="yes"
INTERFACE0_FTP_OUT_ACTIVE[0]="no"
# Network 1 FTP forwarded outgoing client request
NETWORK1_FTP_CLIENT="yes"
NETWORK1_FTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_FTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
NETWORK1_FTP_OUT_PASIVE[0]="yes"
NETWORK1_FTP_OUT_ACTIVE[0]="yes"
# ****************************************************************************
# *
# S S H *
# *
# ****************************************************************************
ACCEPT_SSH="yes"
# ----------------------------------------------------------------------------
# SSH outgoing client request
#
# Interface 0 SSH outgoing client request
INTERFACE0_SSH_CLIENT="yes"
INTERFACE0_SSH_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_SSH_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 SSH forwarded outgoing client request
NETWORK1_SSH_CLIENT="yes"
NETWORK1_SSH_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_SSH_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# SSH incoming client request
#
# Interface 0 SSH incoming client request
INTERFACE0_SSH_SERVER="yes"
INTERFACE0_SSH_IN_SRC_IPADDR[0]=$ANY_IPADDR
INTERFACE0_SSH_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-291-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
292
# Interface 1 SSH incoming client request
INTERFACE1_SSH_SERVER="yes"
INTERFACE1_SSH_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_SSH_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
# ****************************************************************************
# *
# T E L N E T *
# *
# ****************************************************************************
ACCEPT_TELNET="no"
# ----------------------------------------------------------------------------
# TELNET outgoing client request
#
# Interface 0 TELNET outgoing client request
INTERFACE0_TELNET_CLIENT="yes"
INTERFACE0_TELNET_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_TELNET_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 TELNET forwarded outgoing client request
NETWORK1_TELNET_CLIENT="yes"
NETWORK1_TELNET_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_TELNET_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# TELNET incoming client request
#
# Interface 1 TELNET incoming client request
INTERFACE1_TELNET_SERVER="no"
INTERFACE1_TELNET_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_TELNET_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
# ****************************************************************************
# *
# T E L N E T S *
# *
# ****************************************************************************
ACCEPT_TELNETS="no"
# ****************************************************************************
# *
# S M T P *
# *
# ****************************************************************************
ACCEPT_SMTP="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-292-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
293
# ----------------------------------------------------------------------------
# SMTP outgoing client request
#
# Interface 0 SMTP outgoing client request
INTERFACE0_SMTP_CLIENT="yes"
INTERFACE0_SMTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_SMTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 SMTP forwarded outgoing client request
NETWORK1_SMTP_CLIENT="yes"
NETWORK1_SMTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_SMTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# S M T P S *
# *
# ****************************************************************************
ACCEPT_SMTPS="no"
# ----------------------------------------------------------------------------
# SMTPS outgoing client request
#
# Interface 0 SMTPS outgoing client request
INTERFACE0_SMTPS_CLIENT="yes"
INTERFACE0_SMTPS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_SMTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 SMTPS forwarded outgoing client request
NETWORK1_SMTPS_CLIENT="yes"
NETWORK1_SMTPS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_SMTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# P O P 3 *
# *
# ****************************************************************************
ACCEPT_POP3="yes"
# ----------------------------------------------------------------------------
# POP3 outgoing client request
#
# Network 1 POP3 forwarded outgoing client request
NETWORK1_POP3_CLIENT="yes"
NETWORK1_POP3_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_POP3_OUT_DST_IPADDR[0]=$ANY_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-293-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
294
# ****************************************************************************
# *
# P O P 3 S *
# *
# ****************************************************************************
ACCEPT_POP3S="yes"
# ----------------------------------------------------------------------------
# POP3S outging client request
#
# Network 1 POP3S forwarded outging client request
NETWORK1_POP3S_CLIENT="yes"
NETWORK1_POP3S_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_POP3S_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# I M A P *
# *
# ****************************************************************************
ACCEPT_IMAP="no"
# ----------------------------------------------------------------------------
# IMAP outgoing client request
#
# Network 1 IMAP forwarded outgoing client request
NETWORK1_IMAP_CLIENT="yes"
NETWORK1_IMAP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_IMAP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# I M A P S *
# *
# ****************************************************************************
ACCEPT_IMAPS="no"
# ----------------------------------------------------------------------------
# IMAPS outgoing client request
#
# Network 1 IMAPS forwarded outgoing client request
NETWORK1_IMAPS_CLIENT="yes"
NETWORK1_IMAPS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_IMAPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# H T T P *
# *
# ****************************************************************************](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-294-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
295
ACCEPT_HTTP="yes"
# ----------------------------------------------------------------------------
# HTTP outgoing client request
#
# Interface 0 HTTP outgoing client request
INTERFACE0_HTTP_CLIENT="yes"
INTERFACE0_HTTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_HTTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 HTTP forwarded outgoing client request
NETWORK1_HTTP_CLIENT="yes"
NETWORK1_HTTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_HTTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# H T T P S *
# *
# ****************************************************************************
ACCEPT_HTTPS="yes"
# ----------------------------------------------------------------------------
# HTTPS outgoing client request
#
# Interface 0 HTTPS outgoing client request
INTERFACE0_HTTPS_CLIENT="yes"
INTERFACE0_HTTPS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_HTTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 HTTPS forwarded outgoing client request
NETWORK1_HTTPS_CLIENT="yes"
NETWORK1_HTTPS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_HTTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# S Q U I D *
# *
# ****************************************************************************
ACCEPT_SQUID="yes" # Squid in Proxy-Caching Mode
# ----------------------------------------------------------------------------
# SQUID outgoing client request
#
# Interface 0 SQUID outgoing client request
INTERFACE0_SQUID_CLIENT="yes"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-295-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
296
INTERFACE0_SQUID_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_SQUID_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Interface 1 SQUID outgoing client request
INTERFACE1_SQUID_CLIENT="yes"
INTERFACE1_SQUID_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_SQUID_OUT_DST_IPADDR[0]=$NETWORK1
# Network 1 SQUID forwarded outgoing client request
NETWORK1_SQUID_CLIENT="yes"
NETWORK1_SQUID_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_SQUID_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# SQUID incoming client request
#
# Interface 0 SQUID incoming client request
INTERFACE0_SQUID_SERVER="yes"
INTERFACE0_SQUID_IN_SRC_IPADDR[0]=$ANY_IPADDR
INTERFACE0_SQUID_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
# Interface 1 SQUID incoming client request
INTERFACE1_SQUID_SERVER="yes"
INTERFACE1_SQUID_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_SQUID_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR
# ****************************************************************************
# *
# W E B C A C H E *
# *
# ****************************************************************************
ACCEPT_WEBCACHE="no" # Squid in HTTPD-Accelerator Mode
# ****************************************************************************
# *
# N N T P *
# *
# ****************************************************************************
ACCEPT_NNTP="no"
# ----------------------------------------------------------------------------
# NNTP outgoing client request
#
# Network 1 NNTP forwarded outgoing client request
NETWORK1_NNTP_CLIENT="yes"
NETWORK1_NNTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_NNTP_OUT_DST_IPADDR[0]=$ANY_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-296-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
298
INTERFACE0_AUTH_CLIENT="yes"
INTERFACE0_AUTH_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_AUTH_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 AUTH forwarded outgoing client request
NETWORK1_AUTH_CLIENT="yes"
NETWORK1_AUTH_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_AUTH_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# W H O I S *
# *
# ****************************************************************************
ACCEPT_WHOIS="no"
# ----------------------------------------------------------------------------
# WHOIS outgoing client request
#
# Interface 0 WHOIS outgoing client request
INTERFACE0_WHOIS_CLIENT="yes"
INTERFACE0_WHOIS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_WHOIS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 WHOIS forwarded outgoing client request
NETWORK1_WHOIS_CLIENT="yes"
NETWORK1_WHOIS_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_WHOIS_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# F I N G E R *
# *
# ****************************************************************************
ACCEPT_FINGER="no"
# ----------------------------------------------------------------------------
# FINGER outgoing client request
#
# Interface 0 FINGER outgoing client request
INTERFACE0_FINGER_CLIENT="yes"
INTERFACE0_FINGER_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_FINGER_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 FINGER forwarded outgoing client request
NETWORK1_FINGER_CLIENT="yes"
NETWORK1_FINGER_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_FINGER_OUT_DST_IPADDR[0]=$ANY_IPADDR](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-298-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
299
# ****************************************************************************
# *
# N T P *
# *
# ****************************************************************************
ACCEPT_NTP="no"
# ----------------------------------------------------------------------------
# NTP outgoing client request
#
# Interface 0 NTP outgoing client request
INTERFACE0_NTP_CLIENT="yes"
INTERFACE0_NTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_NTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 NTP forwarded outgoing client request
NETWORK1_NTP_CLIENT="no"
NETWORK1_NTP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_NTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ****************************************************************************
# *
# N E T B I O S *
# *
# ****************************************************************************
ACCEPT_NETBIOS="yes"
# ----------------------------------------------------------------------------
# NETBIOS outgoing client request
#
# Interface 1 NETBIOS outgoing client request
INTERFACE1_NETBIOS_CLIENT="yes"
INTERFACE1_NETBIOS_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_NETBIOS_OUT_DST_IPADDR[0]=$NETWORK1
# ----------------------------------------------------------------------------
# NETBIOS incoming client request
#
# Interface 1 NETBIOS incoming client request
INTERFACE1_NETBIOS_SERVER="yes"
INTERFACE1_NETBIOS_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_NETBIOS_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR
# ****************************************************************************
# *
# S Y S L O G *
# *
# ****************************************************************************](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-299-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
300
ACCEPT_SYSLOG="no"
# ----------------------------------------------------------------------------
# SYSLOG outgoing client request
#
# Interface 1 SYSLOG outgoing client request
INTERFACE1_SYSLOG_CLIENT="yes"
INTERFACE1_SYSLOG_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR
INTERFACE1_SYSLOG_OUT_DST_IPADDR[0]=$SYSLOG_SERVER
# ****************************************************************************
# *
# T R A C E R O U T E *
# *
# ****************************************************************************
ACCEPT_TRACEROUTE="yes"
# ----------------------------------------------------------------------------
# TRACEROUTE outgoing client request
#
# Interface 0 TRACEROUTE outgoing client request
INTERFACE0_TRACEROUTE_CLIENT="yes"
INTERFACE0_TRACEROUTE_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_TRACEROUTE_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 TRACEROUTE forwarded outgoing client request
NETWORK1_TRACEROUTE_CLIENT="yes"
NETWORK1_TRACEROUTE_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_TRACEROUTE_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# TRACEROUTE incoming client request
#
# Interface 1 TRACEROUTE incoming client request
INTERFACE1_TRACEROUTE_SERVER="no"
INTERFACE1_TRACEROUTE_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_TRACEROUTE_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
# ****************************************************************************
# *
# I C M P *
# *
# ****************************************************************************
ACCEPT_ICMP="yes"
# ----------------------------------------------------------------------------
# ICMP outgoing client request
#
# Interface 0 ICMP outgoing client request](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-300-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
301
INTERFACE0_ICMP_CLIENT="yes"
INTERFACE0_ICMP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE0_ICMP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# Network 1 ICMP forwarded outgoing client request
NETWORK1_ICMP_CLIENT="yes"
NETWORK1_ICMP_OUT_SRC_IPADDR[0]=$NETWORK1
NETWORK1_ICMP_OUT_DST_IPADDR[0]=$ANY_IPADDR
# ----------------------------------------------------------------------------
# ICMP incoming client request
#
# Interface 1 ICMP incoming client request
INTERFACE1_ICMP_SERVER="no"
INTERFACE1_ICMP_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_ICMP_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
INTERFACE1_ICMP_IN_SRC_IPADDR[1]=$NETWORK1
INTERFACE1_ICMP_IN_DST_IPADDR[1]=$INTERFACE1_IPADDR
# ****************************************************************************
# *
# D H C P *
# *
# ****************************************************************************
ACCEPT_DHCP="yes"
# ----------------------------------------------------------------------------
# DHCP incoming client request
#
# Interface 1 DHCP incoming client request
INTERFACE1_DHCP_SERVER="yes"
# If above option is "yes", do not forget to configure the following
# lines in the "Spoofing and bad addresses" section
# REFUSE_SPOOFING_IPADDR[2]="0.0.0.0/8"
# INTERFACE1_IN_REFUSE_SPOOFING[2]="no"
INTERFACE1_DHCP_IN_SRC_IPADDR[0]=$NETWORK1
INTERFACE1_DHCP_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR
# ****************************************************************************
# *
# E N D *
# *
# ****************************************************************************
DROP_EVERYTHING_FROM_HERE="yes"
# ----------------------------------------------------------------------------
# LOG & DROP everything from here... just in case.
#](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-301-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
302
INTERFACE0_IN_DROP_EVERYTHING_FROM_HERE="yes"
INTERFACE1_IN_DROP_EVERYTHING_FROM_HERE="yes"
NETWORK1_IN_DROP_EVERYTHING_FROM_HERE="yes"
# ----------------------------------------------------------------------------
# End of file
Step 4
Once the configuration file has been configured, it is time to start the firewall on your system.
• To start the firewall on your system, use the following command:
[root@deep /]# /etc/init.d/giptables start
Starting Firewalling Services: [OK]
GIPTables-Firewall Administrative Tools
The commands listed below are some that we use often, but many more exist. Check the manual
page and documentation for more information.
IPTables
The iptables tool is used for the firewall packet filter administration of the system. We can use
it to set up a firewall rules file, as we are doing in this book. Once firewall rules have been created
we can play with its many commands to maintain, and inspect the rules in the kernel.
• To list all rules in the selected chain, use the command:
[root@deep /]# iptables –L
Chain INPUT (policy DROP)
target prot opt source destination
ACCEPT all -- anywhere anywhere
Chain FORWARD (policy DROP)
target prot opt source destination
Chain OUTPUT (policy DROP)
target prot opt source destination
ACCEPT all -- anywhere anywhere
This command will list all rules in the selected chain. If no chain is selected, all chains are listed.
• To list all INPUT rules in the selected chain, use the command:
[root@deep /]# iptables -L INPUT
Chain INPUT (policy DROP)
target prot opt source destination
ACCEPT all -- anywhere anywhere
ACCEPT all -- 192.168.1.0/24 anywhere
DROP all -- 204.254.45.9 anywhere
DROP all -- 187.231.11.5 anywhere
DROP all -- 207.35.78.5 anywhere](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-302-2048.jpg)
![GIPTables Firewall 1
CHAPTER 0
303
• To list all OUTPUT rules in the selected chain, use the command:
[root@deep /]# iptables -L OUTPUT
Chain OUTPUT (policy DROP)
target prot opt source destination
ACCEPT all -- anywhere anywhere
ACCEPT all -- anywhere 192.168.1.0/24
ACCEPT udp -- 207.35.78.5 207.35.78.3 udp
spt:domain dpt:domain
ACCEPT tcp -- 207.35.78.5 207.35.78.3 tcp
spts:1024:65535 dpt:domain
• To list all FORWARD rules in the selected chain, use the command:
[root@deep /]# iptables -L FORWARD
Chain FORWARD (policy DROP)
target prot opt source destination
DROP tcp -- anywhere anywhere tcp
DROP tcp -- anywhere anywhere tcp
DROP all -- !192.168.0.0/24 anywhere
ACCEPT all -- 192.168.0.0/24 anywhere state NEW
ACCEPT all -- !192.168.0.0/24 anywhere state
This of course works only if you have configured Masquerading on your server (for Gateway
servers in general).
• To list all rules in numeric OUTPUT in the selected chain, use the command:
[root@deep /]# iptables –nL
Chain INPUT (policy DROP)
target prot opt source destination
ACCEPT all -- 0.0.0.0/0 0.0.0.0/0
ACCEPT all -- 192.168.1.0/24 0.0.0.0/0
DROP all -- 204.254.45.9 0.0.0.0/0
Chain FORWARD (policy DROP)
target prot opt source destination
Chain OUTPUT (policy DROP)
target prot opt source destination
ACCEPT all -- 0.0.0.0/0 0.0.0.0/0
ACCEPT all -- 0.0.0.0/0 192.168.1.0/24
ACCEPT udp -- 207.35.78.5 207.35.78.5
All the IP addresses and port numbers will be printed in numeric format.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-303-2048.jpg)
![Squid 1
CHAPTER 1
308
These installation instructions assume
Commands are Unix-compatible.
The source path is /var/tmp (note that other paths are possible, as personal discretion).
Installations were tested on OpenNA Linux & Red Hat Linux 7.3.
All steps in the installation will happen using the super-user account “root”.
Whether kernel recompilation may be required: No
Latest Squid version number is 2.4.STABLE6
Packages
The following are based on information listed by Squid as of 2002/03/20. Please regularly check
at www.squid-cache.org for the latest status. We chose to install the required component from
source file because it provides the facility to fine tune the installation.
Source code is available from:
Squid Homepage: http://www.squid-cache.org/
Squid FTP Site: 206.168.0.9
You must be sure to download: squid-2.4.STABLE7-src.tar.gz
Though the procedures given in this chapter are likely to work on all Linux platforms, we have
only tested it on OpenNA Linux and Red Hat Linux.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all files
installed into the system if the package is updated in the future. To solve this problem, it is a good
idea to make a list of files on the system before you install Squid, and one afterwards, and then
compares them using the diff utility of Linux to find out what files are placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Squid1
• And the following one after you install the software:
[root@deep root]# find /* > Squid2
• Then use the following command to get a list of what changed:
[root@deep root]# diff Squid1 Squid2 > Squid-Installed
By doing this, if in the future any upgrade appears, all you have to do is to read the generated list
of what files were added or changed by the program and remove them manually from your
system before installing the new software. We use the /root directory of the system to store all
generated list files.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-308-2048.jpg)
![Squid 1
CHAPTER 1
309
Compiling - Optimizing & Installing Squid
Below are the steps that you must make to configure, compile and optimize the Squid server
software before installing it on your system. First off, we install the program as user 'root' so as to
avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp squid-version-src.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf squid-version-src.tar.gz
Step 2
To avoid security risks, we must create a new user account called “squid” to be the owner of the
Squid database cache files and daemon.
• To create this special Squid user on OpenNA Linux, use the following command:
[root@deep tmp]# groupadd -g 23 squid > /dev/null 2>&1 || :
[root@deep tmp]# useradd -c "Squid Server" -d /var/spool/squid -g 23 -s
/bin/false -u 23 squid > /dev/null 2>&1 || :
• To create this special Squid user on Red Hat Linux, use the following command:
[root@deep tmp]# useradd -c "Squid Server" -u 23 -s /bin/false -r -d
/var/spool/squid squid 2>/dev/null || :
The above command will create a null account, with no password, no valid shell, no files owned-
nothing but a UID and a GID for the program. Remember that Squid daemon does not need to
have a shell account on the server.
Step 3
After that, move into the newly created Squid source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created Squid source directory use the command:
[root@deep tmp]# cd squid-2.4.STABLE7/
Step 4
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to fix some problems and to configure the program for our PATH
environment variable under Linux.
• Edit the acl.c file (vi +651 src/acl.c) and change the line:
*Top = splay_insert(xstrdup(t), *Top, aclDomainCompare);
To read:
*Top = splay_insert(xstrdup(t), *Top, aclHostDomainCompare);](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-309-2048.jpg)
![Squid 1
CHAPTER 1
312
Step 6
Now, we must make a list of all existing files on the system before installing the software, and one
afterwards, then compare them using the diff utility tool of Linux to find out what files are placed
where and finally install the Squid Proxy Server:
[root@deep squid-2.4.STABLE7]# make all
[root@deep squid-2.4.STABLE7]# cd auth_modules
[root@deep auth_modules]# cd NCSA
[root@deep NCSA]# make
[root@deep NCSA]# cd ../PAM
[root@deep PAM]# make
[root@deep PAM]# cd ../SMB
[root@deep SMB]# make SAMBAPREFIX=/usr
[root@deep SMB]# cd ../getpwnam
[root@deep getpwnam]# make
[root@deep getpwnam]# cd
[root@deep root]# find /* > Squid1
[root@deep root]# cd /var/tmp/squid-2.4.STABLE7/
[root@deep squid-2.4.STABLE7]# make install
[root@deep squid-2.4.STABLE7]# cd auth_modules
[root@deep auth_modules]# install -m 4511 PAM/pam_auth /usr/lib/squid/
[root@deep auth_modules]# install -m 0511 NCSA/ncsa_auth /usr/lib/squid/
[root@deep auth_modules]# install -m 0511 SMB/smb_auth /usr/lib/squid/
[root@deep auth_modules]# install –m 0511 getpwnam/getpwnam_auth
/usr/lib/squid/
[root@deep auth_modules]# mkdir -p /var/spool/squid
[root@deep auth_modules]# mkdir -p /var/log/squid
[root@deep auth_modules]# chown -R squid.squid /var/spool/squid/
[root@deep auth_modules]# chown -R squid.squid /var/log/squid/
[root@deep auth_modules]# chmod 0750 /var/spool/squid/
[root@deep auth_modules]# chmod 0750 /var/log/squid/
[root@deep auth_modules]# rm -rf /var/logs/
[root@deep auth_modules]# rm -f /usr/sbin/RunCache
[root@deep auth_modules]# rm -f /usr/sbin/RunAccel
[root@deep auth_modules]# strip /usr/sbin/squid
[root@deep auth_modules]# strip /usr/sbin/client
[root@deep auth_modules]# strip /usr/lib/squid/*
[root@deep auth_modules]# /sbin/ldconfig
[root@deep auth_modules]# cd
[root@deep root]# find /* > Squid2
[root@deep root]# diff Squid1 Squid2 > Squid-Installed
The make all command will compile all source files into executable binaries that can be
installed, and make install will install the binaries and any supporting files into the appropriate
locations. Pay special attention to the authenticator module directory of Squid, we move into this
directory (auth_modules) and compile all authenticator modules that may be needed with
Squid.
Squid authenticator module is required when you want to authorize and authenticate users
before allowing them an access to the Internet or the network. Different authenticator modules
using different techniques are available with Squid. In our compilation, we compile Squid
authenticator modules for PAM, NCSA, SMB, and getpwnam. You don’t need to compile all of them
but only the one that you want to use or nothing if you are not intending to provide user
authentication for Proxy access.
The mkdir command will create two new directories named “squid” under /var/spool and
/var/log directory.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-312-2048.jpg)
![Squid 1
CHAPTER 1
313
The rm command will remove the /var/logs directory since it has been created to handle the
log files for Squid that we have relocated during compile time into the /var/log/squid
directory.
The chown command will change the owner of the /var/spool/squid and /var/log/squid
directories to be owned by the user squid, and the chmod command will make the mode of both
squid directories (0750/drwxr-x---) for security reasons. This means that only squid owner
and group will be able to access these directories and others will not.
Note that we remove the small scripts named “RunCache” and “RunAccel” which take care of
starting Squid in either caching mode or accelerator mode, since we use a better script named
“squid” located under /etc/init.d directory that takes advantage of Linux system V.
Finally, the strip command will reduce the size of the specified binaries for optimum
performance.
Step 7
Once we’ve configured, optimized, compiled, and installed the Squid Proxy Server software, we
can free up some disk space by deleting the program tar archive and the related source directory
since they are no longer needed.
• This is done by using the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf squid-version/
[root@deep tmp]# rm -f squid-version-src.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install Squid. It will also remove the Squid compressed archive from the /var/tmp directory.
Configuring Squid
After Squid has been built and installed successfully on your system, your next step is to
configure and customize all the required parameters in the different Squid configuration files.
Parameters entered into the Squid configuration file (squid.conf) will decide how the Squid
software will run on the server and in which mode (either httpd-accelerator mode or in proxy-
caching mode). This shows us that the installation of Squid under Linux does not care and that
only the configuration of the squid.conf file will decide whether Squid will run in httpd-
accelerator or proxy-caching mode.
/etc/squid/squid.conf: (The Squid Configuration File)
/etc/sysconfig/squid: (The Squid System Configuration File)
/etc/logrotate.d/squid: (The Squid Log Rotation File)
/etc/init.d/squid: (The Squid Initialization File)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-313-2048.jpg)
![Squid 1
CHAPTER 1
324
# via the HTTP, FTP, and gopher protocols) on a system closer to the
# requesting site than to the source. Web browsers can then use the
# local Squid cache as a proxy HTTP server, reducing access time as
# well as bandwidth consumption.
#
# processname: squid
# pidfile: /var/run/squid.pid
# config: /etc/squid/squid.conf
PATH=/usr/bin:/sbin:/bin:/usr/sbin
export PATH
# Source function library.
. /etc/rc.d/init.d/functions
# Source networking configuration.
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
# Check if the squid.conf file is present.
[ -f /etc/squid/squid.conf ] || exit 0
# Source Squid configureation.
if [ -f /etc/sysconfig/squid ]; then
. /etc/sysconfig/squid
else
SQUID_OPTS="-D"
SQUID_SHUTDOWN_TIMEOUT=100
fi
# Determine the name of the squid binary.
[ -f /usr/sbin/squid ] && SQUID=squid
[ -z "$SQUID" ] && exit 0
prog="$SQUID"
# Determine which one is the cache_swap directory
CACHE_SWAP=`sed -e 's/#.*//g' /etc/squid/squid.conf |
grep cache_dir | awk '{ print $3 }'`
[ -z "$CACHE_SWAP" ] && CACHE_SWAP=/var/spool/squid
RETVAL=0
start() {
for adir in $CACHE_SWAP; do
if [ ! -d $adir/00 ]; then
echo -n "init_cache_dir $adir... "
$SQUID -z -F 2>/dev/null
fi
done
echo -n $"Starting $prog: "
$SQUID $SQUID_OPTS 2> /dev/null &
# Trap and prevent certain signals from being sent to the Squid process.
trap '' 1 2 3 18
RETVAL=$?
[ $RETVAL -eq 0 ] && touch /var/lock/subsys/$SQUID
[ $RETVAL -eq 0 ] && echo_success
[ $RETVAL -ne 0 ] && echo_failure
echo
return $RETVAL](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-324-2048.jpg)
![Squid 1
CHAPTER 1
325
}
stop() {
echo -n $"Stopping $prog: "
$SQUID -k check >/dev/null 2>&1
RETVAL=$?
if [ $RETVAL -eq 0 ] ; then
$SQUID -k shutdown &
rm -f /var/lock/subsys/$SQUID
timeout=0
while : ; do
[ -f /var/run/squid.pid ] || break
if [ $timeout -ge $SQUID_SHUTDOWN_TIMEOUT ]; then
echo
return 1
fi
sleep 2 && echo -n "."
timeout=$((timeout+2))
done
echo_success
echo
else
echo_failure
echo
fi
return $RETVAL
}
reload() {
$SQUID $SQUID_OPTS -k reconfigure
}
restart() {
stop
start
}
condrestart() {
[ -e /var/lock/subsys/squid ] && restart || :
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;
reload)
reload
;;
restart)
restart
;;
condrestart)
condrestart
;;
*)
echo $"Usage: $0 {start|stop|reload|restart|condrestart}"
exit 1
esac
exit $?](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-325-2048.jpg)
![Squid 1
CHAPTER 1
326
Step 2
Once the /etc/init.d/squid script file has been created, it is important to make it
executable, change its default permissions, create the necessary links and start it. Making this file
executable will allow the system to run it, changing its default permission is to allow only the root
user to change this file for security reason, and creation of the symbolic links will let the process
control initialization of Linux which is in charge of starting all the normal and authorized processes
that need to run at boot time on your system to start the program automatically for you at each
system reboot.
• To make this script executable and to change its default permissions, use the commands:
[root@deep /]# chmod 700 /etc/init.d/squid
[root@deep /]# chown 0.0 /etc/init.d/squid
• To create the symbolic rc.d links for Squid, use the following commands:
[root@deep /]# chkconfig --add squid
[root@deep /]# chkconfig --level 345 squid on
• To start Squid software manually, use the following command:
[root@deep /]# /etc/init.d/squid start
Starting squid: [OK]
Running Squid with Users Authentication Support
Squid is a complete Proxy software solution that provides many useful features for the
administrator and it is up to us to use and configure them. In this section, we’ll discuss the Proxy
Authentication mechanism that provides a way to authenticate internal users who can use it to
access the Internet. This allows us to add an additional layer of security on which users need to
have rights (the authorization) to access the Internet via the Gateway server in the enterprise.
The Squid source code comes with a few authentication processes. These include:
LDAP: Uses the Lightweight Directory Access Protocol.
NCSA: Uses an NCSA-style username and password file.
MSNT: Uses a Windows NT authentication domain.
PAM: Uses the Linux Pluggable Authentication Modules scheme.
SMB: Uses a SMB server like Windows NT or Samba.
getpwam: Uses the old-fashioned Unix password file.
In order to authenticate users, you need to compile and install one of the above supplied
authentication modules. In our compilation of Squid, we have already included the most
interesting authentication modules, which were NCSA, PAM, SMB, and getpwam.
One problem with all of these authentication modules is the fact that the supplied username and
password are essentially sent in clear text between the browser and the proxy. Therefore,
administrators should not set-up the same username and password that users would use for
account login on the server (if they are allowed) or for email accounts.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-326-2048.jpg)
![Squid 1
CHAPTER 1
329
Step 4
One of the last steps is to create accounts for all users who will be allowed to access the Internet
with Squid after proper authentication with a username and password. Remember, we have to
create null account, with no valid shell for our users.
• To create null user account, use the following command:
[root@deep /]# useradd -s /bin/false gmourani
The above command will create a null account, with no password, no valid shell, no files owned-
nothing but a UID and a GID.
• To set a password for this new user, use the following command:
[root@deep /]# passwd gmourani
Changing password for user gmourani
New UNIX password:
Retype new UNIX password:
passwd: all authentication tokens updated successfully
NOTE: It is NOT important to create a home directory for the users (i.e. /home/gmourani).
Squid with Users Authentication Support can run even if home directories are not created for the
users. All we need for authentication is a username and password. Therefore, a home directory is
futile and since we do not give shell access, there is really no reason for users to have a home
directory on the system.
Step 5
Finally, open your favorite browser and enter the username and password to access the Internet
with Squid as your Proxy Caching Gateway Server.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-329-2048.jpg)
![Squid 1
CHAPTER 1
330
Securing Squid
This section deals specifically with actions we can take to improve and tighten security under
Squid. As with the other chapters, the interesting points here are that we refer to the features
available within the base installed program and not any additional software.
More control on mounting the cache directory of Squid
If you have created the cache directory of Squid in a separate partition your system (i.e.
/var/spool), like we have done during the initial set-up of Linux, then you can use the noexec,
nodev, and nosuid features of Linux to improve and consolidate the cache security.
These features can be set up in the /etc/fstab file to inform the system to not allow execution
of any binaries (noexec), to not interpret character or block special devices (nodev), and to not
allow set-user-identifier or set-group-identifier bits to take effect (nosuid) on the mounted file
system (/var/spool in our example).
Applying this procedure on the partition where the Squid Cache resides will help to eliminate the
possibility of DEV, SUID/SGID, and execution of any binaries that may be in the Squid cache.
Step 1
• Edit the fstab file (vi /etc/fstab) and add in the line that refer to /var/spool file
system the following options after the defaults option as show below:
LABEL=/var/spool /var/spool ext3 defaults,noexec,nodev,nosuid 1 2
Step 2
Once you have made the necessary adjustments to the /etc/fstab file, it is time to inform the
system about the modification.
• This can be accomplished with the following commands:
[root@deep /]# mount /var/spool -oremount
Each file system that has been modified must be remounted with the command as shown
previously. In our example we have modified the /var/spool file system and it is for this reason
that we remount this file system with the above command.
NOTE: If /var/spool is not a file system but just a directory, then the above command obviously
will not work. The ‘-oremount’ option of the Linux ‘mount’ command is used to remount a file
system, which resides on its own partition on your computer.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-330-2048.jpg)
![Squid 1
CHAPTER 1
331
Step 3
• You can verify if the modifications have been correctly applied to the Linux system with
the following command:
[root@deep /]# cat /proc/mounts
/dev/root / ext2 rw 0 0
/proc /proc proc rw 0 0
/dev/sda1 /boot ext3 rw 0 0
/dev/sda9 /chroot ext3 rw 0 0
/dev/sda8 /home ext3 rw 0 0
/dev/sda13 /tmp ext3 rw 0 0
/dev/sda7 /usr ext3 rw 0 0
/dev/sda11 /var ext3 rw 0 0
/dev/sda12 /var/spool ext3 rw,noexec,nodev,nosuid 0 0
none /dev/pts devpts rw 0 0
This command will show you all file system in your Linux server with parameters applied to them.
If you see something like the following, congratulations!
/var/spool /var/spool ext3 rw,noexec,nodev,nosuid 0 0
Immunize the Squid configuration file
As we already know, the immutable bit can be used to prevent deletion, overwriting, or creation of
a symbolic link to a file. Once your squid.conf file has been configured, it’s a good idea to
immunize it with the following command:
[root@deep /]# chattr +i /etc/squid/squid.conf
Banning all Destination addresses except one
In the university libraries, we can often see computers available to students that want to, for their
studies, search for a specific author. Administrators have the task of configuring the computer to
allow only searches on one site where the database of all books and authors are stored.
Therefore, they don’t want to give students access to other sites on the Internet but just to the
database site.
With Squid as the Proxy Server, this can be accomplished easily by adding the right ACL to its
existing configuration file. In the next example, we introduce new ACL rules to our Squid
example configuration file to do just this.
• Edit the squid.conf file (vi /etc/squid/squid.conf) and add/change the
following options. Text in bold are what we have added/changed to our default Squid
example configuration file. Below is what we recommend you enter:
icp_port 0
hierarchy_stoplist cgi-bin ?
acl QUERY urlpath_regex cgi-bin ?
no_cache deny QUERY
cache_mem 128 MB
cache_replacement_policy heap GDSF
memory_replacement_policy heap GDSF
cache_dir diskd /var/spool/squid 2000 16 256
cache_store_log none
acl localhost src 127.0.0.1/255.255.255.255
acl Safe_ports port 80 443 1025-65535
acl CONNECT method CONNECT
acl DATABASE dst 207.78.0.1](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-331-2048.jpg)
![Squid 1
CHAPTER 1
333
http_access deny all
cache_mgr sysadmin@openna.com
cache_effective_user squid
cache_effective_group squid
logfile_rotate 0
log_icp_queries off
cachemgr_passwd my-secret-pass all
buffered_logs on
This new ACL configuration allows all internal clients from the private class C 192.168.1.0 to
access the Internet between 08:30 and 17:30. In this way, we limit the time when our staff can
connect to the Internet to the working hours of the company only.
NOTE: Don’t forget to restart your Squid Proxy Server for the changes to take effect. The order in
which each line appears in the Squid configuration file is important and you have to respect
them. You can’t just add ‘acl’ or ‘http_access’ parameters, wherever you want. Because the
program reads and interprets each access line in the order that they appear.
Optimizing Squid
This section deals specifically with the actions we can take to improve and tighten the
performance of Squid. Note that we refer to the features available within the base installed
program only.
The atime and noatime attributes
The atime and noatime attributes of Linux can be used to get a measurable performance gain
in the Squid cache directory (/var/spool/squid). See the chapter related to the kernel in this
book for more information on this issue.
Physical memory
The most important resource for Squid is physical memory. Your processor does not need to be
ultra-fast. Your disk system will be the major bottleneck, so fast disks are also important for high-
volume caches. Therefore, our recommendation is to use a SCSI disk with at least 512 MB of
physical memory.
Squid Administrative Tools
Now you’ve configured Squid, tightened its security and optimized it for maximum performance,
we can start to play with its utilities.
Stopping Squid process immediately
There are some interesting command line options, especially when we want to stop Squid on the
server. Unlike other services that run as daemons on the system, Squid cannot be stopped
directly and we have to wait for existing connections to terminate before Squid shutdowns.
Sometimes this is not appropriate and we have to stop Squid immediately. This is possible with
the following command:
• To stop Squid process immediately, use the following command:
[root@deep /]# /usr/sbin/squid -k kill
This command sends a KILL signal, which causes the Squid process to exit immediately, without
closing any connections or log files.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-333-2048.jpg)
![Squid 1
CHAPTER 1
335
Step 2
Once the correct ACL have been added to your squid.conf file to allow purge requests on the
Proxy Server, we have to use the client program that comes with Squid to purge an object.
• This procedure can be accomplished with the following commands:
[root@deep /]# client -m PURGE http://www.mydomain.com/
Where <www.mydomain.com> is the object that you want to purge. If the purge was successful,
you will see a “200 OK” response. If the object was not found in the cache, you will see a "404
Not Found" response.
NOTE: The PURGE feature of Squid works only when Users Authentication Support is disabled in
the Squid configuration file. The client program of Squid is not capable of using User
Authentication because it doesn’t have the option to specify a username or password through its
command line.
The cachemgr.cgi program utility of Squid
The cachemgr.cgi utility program, which is available by default when you compile and install
Squid on your system, is designed to run through a web interface, and outputs various statistics
about Squid configuration and performance.
Personally, I don’t recommend you use it. The cachemgr.cgi is a buggy utility, which provides
incomprehensible and cryptic results. Connection to its web interface is not always guaranteed
even if you have the proper configuration. I think that more development and a complete revision
of its functionality is required. Especially when we want to make a remote connection to its web
interface. If you really want to use it, then here are the correct steps you must follow.
This program, by default, is located under the /usr/lib/squid directory, and you have to put it
in your “cgi-bin” directory (eg, /home/httpd/cgi-bin) to be able to use it. Follow the simple
steps below to use this program.
Step 1
The first step will be to move the “cachemgr.cgi” CGI file from the /usr/lib/squid directory
to your /home/httpd/cgi-bin directory.
• This procedure can be accomplished with the following command:
[root@deep /]# mv /usr/lib/squid/cachemgr.cgi /home/httpd/cgi-bin/
Step 2
Once you’ve put the “cachemgr.cgi” program into your /cgi-bin directory, it is time to
change its default mode permission and owner.
• These procedures can be accomplished with the following commands:
[root@deep /]# cd /home/httpd/cgi-bin/
[root@deep cgi-bin]# chown 0.0 cachemgr.cgi
[root@deep cgi-bin]# chmod 0511 cachemgr.cgi](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-335-2048.jpg)
![SquidGuard 1
CHAPTER 2
340
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install SquidGuard, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > SquidGuard1
• And the following one after you install the software:
[root@deep root]# find /* > SquidGuard2
• Then use this command to get a list of what changed:
[root@deep root]# diff SquidGuard1 SquidGuard2 > SquidGuard-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing SquidGuard
Below are the steps that you must take to configure, compile and optimize the SquidGuard
software before installing it onto your system.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• These procedures can be accomplished with the following commands:
[root@deep /]# cp squidGuard-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf squidGuard-version.tar.gz
Step 2
After that, move into the newly created SquidGuard source directory and perform the following
steps to configure and optimize the software for your system.
• To move into the newly created SquidGuard source directory use the command:
[root@deep tmp]# cd squidGuard-1.2.0/
• To configure and optimize SquidGuard use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"
./configure
--prefix=/usr
--sysconfdir=/etc
--localstatedir=/var
--with-sg-config=/etc/squid/squidGuard.conf
--with-sg-logdir=/var/log/squid/squidGuard
--with-sg-dbhome=/var/spool/squid/squidGuard
--with-db-inc=/usr/include
--with-db-lib=/usr/lib](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-340-2048.jpg)
![SquidGuard 1
CHAPTER 2
341
This tells SquidGuard to set itself up for this particular configuration setup with:
- The location of where the squidGuard configuration file must be installed.
- The location of where the squidGuard log file must be installed.
- The location of where the squidGuard database directory must be installed.
- The location of the Berkley DB includes files that SquidGuard need.
- The location of the Berkley DB library that SquidGuard need.
Step 3
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the SquidGuard software:
[root@deep squidGuard-1.2.0]# make
[root@deep squidGuard-1.2.0]# cd
[root@deep root]# find /* > SquidGuard1
[root@deep root]# cd /var/tmp/squidGuard-1.2.0/
[root@deep squidGuard-1.2.0]# make install
[root@deep squidGuard-1.2.0]# cd samples/
[root@deep samples]# install –m 511 squidGuard.cgi /home/httpd/cgi-bin/
[root@deep samples]# cd dest/
[root@deep dest]# mkdir -p /var/spool/squid/squidGuard
[root@deep dest]# chown -R squid.squid /var/spool/squid/squidGuard/
[root@deep dest]# chmod 0750 /var/spool/squid/squidGuard/
[root@deep dest]# chown -R squid.squid /var/log/squid/squidGuard/
[root@deep dest]# chmod 0750 /var/log/squid/squidGuard/
[root@deep dest]# cp blacklists.tar.gz /var/spool/squid/squidGuard/
[root@deep dest]# cd /var/spool/squid/squidGuard/
[root@deep squidGuard]# mkdir -p aggressive
[root@deep squidGuard]# mkdir -p gambling
[root@deep squidGuard]# mkdir -p hacking
[root@deep squidGuard]# mkdir -p porn
[root@deep squidGuard]# chown -R squid.squid aggressive/
[root@deep squidGuard]# chown -R squid.squid gambling/
[root@deep squidGuard]# chown -R squid.squid hacking/
[root@deep squidGuard]# chown -R squid.squid porn/
[root@deep squidGuard]# tar xzpf blacklists.tar.gz
[root@deep squidGuard]# cd blacklists
[root@deep blacklists]# install -m 644 aggressive/domains
../aggressive/
[root@deep blacklists]# install -m 644 aggressive/urls ../aggressive/
[root@deep blacklists]# install -m 644 gambling/domains ../gambling/
[root@deep blacklists]# install -m 644 gambling/urls ../gambling/
[root@deep blacklists]# install -m 644 hacking/domains ../hacking/
[root@deep blacklists]# install -m 644 hacking/urls ../hacking/
[root@deep blacklists]# install -m 644 porn/domains ../porn/
[root@deep blacklists]# install -m 644 porn/urls ../porn/
[root@deep blacklists]# install -m 644 porn/expressions ../porn/
[root@deep blacklists]# cd ..
[root@deep squidGuard]# chown -R squid.squid *
[root@deep squidGuard]# strip /usr/bin/squidGuard
[root@deep squidGuard]# /sbin/ldconfig
[root@deep squidGuard]# rm -rf blacklists blacklists.tar.gz
[root@deep squidGuard]# cd
[root@deep root]# find /* > SquidGuard2
[root@deep root]# diff SquidGuard1 SquidGuard2 > SquidGuard-Installed](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-341-2048.jpg)
![SquidGuard 1
CHAPTER 2
342
The make command will compile all source files into executable binaries that can be installed,
and make install will install the binaries and any supporting files into the appropriate locations.
The “install -m 0511” command will install the CGI program called squidGuard.cgi
(squidGuard.cgi is a small script, which is used to explain to the user that the URL is blocked
and by which rule set) into your cgi-bin directory.
The “mkdir -p” command will create the SquidGuard directory and subdirectories to store
database filter files to run with squidGuard, the “chown and chmod” commands will set the
appropriate mode and ownership permissions to the squidGuard directory and it’s
subdirectories. The “tar” command will untar the blacklists.tar.gz compressed archive
containing all the filter files and the “install -m 644” commands will install the entire filter files
to their appropriate directories.
Finally, the strip command will reduce the size of the specified binaries for optimum
performance and the “rm -rf” commands will remove the blacklists directory and archive
file that we no longer need on our system.
Step 4
Once the configuration, optimization, compilation, and installation of the SquidGuard software
have been accomplished, we can free up some disk space by deleting the program tar archive
and the related source directory since they are no longer needed.
• To delete SquidGuard and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf squidGuard-version/
[root@deep tmp]# rm -f squidGuard-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install SquidGuard. It will also remove the SquidGuard compressed archive from the
/var/tmp directory.
Configuring SquidGuard
After SquidGuard has been built and installed successfully on your system, your next step is to
configure and customize the squidGuard.conf file to suit your needs.
The parameters entered into the squidGuard configuration file (squidGuard.conf) will decide
how the ACL should be applied and on which users, hosts, IP addresses, times, dates,
destination, source, etc.
/etc/squid/squidGuard.conf (The SquidGuard Configuration File)
/home/httpd/cgi-bin/squidGuard.cgi (The squidGuard.cgi File)
/etc/squid/squidGuard.conf: The SquidGuard Configuration File
The /etc/squid/squidGuard.conf file is the main and only configuration file for
squidGuard. It is not difficult to configure or understand but we have to understand the rules
order if we want to have a working SquidGuard configuration file.
The SquidGuard configuration file (squidGuard.conf) has a structure that must be followed
during its configuration. Next we show you the recommended structure for the configuration file
and the order in which declarations are supposed to appear. There are six different possible
declarations where five are optional and one required.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-342-2048.jpg)
![SquidGuard 1
CHAPTER 2
351
NOTE: If you receive an error message here, it is likely because you have forgot to configure the
squidguard.cgi program to fit your domain name information. Please edit the
“/home/httpd/cgi-bin/suidguard.cgi” program and make the appropriate changes.
Optimizing SquidGuard
This section deals specifically with the actions we can take to improve and tighten the
performance of SquidGuard. Note that we refer to the features available within the base
installed program only.
Creating a prebuilt database
Ok, your SquidGuard program is running and you’re happy to see that it works, but wait a
minute; it is possible to make SquidGuard runs faster simply by converting its filter files (where
blacklisted domains and urls reside) into a db filter file.
The default filter files used with SquidGuard are in plain text format, and SquidGuard needs to
parse all the lines inside the filter file to decide if domains/url’s can be allowed or not. There is a
better method that gives the same result and also runs faster by converting all of its filter files into
a db file.
Step1
The first step to accomplish this conversion will be to use the “-C” command of SquidGuard.
This command simply converts the text file into a db file.
• To convert your filter text files into a db file,use the following commands:
[root@deep /]# cd /var/spool/squid/squidGuard/
[root@deep /]# squidGuard -C aggressive/domains
[root@deep /]# squidGuard -C aggressive/urls
[root@deep /]# squidGuard -C gambling/domains
[root@deep /]# squidGuard -C gambling/urls
[root@deep /]# squidGuard -C hacking/domains
[root@deep /]# squidGuard -C hacking/urls
[root@deep /]# squidGuard -C porn/domains
[root@deep /]# squidGuard -C porn/urls
The above commands, will convert a domainlist or urllist from plain text file to a prebuilt database.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-351-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
359
These installation instructions assume
Commands are Unix-compatible.
The source path is /usr/src (note that other paths are possible, at personal discretion).
Installations were tested on OpenNA Linux and Red Hat Linux 7.3.
All steps in the installation will happen using the super-user account “root”.
Whether kernel recompilation may be required: Yes
Latest FreeS/WAN version number is 1.95
The procedures given in this chapter are likely to work on all Linux platforms, but we have only
tested it on OpenNA Linux and Red Hat Linux.
Packages
The following is based on information listed by FreeS/WAN as of 2001/02/04. Please check
http://www.freeswan.org/ regularly for the latest status. We chose to install from source because it
provides the facility to fine tune the installation.
Source code is available from:
FreeS/WAN Homepage Site: http://www.freeswan.org/
FreeS/WAN FTP Site: 194.109.6.26
You must be sure to download: freeswan-1.95.tar.gz
Prerequisites
Linux FreeS/WAN requires that the software below is already installed on your system to be able
to run and work successfully. If this is not the case, you must install it from your Linux CD-ROM or
source archive file. Please make sure you have this program installed on your machine before
you proceed with this chapter.
gmp is required to compile and make FreeS/WAN works in your system.
NOTE: Not installing the GMP library will make pluto fail to compile on your server.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install FreeS/WAN, and
then one afterwards, and then compares them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Freeswan1
• And the following one after you install the software:
[root@deep root]# find /* > Freeswan2
• Then use this command to get a list of what changed:
[root@deep root]# diff Freeswan1 Freeswan2 > Freeswan-Installed](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-359-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
360
Compiling - Optimizing & Installing FreeS/WAN
Below are the required steps that you must make to compile and optimize the FreeS/WAN
software before installing it into your Linux system.
Step 1
The installation of IPSEC FreeS/WAN software requires some modification of your original
kernel since some parts of FreeS/WAN must be included and incorporated in your kernel before
you can use it. For this reason the first step in installing FreeS/WAN is to go to the Linux Kernel
section in this book and follow the instructions on how to install the Linux Kernel on your system
(even if you have already done this before) and come back to this section after you have
executed the “make dep; make clean” commands, but before the “make bzImage”
command in the Linux Kernel chapter.
Step 2
Once your kernel is configured and you download the FreeS/WAN program from the main
software site you must copy it to the /usr/src directory and change to this location before
expanding the archive. Putting FreeS/WAN under /usr/src/linux will confuse the links,
therefore, expand the software under /usr/src and never under /usr/src/linux directory.
• These procedures can be accomplished with the following commands:
[root@deep /]# cp freeswan-version.tar.gz /usr/src/
[root@deep /]# cd /usr/src/
[root@deep src]# tar xzpf freeswan-version.tar.gz
Step 3
After that, move into the newly created FreeS/WAN directory then configure, compile and
optimize it.
• To move into the top-level directory of FreeS/WAN distribution use the command:
[root@deep src]# cd freeswan-1.95/
Step 4
You must modify the Makefile.inc under the FreeS/WAN source directory to specify
installation paths and optimization parameters. We must modify this file to be compliant with
Linux file system structure, add optimization parameters for our specific processor architecture
and install FreeS/WAN files under our PATH environment variable.
• Edit the Makefile.inc file (vi Makefile.inc) and change all of the targeted lines in
the order shown below:
PUBDIR=$(DESTDIR)/usr/local/sbin
To read:
INC_USRLOCAL=/usr
REALPRIVDIR=/usr/local/lib/ipsec
To read:
INC_MANDIR=/share/man](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-360-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
361
MANTREE=$(DESTDIR)/usr/local/man
To read:
USERCOMPILE=-O2
CONFDIR=$(DESTDIR)/etc
To read:
KLIPSCOMPILE=-O2
All of the above changes, will relocate all files related to the FreeS/WAN software to the
destination target directories we have chosen. We also add optimization parameters related to
the type of processor that we use in our system for better performance.
Step 5
Once the modifications have been made to the source file of FreeS/WAN as described in step 4,
we need to patch the pre-configured Linux Kernel to include FreeS/WAN support.
• This procedure can be accomplished with the following command:
[root@deep freeswan-1.95]# make ogo
echo "===============" >>out.kpatch
echo "`date` `cd /usr/src/linux ; pwd`" >>out.kpatch
make _patches2.3 >>out.kpatch
…………
The make ogo command is what we use to patch the kernel. It will automatically start the kernel
configuration part for the second time and will let you answer all kernel configuration questions
before compilation and integration of its component into the kernel.
During the second kernel configuration, be sure that your kernel has been built with FreeS/WAN
support enabled. A new section related to Frees/WAN support named “IPSec options
(FreeS/WAN)” should appear in your kernel configuration after you have patched the kernel with
the FreeS/WAN program as described above. You need ensure that you have answered Y to the
following questions under the new kernel section: IPSec options (FreeS/WAN).
IP Security Protocol (FreeS/WAN IPSEC) (CONFIG_IPSEC) [Y/n/?]
*
* IPSec options (FreeS/WAN)
*
IPSEC: IP-in-IP encapsulation (tunnel mode) (CONFIG_IPSEC_IPIP) [Y/n/?]
IPSEC: Authentication Header (CONFIG_IPSEC_AH) [Y/n/?]
HMAC-MD5 authentication algorithm (CONFIG_IPSEC_AUTH_HMAC_MD5) [Y/n/?]
HMAC-SHA1 authentication algorithm (CONFIG_IPSEC_AUTH_HMAC_SHA1) [Y/n/?]
IPSEC: Encapsulating Security Payload (CONFIG_IPSEC_ESP) [Y/n/?]
3DES encryption algorithm (CONFIG_IPSEC_ENC_3DES) [Y/n/?]
IPSEC: IP Compression (CONFIG_IPSEC_IPCOMP) [Y/n/?]
IPSEC Debugging Option (CONFIG_IPSEC_DEBUG) [Y/n/?]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-361-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
362
NOTE: All the customization you made to your kernel the first time you ran the make config,
make dep, and make clean commands will be preserved, so you don’t need to reconfigure every
part of your kernel; Just the new section added by FreeS/WAN named “IPSec options
(FreeS/WAN)” is required, as shown above.
Some networking options will get turned on automatically, even if you previously turned them off;
This is because IPSEC needs them. Whichever configuration program you are using, you should
pay careful attention to a few issues: in particular, do NOT disable any of the following under the
“Networking Options” of your kernel configuration:
Kernel/User netlink socket (CONFIG_NETLINK) [Y/n/?]
Netlink device emulation (CONFIG_NETLINK_DEV) [Y/n/?]
Step 6
Once the make ogo command is completed, your FreeS/WAN software and Linux kernel with
FreeS/WAN support is ready to be installed on your server. We must make a list of files on the
system before you install the software and the kernel, and one afterwards, then compare them
using the diff utility to find out what files are placed where and finally install FreeS/WAN and
the new kernel with FreeS/WAN support in your server:
[root@deep freeswan-1.95]# cd
[root@deep root]# find /* > Freeswan1
[root@deep root]# cd /usr/src/freeswan-1.95/
[root@deep freeswan-1.95]# make install
[root@deep freeswan-1.95]# cd
[root@deep root]# find /* > Freeswan2
[root@deep root]# diff Freeswan1 Freeswan2 > Freeswan-Installed
The make install command will install all FreeS/WAN and kernel components together to the
appropriated location on your server.
Step 7
At this stage of your installation of FreeS/WAN, you must follow the rest of the instructions in the
Linux Kernel chapter of this book as normal to install the kernel. At this point, after you have
copied and installed your new kernel image, system.map, or modules (if necessary), and set the
lilo.conf file to load the new kernel, you must edit and customize the configuration files
related to FreeS/WAN “ipsec.conf” and “ipsec.secrets” before rebooting your system.
Step 8
Once the compilation, optimization and installation of the software have been finished, we can
free up some disk space by deleting the program tar archive and the related source directory
since they are no longer needed.
• To delete FreeS/WAN and its related source directory, use the following commands:
[root@deep /]# cd /usr/src/
[root@deep src]# rm -rf freeswan-version/
[root@deep src]# rm -f freeswan-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-362-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
366
/etc/ipsec.secrets: The FreeS/WAN File to store Secret Keys
The file ipsec.secrets stores the secrets used by the pluto daemon to authenticate
communication between both gateways. Two different kinds of secrets can be configured in this
file, preshared secrets and RSA private keys. You must check the permissions of this file to be
sure that the super-user “root” owns the file, and its permissions are set to block all access by
others.
Step 1
An example secret is supplied in the ipsec.secrets file by default. You should change it by
creating your own. With automatic keying you may have a shared secret up to 256 bits, which is
then used during the key exchanges to make sure a man in the middle attack does not occur.
• To create a new shared secret, use the following commands:
[root@deep /]# ipsec ranbits 256 > temp
New, random keys are created with the ranbits(8) utility in the file named “temp”. The ranbits
utility may pause for a few seconds if not enough entropy is available immediately. Don’t forget to
delete the temporary file as soon as you are done with it.
Step 2
Now that our new shared secret key has been created in the “temp” file, we must put it in the
/etc/ipsec.secrets file. When editing the ipsec.secrets file, you should see something
like the following appearing in your text editor. Each line has the IP addresses of the two
gateways plus the secret. It should look something like this:
# This file holds shared secrets which are currently the only inter-Pluto
# authentication mechanism. See ipsec_pluto(8) manpage. Each secret is
# (oversimplifying slightly) for one pair of negotiating hosts.
# The shared secrets are arbitrary character strings and should be both
# long and hard to guess.
# Note that all secrets must now be enclosed in quotes, even if they have
# no white space inside them.
10.0.0.1 11.0.0.1 "jxVS1kVUTTulkVRRTnTujSm444jRuU1mlkklku2nkW3nnVu
V2WjjRRnulmlkmU1Run5VSnnRT"
• Edit the ipsec.secrets file (vi /etc/ipsec.secrets) and change the default
secrets keys:
10.0.0.1 11.0.0.1 " jxVS1kVUTTulkVRRTnTujSm444jRuU1mlkklku2nkW3nnVu
V2WjjRRnulmlkmU1Run5VSnnRT "
To read:
208.164.186.1 208.164.186.2
"0x9748cc31_2e99194f_d230589b_cd846b57_dc070b01_74b66f34_19c40a1a_804906ed"
Where “208.164.186.1" and “208.164.186.2" are the IP addresses of the two gateways and
"0x9748cc31_2e99194f_d230589b_cd846b57_dc070b01_74b66f34_19c40a1a_804906
ed" (note that the quotes are required) is the shared secret we have generated above with the
command “ipsec ranbits 256 > temp” in the “temp” file.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-366-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
367
Step 3
The files ipsec.conf, and ipsec.secrets must be copied to the second gateway machine so
as to be identical on both ends. The only exception to this is the ipsec.conf file, which must
have in it a section labeled by the line config setup with the correct interface settings for the
second gateway, if they differ from the first. The ipsec.secrets file, contrary to the RSA
private key, should have the same-shared secrets on the two gateways.
WARNING: The file /etc/ipsec.secrets should have permissions rw------- (600) and be
owned by the super-user “root”. The file /etc/ipsec.conf is installed with permissions rw-r-
-r— (644) and must be owned also by “root”.
Configuring RSA private keys secrets
Recall that currently with FreeSWAN software there are two kinds of secrets: preshared secrets
and RSA private keys. The preshared secrets are what we have configured in our ipsec.conf
and ipsec.secrets example, above. Some people may prefer to use RSA private keys for
authentication by the Pluto daemon of the other hosts. If you are in this situation, you will have
to make some minor modifications to your ipsec.conf and ipsec.secrets files as described
in the following steps:
You need to create a separate RSA key for *each* gateway. Each one gets its private key in its
own ipsec.secrets file, and the public keys go in leftrsasigkey and rightrsasigkey
parameters in the conn description of ipsec.conf file, which goes to both.
Step 1
Create a separate RSA key for *each* gateway:
• On the first gateway (e.i. deep), use the following commands:
[root@deep /]# cd /
[root@deep /]# ipsec rsasigkey --verbose 1024 > deep-keys
computing primes and modulus...
getting 64 random bytes from /dev/random
looking for a prime starting there
found it after 30 tries
getting 64 random bytes from /dev/random
looking for a prime starting there
found it after 230 tries
swapping primes so p is the larger
computing (p-1)*(q-1)...
computing d...
computing exp1, exp1, coeff...
output...
• On the second gateway (e.i. mail), use the following commands:
[root@mail /]# cd /
[root@mail /]# ipsec rsasigkey --verbose 1024 > mail-keys
computing primes and modulus...
getting 64 random bytes from /dev/random
looking for a prime starting there
found it after 30 tries
getting 64 random bytes from /dev/random
looking for a prime starting there
found it after 230 tries
swapping primes so p is the larger
computing (p-1)*(q-1)...](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-367-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
369
RSA keys for gateway deep (deep-keys):
[root@deep /]# cd /
[root@deep /]# vi deep-keys
# 1024 bits, Fri Feb 4 05:05:19 2000
# for signatures only, UNSAFE FOR ENCRYPTION
#pubkey=0x010395daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74
bcfb51a6ecc08890d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb
4d7689f2da7a69199e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16
a76bab772888f1fd71aa08f08502a141b611f
Modulus:
0x95daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74bcfb51a6ecc08890
d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb4d7689f2da7a6919
9e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16a76bab772888f1fd
71aa08f08502a141b611f
PublicExponent: 0x03
# everything after this point is secret
PrivateExponent:
0x63e74967eaea2025c98c69f6ef0753a6a3ff6764157dbdf1f50013471324dd352366f48805b0b
37f232384b2b52ce2ee85d173468b62eaa052381a9588a317b3a1324d01a531a41fa7add6c5efbd
d88f4718feed2bc0246be924e81bb90f03e49ceedf7af0dd48f06f265b519600bd082c6e6bd27ea
a71cc0288df1ecc3b062b
Prime1:
0xc5b471a88b025dd09d4bd7b61840f20d182d9b75bb7c11eb4bd78312209e3aee7ebfe632304db
6df5e211d21af7fee79c5d45546bea3ccc7b744254f6f0b847f
Prime2:
0xc20a99feeafe79767122409b693be75f15e1aef76d098ab12579624aec708e85e2c5dd62080c3
a64363f2f45b0e96cb4aef8918ca333a326d3f6dc2c72b75361
Exponent1:
0x83cda11b0756e935be328fcebad5f6b36573bcf927a80bf2328facb6c0697c9eff2a9976cade7
9ea3ec0be1674fff4512e8d8e2f29c2888524d818df9f5d02ff
Exponent2:
0x815c66a9f1fefba44b6c2b124627ef94b9411f4f9e065c7618fb96dc9da05f03ec83e8ec055d7
c42ced4ca2e75f0f3231f5061086ccd176f37f9e81da1cf8ceb
Coefficient:
0x10d954c9e2b8d11f4db1b233ef37ff0a3cecfffad89ba5d515449b007803f577e3bd7f0183ced
dfd805466d62f767f3f5a5731a73875d30186520f1753a7e325
RSA keys for gateway mail (mail-keys):
[root@mail /]# cd /
[root@mail /]# vi mail-keys
# 1024 bits, Fri Feb 4 04:46:59 2000
# for signatures only, UNSAFE FOR ENCRYPTION
#pubkey=0x01037631b81f00d5e6f888c542d44dbb784cd3646f084ed96f942d341c7c4686c
bd405b805dc728f8697475f11e8b1dd797550153a3f0d4ff0f2b274b70a2ebc88f073748d1c1c88
21dc6be6a2f0064f3be7f8e4549f8ab9af64944f829b014788dd202cf7d2e320cab666f5e7a197e
64efe0bfee94e92ce4dad82d5230c57b89edf
Modulus:
0x7631b81f00d5e6f888c542d44dbb784cd3646f084ed96f942d341c7c4686cbd405b805dc728f8
697475f11e8b1dd797550153a3f0d4ff0f2b274b70a2ebc88f073748d1c1c8821dc6be6a2f0064f
3be7f8e4549f8ab9af64944f829b014788dd202cf7d2e320cab666f5e7a197e64efe0bfee94e92c
e4dad82d5230c57b89edf
PublicExponent: 0x03
# everything after this point is secret
PrivateExponent:
0x4ecbd014ab3944a5b08381e2de7cfadde242f4b03490f50d737812fd8459dd3803d003e84c5fa
f0f84ea0bf07693a64e35637c2a08dff5f721a324b1747db09f62c871d5e11711251b845ae76753
d4ef967c494b0def4f5d0762f65da603bc04c41b4c6cab4c413a72c633b608267ae2889c162a3d5
bc07ee083b1c6e038400b](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-369-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
370
Prime1:
0xc7f7cc8feaaac65039c39333b878bffd8f95b0dc22995c553402a5b287f341012253e9f25b839
83c936f6ca512926bebee3d5403bf9f4557206c6bbfd9aac899
Prime2:
0x975015cb603ac1d488dc876132d8bc83079435d2d3395c03d5386b5c004eadd4d7b01b3d86aad
0a2275d2d6b791a2abe50d7740b7725679811a32ca22db97637
Exponent1:
0x854fddb5471c84357bd7b777d0507ffe5fb92092c1bb92e37801c3cc5aa22b5616e29bf6e7ad1
028624a486e0c619d47f428e2ad2a6a2e3a159d9d2a911c85bb
Exponent2:
0x64e00e87957c81385b3daf9621e5d302050d7937377b92ad38d04792aadf1e8de52012290471e
06c1a3e1e47a61171d435e4f807a4c39a6561177316c9264ecf
Coefficient:
0x6f087591becddc210c2ee0480e30beeb25615a3615203cd3cef65e5a1d476fd9602ca0ef10d9b
858edb22db42c975fb71883a470b43433a7be57df7ace4a0a3f
Extract and copy the public RSA key files of deep and mail to your ipsec.conf files as shown
below. You can locate the line related to the public key by a sentence beginning with the
commented-out: “#pubkey=” line.
# sample connection
conn deep-mail
left=208.164.186.1
leftsubnet=192.168.1.0/24
leftnexthop=205.151.222.250
right=208.164.186.2
rightsubnet=192.168.1.0/24
rightnexthop=205.151.222.251
keyingtries=0
auth=ah
authby=rsasig
leftrsasigkey=0x010395daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801ce
a9cb74bcfb51a6ecc08890d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef9
8a7f29edcb4d7689f2da7a69199e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5d
eac3b19d561c16a76bab772888f1fd71aa08f08502a141b611f
rightrsasigkey=0x01037631b81f00d5e6f888c542d44dbb784cd3646f084ed96f942d341c
7c4686cbd405b805dc728f8697475f11e8b1dd797550153a3f0d4ff0f2b274b70a2ebc88f07
3748d1c1c8821dc6be6a2f0064f3be7f8e4549f8ab9af64944f829b014788dd202cf7d2e320
cab666f5e7a197e64efe0bfee94e92ce4dad82d5230c57b89edf
auto=start
NOTE: Don’t forget that, in this example, the “leftrsasigkey=” parameter contains the public
key of deep and the “rightrsasigkey=” parameter contains the public key of mail.
Step 3
Modify your /etc/ipsec.secrets files to use RSA private keys in *each* gateway:
Edit your original ipsec.secrets file (vi /etc/ipsec.secrets) and add the RSA private
key for authentication on both gateways:
The ipsec.secrets file for gateway deep:
[root@deep /]# vi /etc/ipsec.secrets
208.164.186.1 208.164.186.2
"0x9748cc31_2e99194f_d230589b_cd846b57_dc070b01_74b66f34_19c40a1a_804906ed"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-370-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
371
You must change your original ipsec.secrets file as shown above to look like the following on
both gateways. It is important to note that the private keys are not the same on both gateways,
deep and mail. The private key for deep comes from the RSA key file “deep-keys”, while the
private key for mail comes from the RSA key file “mail-keys”:
208.164.186.1 208.164.186.2: RSA {
Modulus:
0x95daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74bcfb51a6ecc08890
d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb4d7689f2da7a6919
9e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16a76bab772888f1fd
71aa08f08502a141b611f
PublicExponent: 0x03
# everything after this point is secret
PrivateExponent:
0x63e74967eaea2025c98c69f6ef0753a6a3ff6764157dbdf1f50013471324dd352366f48805b0b
37f232384b2b52ce2ee85d173468b62eaa052381a9588a317b3a1324d01a531a41fa7add6c5efbd
d88f4718feed2bc0246be924e81bb90f03e49ceedf7af0dd48f06f265b519600bd082c6e6bd27ea
a71cc0288df1ecc3b062b
Prime1:
0xc5b471a88b025dd09d4bd7b61840f20d182d9b75bb7c11eb4bd78312209e3aee7ebfe632304db
6df5e211d21af7fee79c5d45546bea3ccc7b744254f6f0b847f
Prime2:
0xc20a99feeafe79767122409b693be75f15e1aef76d098ab12579624aec708e85e2c5dd62080c3
a64363f2f45b0e96cb4aef8918ca333a326d3f6dc2c72b75361
Exponent1:
0x83cda11b0756e935be328fcebad5f6b36573bcf927a80bf2328facb6c0697c9eff2a9976cade7
9ea3ec0be1674fff4512e8d8e2f29c2888524d818df9f5d02ff
Exponent2:
0x815c66a9f1fefba44b6c2b124627ef94b9411f4f9e065c7618fb96dc9da05f03ec83e8ec055d7
c42ced4ca2e75f0f3231f5061086ccd176f37f9e81da1cf8ceb
Coefficient:
0x10d954c9e2b8d11f4db1b233ef37ff0a3cecfffad89ba5d515449b007803f577e3bd7f0183ced
dfd805466d62f767f3f5a5731a73875d30186520f1753a7e325
}
The ipsec.secrets file for gateway mail:
[root@mail /]# vi /etc/ipsec.secrets
208.164.186.1 208.164.186.2: RSA {
Modulus:
0x95daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74bcfb51a6ecc08890
d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb4d7689f2da7a6919
9e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16a76bab772888f1fd
71aa08f08502a141b611f
PublicExponent: 0x03
# everything after this point is secret
PrivateExponent:
0x63e74967eaea2025c98c69f6ef0753a6a3ff6764157dbdf1f50013471324dd352366f48805b0b
37f232384b2b52ce2ee85d173468b62eaa052381a9588a317b3a1324d01a531a41fa7add6c5efbd
d88f4718feed2bc0246be924e81bb90f03e49ceedf7af0dd48f06f265b519600bd082c6e6bd27ea
a71cc0288df1ecc3b062b
Prime1:
0xc5b471a88b025dd09d4bd7b61840f20d182d9b75bb7c11eb4bd78312209e3aee7ebfe632304db
6df5e211d21af7fee79c5d45546bea3ccc7b744254f6f0b847f
Prime2:
0xc20a99feeafe79767122409b693be75f15e1aef76d098ab12579624aec708e85e2c5dd62080c3
a64363f2f45b0e96cb4aef8918ca333a326d3f6dc2c72b75361
Exponent1:
0x83cda11b0756e935be328fcebad5f6b36573bcf927a80bf2328facb6c0697c9eff2a9976cade7
9ea3ec0be1674fff4512e8d8e2f29c2888524d818df9f5d02ff](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-371-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
372
Exponent2:
0x815c66a9f1fefba44b6c2b124627ef94b9411f4f9e065c7618fb96dc9da05f03ec83e8ec055d7
c42ced4ca2e75f0f3231f5061086ccd176f37f9e81da1cf8ceb
Coefficient:
0x10d954c9e2b8d11f4db1b233ef37ff0a3cecfffad89ba5d515449b007803f577e3bd7f0183ced
dfd805466d62f767f3f5a5731a73875d30186520f1753a7e325
}
Authentication by RSA Signatures requires that each host have its own private key. The key part
of an entry may start with a token indicating the kind of key. “RSA” signifies RSA private key and
“PSK” (which is the default) signifies PreShared Key. Since “PSK” is the default, we must specify
“RSA”, so that we’ll be able to use RSA private keys in this file (ipsec.secrets). The super-user
“root” should own the file ipsec.secrets, and its permissions should be set to block all access
by others.
Requiring network setup for IPSec
There are some considerations you must ensure are correct before running FreeS/WAN
software. These considerations are important if you don’t want to receive error messages during
start up of your VPN. The following are the steps to follow:
Step1
You will need to enable TCP/IP forwarding on the both gateway servers. In Linux, this is
accomplished by adding the following line:
• To enable IPv4 forwarding on your Linux system, edit the /etc/sysctl.conf file (vi
/etc/sysctl.conf) and add the following line:
# Enable/Disable packet forwarding
net.ipv4.ip_forward = 1
You must restart your network for the change to take effect. The command to restart the network
is the following:
• To restart all network devices manually on your system, use the following command:
[root@deep /]# /etc/init.d/network restart
Setting network parameters [OK]
Bringing up interface lo [OK]
Bringing up interface eth0 [OK]
Bringing up interface eth1 [OK]
Step 2
Recall that automatically keyed connections use keys automatically generated by the Pluto key
negotiation daemon. The pluto daemon will start up, try to connect to the Pluto daemon at the
other end of the tunnel, and establish a connection. For this reason, an IPSEC gateway should
have packet filters rules (in the firewall script file) permitting the following protocols to traverse the
gateway when talking to other IPSEC gateway:
UDP port 500 for IKE implemented by the Pluto daemon
Protocol 50 for ESP encryption and/or authentication
Protocol 51 for AH packet-level authentication
See the GIPTables chapter in this book and the GIPTables manual for the correct rules to add
to your firewall on both gateway machines to allow IPSEC packets to traverse the remote network
gateway to your network gateway and vice versa.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-372-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
373
Step 3
The rp_filter subsystem (related to IP spoofing protection) must be turned off on both
gateways for IPSEC to work properly. This is accomplished by checking if the value 0 (off) is set
in the /proc/sys/net/ipv4/conf/ipsec0/rp_filter and
/proc/sys/net/ipv4/conf/eth0/rp_filter files respectively:
• To check if the value 0 (off) is set in the rp_filter files, use the commands:
[root@deep /]# cat /proc/sys/net/ipv4/conf/ipsec0/rp_filter
0
[root@deep /]# cat /proc/sys/net/ipv4/conf/eth0/rp_filter
0
NOTE: The subdirectory “ipsec0” in our example will be created only after the reboot of your
system. So you may check the value of the “rp_filter” file in the “ipsec0” directory after your
system has been restarted.
• To set the value 0 (off) in the both rp_filter files manually, use the commands:
[root@deep /]# echo 0 > /proc/sys/net/ipv4/conf/ipsec0/rp_filter
[root@deep /]# echo 0 > /proc/sys/net/ipv4/conf/eth0/rp_filter
Also you can put lines like the following in your firewall script files
/etc/rc.d/init.d/iptables on both gateways to automatically set these values to 0 (off)
and avoid making them manually:
# Disable IP spoofing protection to allow IPSEC to work properly
echo 0 > /proc/sys/net/ipv4/conf/ipsec0/rp_filter
echo 0 > /proc/sys/net/ipv4/conf/eth0/rp_filter
NOTE: In the example of the firewall script file above, we assume that eth0 is the interface you
use for your connection. Of course if you use eth1 you must change eth0 to eth1, and so on.
If you forget this step you will receive error messages on your terminal such as the following
during the start up of FreeSWAN IPSEC:
ipsec_setup: WARNING: ipsec0 has route filtering turned on, KLIPS may not work
ipsec_setup: (/proc/sys/net/ipv4/conf/ipsec0/rp_filter = `1', should be 0)
ipsec_setup: WARNING: eth0 has route filtering turned on, KLIPS may not work
ipsec_setup: (/proc/sys/net/ipv4/conf/eth0/rp_filter = `1', should be 0)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-373-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
374
Testing the FreeS/WAN installation
• Reboot the both gateways to get FreeS/WAN started.
• Examine the /var/log/messages file for any signs of trouble. If all goes well you
should see something like this in the /var/log/messages file:
Feb 2 05:22:35 deep ipsec_setup: Starting FreeS/WAN IPSEC
snap2000jan31b...
Feb 2 05:22:35 deep ipsec_setup: KLIPS debug `none'
Feb 2 05:22:35 deep ipsec_setup: KLIPS ipsec0 on eth0
192.168.1.1/255.255.255.0 broadcast 192.168.1.255
Feb 2 05:22:36 deep ipsec_setup: Disabling core dumps:
Feb 2 05:22:36 deep ipsec_setup: Starting Pluto (debug `none'):
Feb 2 05:22:37 deep ipsec_setup: Loading Pluto database `deep-mail':
Feb 2 05:22:37 deep ipsec_setup: Enabling Pluto negotiation:
Feb 2 05:22:37 deep ipsec_setup: Routing for Pluto conns `deep-mail':
Feb 2 05:22:37 deep ipsec_setup: Initiating Pluto tunnel `deep-mail':
Feb 2 05:22:39 deep ipsec_setup: 102 "deep-mail" #1: STATE_MAIN_I1:
initiate
Feb 2 05:22:39 deep ipsec_setup: 104 "deep-mail" #1: STATE_MAIN_I2: from
STATE_MAIN_I1; sent MI2, expecting MR2
Feb 2 05:22:39 deep ipsec_setup: 106 "deep-mail" #1: STATE_MAIN_I3: from
STATE_MAIN_I2; sent MI3, expecting MR3
Feb 2 05:22:39 deep ipsec_setup: 004 "deep-mail" #1: STATE_MAIN_I4: SA
established
Feb 2 05:22:39 deep ipsec_setup: 110 "deep-mail" #2: STATE_QUICK_I1:
initiate
Feb 2 05:22:39 deep ipsec_setup: 004 "deep-mail" #2: STATE_QUICK_I2: SA
established
Feb 2 05:22:39 deep ipsec_setup: ...FreeS/WAN IPSEC started
• Examine the /var/log/secure file for any signs of trouble. If all goes well you should
see something like the following:
Feb 21 14:45:42 deep Pluto[432]: Starting Pluto (FreeS/WAN Version 1.3)
Feb 21 14:45:43 deep Pluto[432]: added connection description "deep-mail"
Feb 21 14:45:43 deep Pluto[432]: listening for IKE messages
Feb 21 14:45:43 deep Pluto[432]: adding interface ipsec0/eth0 192.168.1.1
Feb 21 14:45:43 deep Pluto[432]: loading secrets from
"/etc/ipsec.secrets"
Feb 21 14:45:43 deep Pluto[432]: "deep-mail" #1: initiating Main Mode
Feb 21 14:45:44 deep Pluto[432]: "deep-mail" #1: ISAKMP SA established
Feb 21 14:45:44 deep Pluto[432]: "deep-mail" #2: initiating Quick Mode
POLICY_RSASIG+POLICY_ENCRYPT+POLICY_AUTHENTICATE+POLICY_TUNNEL+POLICY_PFS
Feb 21 14:45:46 deep Pluto[432]: "deep-mail" #2: sent QI2, IPsec SA
established
Feb 21 14:45:47 deep Pluto[432]: "deep-mail" #3: responding to Main Mode
Feb 21 14:45:49 deep Pluto[432]: "deep-mail" #3: sent MR3, ISAKMP SA
established
Feb 21 14:45:49 deep Pluto[432]: "deep-mail" #4: responding to Quick Mode
Feb 21 14:45:50 deep Pluto[432]: "deep-mail" #4: IPsec SA established](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-374-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
375
• On both gateways, the following entries should now exist in the /proc/net/ directory:
[root@deep /]# ls -l /proc/net/ipsec_*
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_eroute
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_klipsdebug
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_spi
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_spigrp
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_spinew
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_tncfg
-r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_version
• The IPSEC interfaces should be attached on top of the specified physical interfaces.
Confirm that with:
[root@deep /]# cat /proc/net/ipsec_tncfg
ipsec0 -> eth0 mtu=16260 -> 1500
ipsec1 -> NULL mtu=0 -> 0
ipsec2 -> NULL mtu=0 -> 0
ipsec3 -> NULL mtu=0 -> 0
• Now execute the following command to show minimal debugging information and see if
the output looks something like this:
[root@deep /]# ipsec look
deep.openna.com Fri Feb 4 17:25:17 EST 2000
============-============
192.168.1.1/32 -> 192.168.1.2/32 => tun0x106@192.168.1.2
esp0x4450894d@192.168.1.2 ah0x4450894c@192.168.1.2
------------=------------
ah0x3350f551@192.168.1.1 AH_HMAC_MD5: dir=in ooowin=32 seq=115
bit=0xffffffff alen=128 aklen=16
life(c,s,h)=bytes(16140,0,0)add(51656,0,0)use(54068,0,0)packets(115,0,0)
idle=499
ah0x4450894c@192.168.1.2 AH_HMAC_MD5: dir=out ooowin=32 seq=2828 alen=128
aklen=16
life(c,s,h)=bytes(449488,0,0)add(51656,0,0)use(51656,0,0)packets(2828,0,0
) idle=6
esp0x3350f552@192.168.1.1 ESP_3DES: dir=in ooowin=32 seq=115
bit=0xffffffff eklen=24
life(c,s,h)=bytes(13380,0,0)add(51656,0,0)use(54068,0,0)packets(115,0,0)
idle=499
esp0x4450894d@192.168.1.2 ESP_3DES: dir=out ooowin=32 seq=2828 eklen=24
life(c,s,h)=bytes(381616,0,0)add(51656,0,0)use(51656,0,0)packets(2828,0,0
) idle=6
tun0x105@192.168.1.1 IPIP: dir=in 192.168.1.2 -> 192.168.1.1
life(c,s,h)=add(51656,0,0)
tun0x106@192.168.1.2 IPIP: dir=out 192.168.1.1 -> 192.168.1.2
life(c,s,h)=bytes(327581,0,0)add(51656,0,0)use(51656,0,0)packets(2828,0,0
) idle=6
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 ipsec0
192.168.1.1 0.0.0.0 255.255.255.255 UH 0 0 0 eth0
192.168.1.2 192.168.1.2 255.255.255.255 UGH 0 0 0 ipsec0
Destination Gateway Genmask Flags MSS Window irtt Iface](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-375-2048.jpg)
![FreeS/WAN VPN 1
CHAPTER 3
376
• Try pinging 192.168.1.2 from the 192.168.1.1 client. If this works then you have set
it up correctly. If it does not work check your network to make sure 208.164.186.1 can
reach 208.164.186.2, and that TCP-IP forwarding is enabled, and make sure that no
firewall rules are blocking the packets, or trying to masquerade them before the rules
allowing IPSec related traffic. For this test to work, it is important to use pings that go
from one subnet to the other.
208.164.186.1 ---- 205.151.222.250 ---- 205.151.222.251 ---- 208.164.186.2
| |
192.168.1.0/24 192.168.1.0/24
| |
192.168.1.1 192.168.1.2
A last note about testing the installation of FreeSWAN IPSEC, if you encounter a problem that
you are unable to resolve, you can use the following command to view a collection of debugging
information (contents of files, selections from logs, etc.) related to the IPSEC
encryption/authentication system that you should send to the Linux-IPSEC Mailing List (linux-
ipsec@clinet.fi) to help you.
• Use the following command to make an output of a collection of debugging information:
[root@deep /]# ipsec barf > result
This command is primarily provided as a convenience for remote debugging; A single command
which packages up (and labels) all information that might be relevant to diagnosing a problem in
IPSEC.
Further documentation
For more details, there are several manual pages about FreeS/WAN that you could read:
$ man ipsec (8) - invoke IPSEC utilities.
$ man ipsec atoaddr, addrtoa (3) - convert Internet addresses to and from ASCII.
$ man ipsec atoasr (3) - convert ASCII to Internet address, subnet, or range.
$ man ipsec atobytes, bytestoa (3) - convert binary data bytes from and to ASCII formats.
$ man ipsec atodata, datatoa (3) - convert binary data from and to ASCII formats.
$ man ipsec atosa, satoa (3) - convert IPSEC Security Association IDs to and from ASCII.
$ man ipsec atosubnet, subnettoa (3) - convert subnet/mask ASCII form to and from addresses.
$ man ipsec atoul, ultoa (3) - convert unsigned-long numbers to and from ASCII.
$ man ipsec auto (8) - control automatically-keyed IPSEC connections.
$ man ipsec barf (8) - spew out collected IPSEC debugging information.
$ man ipsec bitstomask (3) - convert bit count to Internet subnet mask.
$ man ipsec eroute (8) - manipulate IPSEC extended routing tables.
$ man ipsec goodmask (3) - is this Internet subnet mask a valid one?
$ man ipsec hostof (3) - given Internet address and subnet mask, return host part.
$ man ipsec klipsdebug (8) - set Klips (kernel IPSEC support) debug features and level.
$ man ipsec look (8) - show minimal debugging information.
$ man ipsec manual (8) - take manually-keyed IPSEC connections up and down.
$ man ipsec masktobits (3) - convert Internet subnet mask to bit count.
$ man ipsec optionsfrom (3) - read additional ``command-line'' options from file.
$ man ipsec pluto (8) - IPsec IKE keying daemon.
$ man ipsec ranbits (8) - generate random bits in ASCII form.
$ man ipsec rangetoa (3) - convert Internet address range to ASCII.
$ man ipsec rsasigkey (8) - generate RSA signature key.
$ man ipsec setup (8) - control IPSEC subsystem.
$ man ipsec spi (8) - manage IPSEC Security Associations.
$ man ipsec spigrp (8) - group/ungroup IPSEC Security Associations.
$ man ipsec subnetof (3) - given Internet address and subnet mask, return subnet number.
$ man ipsec tncfg (8) - associate IPSEC virtual interface with real interface.
$ man ipsec whack (8) - control interface for IPSEC keying daemon.
$ man ipsec.conf (5) - IPSEC configuration and connections.
$ man ipsec.secrets (5) - secrets for IKE/IPsec authentication.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-376-2048.jpg)
![GnuPG 1
CHAPTER 4
382
Prerequisites
GnuPG requires that the listed software below be already installed on your system to be able to
compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source
archive files. Please make sure you have this program installed on your machine before you
proceed with this chapter.
gettext is required to run GnuPG on your system.
python-1.5 is required to run GnuPG on your system.
expat is required to run GnuPG on your system.
gmp is required to run GnuPG on your system.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed on the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install GnuPG, and then
one afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > GnuPG1
• And the following one after you install the software:
[root@deep root]# find /* > GnuPG2
• Then use the following command to get a list of what changed:
[root@deep root]# diff GnuPG1 GnuPG2 > GnuPG-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing GnuPG
Below are the required steps that you must make to configure, compile and optimize the GnuPG
software before installing it into your Linux system. First off, we install the program as user 'root'
so as to avoid authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp gnupg-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf gnupg-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-382-2048.jpg)
![GnuPG 1
CHAPTER 4
383
Step 2
In order to check that the version of GnuPG, which you are going to install, is an original and
unmodified one, use the commands described below and check the supplied signature. Since we
don’t have GnuPG already installed in the system, we have to verify the MD5 checksum of the
program.
• To verify the MD5 checksum of GnuPG, use the following command:
[root@deep /]# md5sum gnupg-version.tar.gz
This should yield an output similar to this:
d8b36d4dfd213a1a1027b1877acbc897 gnupg-1.0.7.tar.gz
Now check that this checksum is exactly the same as the one published on the GnuPG website at
the following URL: http://www.gnupg.org/download.html
Step 3
Next, move into the newly created GnuPG source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created GnuPG directory use the following command:
[root@deep tmp]# cd gnupg-1.0.7/
• To configure and optimize GnuPG use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--mandir=/usr/share/man
--infodir=/usr/share/info
--enable-shared
--disable-nls
WARNING: Pay special attention to the compile CFLAGS line above. We optimize GnuPG for an
i686 CPU architecture with the parameter “-march=i686”. Please don’t forget to adjust the
CFLAGS line to reflect your own system.
Step 4
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the GnuPG software:
[root@deep gnupg-1.0.7]# make
[root@deep gnupg-1.0.7]# make check
[root@deep gnupg-1.0.7]# cd
[root@deep root]# find /* > GnuPG1
[root@deep root]# cd /var/tmp/gnupg-1.0.7/
[root@deep gnupg-1.0.7]# make install
[root@deep gnupg-1.0.7]# strip /usr/bin/gpg
[root@deep gnupg-1.0.7]# strip /usr/bin/gpgv
[root@deep gnupg-1.0.7]# cd
[root@deep root]# find /* > GnuPG2
[root@deep root]# diff GnuPG1 GnuPG2 > GnuPG-Installed](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-383-2048.jpg)
![GnuPG 1
CHAPTER 4
384
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 5
Once the configuration, optimization, compilation, and installation of the GnuPG software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete GnuPG and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf gnupg-version/
[root@deep tmp]# rm -f gnupg-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install GnuPG. It will also remove the GnuPG compressed archive from the /var/tmp/ directory.
Using GnuPG under Linux terminal
Here we show you how to use GnuPG using a terminal to manage GPG keys. The commands
listed below are ones that we use often, but many more exist. Check the manual page gpg (1)
for more information.
Creating a key-pair:
First of all, we must create a new key-pair (public and private) if this is a first use of the GnuPG
software to be able to use its encryption features.
Step 1
The “--gen-key” option of GnuPG is used to generate a new (public and private) key, we have to
use it every time we need to create a new GnuPG key on the system. When we issue this
command for the first time, GnuPG will create the required directory and options file for us.
• To create a new key-pair, use the following command:
[root@deep /]# gpg --gen-key
gpg (GnuPG) 1.0.7; Copyright (C) 2000 Free Software Foundation, Inc.
This program comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions. See the file COPYING for details.
gpg: /root/.gnupg: directory created
gpg: /root/.gnupg/options: new options file created
gpg: you have to start GnuPG again, so it can read the new options file](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-384-2048.jpg)
![GnuPG 1
CHAPTER 4
385
Step 2
Once the command has been executed, we have to run it again for a second time to create our
public and private keys, because on first utilization, it just creates the required directory and
options file for us. Therefore, it will now create the keys.
• We start GnuPG again with the same command:
[root@deep /]# gpg --gen-key
gpg (GnuPG) 1.0.7; Copyright (C) 2002 Free Software Foundation, Inc.
This program comes with ABSOLUTELY NO WARRANTY.
This is free software, and you are welcome to redistribute it
under certain conditions. See the file COPYING for details.
gpg: keyring `/root/.gnupg/secring.gpg' created
gpg: keyring `/root/.gnupg/pubring.gpg' created
Please select what kind of key you want:
(1) DSA and ElGamal (default)
(2) DSA (sign only)
(4) ElGamal (sign and encrypt)
(5) RSA (sign only)
Your selection? 1
DSA keypair will have 1024 bits.
About to generate a new ELG-E keypair.
minimum keysize is 768 bits
default keysize is 1024 bits
highest suggested keysize is 2048 bits
What keysize do you want? (1024) Press Enter
Requested keysize is 1024 bits
Please specify how long the key should be valid.
0 = key does not expire
<n> = key expires in n days
<n>w = key expires in n weeks
<n>m = key expires in n months
<n>y = key expires in n years
Key is valid for? (0) Press Enter
Key does not expire at all
Is this correct (y/n)? y
You need a User-ID to identify your key; the software constructs the user
id from Real Name, Comment and Email Address in this form:
"Heinrich Heine (Der Dichter) <heinrichh@duesseldorf.de>"
Real name: Gerhard Mourani
Email address: gmourani@openna.com
Comment: Press Enter
You selected this USER-ID:
"Gerhard Mourani <gmourani@openna.com>"
Change (N)ame, (C)omment, (E)mail or (O)kay/(Q)uit? o
You need a Passphrase to protect your secret key.
Enter passphrase: mypassphrase
Repeat passphrase: mypassphrase
We need to generate a lot of random bytes. It is a good idea to perform
some other action (type on the keyboard, move the mouse, utilize the
disks) during the prime generation; this gives the random number
generator a better chance to gain enough entropy.
++++++++++++++++++++.+++++.+++++.+++++++++++++++.+++++++++++++++..+++++++
++++++++...+++++++++++++++.+++++.++++++++++++++++++++++++++++++++++++++++
..+++++](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-385-2048.jpg)
![GnuPG 1
CHAPTER 4
386
<+++++..>+++++.................<..+++++..................................
..........+++++^^^^
public and secret key created and signed.
NOTE: A new key-pair is created (secret and public key) in the “root” home directory ~/root
under the .gnupg subdirectory because we issued this GnuPG command as user “root”. If you
run the above command under other user into the system, then the generated keys will be
located under its home directory on the server.
Exporting GPG key/s for a user:
Once your own key-pair is created, you can expand your horizons by exporting and distributing
your public key over the world (NEVER export you private key). This can be done by publishing it
on your homepage, through an available key server on the Internet, or any other available
method. GnuPG has some useful options to help you publish your public key.
Step 1
First off, we have to extract our public key in ASCII text to be able to distribute it. ASCII text is a
good format to use because it allows people to get it easily. In this way, anyone can just cut and
past your public key and use it when they want to securely communicate with you.
• To extract your public key in ASCII armored output, use the following command:
[root@deep /]# gpg --export –ao UID
As an example:
[root@deep /]# gpg --export –ao Gerhard Mourani
Where “--export” is for extracting Public-key from your pubring encrypted file, “a” is to create
ASCII armored output that you can mail, publish or put it on a web page, “o” to put the result in a
file and UID represents the user key you want to export, which is in our example the user
“Gerhard Mourani” key that we have create previously.
Step 2
Once your public key has been extracted, the resulting output will be a file called “Gerhard”
under the directory where you are issuing the above command, representing the First name of
the user key to extract. In our example, the file is called “Gerhard” because it is the name of the
key we want to export in ASCII text format. Note that the file name will be different for your public
ASCII text format key.
• Edit your public key in ASCII armored output format, and distribute it:
[root@deep /]# vi Gerhard
-----BEGIN PGP PUBLIC KEY BLOCK-----
Version: GnuPG v1.0.7 (GNU/Linux)
Comment: For info see http://www.gnupg.org
mQGiBDzGNQcRBAC+1NrjFMCEtyjcv5lhtFNMLHEQ0VdHObv0CMUdCkiDslJ9QT9v
MtVG1d4r3+0RJan23Z+8fc11E7Q0wRjRO13efRGEbxaIushhRc/p11LsEubWMWC7
E1UCsMmniScEdoZLSq9/myjj7IJqAavgL0a7/VkVHjrX1j/pTTK1wUUsRwCgy0jp
0JzY1+dIK4ElfGxAQ7oHop8D/03MkyVhUZh9asLW4tyGlmMN8exqfRoMdeSv0jnz
ftAAZ71sn8jDdviccvaJvj2eTdZ7J43BIhxALJZ8KMQdEDWQnW62FfV9uGWcB5HL
c869XOD0so9LOJGsgF1XpnMKQhTRXXEIuN0THpGDSLdQtXelBzIusQuSmNBrx7A0
6/5xA/0W3H2NYzvMWnTuENpHUQR8KtIARcmis4bGIH/fEiPQyR7YWIAs9sPOE5Yr
3cQuUpZ3nwGcZ5CGOKm0qRBkhMI49SO25gsoaRVVatNZ1v1o07AaNDimmvE0hhO3
+/LTv9cJYMdm4ijp+XOhssO4zctgdg0bHISsTWqB1AJcSsdAirQpR2VyaGFyZCBN](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-386-2048.jpg)
![GnuPG 1
CHAPTER 4
387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=7bvq
-----END PGP PUBLIC KEY BLOCK-----
WARNING: Never export or distribute your private key to the world. I know, this seem to be a stupid
warning, but I’ve been informed that some people do it.
Importing GPG key/s from a user:
When you receive someone's public key (or some trusted third partly keys) you have to add them
to your key database in order to be able to use his/her keys for future encryption, verification and
authentication. This is often the case, when we install software that has a GPG key available for
verification. Therefore, here is what you should do before installing software that has a GPG key to
your disposal for authenticity.
Step 1
First off, we have to retrieve the GPG public key of the company, organization, etc that we want to
import into our keyring database. In our example, we will retrieve the GPG key that OpenNA uses
to sign RPM packages and other software.
This GPG public key is available from: http://www.openna.com/about/openna.asc. Cut and past it
into a file called “openna.asc” on your server machine where GnuPG is installed.
Step 2
Now, we have to import the OpenNA GPG public key into our database. This procedure should be
done for any GPG public keys that you want to use to verify authenticity of software you want to
install on your server. Most organizations have GPG public keys for you to download.
• To import Public Keys to your keyring database, use the following command:
[root@deep /]# gpg --import filename
As an example:
[root@deep /]# gpg --import openna.asc
gpg: key 3487965A: public key imported
gpg: /root/.gnupg/trustdb.gpg: trustdb created
gpg: Total number processed: 1
gpg: imported: 1
The above command will append the new key “filename” into the keyring database and will
update all already existing keys. It is important to note that GnuPG does not import keys that are
not self-signed (asc).](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-387-2048.jpg)
![GnuPG 1
CHAPTER 4
388
Signing GPG key/s from a user:
When you import keys into your public keyring database and are sure that the trusted third party
is really the person they claim, you can start signing his/her keys. Signing a key certifies that you
know the owner of the keys and this leads to the situation where the signature acknowledges that
the user ID mentioned in the key is actually the owner of that key.
• To sign the key for company OpenNA that we have added into our keyring database
above, use the following command:
[root@deep /]# gpg --sign-key UID
As an example:
[root@deep /]# gpg --sign-key OpenNA
pub 1024D/3487965A created: 2001-07-02 expires: never trust: -/q
sub 1024g/0146F594 created: 2001-07-02 expires: never
(1). OpenNA Inc. <noc@openna.com>
pub 1024D/3487965A created: 2001-07-02 expires: never trust: -/q
Fingerprint: 7A3D 6871 2DF1 9210 8ABE AF36 D460 86D5 3487
965A
OpenNA Inc. <noc@openna.com>
Are you really sure that you want to sign this key
with your key: "Gerhard Mourani <gmourani@openna.com>"
Really sign? y
You need a passphrase to unlock the secret key for
user: "Gerhard Mourani <gmourani@openna.com>"
1024-bit DSA key, ID 2E5378F7, created 2002-04-24
Enter passphrase:
WARNING: You should only sign a key as being authentic when you are ABSOLUTELY SURE that
the key is really authentic! You should never sign a key based on any kind of assumption.
Checking GPG signature:
We have shown above how to sign a key, now we will explain how people can verify if the
signature is really the good one. Once you have extracted your public key and exported it,
everyone who knows or gets your public key should be able to check whether encrypted data
from you is also really signed by you.
• To check the signature of encrypted data, use the following command:
[root@deep /]# gpg --verify Data
The “--verify” option will check the signature where Data is the encrypted data/file you want
to verify.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-388-2048.jpg)
![GnuPG 1
CHAPTER 4
389
Encrypting and decrypting GPG files:
After installing, importing, signing and configuring everything in the way that we want, we can
start encrypting and decrypting our files, software, etc.
• To encrypt and sign data for the user OpenNA that we have added on our keyring
database above, use the following command:
[root@deep /]# gpg -sear OpenNA file
As an example:
[root@deep /]# gpg -sear OpenNA Message-to-OpenNA.txt
You need a passphrase to unlock the secret key for
user: "Gerhard Mourani <gmourani@openna.com>"
1024-bit DSA key, ID 2E5378F7, created 2002-04-24
Enter passphrase:
Of the arguments passed, the “s” is for signing (To avoid the risk that somebody else claims to be
you, it is very useful to sign everything you encrypt), “e” for encrypting, “a” to create ASCII
armored output (“.asc” ready for sending by mail), “r” to encrypt the UID name and “file” is the
message you want to encrypt.
• To decrypt data, use the following command:
[root@deep /]# gpg -d file
For example:
[root@deep /]# gpg -d Message-from-GerhardMourani.asc
You need a passphrase to unlock the secret key for
user: "Gerhard Mourani (Open Network Architecture) <gmourani@openna.com>"
1024-bit DSA key, ID 2E5378F7, created 2002-04-24
Enter passphrase:
Where “d” is for decrypting and “file” is the message you want to decrypt. It is important that
the public key of the sender of the message we want to decrypt be in our public keyring database
or of course nothing will work.
Further documentation
For more details, there are some manual pages about GnuPG that you could read:
$ man gpg (1) - GPG encryption and signing tool.
$ man gpgv (1) - GPGV signature verification tool.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-389-2048.jpg)
![OpenSSL 1
CHAPTER 5
395
These installation instructions assume
Commands are Unix-compatible.
The source path is /var/tmp (note that other paths are possible, at personal discretion).
Installations were tested on OpenNA Linux & Red Hat Linux 7.3.
All steps in the installation will happen using the super-user account “root”.
Whether kernel recompilation may be required: No
Latest OpenSSL version number is 0.9.6d
The procedures given in this chapter are likely to work on all Linux platforms, but we have only
tested it on OpenNA Linux and Red Hat Linux.
Packages
The following is based on information listed by OpenSSL as of 2002/05/09. Please check
http://www.openssl.org/ regularly for the latest status. We chose to install from source because it
provides the facility to fine tune the installation.
Source code is available from:
OpenSSL Homepage: http://www.openssl.org/
OpenSSL FTP Site: 129.132.7.170
You must be sure to download: openssl-0.9.6d.tar.gz
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install OpenSSL, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > OpenSSL1
• And the following one after you install the software:
[root@deep root]# find /* > OpenSSL2
• Then use the following command to get a list of what changed:
[root@deep root]# diff OpenSSL1 OpenSSL2 > OpenSSL-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-395-2048.jpg)
![OpenSSL 1
CHAPTER 5
396
Compiling - Optimizing & Installing OpenSSL
Below are the steps that you must make to configure, compile and optimize the OpenSSL
software before installing it into your Linux system. First off, we install the program as user 'root'
so as to avoid authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp openssl-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf openssl-version.tar.gz
Step 2
Next, move into the newly created OpenSSL source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created OpenSSL directory use the following command:
[root@deep tmp]# cd openssl-0.9.6d/
Step 3
With OpenSSL, the optimization FLAGS should be changed in the “Configure” file of the
program. It is in this file that we define the GCC optimizations we want to use related to the type of
processor running in our system. OpenSSL is cryptography software and there are some
optimization hacks that we can make that can significantly increase the performance of the
program, therefore take the time to modify the “Configure” file of the software. This will be a
benefit for you.
a) Edit the Configure file (vi +337 Configure) and change the following lines:
"linux-elf", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -m486 -Wall::-
D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des}
${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-
fPIC:.so.$(SHLIB_MAJOR).$(SHLIB_MINOR)",
To read:
"linux-elf", "gcc:-DL_ENDIAN -DTERMIO -O3 -march=i686 -funroll-loops -fomit-
frame-pointer -Wall::-D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des}
${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:-
fPIC:.so.$(SHLIB_MAJOR).$(SHLIB_MINOR)",
b) Edit the Configure file (vi +338 Configure) and change the following lines:
"debug-linux-elf","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -
DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -m486 -Wall::-D_REENTRANT:-lefence -
ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn",
To read:
"debug-linux-elf","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -
DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -O3 -march=i686 -funroll-loops -fomit-frame-
pointer -Wall::-D_REENTRANT:-lefence -ldl:BN_LLONG ${x86_gcc_des}
${x86_gcc_opts}:${x86_elf_asm}:dlfcn",](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-396-2048.jpg)
![OpenSSL 1
CHAPTER 5
397
c) Edit the Configure file (vi +339 Configure) and change the following lines:
"debug-linux-elf-noefence","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG
-DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -m486 -Wall::-D_REENTRANT:-ldl:BN_LLONG
${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn",
To read:
"debug-linux-elf-noefence","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG
-DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -O3 -march=i686 -funroll-loops -fomit-frame-
pointer -Wall::-D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des}
${x86_gcc_opts}:${x86_elf_asm}:dlfcn",
Step 4
By default, OpenSSL source files assume that our “perl” binary program is located under
/usr/local/bin/perl. We must change this to reflect our environment variable.
• To point all OpenSSL script files to our “perl” binary, use the following command:
[root@deep openssl-0.9.6d]# perl util/perlpath.pl /usr/bin/perl
Step 5
At this stage, it is time to configure and compile OpenSSL for our system.
• To configure and optimize OpenSSL use the following compilation lines:
./Configure linux-elf no-asm shared
--prefix=/usr
--openssldir=/usr/share/ssl
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the OpenSSL software:
[root@deep openssl-0.9.6d]# LD_LIBRARY_PATH=`pwd` make all build-shared
[root@deep openssl-0.9.6d]# LD_LIBRARY_PATH=`pwd` make test apps tests
[root@deep openssl-0.9.6d]# cd
[root@deep root]# find /* > OpenSSL1
[root@deep root]# cd /var/tmp/openssl-0.9.6d/
[root@deep openssl-0.9.6d]# make install build-shared
[root@deep openssl-0.9.6d]# cd /usr/lib/
[root@deep lib]# mv libcrypto.so.0.9.6 ../../lib/
[root@deep lib]# mv libssl.so.0.9.6 ../../lib/
[root@deep lib]# ln -sf ../../lib/libcrypto.so.0.9.6 libcrypto.so
[root@deep lib]# ln -sf ../../lib/libcrypto.so.0.9.6 libcrypto.so.0
[root@deep lib]# ln -sf ../../lib/libssl.so.0.9.6 libssl.so
[root@deep lib]# ln -sf ../../lib/libssl.so.0.9.6 libssl.so.0
[root@deep lib]# mv /usr/share/ssl/man/man1/* /usr/share/man/man1/
[root@deep lib]# mv /usr/share/ssl/man/man3/* /usr/share/man/man3/
[root@deep lib]# mv /usr/share/ssl/man/man5/* /usr/share/man/man5/
[root@deep lib]# mv /usr/share/ssl/man/man7/* /usr/share/man/man7/
[root@deep lib]# rm -rf /usr/share/ssl/man/
[root@deep lib]# rm -rf /usr/share/ssl/lib/
[root@deep lib]# strip /usr/bin/openssl
[root@deep lib]# mkdir -p /usr/share/ssl/crl
[root@deep lib]# cd
[root@deep root]# find /* > OpenSSL2
[root@deep root]# diff OpenSSL1 OpenSSL2 > OpenSSL-Installed](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-397-2048.jpg)
![OpenSSL 1
CHAPTER 5
398
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then test the OpenSSL libraries to finally install the binaries and any supporting files into the
appropriate locations.
Step 7
Once the configuration, optimization, compilation, and installation of the OpenSSL software has
completed, we can free up some disk space by deleting the program tar archive and the related
source directory since they are no longer needed.
• To delete OpenSSL and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf openssl-version/
[root@deep tmp]# rm -f openssl-version.tar.gz
Configuring OpenSSL
After OpenSSL has been built and installed successfully on your system, your next step is to
configure and customize the openssl.cnf and sign files to suit your needs.
/usr/shared/ssl/openssl.cnf (The OpenSSL Configuration File)
/usr/shared/ssl/misc/sign (A CA script file to sign certificates)
/usr/shared/ssl/openssl.cnf: The OpenSSL Configuration File
This is the general configuration file for OpenSSL, where you can configure the expiration date of
your keys, the name of your organization, address and so on.
The most important parameters you may need to change will be in the [ CA_default ] and
especially the [ req_distinguished_name ] sections of the file. We must change the
default one to fit our requirements and operating system. The text in bold is the parts of the
configuration file that must be customized and adjusted to satisfy our needs.
• Edit the openssl.cnf file (vi /usr/share/ssl/openssl.cnf) and set your needs.
#
# OpenSSL example configuration file.
# This is mostly being used for generation of certificate requests.
#
# This definition stops the following lines choking if HOME isn't
# defined.
HOME = .
RANDFILE = $ENV::HOME/.rnd
# Extra OBJECT IDENTIFIER info:
#oid_file = $ENV::HOME/.oid
oid_section = new_oids
# To use this configuration file with the "-extfile" option of the
# "openssl x509" utility, name here the section containing the
# X.509v3 extensions to use:
# extensions =
# (Alternatively, use a configuration file that has only
# X.509v3 extensions in its main [= default] section.)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-398-2048.jpg)
![OpenSSL 1
CHAPTER 5
399
[ new_oids ]
# We can add new OIDs in here for use by 'ca' and 'req'.
# Add a simple OID like this:
# testoid1=1.2.3.4
# Or use config file substitution like this:
# testoid2=${testoid1}.5.6
####################################################################
[ ca ]
default_ca = CA_default # The default ca section
####################################################################
[ CA_default ]
dir = /usr/share/ssl # Where everything is kept
certs = $dir/certs # Where the issued certs are kept
crl_dir = $dir/crl # Where the issued crl are kept
database = $dir/ca.db.index # database index file.
new_certs_dir = $dir/ca.db.certs # default place for new certs.
certificate = $dir/certs/ca.crt # The CA certificate
serial = $dir/ca.db.serial # The current serial number
crl = $dir/crl.pem # The current CRL
private_key = $dir/private/ca.key # The private key
RANDFILE = $dir/ca.db.rand # private random number file
x509_extensions = usr_cert # The extentions to add to the cert
# Extensions to add to a CRL. Note: Netscape communicator chokes on V2 CRLs
# so this is commented out by default to leave a V1 CRL.
# crl_extensions = crl_ext
default_days = 365 # how long to certify for
default_crl_days= 30 # how long before next CRL
default_md = md5 # which md to use.
preserve = no # keep passed DN ordering
# A few difference way of specifying how similar the request should look
# For type CA, the listed attributes must be the same, and the optional
# and supplied fields are just that :-)
policy = policy_match
# For the CA policy
[ policy_match ]
countryName = match
stateOrProvinceName = match
organizationName = match
organizationalUnitName = optional
commonName = supplied
emailAddress = optional
# For the 'anything' policy
# At this point in time, you must list all acceptable 'object'
# types.
[ policy_anything ]
countryName = optional
stateOrProvinceName = optional
localityName = optional
organizationName = optional
organizationalUnitName = optional
commonName = supplied
emailAddress = optional](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-399-2048.jpg)
![OpenSSL 1
CHAPTER 5
400
####################################################################
[ req ]
default_bits = 1024
default_keyfile = privkey.pem
distinguished_name = req_distinguished_name
attributes = req_attributes
x509_extensions = v3_ca # The extentions to add to the self signed cert
# Passwords for private keys if not present they will be prompted for
# input_password = secret
# output_password = secret
# This sets a mask for permitted string types. There are several options.
# default: PrintableString, T61String, BMPString.
# pkix : PrintableString, BMPString.
# utf8only: only UTF8Strings.
# nombstr : PrintableString, T61String (no BMPStrings or UTF8Strings).
# MASK:XXXX a literal mask value.
# WARNING: current versions of Netscape crash on BMPStrings or UTF8Strings
# so use this option with caution!
string_mask = nombstr
# req_extensions = v3_req # The extensions to add to a certificate request
[ req_distinguished_name ]
countryName = Country Name (2 letter code)
countryName_default = CA
countryName_min = 2
countryName_max = 2
stateOrProvinceName = State or Province Name (full name)
stateOrProvinceName_default = Quebec
localityName = Locality Name (eg, city)
localityName_default = Montreal
0.organizationName = Organization Name (eg, company)
0.organizationName_default = OpenNA, Inc.
# we can do this but it is not needed normally :-)
#1.organizationName = Second Organization Name (eg, company)
#1.organizationName_default = World Wide Web Pty Ltd
organizationalUnitName = Organizational Unit Name (eg, section)
organizationalUnitName_default = Open Network Architecture
commonName = Common Name (eg, YOUR name)
commonName_default = www.openna.com
commonName_max = 64
emailAddress = Email Address
emailAddress_default = noc@openna.com
emailAddress_max = 40
# SET-ex3 = SET extension number 3
[ req_attributes ]
challengePassword = A challenge password
challengePassword_min = 8
challengePassword_max = 20
unstructuredName = An optional company name](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-400-2048.jpg)
![OpenSSL 1
CHAPTER 5
401
[ usr_cert ]
# These extensions are added when 'ca' signs a request.
# This goes against PKIX guidelines but some CAs do it and some software
# requires this to avoid interpreting an end user certificate as a CA.
basicConstraints=CA:FALSE
# Here are some examples of the usage of nsCertType. If it is omitted
# the certificate can be used for anything *except* object signing.
# This is OK for an SSL server.
# nsCertType = server
# For an object signing certificate this would be used.
# nsCertType = objsign
# For normal client use this is typical
# nsCertType = client, email
# and for everything including object signing:
# nsCertType = client, email, objsign
# This is typical in keyUsage for a client certificate.
# keyUsage = nonRepudiation, digitalSignature, keyEncipherment
# This will be displayed in Netscape's comment listbox.
nsComment = "OpenSSL Generated Certificate"
# PKIX recommendations harmless if included in all certificates.
subjectKeyIdentifier=hash
authorityKeyIdentifier=keyid,issuer:always
# This stuff is for subjectAltName and issuerAltname.
# Import the email address.
# subjectAltName=email:copy
# Copy subject details
# issuerAltName=issuer:copy
#nsCaRevocationUrl = http://www.domain.dom/ca-crl.pem
#nsBaseUrl
#nsRevocationUrl
#nsRenewalUrl
#nsCaPolicyUrl
#nsSslServerName
[ v3_req ]
# Extensions to add to a certificate request
basicConstraints = CA:FALSE
keyUsage = nonRepudiation, digitalSignature, keyEncipherment
[ v3_ca ]
# Extensions for a typical CA
# PKIX recommendation.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-401-2048.jpg)
![OpenSSL 1
CHAPTER 5
402
subjectKeyIdentifier=hash
authorityKeyIdentifier=keyid:always,issuer:always
# This is what PKIX recommends but some broken software chokes on critical
# extensions.
#basicConstraints = critical,CA:true
# So we do this instead.
basicConstraints = CA:true
# Key usage: this is typical for a CA certificate. However since it will
# prevent it being used as an test self-signed certificate it is best
# left out by default.
# keyUsage = cRLSign, keyCertSign
# Some might want this also
# nsCertType = sslCA, emailCA
# Include email address in subject alt name: another PKIX recommendation
# subjectAltName=email:copy
# Copy issuer details
# issuerAltName=issuer:copy
# DER hex encoding of an extension: beware experts only!
# obj=DER:02:03
# Where 'obj' is a standard or added object
# You can even override a supported extension:
# basicConstraints= critical, DER:30:03:01:01:FF
[ crl_ext ]
# CRL extensions.
# Only issuerAltName and authorityKeyIdentifier make any sense in a CRL.
# issuerAltName=issuer:copy
authorityKeyIdentifier=keyid:always,issuer:always
WARNING: You don’t need to change all of the default options set in the file openssl.cnf; The
configurations you usually change will be into the [ CA_default ] and [
req_distinguished_name ] sections of the file.
/usr/share/ssl/misc/sign: The CA Script File to Sign Certificates
OpenSSL CA command has some strange requirements and the default OpenSSL config doesn't
allow one to easily use OpenSSL CA directly. It is for this reason that we don’t use the files CA.pl
or CA.sh to sign certificates.
Step 1
To solve the problem, we’ll create and customize the sign script file below to replace them. Text
in bold are the parts of the script that must be customized and adjusted to satisfy our needs.
• Create the sign script file (touch /usr/share/ssl/misc/sign) and add the
following lines:
#!/bin/sh
##](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-402-2048.jpg)
![OpenSSL 1
CHAPTER 5
403
## sign.sh -- Sign a SSL Certificate Request (CSR)
## Copyright (c) 1998-1999 Ralf S. Engelschall, All Rights Reserved.
##
# argument line handling
CSR=$1
if [ $# -ne 1 ]; then
echo "Usage: sign.sign <whatever>.csr"; exit 1
fi
if [ ! -f $CSR ]; then
echo "CSR not found: $CSR"; exit 1
fi
case $CSR in
*.csr ) CERT="`echo $CSR | sed -e 's/.csr/.crt/'`" ;;
* ) CERT="$CSR.crt" ;;
esac
# make sure environment exists
if [ ! -d ca.db.certs ]; then
mkdir ca.db.certs
fi
if [ ! -f ca.db.serial ]; then
echo '01' >ca.db.serial
fi
if [ ! -f ca.db.index ]; then
cp /dev/null ca.db.index
fi
# create an own SSLeay config
cat >ca.config <<EOT
[ ca ]
default_ca = CA_own
[ CA_own ]
dir = /usr/share/ssl
certs = /usr/share/ssl/certs
new_certs_dir = /usr/share/ssl/ca.db.certs
database = /usr/share/ssl/ca.db.index
serial = /usr/share/ssl/ca.db.serial
RANDFILE = /usr/share/ssl/ca.db.rand
certificate = /usr/share/ssl/certs/ca.crt
private_key = /usr/share/ssl/private/ca.key
default_days = 365
default_crl_days = 30
default_md = md5
preserve = no
policy = policy_anything
[ policy_anything ]
countryName = optional
stateOrProvinceName = optional
localityName = optional
organizationName = optional
organizationalUnitName = optional
commonName = supplied
emailAddress = optional
EOT
# sign the certificate
echo "CA signing: $CSR -> $CERT:"
openssl ca -config ca.config -out $CERT -infiles $CSR
echo "CA verifying: $CERT <-> CA cert"
openssl verify -CAfile /usr/share/ssl/certs/ca.crt $CERT
# cleanup after SSLeay](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-403-2048.jpg)
![OpenSSL 1
CHAPTER 5
404
rm -f ca.config
rm -f ca.db.serial.old
rm -f ca.db.index.old
# die gracefully
exit 0
Step 2
Once the script file has been created, it is important to make it executable and change its default
permissions. Making this file executable will allow the system to run it, changing its default
permission is to allow only the root user to change this file for security reason.
• To make this script executable and to change its default permissions, use the commands:
[root@deep /]# chmod 700 /usr/share/ssl/misc/sign
[root@deep /]# chown 0.0 /usr/share/ssl/misc/sign
OpenSSL Administrative Tools
Once your configuration options have been set in the openssl.cnf file, we can play with the
OpenSSL utility. As an example, we’ll show you how to create certificates for the Apache web
server and your own CA (Certifying Authority) to sign your “Certificate Signing Request” yourself.
All commands listed below are to be made in the /usr/share/ssl directory.
The Apache Key & CSR Generation:
The utility “openssl” that you use to generate RSA Private Keys (Key) and Certificate Signing
Requests (CSR) comes with OpenSSL and is usually installed under the directory /usr/bin on
our Linux distribution. Below are the steps to create certificates for Apache.
Step 1
First you have to know the Fully Qualified Domain Name (FQDN) of the website/server for which
you want to request a certificate. When you want to access your website/server through
https://www.mydomain.com/ then the FQDN of your website is www.mydomain.com.
Step 2
Second, select five large and relatively random files from your hard drive (compressed log files
are a good start) and put them under your /usr/share/ssl directory. These will act as your
random seed enhancers. We refer to them as random1: random2:...: random5 below.
• To select five random files and put them under /usr/share/ssl, use the commands:
[root@deep /]# cp /var/log/boot.log /usr/share/ssl/random1
[root@deep /]# cp /var/log/cron /usr/share/ssl/random2
[root@deep /]# cp /var/log/dmesg /usr/share/ssl/random3
[root@deep /]# cp /var/log/messages /usr/share/ssl/random4
[root@deep /]# cp /var/log/secure /usr/share/ssl/random5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-404-2048.jpg)
![OpenSSL 1
CHAPTER 5
405
Step 3
Third, create the RSA private key protected with a pass-phrase for your web server. The
command below will generate 1024 bit RSA Private Key and store it in
www.mydomain.com.key.
It will ask you for a pass-phrase: use something secure and remember it. Your certificate will be
useless without the key. If you don't want to protect your key with a pass-phrase (only if you
absolutely trust that server machine, and you make sure the permissions are carefully set so only
you can read that key) you can leave out the -des3 option below.
• To generate the Key, use the following command:
[root@deep /]# cd /usr/share/ssl/
[root@deep ssl]# openssl genrsa -des3 –rand
random1:random2:random3:random4:random5 -out www.mydomain.com.key 1024
123600 semi-random bytes loaded
Generating RSA private key, 1024 bit long modulus
......................+++++
.....+++++
e is 65537 (0x10001)
Enter PEM pass phrase:
Verifying password - Enter PEM pass phrase:
WARNING: Please backup your www.mydomain.com.key file and remember the pass-phrase you
had to enter at a secure location. A good choice is to backup this information onto a diskette or
other removable media.
Step 4
Finally, generate a Certificate Signing Request (CSR) with the server RSA private key. The
command below will prompt you for the X.509 attributes of your certificate. Remember to give
the name “www.mydomain.com” when prompted for ‘Common Name'. Do not enter your personal
name here. We are requesting a certificate for a web server, so the ‘Common Name’ has to match
the FQDN of your website (a requirement of the browsers).
• To generate the CSR, use the following command:
[root@deep ssl]# openssl req -new -key www.mydomain.com.key -out
www.mydomain.com.csr
Using configuration from /usr/share/ssl/openssl.cnf
Enter PEM pass phrase:
You are about to be asked to enter information that will be incorporated
into your certificate request.
What you are about to enter is what is called a Distinguished Name or a
DN.
There are quite a few fields but you can leave some blank
For some fields there will be a default value,
If you enter '.', the field will be left blank.
-----
Country Name (2 letter code) [CA]:
State or Province Name (full name) [Quebec]:
Locality Name (eg, city) [Montreal]:
Organization Name (eg, company) [OpenNA, Inc.]:
Organizational Unit Name (eg, section) [Open Network Architecture]:
Common Name (eg, YOUR name) [www.openna.com]:
Email Address [noc@openna.com]:
Please enter the following 'extra' attributes](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-405-2048.jpg)
![OpenSSL 1
CHAPTER 5
406
to be sent with your certificate request
A challenge password []:.
An optional company name []:.
WARNING: Make sure you enter the FQDN (Fully Qualified Domain Name) of the server when
OpenSSL prompts you for the “Common Name” (i.e. when you generate a CSR for a website which
will be later accessed via https://www.mydomain.com/, enter “www.mydomain.com” here).
After the generation of your Certificate Signing Request (CSR), you must send this certificate to a
commercial Certifying Authority (CA) like Thawte or Verisign for signing. You usually have to post
the CSR into a web form, pay for the signing, await the signed certificate and store it into a
“www.mydomain.com.crt” file. The result is then a real certificate, which can be used with
Apache.
The CA Key & CRT Generation:
If you don’t want to pay a commercial Certifying Authority (CA) to sign you certificates, you can
use your own CA and now have to sign the CSR yourself by this CA. This solution is economical,
and allows an organization to host their own CA server and generate as many certificates as they
need for internal use without paying a cent to a commercial CA.
Unfortunately using your own CA to generate certificates causes problems in electronic
commerce, because customers need to have some trust in your organization by the use of a
recognized commercial CA. See below on how to sign a CSR with your CA yourself.
Step 1
As for the Apache web server above, the first step is to create the RSA private key protected with
a pass-phrase for your CA. The command below will generate 1024 bit RSA Private Key and
stores it in the file “ca.key”. It will ask you for a pass-phrase: use something secure and
remember it. Your certificate will be useless without the key.
• To create the RSA private key for your (CA), use the following command:
[root@deep /]# cd /usr/share/ssl/
[root@deep ssl]# openssl genrsa -des3 -out ca.key 1024
Generating RSA private key, 1024 bit long modulus
...........................+++++
............................................+++++
e is 65537 (0x10001)
Enter PEM pass phrase:
Verifying password - Enter PEM pass phrase:
WARNING: Please backup your “ca.key” file and remember the pass-phrase you had to enter at a
secure location. A good choice is to backup this information onto a diskette or other removable
media.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-406-2048.jpg)
![OpenSSL 1
CHAPTER 5
407
Step 2
Now, we must create a self-signed (CA) certificate (x509 structure) with the RSA key of the CA.
The req command creates a self-signed certificate when the -x509 switch is used.
• To create a self-signed (CA) certificate, use the following command:
[root@deep ssl]# openssl req -new -x509 -days 365 -key ca.key -out ca.crt
Using configuration from /usr/share/ssl/openssl.cnf
Enter PEM pass phrase:
You are about to be asked to enter information that will be incorporated
into your certificate request.
What you are about to enter is what is called a Distinguished Name or a
DN.
There are quite a few fields but you can leave some blank
For some fields there will be a default value,
If you enter '.', the field will be left blank.
-----
Country Name (2 letter code) [CA]:
State or Province Name (full name) [Quebec]:
Locality Name (eg, city) [Montreal]:
Organization Name (eg, company) [OpenNA, Inc.]:
Organizational Unit Name (eg, section) [Open Network]:Sales Dept
Common Name (eg, YOUR name) [www.openna.com]:
Email Address [noc@openna.com]:sales@openna.com
Step 3
Once the self-signed (CA) certificate has been created, we must place all certificates and CA files
into their appropriate directories.
• To place the files into their appropriate directories, use the following commands:
[root@deep ssl]# mv www.mydomain.com.key private/
[root@deep ssl]# mv ca.key private/
[root@deep ssl]# mv ca.crt certs/
Step 4
Finally, you can use this CA to sign all the servers CSR's in order to create real SSL Certificates
for use inside the web server (assuming you already have a www.mydomain.com.csr at hand).
We must also prepare the script “sign” for signing.
• To sign server CSR's in order to create real SSL Certificates, use the following command:
[root@deep ssl]# /usr/share/ssl/misc/sign www.mydomain.com.csr
CA signing: www.mydomain.com.csr -> www.mydomain.com.crt:
Using configuration from ca.config
Enter PEM pass phrase:
Check that the request matches the signature
Signature ok
The Subjects Distinguished Name is as follows
countryName :PRINTABLE:'CA'
stateOrProvinceName :PRINTABLE:'Quebec'
localityName :PRINTABLE:'Montreal'
organizationName :PRINTABLE:'OpenNA, Inc.'
organizationalUnitName :PRINTABLE:'Open Network Architecture'
commonName :PRINTABLE:'www.openna.com'
emailAddress :IA5STRING:'noc@openna.com'
Certificate is to be certified until Oct 18 14:59:29 2001 GMT (365 days)
Sign the certificate? [y/n]:y](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-407-2048.jpg)
![OpenSSL 1
CHAPTER 5
408
1 out of 1 certificate requests certified, commit? [y/n]y
Write out database with 1 new entries
Data Base Updated
CA verifying: www.mydomain.com.crt <-> CA cert
www.mydomain.com.crt: OK
This signs the CSR and results in a “www.mydomain.com.crt” file. Move this file to its
appropriate directory as follows.
• To move the CRT file to its appropriate directory, use the following command:
[root@deep ssl]# mv www.mydomain.com.crt certs/
Now you have two files: “www.mydomain.com.key” and “www.mydomain.com.crt”. These
can now, for example, be used as follows, inside the virtual host section of your Apache server's
httpd.conf file:
SSLCertificateFile /usr/share/ssl/certs/www.mydomain.com.crt
SSLCertificateKeyFile /usr/share/ssl/private/www.mydomain.com.key
In this example, www.mydomain.com.crt is our web server Certificate Signing Request Public
Key, and www.mydomain.com.key is our web server RSA Private Key.
The www.mydomain.com.csr file is no longer needed; we can remove it from the system.
• To remove this file from the system, use the following command:
[root@deep ssl]# rm -f www.mydomain.com.csr
WARNING: If you receive an error message during the signing of the certificate, it’s probably
because you’ve entered the wrong FQDN (Fully Qualified Domain Name) for the server when
OpenSSL prompted you for the “Common Name”; the “Common Name” must be something like
“www.mydomain.com” and not “mydomain.com”. Also, since you generate both the certificate
and the CA certificate, it’s important that at least ONE piece of information differs between both
files, or you may encounter problems during the signature of the certificate request.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-408-2048.jpg)
![OpenSSL 1
CHAPTER 5
409
Securing OpenSSL
This small section deals specifically with actions we can take to improve and tighten security
under OpenSSL. It’s important to note that we refer to the features available within the base
installed program and not to any additional software.
Changing the default mode of OpenSSL keys:
Make your keys “Read and Write” only by the super-user “root”. This is important because no
one needs to touch these files.
• To make your keys “read and Write” only by “root”, use the following commands:
[root@deep /]# chmod 750 /usr/share/ssl/private/
[root@deep /]# chmod 400 /usr/share/ssl/certs/ca.crt
[root@deep /]# chmod 400 /usr/share/ssl/certs/www.mydomain.com.crt
[root@deep /]# chmod 400 /usr/share/ssl/private/ca.key
[root@deep /]# chmod 400 /usr/share/ssl/private/www.mydomain.com.key
Some possible uses of OpenSSL software
OpenSSL can be used to:
1. Creation of your own Certifying Authority Server.
2. Creation of RSA, DH and DSA key parameters.
3. Creation of X.509 certificates, CSRs and CRLs.
4. Calculation of Message Digest.
5. Encryption and Descryptiion with Ciphers.
6. SSL/TLS Client and Server Tests.
7. Handling of S/MIME signed or encrypted mail.
8. Provide data confidentiality, integrity, authentication, and electronic signature in
transmission for the users.
9. Secure electronic commerce transactions.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-409-2048.jpg)
![OpenSSH 1
CHAPTER 6
414
Prerequisites
OpenSSH requires that the listed software below be already installed on your system to be able to
compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source
archive files. Please make sure you have this program installed on your machine before you
proceed with this chapter.
OpenSSL is required to run OpenSSH on your system.
NOTE: For more information on OpenSSL software, see its related chapter in this book. Even if
you don’t need to use OpenSSL software to create or hold encrypted key files, it’s important to
note that OpenSSH requires its libraries files to be able to work.
Pristine source
As we don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install OpenSSH, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > OpenSSH1
• And the following one after you install the software:
[root@deep root]# find /* > OpenSSH2
• Then use the following command to get a list of what changed:
[root@deep root]# diff OpenSSH1 OpenSSH2 > OpenSSH-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing OpenSSH
Below are the steps that you must make to configure, compile and optimize the OpenSSH server
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp openssh-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf openssh-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-414-2048.jpg)
![OpenSSH 1
CHAPTER 6
415
Step 2
In order to check that the version of OpenSSH, which you are, going to install, is an original and
unmodified one, please check the supplied signature with the GPG key of OpenSSH available on
the OpenSSH website.
To get a GPG key copy of OpenSSH, please point your browser to the following URL:
ftp://ftp.openbsd.org/pub/OpenBSD/OpenSSH/portable/openssh-3.4p1.tar.gz.sig. For more
information about how to use this key for verification, see the GnuPG chapter in this book.
Step 3
OpenSSH needs a UID and GID to properly run on the system but this UID/GID cannot run as
super-user root; for this reason we must create a special user with no shell privileges on the
system for running sshd daemon. This is required by the privilege separation feature of OpenSSH
by which operations that require root privilege are performed by a separate privileged monitor
process.
• To create this special OpenSSH user on OpenNA Linux, use the following command:
[root@deep tmp]# groupadd -g 39 sshd > /dev/null 2>&1 || :
[root@deep tmp]# useradd -c "SSH Server" -d /var/empty -g 39 -s
/bin/false -u 39 sshd > /dev/null 2>&1 || :
• To create this special OpenSSH user on Red Hat Linux, use the following command:
[root@deep tmp]# groupadd -g 39 sshd > /dev/null 2>&1 || :
[root@deep tmp]# useradd -u 39 -g 39 -s /bin/false -M -r -d /var/empty
sshd > /dev/null 2>&1 || :
The above command will create a null account, with no password, no valid shell, no files owned-
nothing but a UID and a GID for the program. Remember that OpenSSH daemon does not need
to have a shell account on the server.
Step 4
Now, edit the shells file (vi /etc/shells) and add a non-existent shell name
“/bin/false”, which is the one we used in the useradd command above.
[root@deep tmp]# vi /etc/shells
/bin/bash2
/bin/bash
/bin/sh
/bin/false This is our added no-existent shell
Patching OpenSSH to run in chroot jail mode for some users:
There is an external patch available for OpenSSH that allow us to compile OpenSSH with chroot
jail support. If you are interested in compiling OpenSSH to run in chroot jail environment for some
of your users, then I recommend you to follow these steps. If you don’t want to compile OpenSSH
with chroot jail support, you can simply skip these steps and go directly to next section where we
will compile the software for our system.
For OpenSSH to run and work in chroot jail mode, you have to be sure that you have recompiled
your Linux kernel without the Grsecurity option that allows us to enable chroot jail restrictions
protection on the system. You should be sure that “Chroot jail restrictions
(CONFIG_GRKERNSEC_CHROOT) [N/y/?]” is NOT enable or nothing will work.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-415-2048.jpg)
![OpenSSH 1
CHAPTER 6
416
Step 1
First of, we have to retrieve the OpenSSH chroot patch available on the Internet. This patch can
be downloaded from the following location: http://chrootssh.sourceforge.net/
Step 2
Once you have a copy of this patch, you should move it under the /var/tmp directory and patch
your OpenSSH source files.
• This can be done with the following commands:
[root@deep /]# mv osshChroot-3.4.diff /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# patch -p0 < osshChroot-3.4.diff
NOTE: It’s important to note that the version number of the OpenSSH chroot patch that you have to
download from the Internet must match the version number of the OpenSSH software you
intended to install. For example, if the version number of OpenSSH is 3.4p1, you should
download the newer OpenSSH chroot patch that matches this number.
Step 3
After that, move into the newly created OpenSSH source directory and perform the following steps
to configure and optimize the software for your system.
• To move into the newly created OpenSSH directory use the following command:
[root@deep tmp]# cd openssh-3.4p1/
• To configure and optimize OpenSSH use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--sysconfdir=/etc/ssh
--libexecdir=/usr/libexec/openssh
--mandir=/usr/share/man
--with-pam
--with-ipaddr-display
--with-ipv4-default
--with-md5-passwords
--with-zlib
This tells OpenSSH to set itself up for this particular configuration setup with:
- Enable PAM support.
- Use the IP address instead of hostname.
- Use IPv4 by connections.
- Enable use of MD5 passwords.
- Use zlib for transport compression.
NOTE: Pay special attention to the compile CFLAGS line above. We optimize OpenSSH for an i686
CPU architecture with the parameter “-march=i686”. Please don’t forget to adjust this CFLAGS
line to reflect your own system and architecture.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-416-2048.jpg)
![OpenSSH 1
CHAPTER 6
417
Step 4
Now, we must make a list of all existing files on the system before installing the software and one
afterwards then compare them using the diff utility to find out what files are placed where and
finally install the OpenSSH Server:
[root@deep openssh-3.4p1]# make
[root@deep openssh-3.4p1]# cd
[root@deep root]# find /* > OpenSSH1
[root@deep root]# cd /var/tmp/openssh-3.4p1/
[root@deep openssh-3.4p1]# make install
[root@deep openssh-3.4p1]# mkdir /var/empty
[root@deep openssh-3.4p1]# chown root.sys /var/empty
[root@deep openssh-3.4p1]# cd
[root@deep root]# find /* > OpenSSH2
[root@deep root]# diff OpenSSH1 OpenSSH2 > OpenSSH-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 5
Once the configuration, optimization, compilation, and installation of the OpenSSH software has
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete OpenSSH and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf openssh-version/
[root@deep tmp]# rm -f openssh-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install OpenSSH. It will also remove the OpenSSH compressed archive from the /var/tmp
directory.
Configuring OpenSSH
After OpenSSH has been built and installed successfully in your system, your next step is to
configure and customize its configuration files to fit your needs.
/etc/ssh/sshd_config: (The OpenSSH Server Configuration File)
/etc/ssh/ssh_config: (The OpenSSH Client Configuration File)
/etc/pam.d/sshd: (The OpenSSH PAM Support Configuration File)
/etc/init.d/sshd: (The OpenSSH Initialization File)
/etc/ssh/sshd_config: The OpenSSH Server Configuration File
The sshd_config file is the system-wide server configuration file for OpenSSH which allows you
to set options that modify the operation of the sshd daemon. This file contains keyword-value
pairs, one per line, with keywords being case insensitive.
Here are the most important keywords to configure your sshd server for maximum security; a
complete listing and/or special requirements are available in the manual page for sshd (8). We
must change the default one to fit our requirements and operating system. The text in bold are
the parts of the configuration file that must be customized and adjusted to satisfy our needs.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-417-2048.jpg)
![OpenSSH 1
CHAPTER 6
423
IdentityFile ~/.ssh/identity
IdentityFile ~/.ssh/id_dsa
IdentityFile ~/.ssh/id_rsa
These options specify alternate multiple authentication identity files to read.
Port 22
This option “Port” specifies on which port number ssh connects to on the remote host. The
default port is 22.
Protocol 2,1
This option “Protocol” specifies the protocol versions ssh should support in order of
preference. In our configuration the default is “2,1”. This means that ssh tries version 2 and falls
back to version 1 if version 2 is not available. Depending of the ssh client version you use to
connect, you may need to invert this order but you can connect with ssh client version 1 even if
the order is “2,1”.
Cipher blowfish
This option “Cipher” specifies what cipher should be used for encrypting sessions. The
blowfish use 64-bit blocks and keys of up to 448 bits.
EscapeChar ~
This option “EscapeChar” specifies the session escape character for suspension.
/etc/pam.d/sshd: The OpenSSH PAM Support Configuration File
For increased security of OpenSSH, we have compiled it to use the PAM mechanism for password
authentication.
Step 1
To be able to use this feature, we must create the /etc/pam.d/sshd file and add the following
parameters inside it.
• Create the sshd file (touch /etc/pam.d/sshd) and add the following lines:
#%PAM-1.0
auth required /lib/security/pam_stack.so service=system-auth
auth required /lib/security/pam_nologin.so
account required /lib/security/pam_stack.so service=system-auth
account required /lib/security/pam_access.so
account required /lib/security/pam_time.so
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_limits.so
Step2
Now, set the permissions of the sshd file to be (0640/-rw-r-----) and owned by the super-
user ‘root’ for security reasons.
• To change the permissions and ownership of the sshd file, use the commands:
[root@deep /]# chmod 640 /etc/pam.d/sshd
[root@deep /]# chown 0.0 /etc/pam.d/sshd](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-423-2048.jpg)
![OpenSSH 1
CHAPTER 6
424
/etc/init.d/sshd: The OpenSSH Initialization File
The /etc/init.d/sshd script file is responsible to automatically starting and stopping the
OpenSSH server on your Linux system. Loading the sshd daemon as a standalone daemon will
eliminate load time and will even reduce swapping since non-library code will be shared.
Please note that the following script is suitable for Linux operating systems that use SystemV. If
you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to
make it work for you.
Step 1
Create the sshd script file (touch /etc/init.d/sshd) and add the following lines:
#!/bin/bash
# This shell script takes care of starting and stopping OpenSSH.
#
# chkconfig: 2345 55 25
# description: OpenSSH is a program thas allows to establish a secure remote
# connection to a server.
#
# processname: sshd
# config: /etc/ssh/ssh_host_key
# config: /etc/ssh/ssh_host_key.pub
# config: /etc/ssh/ssh_random_seed
# config: /etc/ssh/sshd_config
# pidfile: /var/run/sshd.pid
# Source function library.
. /etc/init.d/functions
# Source networking configuration.
. /etc/sysconfig/network
# Source OpenSSH configureation.
if [ -f /etc/sysconfig/sshd ] ; then
. /etc/sysconfig/sshd
fi
RETVAL=0
# Some functions to make the below more readable.
KEYGEN=/usr/bin/ssh-keygen
RSA1_KEY=/etc/ssh/ssh_host_key
RSA_KEY=/etc/ssh/ssh_host_rsa_key
DSA_KEY=/etc/ssh/ssh_host_dsa_key
PID_FILE=/var/run/sshd.pid
my_success() {
local msg
if [ $# -gt 1 ]; then
msg="$2"
else
msg="done"
fi
case "`type -type success`" in
function)
success "$1"
;;
*)
echo -n "${msg}"
;;
esac](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-424-2048.jpg)
![OpenSSH 1
CHAPTER 6
425
}
my_failure() {
local msg
if [ $# -gt 1 ]; then
msg="$2"
else
msg="FAILED"
fi
case "`type -type failure`" in
function)
failure "$1"
;;
*)
echo -n "${msg}"
;;
esac
}
do_rsa1_keygen() {
if ! test -f $RSA1_KEY ; then
echo -n "Generating SSH1 RSA host key: "
if $KEYGEN -q -t rsa1 -f $RSA1_KEY -C '' -N '' >&/dev/null;
then
my_success "RSA1 key generation"
echo
else
my_failure "RSA1 key generation"
echo
exit 1
fi
fi
}
do_rsa_keygen() {
if ! test -f $RSA_KEY ; then
echo -n "Generating SSH2 RSA host key: "
if $KEYGEN -q -t rsa -f $RSA_KEY -C '' -N '' >&/dev/null; then
my_success "RSA key generation"
echo
else
my_failure "RSA key generation"
echo
exit 1
fi
fi
}
do_dsa_keygen() {
if ! test -f $DSA_KEY ; then
echo -n "Generating SSH2 DSA host key: "
if $KEYGEN -q -t dsa -f $DSA_KEY -C '' -N '' >&/dev/null; then
my_success "DSA key generation"
echo
else
my_failure "DSA key generation"
echo
exit 1
fi
fi
}
do_restart_sanity_check() {
sshd -t
RETVAL=$?
if [ ! "$RETVAL" = 0 ]; then
my_failure "Configuration file or keys"
echo](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-425-2048.jpg)
![OpenSSH 1
CHAPTER 6
426
exit $RETVAL
fi
}
case "$1" in
start)
# Create keys if necessary
do_rsa1_keygen;
do_rsa_keygen;
do_dsa_keygen;
echo -n "Starting SSHD: "
if [ ! -f $PID_FILE ] ; then
sshd $OPTIONS
RETVAL=$?
if [ "$RETVAL" = "0" ] ; then
my_success "sshd startup" "sshd"
touch /var/lock/subsys/sshd
else
my_failure "sshd startup" ""
fi
fi
echo
;;
stop)
echo -n "Shutting down SSHD: "
if [ -f $PID_FILE ] ; then
killproc sshd
RETVAL=$?
[ $RETVAL -eq 0 ] && rm -f /var/lock/subsys/sshd
fi
echo
;;
restart)
do_restart_sanity_check
$0 stop
$0 start
RETVAL=$?
;;
condrestart)
if [ -f /var/lock/subsys/sshd ] ; then
do_restart_sanity_check
$0 stop
$0 start
RETVAL=$?
fi
;;
status)
status sshd
RETVAL=$?
;;
*)
echo "Usage: sshd {start|stop|restart|status|condrestart}"
exit 1
;;
esac
exit $RETVAL](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-426-2048.jpg)
![OpenSSH 1
CHAPTER 6
427
Step 2
Once the /etc/init.d/sshd script file has been created, it is important to make it executable,
change its default permissions, create the necessary links and start it. Making this file executable
will allow the system to run it, changing its default permission is to allow only the root user to
change this file for security reason, and creation of the symbolic links will let the process control
initialization of Linux to start the program automatically for you at each system boot.
• To make this script executable and to change its default permissions, use the commands:
[root@deep /]# chmod 700 /etc/init.d/sshd
[root@deep /]# chown 0.0 /etc/init.d/sshd
• To create the symbolic rc.d links for OpenSSH, use the following commands:
[root@deep /]# chkconfig --add sshd
[root@deep /]# chkconfig --level 2345 sshd on
• To start OpenSSH software manually, use the following command:
[root@deep /]# /etc/init.d/sshd start
Starting SSHD: [OK]
Running OpenSSH in a chroot jail
This section applies only if you want to run OpenSSH in chroot jail environment for some of your
users. This kind of setup is useful for web hosting companies that want to provide shell access
via remote secure connection with OpenSSH but don’t want to allow full access to the server and
just limit users to their own web directory. By default, OpenSSH does not support the chroot jail
mode and we have to compile it with an external patch to enable the chroot mode extensions.
The patch is available from the following site: http://chrootssh.sourceforge.net/
Remember that you have to download the version number equal to the OpenSSH version number
you use in order for chroot jail support to work. At the beginning of this chapter, we have already
patched the software with the chroot jail mode extensions patch, therefore, we only need to
create the required skeleton environment and copy the necessary tools into this chroot jail
directory to enable chroot jail support. Below are the steps to follow if you want to run OpenSSH
with chroot jail support for the specified users.
The main benefit of a chroot jail is that the jail will limit the portion of the file system the daemon
can see to the root directory of the jail. Additionally, since the jail only needs to support OpenSSH,
the programs available in the jail can be extremely limited. More importantly, there is no need for
setuid-root programs, which can be used to gain root access and break out of the jail.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-427-2048.jpg)
![OpenSSH 1
CHAPTER 6
429
Necessary steps to run OpenSSH in a chroot jail:
What you're essentially doing is creating a skeleton root file system with enough components
(binaries, libraries, etc.) to allow Unix to do a chroot when the user logs in.
Step 1
With OpenSSH, it’s important to give to your strictly SSH users a real shell account on the Linux
system because we want to allow remote shell access, even if limited to running just a few
commands on the server.
First, create the new users for this purpose; these users will be the users allowed to connect to
your OpenSSH server running in chroot jail mode. These have to be separate from regular user
accounts with unlimited access because of how the "chroot" environment works. Chroot makes it
appear from the user's perspective as if the level of the file system you've placed them in is the
top level of the file system.
Here we create a new SSH user called “gmourani” as an example and set it’s the home directory
under the /home/httpd/gmourani directory since it is the place where it’s the users web
directory and web pages will be located.
• Use the following command to create a new SSH user. This step must be done for each
additional new user you allow to access your OpenSSH server on OpenNA Linux.
[root@deep /]# useradd -m -d /home/httpd/gmourani gmourani
[root@deep /]# passwd gmourani
Changing password for user gmourani
New UNIX password:
Retype new UNIX password:
passwd: all authentication tokens updated successfully
• Use the following command to create a new SSH user. This step must be done for each
additional new user you allow to access your OpenSSH server on Red Hat Linux.
[root@deep /]# useradd -g users -d /home/httpd/gmourani gmourani
[root@deep /]# passwd gmourani
Changing password for user gmourani
New UNIX password:
Retype new UNIX password:
passwd: all authentication tokens updated successfully
The useradd command will add the new SSH user called “gmourani” to our Linux server and
will set it’s the home directory under the /home/httpd/gmourani directory on the system since
it is a useful location for remote clients to maintain their web accounts. The passwd command
will set the password for this user “gmourani”.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-429-2048.jpg)
![OpenSSH 1
CHAPTER 6
430
Step 2
Once the new SSH users have been created, we must edit the /etc/passwd file and make the
appropriated changes to the accounts to allow OpenSSH to chroot when the users login on the
system. In general, the sshd daemon will chroot when it encounters the magic token '/./' in a
users home directory. Therefore this is what we will add to the passwd file for the SSH user in
question.
• Edit the passwd file (vi /etc/passwd) and change the following line:
gmourani:x:501:100::/home/httpd/gmourani:/bin/bash
To read:
gmourani:x:501:100::/home/httpd/gmourani/./:/bin/bash
NOTE: Don’t forget to make the same modification for each additional SSH user for whom you want
to chroot.
Step 3
Now, we have to create all the necessary chrooted environment subdirectories where we will
copy tools we want to allow this SSH user to use on the system.
• Use the following command to create all the necessary chroot subdirectories.
[root@deep /]# mkdir /home/httpd/gmourani/bin
[root@deep /]# mkdir /home/httpd/gmourani/dev
[root@deep /]# mkdir /home/httpd/gmourani/etc
[root@deep /]# mkdir /home/httpd/gmourani/lib
[root@deep /]# mkdir /home/httpd/gmourani/usr
[root@deep /]# mkdir /home/httpd/gmourani/usr/bin
[root@deep /]# mkdir /home/httpd/gmourani/usr/lib
• For Red Hat Linux 7.3 users, you should create the following additional directory:
[root@deep /]# mkdir /home/httpd/gmourani/lib/i686
Step 4
Next, we must change the permissions on all the chroot glue subdirectories to mode (0111/d--
x--x--x) for security reasons.
• Use the following command to change the permissions of all the subdirectories.
[root@deep /]# chmod -R 0111 /home/httpd/gmourani/*](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-430-2048.jpg)
![OpenSSH 1
CHAPTER 6
431
Step 5
Once all permissions of the supporting glues have been changed, it is time to copy the required
binary programs to the related subdirectories in the chroot area for OpenSSH to work.
These programs are necessary to allow the SSH users to list, create directory, copy, remove, and
edit files on the SSH chroot jail directory. If there are features you don’t want the user to be able to
use, then don’t copy them to the chroot area.
• Use the following commands to copy the require binaries programs into the chroot area.
[root@deep /]# cp /bin/bash /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/cp /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/ls /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/mkdir /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/grep /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/rm /home/httpd/gmourani/bin/
[root@deep /]# cp /bin/vi /home/httpd/gmourani/bin/
[root@deep /]# cp /usr/bin/dircolors /home/httpd/gmourani/usr/bin/
[root@deep /]# chmod 0111 /home/httpd/gmourani/bin/*
[root@deep /]# chmod 0111 /home/httpd/gmourani/usr/bin/*
NOTE: The above chmod commands will change default permissions of those programs under the
/bin directories of the chroot jail area to be (0111 ---x--x—x) because we don’t want users to
be able to modify or read binaries in the chroot area but just to execute them if necessary.
Step 6
The binaries we have copied into the chroot area have been compiled with shared libraries by
default and for this reason it is important to find the shared libraries dependencies associated with
them and copy them into the /lib subdirectory in the chroot jail area that we created earlier.
To find the shared library dependencies of binaries, you have to use the ldd command of Linux.
You must copy all the libraries below to the /home/httpd/gmourani/lib directory of the
chroot area. These libraries are part of libc, and needed by various programs.
• Use the following commands to copy the require libraries into the chroot area.
[root@deep /]# cp /lib/libtermcap.so.2 /home/httpd/gmourani/lib/
[root@deep /]# cp /lib/libdl.so.2 /home/httpd/gmourani/lib/
[root@deep /]# cp /lib/libc.so.6 /home/httpd/gmourani/lib/
[root@deep /]# cp /lib/libgcc_s.so.1 /home/httpd/gmourani/lib/
[root@deep /]# cp /lib/ld-linux.so.2 /home/httpd/gmourani/lib/
[root@deep /]# cp /usr/lib/libncurses.so.5 /home/httpd/gmourani/usr/lib/
[root@deep /]# strip -R .comment /home/httpd/gmourani/lib/*
• For Red Hat Linux 7.3 users, you should copy the following additional library:
[root@deep /]# cp /lib/i686/libc.so.6 /home/httpd/gmourani/lib/i686/](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-431-2048.jpg)
![OpenSSH 1
CHAPTER 6
432
WARNING: Depending on what has been compiled, the required shared libraries may be different
than the ones illustrated above. Please use the ldd command on each binary under /bin
subdirectories of the chroot jail to find out the ones you need and copy them to the /lib
subdirectory of the chroot area.
The “strip -R .comment” command will remove all the named section “.comment” from the
libraries files under the /lib subdirectory and will make them smaller in size and can help in the
performance of them.
Step 7
One of the last step to do, is to make a copy of the “DIR_COLORS” and “passwd” files located
under the /etc directory to our chroot jail for SSH to be able to find it.
• Use the following commands to copy the file into the chroot area.
[root@deep /]# cp /etc/DIR_COLORS /home/httpd/gmourani/etc/
[root@deep /]# cp /etc/passwd /home/httpd/gmourani/etc/
Step 8
Finally, we have to create the /home/httpd/gmourani/dev/null device file and set its mode
appropriately.
• Use the following commands to create the null device into the chroot area.
[root@deep /]# mknod /home/httpd/gmourani/dev/null c 1 3
[root@deep /]# chmod 666 /home/httpd/gmourani/dev/null
Creating OpenSSH private & public keys
This section deals with actions that need to be performed to create new private/public keys for
our users to establish a secure connection on the server. There are cryptosystems where
encryption and decryption are done using separate keys, and it is not possible to derive the
decryption key from the encryption key. The idea is that each user creates a public/private key
pair for authentication purposes. The server knows the public key, and only the user knows the
private key.
The file $HOME/.ssh/authorized_keys lists the public keys that are permitted for logging in.
When the user logs in, the ssh program tells the server which key pair it would like to use for
authentication. The server checks if this key is permitted, and if so, sends the user (actually the
ssh program running on behalf of the user) a challenge, a random number, encrypted by the
user's public key. The challenge can only be decrypted using the proper private key. The user's
client then decrypts the challenge using the private key, proving that he/she knows the private
key but without disclosing it to the server.
Step 1
Below, are the steps to follow to create a new SSH private & public key for one user. This
example assumes that secure encrypted connections will be made between Linux servers.
• To create your (RSA) private & public keys for SSH2 of LOCAL, use the commands:
[root@deep /]# su gmourani
[gmourani@deep /]$ ssh-keygen -t rsa
Generating public/private rsa key pair.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-432-2048.jpg)
![OpenSSH 1
CHAPTER 6
433
Enter file in which to save the key (/home/gmourani/.ssh/id_rsa):
Created directory '/home/gmourani/.ssh'.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /home/gmourani/.ssh/id_rsa.
Your public key has been saved in /home/gmourani/.ssh/id_rsa.pub.
The key fingerprint is:
ba:0c:08:6d:9d:51:4f:b3:32:68:9b:0d:83:ce:be:bd gmourani@deep
WARNING: The above example assumes that you want to generate (RSA) private & public keys for
SSH protocol 2 (highly recommended). If you want to generate (RSA) private & public keys for SSH
protocol 1, then you must use the ‘-t rsa1’ option to the key generation command as follows:
[root@deep /]# su gmourani
[gmourani@deep /]$ ssh-keygen -t rsa1
Using the ‘-t rsa1’ option will generate SSH1 instead of SSH2 private & public keys. The SSH1
private key will be named ”identity” and the public key will be “identity.pub”. The '-t'
option is used to specify the type of the key to create. The possible values are "rsa1" for protocol
version 1 and "rsa" or "dsa" for protocol version 2.
If you have multiple accounts you might want to create a separate key on each of them. You may
want to have separate keys for:
• Your server (1)
• Your server (2)
• Your server (3)
This allows you to limit access between these servers, e.g. not allowing the first server (1)
account to access your second server (2) account or the third server (3). This enhances the
overall security in the case any of your authentication keys are compromised for any reason.
Step 2
Copy your local public key id_rsa.pub for SSH2 or identity.pub for SSH1 from the
/home/gmourani/.ssh directory remotely under the name, say, “authorized_keys”. One
way to copy the file is to use the ftp command or you might need to send your public key in
electronic mail to the administrator of the other system. Just include the contents of the
~/.ssh/id_rsa.pub or ~/.ssh/identity.pub file in the message.
To resume the required steps:
1) The user creates his/her RSA keys pair by running ssh-keygen. This stores the private
key in id_rsa (SSH2) or in identity (SSH1) and the public key in id_rsa.pub (SSH2)
or in identity.pub (SSH1) into the user's home directory on the LOCAL machine.
2) The user should then copy the id_rsa.pub key (SSH2) or identity.pub key (SSH1)
to $HOME/.ssh/authorized_keys into his/her home directory on the REMOTE
machine.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-433-2048.jpg)
![OpenSSH 1
CHAPTER 6
434
------------------- -------------------
| | | |
| | | |
| Server 1 | | Server 2 |
| | | |
| | | |
------------------- -------------------
User: gmourani User: gmourani
Pass-phrase: qwerty1 Pass-phrase: qwerty2
Private key: id_rsa Private key: id_rsa
Public key: id_rsa.pub --------------- authorized_keys
authorized_keys ----------------------- Public key: id_rsa.pub
Public key of user gmourani on the first server (1) is sending to the second server (2) under the
$HOME directory of user gmourani and become ‘authorized_keys’; the same action is made
on the second server (2). The public key of user gmourani on server (2) is sending to server (1)
under the $HOME directory of user gmourani and become ‘authorized_keys’.
NOTE: OpenSSH's public key is a one-line string. Adding public keys from commercial SSH tools
which stretch the public key over several lines, will not be recognized by OpenSSH.
OpenSSH Users Tools
The commands listed below are some that we use regularly, but many more exist, and you
should check the manual pages and documentation of OpenSSH for more details.
ssh
The ssh (Secure Shell) command provides secure encrypted communications between two
untrusted hosts over an insecure network. It is a program for securely logging into a remote
machine and executing commands from there. It is a suitable replacement for insecure programs
like telnet, rlogin, rcp, rdist, and rsh.
• To login to a remote machine, use the following command:
[root@deep /]# ssh -l <login_name> <hostname>
For example:
[root@deep /]# ssh -l gmourani deep.openna.com
gmourani@deep.openna.com’s password:
Last login: Tue Oct 19 1999 18:13:00 -0400 from deep.openna.com
No mail.
[gmourani@deep gmourani]$
Where <login_name> is the name you use to connect to the ssh server and <hostname> is
the remote address (you can use IP address here) of your ssh server.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-434-2048.jpg)
![OpenSSH 1
CHAPTER 6
435
scp
The scp (Secure Copy) utility copies files from the local system to a remote system or vice versa,
or even between two remote systems using the scp command.
• To copy files from remote to local system, use the following commands:
[root@deep /]# su gmourani
[gmourani@deep /]$ scp -p <login_name@hostname>:/dir/for/file
localdir/to/filelocation
For example:
[gmourani@deep /]$ scp -p gmourani@mail:/etc/test1 /tmp
Enter passphrase for RSA key 'gmourani@mail.openna.com':
test1 | 2 KB | 2.0 kB/s | ETA: 00:00:00 | 100%
• To copy files from local to remote system, use the following commands:
[root@deep /]# su gmourani
[gmourani@deep /]$ scp -p localdir/to/filelocation
<username@hostname>:/dir/for/file
For example:
[gmourani@deep /]$ scp -p /usr/bin/test2 gmourani@mail:/var/tmp
gmourani@mail's password:
test2 | 7 KB | 7.9 kB/s | ETA: 00:00:00 | 100%
WARNING: The “-p” option indicates that the modification and access times, as well as modes of
the source file, should be preserved on the copy. Usually this is desirable. Please check the
chapter related to backups in this book for more information about other possible uses of SSH
technology with Linux.
Changing your pass-phrase
You can change the pass-phrase at any time by using the -p option of ssh-keygen.
• To change the pass-phrase, use the following commands:
[root@deep /]# su gmourani
[gmourani@deep /]$ ssh-keygen -p
Enter file in which the key is (/home/gmourani/.ssh/id_rsa):
Enter old passphrase:
Key has comment '/home/gmourani/.ssh/id_rsa'
Enter new passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved with the new passphrase.
Further documentation
For more details, there are several manual pages about OpenSSH that you can read:
$ man ssh (1) - OpenSSH secure shell client (remote login program).
$ man ssh [slogin] (1) - OpenSSH secure shell client (remote login program).
$ man ssh-add (1) - Adds identities for the authentication agent.
$ man ssh-agent (1) - Authentication agent.
$ man ssh-keygen (1) - Authentication key generation.
$ man sshd (8) - Secure shell daemon.
$ sftp-server (8) - SFTP server subsystem.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-435-2048.jpg)
![Sudo 1
CHAPTER 7
440
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install Sudo, and then
one afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Sudo1
• And the following one after you install the software:
[root@deep root]# find /* > Sudo2
• Then use this command to get a list of what changed:
[root@deep root]# diff Sudo1 Sudo2 > Sudo-Installed
Compiling - Optimizing & Installing Sudo
Below are the steps that you must make to configure, compile and optimize the Sudo software
before installing it on your system. First off, we install the program as user 'root' so as to avoid
any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp sudo-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf sudo-version.tar.gz
Step 2
Next, move into the newly created Sudo source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created Sudo source directory use the command:
[root@deep tmp]# cd sudo-1.6.6/
• To configure and optimize Sudo use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--sbindir=/usr/sbin
--with-logging=syslog
--with-logfac=authpriv
--with-pam
--with-env-editor
--with-ignore-dot
--with-tty-tickets
--disable-root-mailer
--disable-root-sudo
--disable-path-info
--with-mail-if-noperms
--with-mailsubject="*** Sudo SECURITY information for %h ***"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-440-2048.jpg)
![Sudo 1
CHAPTER 7
441
This tells Sudo to set itself up for this particular configuration setup with:
- Log sudo messages via syslog.
- The syslog facility to log with is authpriv.
- Enable PAM support.
- Use the environment variable EDITOR for visudo.
- Ignore '.' in the PATH.
- Use a different ticket file for each user/tty combo.
- Don't run the mailer as root, run as the user since it’s safer.
- Don't allow root to run sudo command.
- Print 'command not allowed' instead of 'command not found'.
- Send mail to sysadmin if user not allowed to runs command.
- Change subject of sudo mail result.
Step 3
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the Sudo software:
[root@deep sudo-1.6.6]# make
[root@deep sudo-1.6.6]# cd
[root@deep root]# find /* > Sudo1
[root@deep root]# cd /var/tmp/sudo-1.6.6/
[root@deep sudo-1.6.6]# make install
[root@deep sudo-1.6.6]# strip /usr/bin/sudo
[root@deep sudo-1.6.6]# strip /usr/sbin/visudo
[root@deep sudo-1.6.6]# mkdir –p –m0700 /var/run/sudo
[root@deep sudo-1.6.6]# cd
[root@deep root]# find /* > Sudo2
[root@deep root]# diff Sudo1 Sudo2 > Sudo-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 4
Once the configuration, optimization, compilation, and installation of the Sudo software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete Sudo and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf sudo-version/
[root@deep tmp]# rm -f sudo-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-441-2048.jpg)
![Sudo 1
CHAPTER 7
443
This tells the sudoers file to set itself up for this particular configuration with:
Defaults rootpw
With sudo, certain configuration options may be changed from their default values at runtime via
one or more “Default_Entry” options defined in the sudoers file. This is what we do here. In
our case, we inform sudo with the “Defaults rootpw” option that we want it to prompt any
allowed user who wants to sudo to super-user “root” to enter the “root” password instead of
the password of the invoking user.
By default sudo asks for the users password instead of super-user password when someone
wants to sudo to “root” user. Because in this configuration file we want to allow full “root”
access for users that are members of the “wheel” group and because we trust these users, we
decide to change the default sudo setting and ask for “root” password before having access to
“root” privilege on the server.
This setting is useful when you make secure remote connections on the server with SSH software
and want to sudo to “root” user.
root ALL=(ALL) ALL
Here we inform sudo that the super-user “root” can run anything as any user on the server. This
option is required for the software to work or there is no need to use sudo. The “ALL=(ALL)
ALL” parameter means that everything is allowed for the super-user “root”.
%wheel ALL=(ALL) ALL
Here we allow every user who is a member of the group “wheel” to run all commands as super-
user “root” on the system. This is the equivalence of what we achieve with the PAM security
feature, but in a most efficient and secure way.
Step 2
Once the sudoers file has been configured, it is time to add some users who will be allowed to
sudo to “root” account.
• If you want to make, for example, the user “sysadmin” a member of the “wheel” group,
and thus be able to sudo to “root”, use the following command:
[root@deep /]# usermod -G10 sysadmin
This means “G” is a list of supplementary groups that the user is also a member of. “10” is the
numeric value of the user’s ID “wheel”, and “sysadmin” is the user we want to add to the
“wheel” group. Use the same command above for all users on your system you want to be able
to sudo to “root” account.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-443-2048.jpg)
![Sudo 1
CHAPTER 7
444
/etc/pam.d/sudo: The Sudo PAM Support Configuration File
For better security of Sudo, we have compiled it to use the PAM mechanism for password
authentication.
Step 1
To be able to use this feature, we must create the /etc/pam.d/sudo file and add the following
parameters inside it.
• Create the sudo file (touch /etc/pam.d/sudo) and add the following lines:
#%PAM-1.0
auth required /lib/security/pam_stack.so service=system-auth
account required /lib/security/pam_stack.so service=system-auth
password required /lib/security/pam_stack.so service=system-auth
session required /lib/security/pam_stack.so service=system-auth
Step2
Now, set the permissions of the sudo file to be (0640/-rw-r-----) and owned by the super-
user ‘root’ for security reasons.
• To change the permissions and ownership of the sudo file, use the commands:
[root@deep /]# chmod 640 /etc/pam.d/sudo
[root@deep /]# chown 0.0 /etc/pam.d/sudo
A more complex sudoers configuration file
For those who want to have complete control on who can use the “root” account on the server
when there are many users with shell access doing different tasks, here is the sudoers
configuration file to go with. As you’ll see it is more complex and covers some basic definitions.
It is important to understand how the sudo policy works. To be as clear as possible, I will simply
say that when you allow full access to user with the “ALL” definition, you cannot deny other
access or privileges. Therefore, the best way is to allow only what you want user to be able to run
with “root” privilege through the “Cmnd alias specification” part of the configuration file
and use the defined aliases rules under the “User privilege specification” part of the
configuration file. Here is a working sudoers configuration file to better understand what I mean.
• Edit the sudoers file (visudo) and set your needs. Below is what we recommend you
use for servers that have many users with shell access:
# /etc/sudoers: OpenNA, Inc. (last updated 2002 Apr 19)
# This file MUST be edited with the 'visudo' command as root.
# User alias specification
User_Alias FULLTIME_USERS = sysadmin, gmourani
User_Alias PARTTIME_USERS = zeljko, mary
# Cmnd alias specification
Cmnd_Alias HTTP = /etc/init.d/httpd, /bin/vi /etc/httpd/*
Cmnd_Alias FTP = /etc/init.d/proftpd, /bin/vi /etc/proftpd.conf
Cmnd_Alias SMTP = /etc/init.d/exim, /bin/vi /etc/mail/*
Cmnd_Alias SQL = /etc/init.d/mysqld, /usr/bin/mysql,
/usr/bin/mysqladmin
Cmnd_Alias BIND = /etc/init.d/named, /bin/vi /chroot/named/*](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-444-2048.jpg)
![Sudo 1
CHAPTER 7
445
# Defaults specification
Defaults:FULLTIME_USERS rootpw
Defaults:FULLTIME_USERS !lecture
# User privilege specification
# Super-user root can run anything as any user.
root ALL=(ALL) ALL
# Every users member of the group wheel will be allowed
# to run all commands as super-user root.
%wheel ALL=(ALL) ALL
# Full time users may run anything but need a password.
FULLTIME_USERS ALL = ALL
# Part time users may administrate httpd, ftpd, smtp, sql, and bind
servers.
PARTTIME_USERS ALL = HTTP, FTP, SMTP, SQL, BIND
This tells the sudoers file to set itself up for this particular configuration with:
User_Alias FULLTIME_USERS = sysadmin, gmourani
User_Alias PARTTIME_USERS = zeljko, mary
The “User_Alias” definition lines are used to define local users who may sudo to the “root”
account on the server. The definition works on the following way:
Alias_Type NAME = item1, item2, ...
Where “Alias_Type” is the type of alias to use, in our case, we use “User_Alias” to define
local users aliases on the system. A NAME is a string of uppercase letters, numbers, or
underscores characters ('_'). A NAME must always start with an uppercase letter. You can use as
any name as you like to define the NAME.
In our example, we use “FULLTIME_USERS” to define local users on the system who have full
time access to the server and “PARTTIME_USERS” to define local users on the server who have
part time access to the server for different reason. Item represents usernames to add in each
category separated by [,].
NOTE: It is important to note that users added to the “User_Alias” definition will be able to sudo
to super-user “root” account even if their names do not appear under the group “wheel”.
Cmnd_Alias HTTP = /etc/init.d/httpd, /bin/vi /etc/httpd/*
Cmnd_Alias FTP = /etc/init.d/proftpd, /bin/vi /etc/proftpd.conf
Cmnd_Alias SMTP = /etc/init.d/exim, /bin/vi /etc/mail/*
Cmnd_Alias SQL = /etc/init.d/mysqld, /usr/bin/mysql, /usr/bin/mysqladmin
Cmnd_Alias BIND = /etc/init.d/named, /bin/vi /chroot/named/*
Here we use another “Alias_Type” definition. This type of alias is used to define commands,
programs, files, and directories that we want to allow certain users to be able to use when issuing
the sudo command on the system. The definition works on the following way:
Alias_Type NAME = item1, item2, ...](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-445-2048.jpg)
![Sudo 1
CHAPTER 7
446
Where “Alias_Type” is the type of alias to use, in our case, we use “Cmnd_Alias” to define
command aliases on the system. A NAME is a string of uppercase letters, numbers, or
underscores characters ('_'). A NAME must always start with an uppercase letter. You can use any
name you like to define the NAME.
In our example, we use “HTTP, FTP, SMTP, SQL, and BIND” to define command names aliases
associated with the commands we will allow local users to run when issuing the sudo command
on the system. Item represent the commands, files, programs, or directories to add in each
category separated by [,].
Defaults:FULLTIME_USERS rootpw
Defaults:FULLTIME_USERS !lecture
With sudo, certain configuration options may be changed from their default values at runtime via
one or more “Default_Entry” options. Again, this is what we do here. In our case, we inform
sudo with the “Defaults:FULLTIME_USERS rootpw” option that we want it to prompt any
users with full time access on the server “FULLTIME_USERS” to enter the “root” password
instead of their own password.
Remember that by default sudo asks for the users password instead of super-user password
when someone wants to sudo to “root”. Because we allow full “root” access for users under
the “FULLTIME_USERS” alias, we decide to change the default sudo setting and ask for the
“root” password before giving access to “root” privileges on the server. This also means that
users under the “PARTTIME_USERS” alias will have to enter their own password and not the
“root” password.
This is a security feature to separate trusted users with full time access on the server from semi-
trusted users with part time access on the system. Users having part time access could be
students, limited administrators, or anything else that you think about. In this way, users with part
time access do not know the “root” password since they enter their own passwords to get some
“root” privileges and we don’t need to change “root” password every time these users leave
the company.
The second “Default_Entry” option “Defaults:FULLTIME_USERS !lecture”, simply
informs sudo to not send a short lecture message about the policy use of the program to users
with full time access on the server.
FULLTIME_USERS ALL = ALL
Here we allow every user who is listed in the “FULLTIME_USERS” alias of the configuration file to
run all commands as super-user “root” on the system. This is the equivalence of what we
achieve with the PAM security feature.
PARTTIME_USERS ALL = HTTP, FTP, SMTP, SQL, BIND
Here we allow every user who is listed in the “PARTTIME_USERS” alias of the configuration file to
only run the specified commands as super-user “root”. These users will be able to edit, run and
restart the specified services and nothing else. Therefore, they only have limited “root” access
on the server to accomplish their tasks.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-446-2048.jpg)
![Sudo 1
CHAPTER 7
447
Securing Sudo
This section deals specifically with actions we can take to improve and tighten security under
Sudo. Sudo is very good and well-written software with high security in mind. Once properly
compiled, installed, and configured, there are only some little things that we can do to better
secure it. Most important of all the security measures are already made within the software.
Disabling su access on the server
Once sudo is configured and running on your server, you really don’t need to continue to use ‘su’
to get “root” access on the system. Here, I show you how to disable ‘su’ command to provide
“root” access. The simpler approach will be to remove the SUID bit on the ‘su’ program. In this
way, no one will be able to use it to gain “root” anymore.
• To remove the SUID bit on the ‘su’ command, use the following command:
[root@deep /]# chmod 0511 /bin/su
Sudo Users Tools
The commands and options listed bellows are some that we use regularly, but many more exist,
and you should check the manual pages and documentation of Sudo for more details.
Running sudo for user with full “root” access
Basically, sudo is used by prepending "sudo" (followed by a space) to your command. It then
prompts you for your personal password or root password depending of the configuration and
then checks the /etc/sudoers configuration file to make sure you have "sudo" permission to
run that command on the server. Sudo runs the command as root (or another user) and logs the
details to syslog. It also logs problems and other invalid uses.
When you have full “root” privileges on the system because you are listed in the sudoers file
as one user with all “root” access right, you can run the sudo command with the -s (shell)
option to become the super-user “root” on the server.
• To sudo as super-user “root” with shell access, use the following command:
[sysadmin@deep /]# sudo –s
Password:
To be able to use the above command, you should have all “root” access rights in the sudoers
configuration file. Please note that, in this example, you have to enter the super-user “root”
password and not the password of the user “sysadmin”.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-447-2048.jpg)
![Sudo 1
CHAPTER 7
448
Running sudo for user with limited “root” access
If you are a user with limited “root” access rights on the server, you cannot use the previous
sudo command to sudo as super-user “root” on the server. Instead, you should prepend the
sudo command with the command you want to run as “root” user. This supposes that
commands you expect to run are listed as allowed commands to use inside the sudoers
configuration file.
• To sudo as super-user “root” with limited access, use the following command:
[mary@deep /]# sudo /etc/init.d/httpd restart
Password:
The above command will restart the httpd web server daemon on the system even if you are the
user called “mary” because the sudoers file allows you to do it as super-user “root”. Please
note that in this example you have to enter the password of user “mary” and not the password of
the super-user “root”.
Further documentation
For more details, there are some manual pages you can read:
$ man sudoers (5) - List of which users may execute what.
$ man sudo (8) - Execute a command as another user.
$ man visudo (8) - Used to edit the sudoers file.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-448-2048.jpg)
![sXid 1
CHAPTER 8
453
Linux sXid
Abstract
SUID/SGID files can be a security hazard. To reduce the risks, we have previously removed the
's' bits from root-owned programs that won't require such privileges (See chapter related to
General System Security), but future and existing files may be set with these ‘s’ bits enabled
without you being notified.
sXid is an all in one suid/sgid monitoring program designed to be run by “cron” on a regular
basis. Basically, it tracks any changes in your s[ug]id files and folders. If there are any new ones,
ones that aren't set any more, or they have changed bits or other modes then it reports the
changes in an easy to read format via email or on the command line. sXid will automate the task
to find all SUID/SGID on your server and report them to you. Once installed you can forget it and
it will do the job for you.
These installation instructions assume
Commands are Unix-compatible.
The source path is /var/tmp (note that other paths are possible, as personal discretion).
Installations were tested on OpenNA Linux & Red Hat Linux 7.3.
All steps in the installation will happen using the super-user account “root”.
Whether kernel recompilation may be required: No
Latest sXid version number is 4.0.2
The procedures given in this chapter are likely to work on all Linux platforms, but we have only
tested it on OpenNA Linux and Red Hat Linux.
Packages
The following is based on information listed by sXid as of 2002/06/24. Please check
ftp://marcus.seva.net/pub/sxid/ regularly for the latest status. We chose to install from source
because it provides the facility to fine tune the installation.
Source code is available from:
sXid Homepage: ftp://marcus.seva.net/pub/sxid/
sXid FTP Site: 137.155.111.51
You must be sure to download: sxid_4.0.2.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-453-2048.jpg)
![sXid 1
CHAPTER 8
454
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install sXid, and then
one afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > sXid1
• And the following one after you install the software:
[root@deep root]# find /* > sXid2
• Then use the following command to get a list of what changed:
[root@deep root]# diff sXid1 sXid2 > sXid-Installed
Compiling - Optimizing & Installing sXid
Below are the steps that you must make to configure, compile and optimize the sXid software
before installing it on your system. First off, we install the program as user 'root' so as to avoid
any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp sxid_version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf sxid_version.tar.gz
Step 2
Now move into the newly created sXid source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created sXid directory use the following command:
[root@deep tmp]# cd sxid-4.0.2/
• To configure and optimize sXid use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--sysconfdir=/etc
--mandir=/usr/share/man
WARNING: Pay special attention to the compile CFLAGS line above. We optimize sXid for an i686
CPU architecture with the parameter “-march=i686”. Please don’t forget to adjust this CFLAGS
line to reflect your own system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-454-2048.jpg)
![sXid 1
CHAPTER 8
455
Step 3
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the sXid software:
[root@deep sXid-4.0.2]# cd
[root@deep root]# find /* > sXid1
[root@deep root]# cd /var/tmp/sxid-4.0.2/
[root@deep sxid-4.0.2]# make install
[root@deep sxid-4.0.2]# cd
[root@deep root]# find /* > sXid2
[root@deep root]# diff sXid1 sXid2 > sXid-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 4
Once the configuration, optimization, compilation, and installation of the sXid software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete sXid and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf sxid-version/
[root@deep tmp]# rm -f sxid_version_tar.gz
The rm command as used above will remove all the source files we have used to compile and
install sXid. It will also remove the sXid compressed archive from the /var/tmp directory.
Configuring sXid
After sXid has been built and installed successfully in your system, your next step is to configure
and customize its configuration files to fit your needs.
/etc/sxid.conf: (The sXid Configuration File)
/etc/cron.daily/sxid: (The sXid Cron File)
/etc/sxid.conf: The sXid Configuration File
The configuration file for sXid allows you to set options that modify the operation of the program.
It is well commented and very basic.
Step 1
We must change the default one to fit our requirements and operating system. The text in bold
are the parts of the configuration file that must be customized and adjusted to satisfy our needs.
• Edit the sxid.conf file (vi /etc/sxid.conf) and set your needs. Below is what we
recommend you.
# Configuration file for sXid
# Note that all directories must be absolute with no trailing /'s
# Where to begin our file search
SEARCH = "/"](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-455-2048.jpg)
![sXid 1
CHAPTER 8
456
# Which subdirectories to exclude from searching
EXCLUDE = "/proc /mnt /cdrom /floppy"
# Who to send reports to
EMAIL = "root"
# Always send reports, even when there are no changes?
ALWAYS_NOTIFY = "no"
# Where to keep interim logs. This will rotate 'x' number of
# times based on KEEP_LOGS below
LOG_FILE = "/var/log/sxid.log"
# How many logs to keep
KEEP_LOGS = "5"
# Rotate the logs even when there are no changes?
ALWAYS_ROTATE = "no"
# Directories where +s is forbidden (these are searched
# even if not explicitly in SEARCH), EXCLUDE rules apply
FORBIDDEN = "/home /tmp"
# Remove (-s) files found in forbidden directories?
ENFORCE = "yes"
# This implies ALWAYS_NOTIFY. It will send a full list of
# entries along with the changes
LISTALL = "no"
# Ignore entries for directories in these paths
# (this means that only files will be recorded, you
# can effectively ignore all directory entries by
# setting this to "/"). The default is /home since
# some systems have /home g+s.
IGNORE_DIRS = "/home"
# File that contains a list of (each on it's own line)
# other files that sxid should monitor. This is useful
# for files that aren't +s, but relate to system
# integrity (tcpd, inetd, apache...).
# EXTRA_LIST = "/etc/sxid.list"
# Mail program. This changes the default compiled in
# mailer for reports. You only need this if you have changed
# it's location and don't want to recompile sxid.
MAIL_PROG = "/bin/mail"
Step 2
Now, for security reasons, change the mode of this file to be 0400.
• This can be done with the following command:
[root@deep /]# chmod 400 /etc/sxid.conf](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-456-2048.jpg)
![sXid 1
CHAPTER 8
457
/etc/cron.daily/sxid: The sXid Cron File
The sxid file is a small script executed automatically by the “crond” program of your server
each day to tracks any changes in your s[ug]id files and folders.
Step 1
If there are any new ones, ones that aren't set any more, or they have changed bits or other
modes then it reports the changes. If you intend to automate this task, follow the simple steps
below.
• Create the sxid script file (touch /etc/cron.daily/sxid) and add the following
lines:
#!/bin/sh
SXID_OPTS=
if [ -x /usr/bin/sxid ]; then
/usr/bin/sxid ${SXID_OPTS}
fi
Step2
Now, make this script executable and change its permissions to be 0510.
• This can be done with the following command:
[root@deep /]# chmod 510 /etc/cron.daily/sxid
sXid Administrative Tools
After your desired configuration options have been set and the program is running, we can play
with its utility. The sXid software is meant to run as a cronjob. It must run once a day, but busy
shell boxes may want to run it twice a day. You can also run this manually for spot-checking.
• To run sXid manually, use the command:
[root@deep /]# sxid -k
sXid Vers : 4.0.2
Check run : Thu Apr 25 19:35:36 2002
This host : deep.openna.com
Spotcheck : /root
Excluding : /proc /mnt /cdrom /floppy
Ignore Dirs: /home
Forbidden : /home /tmp
(enforcing removal of s[ug]id bits in forbidden paths)
No changes found
This checks for changes by recursing the current working directory. Log files will not be rotated
and no email sent. All output will go to stdout.
Further documentation
For more details, there are some manual pages you can read:
$ man sxid.conf (5) - Configuration settings for sxid.
$ man sxid (1) - Check for changes in s[ug]id files and directories.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-457-2048.jpg)
![LogSentry 1
CHAPTER 9
462
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install LogSentry, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > LogSentry1
• And the following one after you install the software:
[root@deep root]# find /* > LogSentry2
• Then use the following command to get a list of what changed:
[root@deep root]# diff LogSentry1 LogSentry2 > LogSentry-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing LogSentry
Below are the steps that you must make to configure, compile and optimize the LogSentry
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp logsentry-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf logsentry-version.tar.gz
Step 2
In order to check that the version of LogSentry, which you are going to install, is an original and
unmodified one, use the command described below to check its MD5 hashes checksum.
• To verify the MD5 checksum of LogSentry, use the following command:
[root@deep tmp]# md5sum logsentry-1.1.1.tar.gz
This should yield an output similar to this:
e97c2f096e219e20310c1b80e9e1bc29 logsentry-1.1.1.tar.gz
Now check that this checksum is exactly the same as the one published on the LogSentry
website at the following URL: http://www.psionic.com/downloads/checksums.md5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-462-2048.jpg)
![LogSentry 1
CHAPTER 9
463
Step 3
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to configure the program for our PATH environment variable under
Linux. Therefore, move into the newly created LogSentry source directory and perform the
following steps to configure and optimize the software for your system.
• To move into the newly created LogSentry directory use the following command:
[root@deep tmp]# cd logcheck-1.1.1/
Step 4
Here, we have to change default locations of different LogSentry configuration files on the
system. To archive these modifications, we must edit the logcheck.sh script file as follow.
• Edit the logcheck.sh file and change all of the targeted lines in the order shown below:
a) vi +47 systems/linux/logcheck.sh and change the line:
LOGTAIL=/usr/local/bin/logtail
To read:
LOGTAIL=/usr/bin/logtail
b) vi +55 systems/linux/logcheck.sh and change the line:
TMPDIR=/usr/local/etc/tmp
To read:
TMPDIR=/var/logsentry
c) vi +92 systems/linux/logcheck.sh and change the line:
HACKING_FILE=/usr/local/etc/logcheck.hacking
To read:
HACKING_FILE=/etc/logsentry/hacking
d) vi +101 systems/linux/logcheck.sh and change the line:
VIOLATIONS_FILE=/usr/local/etc/logcheck.violations
To read:
VIOLATIONS_FILE=/etc/logsentry/violations](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-463-2048.jpg)
![LogSentry 1
CHAPTER 9
465
d) vi +30 Makefile and change the line:
INSTALLDIR_SH = /usr/local/etc
To read:
INSTALLDIR_SH = /usr/sbin
e) vi +30 Makefile and change the line:
TMPDIR = /usr/local/etc/tmp
To read:
TMPDIR = /var/logsentry
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the LogSentry software:
[root@deep logcheck-1.1.1]# cd
[root@deep root]# find /* > LogSentry1
[root@deep root]# cd /var/tmp/logcheck-1.1.1/
[root@deep logcheck-1.1.1]# mkdir –m0700 /etc/logsentry
[root@deep logcheck-1.1.1]# make linux
[root@deep logcheck-1.1.1]# strip /usr/bin/logtail
[root@deep logcheck-1.1.1]# cd /etc/logsentry/
[root@deep logsentry]# mv logcheck.hacking hacking
[root@deep logsentry]# mv logcheck.violations violations
[root@deep logsentry]# mv logcheck.violations.ignore violations.ignore
[root@deep logsentry]# mv logcheck.ignore ignore
[root@deep logsentry]# cd
[root@deep root]# find /* > LogSentry2
[root@deep root]# diff LogSentry1 LogSentry2 > LogSentry-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 7
Once the configuration, optimization, compilation, and installation of the LogSentry software
have been accomplished, we can free up some disk space by deleting the program tar archive
and the related source directory since they are no longer needed.
• To delete LogSentry and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf logcheck-version/
[root@deep tmp]# rm -f logsentry-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install LogSentry. It will also remove the LogSentry compressed archive from the /var/tmp
directory.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-465-2048.jpg)
![LogSentry 1
CHAPTER 9
466
Configuring LogSentry
After LogSentry has been built and installed successfully in your system, your next step is to
check its configuration files to see if they fit your needs.
/etc/logsentry/hacking: (The LogSentry Hacking File)
/etc/logsentry/ignore: (The LogSentry Ignore File)
/etc/logsentry/violations: (The LogSentry Violation File)
/etc/logsentry/violations.ignore: (The LogSentry Violation Ignore File)
From the default install, there are no LogSentry configuration files to modify, the default entries
look fine and if you want to make some personal adjustment, all you have to do is to edit the
related LogSentry configuration files located under /etc/logsentry directory.
More information about the operation of each one is contained in the INSTALL file of LogSentry
under its uncompressed source directory.
Although the fact that there is no LogSentry configuration files to change, the last action to
make before using the program is to automate it.
Step 1
Create a file called logsentry under the /etc/cron.daily directory and add the following
lines to set LogSentry to run once per day.
• To create the logsentry file under /etc/cron.daily directory, type the following
lines in your terminal (as root):
cat <<EOF > /etc/cron.daily/logsentry
#!/bin/sh
# Hourly check Log files for security violations and unusual activity.
/usr/bin/logcheck.sh
EOF
Step 2
Now, make this script executable and change its permissions to be 0510.
• This can be done with the following commands:
[root@deep /]# chmod 510 /etc/cron.daily/logsentry
WARNING: Remember, in our configuration and installation, LogSentry does not report anything
via email if it has nothing useful to say.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-466-2048.jpg)
![HostSentry 2
CHAPTER 0
470
Prerequisites
HostSentry requires that the listed software below be already installed on your system to be
able to compile successfully. If this is not the case, you must install it from your Linux CD-ROM or
source archive files. Please make sure you have this program installed on your machine before
you proceed with this chapter.
Python, which allows HostSentry to run, must already be installed on your system to
be able to compile and use the HostSentry software.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install HostSentry, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > HostSentry1
• And the following one after you install the software:
[root@deep root]# find /* > HostSentry2
• Then use the following command to get a list of what changed:
[root@deep root]# diff HostSentry1 HostSentry2 > HostSentry-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing HostSentry
Below are the steps that you must make to configure, compile and optimize the HostSentry
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp hostsentry-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf hostsentry-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-470-2048.jpg)
![HostSentry 2
CHAPTER 0
471
Step 2
In order to check that the version of HostSentry, which you are going to install, is an original
and unmodified one, use the command described below to check its MD5 hashes checksum.
• To verify the MD5 checksum of HostSentry, use the following command:
[root@deep tmp]# md5sum hostsentry-0.02.tar.gz
This should yield an output similar to this:
3de0bbb7d456bb53683de56dfdf98362 hostsentry-0.02.tar.gz
Now check that this checksum is exactly the same as the one published on the HostSentry
website at the following URL: http://www.psionic.com/downloads/checksums.md5
Step 3
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to configure the program for our PATH environment variable under
Linux. Therefore, move into the newly created HostSentry source directory and perform the
following steps to configure and optimize the software for your system.
• To move into the newly created HostSentry directory use the following command:
[root@deep tmp]# cd hostsentry-0.02/
Step 4
First, we have to define directories where we want HostSentry to be installed on our system.
Editing the Makefile script file as follows does this:
• Edit the Makefile file and change all of the targeted lines in the order shown below:
a) vi +7 Makefile and change the line:
INSTALLDIR = /usr/local/abacus/hostsentry
To read:
INSTALLDIR = /etc/hostsentry
LIBDIR= /usr/lib/hostsentry
b) vi +21 Makefile and change the lines:
@echo "Installing HostSentry in: $(INSTALLDIR)"
install -d -g 0 -o root -m 0700 $(INSTALLDIR)
install -d -g 0 -o root -m 0700 $(INSTALLDIR)/modules
install -g 0 -o root -m 0700 host* $(INSTALLDIR)
install -g 0 -o root -m 0700 module* $(INSTALLDIR)/modules
To read:
install -d -m 0700 $(INSTALLDIR)
install -d -m 0700 $(LIBDIR)/modules
install -m 0400 host* $(LIBDIR)
install -m 0400 module* $(LIBDIR)/modules](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-471-2048.jpg)
![HostSentry 2
CHAPTER 0
473
• Edit the moduleMultipleLogins.py file (vi +78 moduleMultipleLogins.py)
and change the line:
allowPath = config.parseToken('MODULE_PATH')
To read:
allowPath = '/etc/hostsentry/'
Step 6
Finally, we have to edit the hostsentry.py file and add a new line at the BEGINNING of the file
to set the environment variable of the python binary for the program to find and use it when it
runs.
• Edit the hostsentry.py file (vi hostsentry.py) and add the line:
#!/usr/bin/env python
Step 7
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the HostSentry software:
[root@deep hostsentry-0.02]# cd
[root@deep root]# find /* > HostSentry1
[root@deep root]# cd /var/tmp/hostsentry-0.02/
[root@deep hostsentry-0.02]# make install
[root@deep hostsentry-0.02]# mkdir -m0700 /var/hostsentry
[root@deep logsentry]# cd
[root@deep root]# find /* > HostSentry2
[root@deep root]# diff HostSentry1 HostSentry2 > HostSentry-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 8
Once the configuration, optimization, compilation, and installation of the HostSentry software
have been accomplished, we can free up some disk space by deleting the program tar archive
and the related source directory since they are no longer needed.
• To delete HostSentry and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf hostsentry-version/
[root@deep tmp]# rm -f hostsentry-version.tar.gz](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-473-2048.jpg)
![HostSentry 2
CHAPTER 0
474
Configuring HostSentry
After HostSentry has been built and installed successfully in your system, your next step is to
configure and customize its configuration files.
/etc/hostsentry/hostsentry.conf: (The HostSentry Configuration File)
/etc/hostentry/hostsentry.ignore: (The HostSentry Ignore File)
/etc/hostentry/hostsentry.action: (The HostSentry Action File)
/etc/hostentry/hostsentry.modules: (The HostSentry Modules File)
/etc/hostentry/moduleForeignDomain.allow
/etc/hostentry/moduleMultipleLogins.allow
/etc/init.d/hostsentry: (The HostSentry Initialization File)
/etc/hostsentry/hostsentry.conf: The HostSentry Config File
The hostsentry.conf file is the main configuration file for HostSentry. It is in this file that
HostSentry gets all of its information about the way it should run on your system.
Step 1
By default, the hostsentry.conf file does not exist after installation and we have to create it.
• Create the hostsentry.conf file (touch /etc/hostsentry/hostsentry.conf)
and set your needs. Below is what we recommend you have in your file.
IGNORE_FILE = "/etc/hostsentry/hostsentry.ignore"
ACTION_FILE = "/etc/hostsentry/hostsentry.action"
MODULE_FILE = "/etc/hostsentry/hostsentry.modules"
MODULE_PATH = "/usr/lib/hostsentry/modules"
WTMP_FILE = "/var/log/wtmp"
DB_FILE = "/var/hostsentry/hostsentry.db"
DB_TTY_FILE = "/var/hostsentry/hostsentry.tty.db"
WTMP_FORMAT = "384/8:32/44:32/76:256"
Step2
Now, set the permissions of the hostsentry.conf file to be (0600/-rw-------) and owned
by the super-user ‘root’ for security reasons.
• To change the permissions and ownership of the hostsentry.conf file, use:
[root@deep /]# chmod 600 /etc/hostsentry/hostsentry.conf
[root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.conf](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-474-2048.jpg)
![HostSentry 2
CHAPTER 0
475
/etc/hostsentry/hostsentry.ignore: The HostSentry Ignore File
The hostsentry.ignore file contains a list of users you want HostSentry to never process
or never take action against with the modules. This is useful for users such as "ftp" who show
up in wtmp but would cause a large number of false alarms because of the anonymous access. It
is important to note that each user must be placed one per line.
Step 1
By default, the hostsentry.ignore file doesn’t exist after installation and we have to create it.
• Create the hostsentry.ignore file (touch /etc/hostsentry/hostsentry.ignore)
and put in any user account you want to be ignored if it logs in to the system. Below is
what we recommend you put in the file. By default, we have no users defined.
# Place usernames in this file that you want to ignore (ftp, etc.)
Step2
Now, set the permissions of the hostsentry.ignore file to be (0600/-rw-------) and
owned by the super-user ‘root’ for security reasons.
• To change the permissions and ownership of the hostsentry.ignore file, use:
[root@deep /]# chmod 600 /etc/hostsentry/hostsentry.ignore
[root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.ignore
/etc/hostsentry/hostsentry.action: The HostSentry Action File
The hostsentry.action file is used to define actions we want HostSentry to take on the
specified module. In our example we inform it to logs, blocks the route connection, blocks the TCP
connection and disables the user access. It seems that this feature is not implemented but we
define and configure it for future version.
Step 1
By default, the hostsentry.action file doesn’t exist after installation, so we have to create it
manually.
• Create the hostsentry.action file (touch /etc/hostsentry/hostsentry.action)
and add in any action you want to be taken. Below is what we recommend you enter.
moduleFirstLogin=log,blockRoute,blockTCP,disable
Step2
Now, set the permissions of the hostsentry.action file to be (0600/-rw-------) and
owned by the super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the hostsentry.action file, use:
[root@deep /]# chmod 600 /etc/hostsentry/hostsentry.action
[root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.action](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-475-2048.jpg)
![HostSentry 2
CHAPTER 0
476
/etc/hostsentry/hostsentry.modules: The HostSentry Modules File
The hostsentry.modules file tells HostSentry what modules to execute on login/logout and
in what order. If you don't want a particular module to run for whatever reason (false alarms, not
interested, etc.) then delete it here.
Step 1
By default, the hostsentry.modules file doesn’t exist after installation, so we have to create it.
• Create hostsentry.modules file (touch /etc/hostsentry/hostsentry.modules)
and add the following module lines to the file. Below is what we recommend.
moduleLoginLogout
moduleFirstLogin
moduleForeignDomain
moduleMultipleLogins
moduleRhostsCheck
moduleHistoryTruncated
moduleOddDirnames
Step2
Now, set the permissions of the hostsentry.modules file to be (0600/-rw-------) and
owned by the super-user ‘root’ for security reasons.
• To change the permission and ownership of the hostsentry.modules file, use:
[root@deep /]# chmod 600 /etc/hostsentry/hostsentry.modules
[root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.modules
/etc/hostsentry/moduleForeignDomain.allow
The moduleForeignDomain.allow file is used to list all domains from which we don’t want an
alert to be sent to the administrator when they log in to the system. Every domain listed in this file
will be processed as allowed domains. I recommend you only add your localhost to this file.
Step 1
By default, the moduleForeignDomain.allow file doesn’t exist after installation and we have
to create it.
• Create the moduleForeignDomain.allow file (touch
/etc/hostsentry/moduleForeignDomain.allow) and add the following line.
:0.0
Step2
Now, set the permissions of the moduleForeignDomain.allow file to be (0600/-rw-------)
and owned by the super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the moduleForeignDomain.allow
file, use the following commands:
[root@deep /]# chmod 600 /etc/hostsentry/moduleForeignDomain.allow
[root@deep /]# chown 0.0 /etc/hostsentry/moduleForeignDomain.allow](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-476-2048.jpg)
![HostSentry 2
CHAPTER 0
477
/etc/hostsentry/moduleMultipleLogins.allow
The moduleMultipleLogins.allow file is used to list all hosts from which multiple loggings
are allowed. This mean that all hosts listed in this file will be allowed to make multiple connections
from different place without an alert to be send to the administrator. Again, I recommend you to
only add your localhost to this file as we do below.
Step 1
By default, the moduleMultipleLogins.allow file does not exist after installation; therefore
we have to create it.
• Create the moduleMultipleLogins.allow file (touch
/etc/hostsentry/moduleMultipleLogins.allow) and add the following line.
Below is what we recommend.
# Place hosts in here you want this module to disregard logins from.
localhost
Step2
Now, set the permissions of the moduleMultipleLogins.allow file to be (0600/-rw-------)
and owned by the super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the moduleMultipleLogins.allow
file, use the following commands:
[root@deep /]# chmod 600 /etc/hostsentry/moduleMultipleLogins.allow
[root@deep /]# chown 0.0 /etc/hostsentry/moduleMultipleLogins.allow
/etc/init.d/hostsentry: The HostSentry Initialization File
The /etc/init.d/hostsentry script file is responsible to automatically starting and stopping
the HostSentry server on your Linux system.
Please note that the following script is suitable for Linux operating systems that use SystemV. If
you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to
make it work for you.
Step 1
Create the hostsentry script file (touch /etc/init.d/hostsentry) and add the following
lines inside it:
#!/bin/bash
# This shell script takes care of starting and stopping HostSentry.
#
# chkconfig: 345 98 85
# description: HostSentry is a host based intrusion detection tool that
# performs Login Anomaly Detection (LAD). This tool allows
# administrators to spot strange login behavior and quickly
# respond to compromised accounts and unusual behavior.
#
# processname: hostsentry
# config: /etc/hostsentry/hostsentry.conf
# pidfile: /var/run/hostsentry.pid
# Source function library.
. /etc/init.d/functions](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-477-2048.jpg)
![HostSentry 2
CHAPTER 0
478
# Source networking configuration.
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
RETVAL=0
prog="HostSentry"
start() {
if [ -f /var/run/hostsentry.pid ] ; then
pid=`cat /var/run/hostsentry.pid`
if [ "$pid" != "" ] ; then
echo $"HostSentry is already running"
exit 0
fi
fi
echo -n $"Starting $prog: "
cd /usr/lib/hostsentry
daemon python hostsentry.py
RETVAL=$?
echo
echo `ps aux | grep "python hostsentry.py" | cut --delimiter=" " -f 7`
>
/var/run/hostsentry.pid
[ $RETVAL -eq 0 ] && touch /var/lock/subsys/hostsentry
return $RETVAL
}
stop() {
echo -n $"Shutting down $prog: "
cd /usr/lib/hostsentry
killproc python hostsentry.py
RETVAL=$?
echo
[ $RETVAL -eq 0 ] && rm -f /var/lock/subsys/hostsentry && rm -f
/var/run
/hostsentry.pid
return $RETVAL
}
restart() {
stop
start
}
condrestart() {
if [ -f /var/lock/subsys/hostsentry ]; then
restart
fi
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;
restart)
restart](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-478-2048.jpg)
![HostSentry 2
CHAPTER 0
479
;;
condrestart)
condrestart
;;
*)
echo $"Usage: $0 {start|stop|restart|condrestart}"
exit 1
;;
esac
Step 2
Once the /etc/init.d/hostsentry script file has been created, it is important to make it
executable, change its default permissions, create the necessary links and start it. Making this file
executable will allow the system to run it, changing its default permission is to allow only the root
user to change this file for security reason, and creation of the symbolic links will let your system
start the program automatically for you at each system boot.
• To make this script executable and to change its default permissions, use the commands:
[root@deep /]# chmod 700 /etc/init.d/hostsentry
[root@deep /]# chown 0.0 /etc/init.d/hostsentry
• To create the symbolic rc.d links for HostSentry, use the following commands:
[root@deep /]# chkconfig --add hostsentry
[root@deep /]# chkconfig --level 345 hostsentry on
• To start HostSentry software manually, use the following command:
[root@deep /]# /etc/init.d/hostsentry start
Starting HostSentry: [OK]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-479-2048.jpg)
![PortSentry 2
CHAPTER 1
484
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install PortSentry, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > PortSentry1
• And the following one after you install the software:
[root@deep root]# find /* > PortSentry2
• Then use the following command to get a list of what changed:
[root@deep root]# diff PortSentry1 PortSentry2 > PortSentry-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing PortSentry
Below are the steps that you must make to configure, compile and optimize the PortSentry
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp portsentry-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf portsentry-version.tar.gz
Step 2
In order to check that the version of PortSentry, which you are going to install, is an original
and unmodified one, use the command described below to check its MD5 hashes checksum.
• To verify the MD5 checksum of PortSentry, use the following command:
[root@deep tmp]# md5sum portsentry-1.1.tar.gz
This should yield an output similar to this:
782839446b7eca554bb1880ef0882670 portsentry-1.1.tar.gz
Now check that this checksum is exactly the same as the one published on the PortSentry
website at the following URL: http://www.psionic.com/downloads/checksums.md5](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-484-2048.jpg)
![PortSentry 2
CHAPTER 1
485
Step 3
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to configure the program for our PATH environment variable under
Linux. Therefore, move into the newly created PortSentry source directory and perform the
following steps to configure and optimize the software for your system.
• To move into the newly created PortSentry directory use the following command:
[root@deep tmp]# cd portsentry-1.1/
Step 4
Here, we have to change default locations of different PortSentry configuration files on the
system and add the required optimization FLAGS for our CPU architecture. To make these
modifications, we must edit the Makefile script file as follows.
• Edit the Makefile file and change all of the targeted lines in the order shown below:
a) vi +29 Makefile and change the line:
CFLAGS = -O -Wall
To read:
CFLAGS = -O2 -march=i686 -funroll-loops -Wall
b) vi +40 Makefile and change the lines:
INSTALLDIR = /usr/local/psionic
CHILDDIR=/portsentry
To read:
INSTALLDIR = /usr/sbin
CONFIGDIR = /etc/portsentry
c) vi +71 Makefile and change the line:
@if [ ! -d $(INSTALLDIR) ]; then /bin/mkdir $(INSTALLDIR); fi
To read:
@if [ ! -d $(INSTALLDIR) ]; then /bin/mkdir -p $(INSTALLDIR); fi
d) vi +73 Makefile and change the lines:
@if [ "$(INSTALLDIR)" = "/usr/local/psionic" ]; then /bin/chmod 700 $(INSTALLDIR) ; fi
@echo "Creating portsentry directory $(INSTALLDIR)$(CHILDDIR)"
@if [ ! -d $(INSTALLDIR)$(CHILDDIR) ]; then /bin/mkdir $(INSTALLDIR)$(CHILDDIR); fi
To read:
@if [ "$(INSTALLDIR)" = "/usr/sbin" ]; then /bin/chmod 700 $(INSTALLDIR) ; fi
@echo "Creating portsentry directory $(CONFIGDIR)"
@if [ ! -d $(CONFIGDIR) ]; then /bin/mkdir -p $(CONFIGDIR); fi](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-485-2048.jpg)
![PortSentry 2
CHAPTER 1
487
Step 5
The second file that we will modify is the portsentry_config.h header file. In this file, we will
change the default install location of the configuration file for PortSentry.
• Edit the portsentry_config.h file (vi +32 portsentry_config.h) and change
the following line:
#define CONFIG_FILE "/usr/local/psionic/portsentry/portsentry.conf"
To read:
#define CONFIG_FILE "/etc/portsentry/portsentry.conf"
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the PortSentry software:
[root@deep portsentry-1.1]# cd
[root@deep root]# find /* > PortSentry1
[root@deep root]# cd /var/tmp/portsentry-1.1/
[root@deep portsentry-1.1]# make linux
[root@deep portsentry-1.1]# make install
[root@deep portsentry-1.1]# strip /usr/sbin/portsentry
[root@deep portsentry-1.1]# mkdir -m0700 /var/portsentry
[root@deep portsentry-1.1]# cd
[root@deep root]# find /* > PortSentry2
[root@deep root]# diff PortSentry1 PortSentry2 > PortSentry-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 7
Once the configuration, optimization, compilation, and installation of the PortSentry software
have been accomplished, we can free up some disk space by deleting the program tar archive
and the related source directory since they are no longer needed.
• To delete PortSentry and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf portsentry-version/
[root@deep tmp]# rm -f portsentry-version.tar.gz
Configuring PortSentry
After PortSentry has been built and installed successfully on your system, your next step is to
configure and customize its configuration files to fit your needs.
/etc/portsentry/portsentry.conf: (The PortSentry Configuration File)
/etc/portsentry/portsentry.ignore: (The PortSentry Ignore File)
/etc/portsentry/portsentry.modes: (The PortSentry Modes File)
/etc/init.d/portsentry: (The PortSentry Initialization File)](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-487-2048.jpg)
![PortSentry 2
CHAPTER 1
492
Step2
Now, set the permissions of the portsentry.modes file to be (0600/-rw-------) and owned
by the super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the portsentry.modes file, use:
[root@deep /]# chmod 600 /etc/portsentry/portsentry.modes
[root@deep /]# chown 0.0 /etc/portsentry/portsentry.modes
/etc/init.d/portsentry: The PortSentry Initialization File
The /etc/init.d/portsentry script file is responsible to automatically starting and stopping
the PortSentry server on your Linux system.
Please note that the following script is suitable for Linux operating systems that use SystemV. If
you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to
make it work for you.
Step 1
Create the portsentry script file (touch /etc/init.d/portsentry) and add the following
lines inside it:
#!/bin/bash
# This shell script takes care of starting and stopping the Port Scan Detector.
#
# chkconfig: 345 98 05
# description: PortSentry Port Scan Detector is part of the Abacus Project
# suite of tools. The Abacus Project is an initiative to release
# low-maintenance, generic, and reliable host based intrusion
# detection software to the Internet community.
#
# processname: portsentry
# config: /etc/portsentry/portsentry.conf
# pidfile: /var/run/portsentry.pid
# Source function library.
. /etc/rc.d/init.d/functions
# Source networking configuration.
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
RETVAL=0
prog="PortSentry"
start() {
SENTRYDIR=/etc/portsentry
if [ -s $SENTRYDIR/portsentry.modes ] ; then
modes=`cut -d "#" -f 1 $SENTRYDIR/portsentry.modes`
else
modes="tcp udp"
fi
for i in $modes ; do
action "Starting $prog -$i: " /usr/sbin/portsentry -$i
RETVAL=$?](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-492-2048.jpg)
![PortSentry 2
CHAPTER 1
493
done
echo
[ $RETVAL = 0 ] && touch /var/lock/subsys/portsentry
return $RETVAL
}
stop() {
echo -n "Shutting down $prog: "
killproc portsentry
RETVAL=$?
echo
[ $RETVAL = 0 ] && rm -f /var/lock/subsys/portsentry
return $RETVAL
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;
restart|reload)
stop
start
RETVAL=$?
;;
condrestart)
[ -f /var/lock/subsys/portsentry ] && restart || :
;;
status)
status portsentry
;;
*)
echo "Usage: portsentry {start|stop|restart|reload|condrestart|status}"
exit 1
esac
Step 2
Once the /etc/init.d/portsentry script file has been created, it is important to make it
executable, change its default permissions, create the necessary links and start it. Making this file
executable will allow the system to run it, changing its default permission is to allow only the root
user to change this file for security reasons, and creation of the symbolic links to start the
program automatically for you at each system boot.
• To make this script executable and to change its default permissions, use the command:
[root@deep /]# chmod 700 /etc/init.d/portsentry
[root@deep /]# chown 0.0 /etc/init.d/portsentry
• To create the symbolic rc.d links for PortSentry, use the following command:
[root@deep /]# chkconfig --add portsentry
[root@deep /]# chkconfig --level 345 portsentry on
• To start PortSentry software manually, use the following command:
[root@deep /]# /etc/init.d/portsentry start
Starting PortSentry: [OK]](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-493-2048.jpg)
![PortSentry 2
CHAPTER 1
494
Removing hosts that have been blocked by PortSentry
When PortSentry want to blocks attackers it uses the “route” command of Linux to do it.
Every host that has been blocked by PortSentry will appear under the ‘route” command.
Below, I show you how to remove hosts that have been blocked by PortSentry from your
system.
Step 1
We have to use the “route” command to list which hosts are presently blocked by the program.
The “route” command also lists other important information about your network routing but we
use it in this example to get the list of blocked hosts and to unlock them from the system.
• To list which hosts are presently blocked by PortSentry, use the command:
[root@deep /]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
www.hack.com – 255.255.255.255 !H 0 – 0 -
207.35.78.0 * 255.255.255.224 U 0 0 0 eth0
127.0.0.0 * 255.0.0.0 U 0 0 0 lo
default rt.openna.c 0.0.0.0 UG 0 0 0 eth0
In the above example, we can see that “www.hack.com” is listed into our routing table as a
domain that has been blocked by PortSentry because it tried to scan our system. The “-”
string inform us about the fact that this host is locked. Every host in the routing table with this
string “-” is marked as blocked by the system.
Step2
Now that we know about the host that has been blocked by PortSentry, we can decide to
remove it from the list of blocked hosts on our system.
• To remove the blocked host in question, use the following command:
[root@deep /]# route del –host www.hack.com reject
The above command will remove www.hack.com from the list of blocked hosts in the routing table
of our system. The option “del” in the “route” command is what makes it possible to remove the
host from the list. Your have to use the above command for any additional hosts that you want to
remove from the routing table.
Step 3
Finally, we have to edit the portsentry.history file and remove the line corresponding to
www.hack.com from the file. This is important for PortSentry to be able to add the site into the
list of blocked host in the event that the corresponding host tries to scan your system again.
• Edit the portsentry.history file and remove the line corresponding to the host:
[root@deep /]# vi /var/portsentry/portsentry.history
1020371099 - 05/02/2002 16:24:59 Host: 1.2.3.4/1.2.3.4 Port: 80 TCP Blocked](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-494-2048.jpg)
![Snort 2
CHAPTER 2
498
Packages
The following is based on information listed by Snort as of 2002/07/08. Please check
http://www.snort.org/ regularly for the latest status. We chose to install from source because it
provides the facility to fine tune the installation.
Source code is available from:
Snort Homepage: http://www.snort.org/
You must be sure to download: snort-1.8.7.tar.gz
Prerequisites
Snort requires that the listed software below be already installed on your system to be able to
compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source
archive files. Please make sure you have this program installed on your machine before you
proceed with this chapter.
Libpcap, which is used extensively by Snort, must already be installed on your system.
Tcpdump, which allows some additional features with Snort.
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install Snort, and then
one afterwards, and then compare them using the diff utility to find out what files were placed
where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Snort1
• And the following one after you install the software:
[root@deep root]# find /* > Snort2
• Then use the following command to get a list of what changed:
[root@deep root]# diff Snort1 Snort2 > Snort-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-498-2048.jpg)
![Snort 2
CHAPTER 2
499
Compiling - Optimizing & Installing Snort
Below are the steps that you must make to configure, compile and optimize the Snort software
before installing it on your system. First off, we install the program as user 'root' so as to avoid
any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp snort-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf snort-version.tar.gz
Step 2
In order to check that the version of Snort, which you are going to install, is an original and
unmodified one, use the command described below to check its MD5 hashes checksum.
• To verify the MD5 checksum of Snort, use the following command:
[root@deep tmp]# md5sum snort-1.8.7.tar.gz
This should yield an output similar to this:
29c81d0bc243edb21ba4ab33ee80457e snort-1.8.7.tar.gz
Now check that this checksum is exactly the same as the one published on the Snort website at
the following URL: http://www.snort.org/dl/snort-1.8.7.tar.gz.md5
Step 3
Snort needs a UID and GID to properly run on the system but this UID/GID cannot run as
super-user root; therefore we must create a special user with no shell privileges on the system for
running Snort daemon.
• To create this special Snort user on OpenNA Linux, use the following command:
[root@deep tmp]# groupadd -g 69 snort > /dev/null 2>&1 || :
[root@deep tmp]# useradd -c "Snort NIDS" -d /var/log/snort -g 69 -s
/bin/false -u 69 snort > /dev/null 2>&1 || :
• To create this special Snort user on Red Hat Linux, use the following command:
[root@deep tmp]# groupadd -g 69 snort > /dev/null 2>&1 || :
[root@deep tmp]# useradd -u 69 -g 69 -s /bin/false -M -r -d
/var/log/snort snort > /dev/null 2>&1 || :
The above command will create a null account, with no password, no valid shell, no files owned-
nothing but a UID and a GID for the program. Remember that Snort daemon does not need to
have a shell account on the server.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-499-2048.jpg)
![Snort 2
CHAPTER 2
500
Step 4
Now, edit the shells file (vi /etc/shells) and add a non-existent shell name
“/bin/false”, which is the one we used in the passwd command above.
[root@deep tmp]# vi /etc/shells
/bin/bash2
/bin/bash
/bin/sh
/bin/false This is our added no-existent shell
Step 5
Next, move into the newly created Snort source directory and perform the following steps to
configure and optimize the software for your system.
• To move into the newly created Snort directory use the following command:
[root@deep tmp]# cd snort-1.8.7/
• To configure and optimize Snort use the following compilation lines:
CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS
./configure
--prefix=/usr
--sysconfdir=/etc
--localstatedir=/var
--mandir=/usr/share/man
--with-openssl
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the Snort software:
[root@deep snort-1.8.7]# make
[root@deep snort-1.8.7]# cd
[root@deep root]# find /* > Snort1
[root@deep root]# cd /var/tmp/snort-1.8.7/
[root@deep snort-1.8.7]# make install
[root@deep snort-1.8.7]# mkdir -p /var/log/snort
[root@deep snort-1.8.7]# mkdir -p /etc/snort
[root@deep snort-1.8.7]# chown -R snort.snort /var/log/snort/
[root@deep snort-1.8.7]# install classification.config /etc/snort/
[root@deep snort-1.8.7]# install snort.conf *.rules /etc/snort/
[root@deep snort-1.8.7]# chmod 0644 /etc/snort/*
[root@deep snort-1.8.7]# strip /usr/bin/snort
[root@deep snort-1.8.7]# cd
[root@deep root]# find /* > Snort2
[root@deep root]# diff Snort1 Snort2 > Snort-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-500-2048.jpg)
![Snort 2
CHAPTER 2
501
Step 7
Once the configuration, optimization, compilation, and installation of the Snort software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete Snort and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf snort-version/
[root@deep tmp]# rm -f snort-version.tar.gz
Configuring Snort
After Snort has been built and installed successfully in your system, your next step is to
configure and customize its configuration files to fit your needs.
/etc/snort/snort.conf: (The Snort Configuration File)
/etc/init.d/snort: (The Snort Initialization File)
/etc/snort/snort.conf: The Snort Config File
The snort.conf file is the main configuration file for Snort. It is in this file that Snort gets all
of its startup information and the way it should run on your system. You can edit it to specify your
network variables, preprocessors parameters, output plugging to use and Snort rules files.
The Snort configuration file is divided into four different sections. The first section is used to
define network variables, the second section is used to configure the preprocessor parameters
that Snort should use, the third section is used to configure output plugging to activate, and the
last section is used to enable specific Snort rule set. Below, we will explain each section and
how you should use them to configure Snort for your server.
• Edit the snort.conf file (vi /etc/snort/snort.conf) and set your needs. Below
is what we recommend you.
var HOME_NET $eth0_ADDRESS
var EXTERNAL_NET any
var SMTP $HOME_NET
var HTTP_SERVERS $HOME_NET
var SQL_SERVERS $HOME_NET
var DNS_SERVERS $HOME_NET
var RULE_PATH ./
preprocessor frag2
preprocessor stream4: detect_scans,detect_state_problems
preprocessor stream4_reassemble: both,ports all
preprocessor http_decode: 80
preprocessor rpc_decode: 111 32771
preprocessor bo
preprocessor telnet_decode
preprocessor portscan: $HOME_NET 4 3 portscan.log
preprocessor portscan-ignorehosts: 207.35.78.40 207.35.78.41
output alert_syslog: LOG_AUTH LOG_ALERT
include classification.config
This tells the snort.conf file to set itself up for this particular configuration with:](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-501-2048.jpg)
![Snort 2
CHAPTER 2
506
/etc/init.d/snort: The Snort Initialization File
The /etc/init.d/snort script file is responsible to automatically starting and stopping the
Snort server on your Linux system. Please note that the following script is suitable for Linux
operating systems that use SystemV. If you Linux system use some other methods like BSD,
you’ll have to adjust the script bellow to make it work for you.
Step 1
Create the snort script file (touch /etc/init.d/snort) and add the following lines inside it:
#!/bin/bash
# This shell script takes care of starting and stopping the snort IDS daemon.
#
# chkconfig: 2345 40 60
# description: Snort is a lightweight network intrusion detection tool that
# currently detects more than 1100 host and network
# vulnerabilities, portscans, backdoors, and more.
# Source function library.
. /etc/init.d/functions
# Source networking configuration.
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
# Specify your network interface here
INTERFACE=eth0
RETVAL=0
prog="Snort"
start() {
echo -n $"Starting $prog: "
daemon /usr/bin/snort -A fast -u snort -g snort -b -s -z -d -D
-i $INTERFACE -c /etc/snort/snort.conf
RETVAL=$?
echo
[ $RETVAL = 0 ] && touch /var/lock/subsys/snort
return $RETVAL
}
stop() {
echo -n $"Shutting down $prog: "
killproc snort
RETVAL=$?
echo
[ $RETVAL = 0 ] && rm -f /var/lock/subsys/snort
return $RETVAL
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-506-2048.jpg)
![Snort 2
CHAPTER 2
507
status)
status snort
;;
restart)
stop
start
;;
condrestart)
[ -f /var/lock/subsys/snort ] && restart
;;
*)
echo $"Usage: $prog {start|stop|status|restart|condrestart}"
exit 1
esac
exit $RETVAL
Step 2
Once the /etc/init.d/snort script file has been created, it is important to make it
executable, change its default permissions, create the necessary links and start it. Making this file
executable will allow the system to run it, changing its default permission to allow only the root
user to change this file for security reasons, and creation of the symbolic links will let the process
control initialization of Linux start the program automatically for you at each system boot.
• To make this script executable and to change its default permissions, use the command:
[root@deep /]# chmod 700 /etc/init.d/snort
[root@deep /]# chown 0.0 /etc/init.d/snort
• To create the symbolic rc.d links for Snort, use the following command:
[root@deep /]# chkconfig --add snort
[root@deep /]# chkconfig --level 2345 snort on
• To start Snort software manually, use the following command:
[root@deep /]# /etc/init.d/snort start
Starting Snort: [OK]
Running Snort in a chroot jail
This section applies only if you want to run Snort in chroot jail environment. To do it, we need to
create the required skeleton environment and copy necessary files into this chroot jail. Below are
the steps to follow if you want to run Snort with chroot jail support.
The main benefit of a chroot jail is that the jail will limit the portion of the file system the daemon
can see to the root directory of the jail. Additionally, since the jail only needs to support Snort,
the programs available in the jail can be extremely limited. Most importantly, there is no need for
setuid-root programs, which can be used to gain root access and break out of the jail.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-507-2048.jpg)
![Snort 2
CHAPTER 2
508
Necessary steps to run Snort in a chroot jail:
What you're essentially doing is creating a skeleton root file system with just enough components
necessary (files, directories) to allow UNIX to do a chroot when Snort starts. The procedures to
run Snort in chroot jail are really easy to accomplish as follows.
Step 1
First, we have to create all the necessary chrooted environment subdirectories where we will
move Snort files and directories.
• Use the following command to create all the necessary chroot subdirectories.
[root@deep /]# mkdir -p /chroot/snort/etc/snort
[root@deep /]# mkdir -p /chroot/snort/var/log/snort
[root@deep /]# chown -R snort.snort /chroot/snort/var/log/snort
Step 2
Now, it is time to move the required Snort files to the related subdirectories in the chroot area for
Snort to work. We can copy these files to the chroot jail but it’s better to move them to avoid
unnecessary duplication of Snort files on the server.
• Use the following commands to move the require files into the chroot area.
[root@deep /]# mv /etc/snort/* /chroot/snort/etc/snort/
[root@deep /]# chmod 0644 /chroot/snort/etc/snort/*
Step 3
Once the Snort files have been moved to the chroot location, we can remove the old Snort
directories from the system since they are no longer required.
• This can be done with the following commands.
[root@deep /]# rm -rf /etc/snort/
[root@deep /]# rm -rf /var/log/snort/
Step 4
Next, we have to recreate a new snort initialization script file which starts Snort in the chroot
environment. Please note that the following script is suitable for Linux operating systems that use
SystemV. If you Linux system use some other method like BSD, you’ll have to adjust the script
below to make it work for you. The only difference with the previous Snort initialization script file
is that we use the “-t” option of Snort to specify the chroot location.
Edit the snort script file (vi /etc/init.d/snort) and add the following lines inside it:
#!/bin/bash
# This shell script takes care of starting and stopping the snort IDS daemon.
#
# chkconfig: 2345 40 60
# description: Snort is a lightweight network intrusion detection tool that
# currently detects more than 1100 host and network
# vulnerabilities, portscans, backdoors, and more.
# Source function library.
. /etc/init.d/functions
# Source networking configuration.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-508-2048.jpg)
![Snort 2
CHAPTER 2
509
. /etc/sysconfig/network
# Check that networking is up.
[ ${NETWORKING} = "no" ] && exit 0
# Specify your network interface here
INTERFACE=eth0
RETVAL=0
prog="Snort"
start() {
echo -n $"Starting $prog: "
daemon /usr/bin/snort -A fast -u snort -g snort -b -s -z -d -D
-i $INTERFACE -c /etc/snort/snort.conf -t /chroot/snort/
RETVAL=$?
echo
[ $RETVAL = 0 ] && touch /var/lock/subsys/snort
return $RETVAL
}
stop() {
echo -n $"Shutting down $prog: "
killproc snort
RETVAL=$?
echo
[ $RETVAL = 0 ] && rm -f /var/lock/subsys/snort
return $RETVAL
}
# See how we were called.
case "$1" in
start)
start
;;
stop)
stop
;;
status)
status snort
;;
restart)
stop
start
;;
condrestart)
[ -f /var/lock/subsys/snort ] && restart
;;
*)
echo $"Usage: $prog {start|stop|status|restart|condrestart}"
exit 1
esac
exit $RETVAL](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-509-2048.jpg)
![Snort 2
CHAPTER 2
510
Step 5
Finally, we must test the new chrooted jail configuration of our Snort program.
• Start the new chrooted jail Snort with the following command:
[root@deep /]# /etc/init.d/snort start
Starting Snort: [OK]
• If you don't get any errors, do a ps ax | grep snort and see if we're running:
[root@deep /]# ps ax | grep snort
16295 ? R 0:38 /usr/bin/snort -A fast -u snort -g snort -b -s -z -d
If so, lets check to make sure it's chrooted by picking its process number and doing ls -la
/proc/that_process_number/root/.
[root@deep /]# ls -la /proc/16295/root/
If you see something like:
total 4
drwxr-xr-x 4 root root 4096 May 7 19:10 ./
drwxr-xr-x 5 root root 4096 May 7 19:12 ../
drwxr-xr-x 3 root root 4096 May 7 19:12 etc/
drwxr-xr-x 3 root root 4096 May 7 19:12 var/
Congratulations! Your Snort in a chroot jail is working.
Further documentation
For more details, there is one manual page about Snort that you should read:
$ man snort (8) - Open source network intrusion detection system.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-510-2048.jpg)
![Tripwire 2
CHAPTER 3
514
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
files installed onto the system if you want to update the package in the future. To solve this
problem, it’s a good idea to make a list of files on the system before you install Tripwire, and
then one afterwards, and then compare them using the diff utility to find out what files were
placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > Tripwire1
• And the following one after you install the software:
[root@deep root]# find /* > Tripwire2
• Then use the following command to get a list of what changed:
[root@deep root]# diff Tripwire1 Tripwire2 > Tripwire-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were added or changed by the program and remove them manually from your system
before installing the new software. In our example above, we use the /root directory of the
system to store all the generated file lists.
Compiling - Optimizing & Installing Tripwire
Below are the steps that you must make to configure, compile and optimize the Tripwire
software before installing it on your system. First off, we install the program as user 'root' so as
to avoid any authorization problems.
Step 1
Once you get the program from the main software site you must copy it to the /var/tmp
directory and change to this location before expanding the archive.
• This can be done with the following commands:
[root@deep /]# cp tripwire-version.tar.gz /var/tmp/
[root@deep /]# cd /var/tmp/
[root@deep tmp]# tar xzpf tripwire-version.tar.gz
Step 2
There are some source files to modify before going into the configuration and compilation of the
program; the changes allow us to fix many bugs with Tripwire. Therefore, move into the newly
created Tripwire source directory and perform the following steps to configure and optimize the
software for your system.
• To move into the newly created Tripwire directory use the following command:
[root@deep tmp]# cd tripwire-2.3.1-2/](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-514-2048.jpg)
![Tripwire 2
CHAPTER 3
515
Step 3
The first source file to modify is called “mailmessage.cpp”.
• Edit the mailmessage.cpp file (vi +244 src/tripwire/mailmessage.cpp) and
change:
const TCHAR* szFormat = _T("%a %d %b %Y %H:%M:%S %z");
To read:
const TCHAR* szFormat = _T("%a, %d %b %Y %H:%M:%S %z");
Step 4
The second file is called “platform.h” and we have to edit it and add a new line as follows.
• Edit the platform.h file (vi +294 src/core/platform.h) and change the line:
#define USES_FHS IS_LINUX
To read:
#define USE_FHS IS_LINUX
Step 5
The last file to modify is very important for Linux systems with GCC version 3; which should be the
default compiler for most Linux system now. The modifications are important to allow Tripwire
to compile with GCC v3. There is one problem, the modifications are too big to be listed in a book
and we have to retrieve it from the Internet as a patch file and patch our sources code.
The patch is available from the OpenNA website at the following URL:
http://www.openna.com/products/books/securing-optimizing-linux/3rdedition/index.htm
Please, download the patch and patch your Tripwire source codes as follow:
• To patch your Tripwire source codes, use the command:
[root@deep /]# cp tripwire-gcc3.patch /var/tmp/
[root@deep /]# cd /var/tmp/tripwire-2.3.1-2/
[root@deep tripwire-2.3.1-2]# patch -p1 < ../tripwire-gcc3.patch](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-515-2048.jpg)
![Tripwire 2
CHAPTER 3
516
Step 6
Now, we must make a list of all files on the system before installing the software, and one
afterwards, then compare them using the diff utility to find out what files are placed where and
finally we install the Tripwire software:
[root@deep tripwire-2.3.1-2]# cd src/
[root@deep src]# rm -rf STLport*
[root@deep src]# touch STLport_r STLport_d
[root@deep src]# export CXXFLAGS="-O2 -march=i686 -funroll-loops"
[root@deep src]# make release
[root@deep src]# cd
[root@deep root]# find /* > Tripwire1
[root@deep root]# cd /var/tmp/tripwire-2.3.1-2/bin/i686-pc-linux_r/
[root@deep i686-pc-linux_r]# install -m0500 siggen /usr/sbin/
[root@deep i686-pc-linux_r]# install -m0500 tripwire /usr/sbin/
[root@deep i686-pc-linux_r]# install -m0500 twadmin /usr/sbin/
[root@deep i686-pc-linux_r]# install -m0500 twprint /usr/sbin/
[root@deep i686-pc-linux_r]# cd ../../man/
[root@deep man]# install -m0440 man4/*.4 /usr/share/man/man4/
[root@deep man]# install -m0440 man5/*.5 /usr/share/man/man5/
[root@deep man]# install -m0440 man8/*.8 /usr/share/man/man8/
[root@deep man]# mkdir -m0700 /etc/tripwire
[root@deep man]# mkdir -p /var/lib/tripwire/report
[root@deep man]# chmod -R 0700 /var/lib/tripwire/
[root@deep man]# cd
[root@deep root]# find /* > Tripwire2
[root@deep root]# diff Tripwire1 Tripwire2 > Tripwire-Installed
The above commands will configure the software to ensure your system has the necessary
libraries to successfully compile the package, compile all source files into executable binaries,
and then install the binaries and any supporting files into the appropriate locations.
Step 7
Once the configuration, optimization, compilation, and installation of the Tripwire software have
been accomplished, we can free up some disk space by deleting the program tar archive and the
related source directory since they are no longer needed.
• To delete Tripwire and its related source directory, use the following commands:
[root@deep /]# cd /var/tmp/
[root@deep tmp]# rm -rf tripwire-version/
[root@deep tmp]# rm -f tripwire-version.tar.gz
The rm command as used above will remove all the source files we have used to compile and
install Tripwire. It will also remove the Tripwire compressed archive from the /var/tmp
directory.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-516-2048.jpg)
![Tripwire 2
CHAPTER 3
517
Configuring Tripwire
After Tripwire has been built and installed successfully in your system, your next step is to
configure and customize its configuration files and policy file to fit your needs.
/etc/tripwire/twcfg.txt: (The Tripwire Configuration File)
/etc/tripwire/twpol.txt: (The Tripwire Policy File)
/etc/tripwire/twinstall.sh: (The Tripwire Cryptographic File)
/etc/cron.weekly/tripwire.cron: (The Tripwire Cron File)
/etc/tripwire/twcfg.txt: The Tripwire Configuration File
The twcfg.txt file is a small Tripwire configuration file used by the program the first time you
run it to get information about locations of different files and the way Tripwire runs and reports
on the integrity of the system. It stores system-specific information, such as the location of
Tripwire data files. In general, we use this file ONE time and REMOVE it from the server once
Tripwire is configured.
Step 1
By default, the twcfg.txt file do not exist after installation, we have to create it as follow.
• Create the twcfg.txt file (touch /etc/tripwire/twcfg.txt) and add the
following lines inside the file. Below is what we recommend you.
ROOT =/usr/sbin
POLFILE =/etc/tripwire/tw.pol
DBFILE =/var/lib/tripwire/$(HOSTNAME).twd
REPORTFILE =/var/lib/tripwire/report/$(HOSTNAME)-$(DATE).twr
SITEKEYFILE =/etc/tripwire/site.key
LOCALKEYFILE =/etc/tripwire/$(HOSTNAME)-local.key
EDITOR =/bin/vi
LATEPROMPTING =true
LOOSEDIRECTORYCHECKING =true
MAILNOVIOLATIONS =false
EMAILREPORTLEVEL =3
REPORTLEVEL =3
MAILMETHOD =SENDMAIL
SYSLOGREPORTING =true
MAILPROGRAM =/usr/sbin/sendmail -oi -t
Step2
Now, set the permissions of the twcfg.txt file to be (0640/-rw-r-----) and owned by the
super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the twcfg.txt file, use:
[root@deep /]# chmod 640 /etc/tripwire/twcfg.txt
[root@deep /]# chown 0.0 /etc/tripwire/twcfg.txt](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-517-2048.jpg)
![Tripwire 2
CHAPTER 3
522
(
rulename = "/chroot directory",
severity = $(SIG_HI)
)
{
/chroot -> $(SEC_CRIT) (recurse = -1) ;
}
Step2
Now, set the permissions of the twpol.txt file to be (0640/-rw-r-----) and owned by the
super-user ‘root’ for security reasons.
• To change the permission mode and ownership of the twpol.txt file, use:
[root@deep /]# chmod 640 /etc/tripwire/twpol.txt
[root@deep /]# chown 0.0 /etc/tripwire/twpol.txt
NOTE: Please, add to the above policy file all files, binaries, directories that you want the software
to monitor for you. Remove any files, binaries, directories that you don’t want the software to
monitor for you and don’t send emails to the mailing list if you receive error messages about the
fact that some files, binaries, directories don’t exist on your system. Instead, review your policy
file and make the changes related to the error that you received, because in most cases, this is
why you have this kind of errors messages. Finally, reads the twpolicy manual page for more
information on the parameters of this policy file.
/etc/tripwire/twinstall.sh: The Tripwire Cryptographic File
The twinstall.sh file is a script file used by Tripwire to configure, install and generate
cryptographic keys used by Tripwire during operation and verification.
Step 1
This script file asks you the passpharse that you want to run with Tripwire as well as other
operations related to the Tripwire database generation and policies. By default, the
twinstall.sh file does not exist after the installation; we have to create it as follows.
• Create the twinstall.sh file (touch /etc/tripwire/twinstall.sh) and add the
following lines inside the file.
#!/bin/sh
HOST_NAME='localhost'
if uname -n > /dev/null 2> /dev/null ; then
HOST_NAME=`uname -n`
fi
# Site Passphrase variable
TW_SITE_PASS=""
# Complete path to site key
SITE_KEY="/etc/tripwire/site.key"
# Local Passphrase variable
TW_LOCAL_PASS=""
# Complete path to local key](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-522-2048.jpg)
![Tripwire 2
CHAPTER 3
523
LOCAL_KEY="/etc/tripwire/${HOST_NAME}-local.key"
# If clobber==true, overwrite files; if false, do not overwrite files.
CLOBBER="false"
# If prompt==true, ask for confirmation before continuing with install.
PROMPT="true"
# Name of twadmin executeable
TWADMIN="twadmin"
# Path to twadmin executeable
TWADMPATH=@sbindir@
# Path to configuration directory
CONF_PATH="/etc/tripwire"
# Name of clear text policy file
TXT_POL=$CONF_PATH/twpol.txt
# Name of clear text configuration file
TXT_CFG=$CONF_PATH/twcfg.txt
# Name of encrypted configuration file
CONFIG_FILE=$CONF_PATH/tw.cfg
# Path of the final Tripwire policy file (signed)
SIGNED_POL=`grep POLFILE $TXT_CFG | sed -e 's/^.*=(.*)/1/'`
if [ -z "$TW_SITE_PASS" ] || [ -z "$TW_LOCAL_PASS" ]; then
cat << END_OF_TEXT
----------------------------------------------
The Tripwire site and local passphrases are used to
sign a variety of files, such as the configuration,
policy, and database files.
Passphrases should be at least 8 characters in length
and contain both letters and numbers.
See the Tripwire manual for more information.
END_OF_TEXT
fi
echo
echo "----------------------------------------------"
echo "Creating key files..."
if [ "$CLOBBER" = "true" ] && [ "$PROMPT" = "false" ] && [ -f "$SITE_KEY" ] ;
then
rm -f "$SITE_KEY"
fi
if [ -f "$SITE_KEY" ] && [ "$CLOBBER" = "false" ] ; then
echo "The site key file "$SITE_KEY""
echo 'exists and will not be overwritten.'
else
cmdargs="--generate-keys --site-keyfile "$SITE_KEY""
if [ -n "$TW_SITE_PASS" ] ; then
cmdargs="$cmdargs --site-passphrase "$TW_SITE_PASS""
fi
eval ""$TWADMPATH/$TWADMIN" $cmdargs"
if [ $? -ne 0 ] ; then](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-523-2048.jpg)
![Tripwire 2
CHAPTER 3
524
echo "Error: site key generation failed"
exit 1
else chmod 640 "$SITE_KEY"
fi
fi
if [ "$CLOBBER" = "true" ] && [ "$PROMPT" = "false" ] && [ -f "$LOCAL_KEY" ] ;
then
rm -f "$LOCAL_KEY"
fi
if [ -f "$LOCAL_KEY" ] && [ "$CLOBBER" = "false" ] ; then
echo "The site key file "$LOCAL_KEY""
echo 'exists and will not be overwritten.'
else
cmdargs="--generate-keys --local-keyfile "$LOCAL_KEY""
if [ -n "$TW_LOCAL_PASS" ] ; then
cmdargs="$cmdargs --local-passphrase "$TW_LOCAL_PASS""
fi
eval ""$TWADMPATH/$TWADMIN" $cmdargs"
if [ $? -ne 0 ] ; then
echo "Error: local key generation failed"
exit 1
else chmod 640 "$LOCAL_KEY"
fi
fi
echo
echo "----------------------------------------------"
echo "Signing configuration file..."
if [ "$CLOBBER" = "false" ] && [ -s "$CONFIG_FILE" ] ; then
backup="${CONFIG_FILE}.$$.bak"
echo "Backing up $CONFIG_FILE"
echo " to $backup"
`mv "$CONFIG_FILE" "$backup"`
if [ $? -ne 0 ] ; then
echo "Error: backup of configuration file failed."
exit 1
fi
fi
cmdargs="--create-cfgfile"
cmdargs="$cmdargs --cfgfile "$CONFIG_FILE""
cmdargs="$cmdargs --site-keyfile "$SITE_KEY""
if [ -n "$TW_SITE_PASS" ] ; then
cmdargs="$cmdargs --site-passphrase "$TW_SITE_PASS""
fi
eval ""$TWADMPATH/$TWADMIN" $cmdargs "$TXT_CFG""
if [ $? -ne 0 ] ; then
echo "Error: signing of configuration file failed."
exit 1
fi
# Set the rights properly
chmod 640 "$CONFIG_FILE"
cat << END_OF_TEXT
A clear-text version of the Tripwire configuration file
$TXT_CFG
has been preserved for your inspection. It is recommended](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-524-2048.jpg)
![Tripwire 2
CHAPTER 3
525
that you delete this file manually after you have examined it.
END_OF_TEXT
echo
echo "----------------------------------------------"
echo "Signing policy file..."
if [ "$CLOBBER" = "false" ] && [ -s "$POLICY_FILE" ] ; then
backup="${POLICY_FILE}.$$.bak"
echo "Backing up $POLICY_FILE"
echo " to $backup"
mv "$POLICY_FILE" "$backup"
if [ $? -ne 0 ] ; then
echo "Error: backup of policy file failed."
exit 1
fi
fi
cmdargs="--create-polfile"
cmdargs="$cmdargs --cfgfile "$CONFIG_FILE""
cmdargs="$cmdargs --site-keyfile "$SITE_KEY""
if [ -n "$TW_SITE_PASS" ] ; then
cmdargs="$cmdargs --site-passphrase "$TW_SITE_PASS""
fi
eval ""$TWADMPATH/$TWADMIN" $cmdargs "$TXT_POL""
if [ $? -ne 0 ] ; then
echo "Error: signing of policy file failed."
exit 1
fi
# Set the proper rights on the newly signed policy file.
chmod 0640 "$SIGNED_POL"
cat << END_OF_TEXT
A clear-text version of the Tripwire policy file
$TXT_POL
has been preserved for your inspection. This implements
a minimal policy, intended only to test essential
Tripwire functionality. You should edit the policy file
to describe your system, and then use twadmin to generate
a new signed copy of the Tripwire policy.
END_OF_TEXT
Step 2
Now, set the permissions of the twinstall.sh file to be (0500/---x------) and owned by
the super-user ‘root’ for security reasons.
• This procedure can be accomplished with the following command:
[root@deep /]# chmod 500 /etc/tripwire/twinstall.sh
[root@deep /]# chown 0.0 /etc/tripwire/twinstall.sh
NOTE: The above script file can also be retrieved from the following URL:
http://www.openna.com/products/books/securing-optimizing-linux/3rdedition/index.htm](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-525-2048.jpg)
![Tripwire 2
CHAPTER 3
526
/etc/cron.weekly/tripwire.cron: The Tripwire Cron File
The tripwire.cron file is a small script executed automatically by the crond program each
week to scan your hard disk for possible changes to files or directories and mail the results to the
system administrator.
Step 1
This script will automate the procedure of integrity checking for you. If you want to automate this
task, follow the simple steps below.
• Create the tripwire.cron script file (touch /etc/cron.weekly/tripwire.cron)
and add the following lines:
#!/bin/sh
HOST_NAME=`uname -n`
if [ ! -e /var/lib/tripwire/${HOST_NAME}.twd ] ; then
echo "**** Error: Tripwire database for ${HOST_NAME} not found. ****"
echo "**** Run "/etc/tripwire/twinstall.sh" and/or "tripwire --init". ****"
else
test -f /etc/tripwire/tw.cfg && /usr/sbin/tripwire --check
fi
Step 2
Now, set the permissions of the tripwire.cron file to be (0500/---x------) and owned by
the super-user ‘root’ for security reasons.
• This procedure can be accomplished with the following command:
[root@deep /]# chmod 500 /etc/cron.weekly/tripwire.cron
[root@deep /]# chown 0.0 /etc/cron.weekly/tripwire.cron
Running Tripwire for the first time
Once all the files are installed and configured on your server, it’s time to run Tripwire to
generate the cryptography key, passphrases and database files. These procedures should be
made the first time you install the software and ONLY the first time.
Step 1
Here we begin by running the twinstall.sh script file which will generate the cryptography
keys and will ask us to enter our passphrase (password) which is required each time we want to
update and accept Tripwire integrity reports.
• To run the twinstall.sh file use the following command:
[root@deep /]# /etc/tripwire/twinstall.sh
----------------------------------------------
The Tripwire site and local passphrases are used to
sign a variety of files, such as the configuration,
policy, and database files.
Passphrases should be at least 8 characters in length
and contain both letters and numbers.
See the Tripwire manual for more information.
----------------------------------------------](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-526-2048.jpg)
![Tripwire 2
CHAPTER 3
527
Creating key files...
(When selecting a passphrase, keep in mind that good passphrases
typically have upper and lower case letters, digits and punctuation
marks, and are at least 8 characters in length.)
Enter the site keyfile passphrase: Your site keyfile passphrase
Verify the site keyfile passphrase: Your site keyfile passphrse again
Generating key (this may take several minutes)...Key generation complete.
(When selecting a passphrase, keep in mind that good passphrases
typically have upper and lower case letters, digits and punctuation
marks, and are at least 8 characters in length.)
Enter the local keyfile passphrase: Your local keyfile passphrase
Verify the local keyfile passphrase: Your local keyfile passphrase again
Generating key (this may take several minutes)...Key generation complete.
----------------------------------------------
Signing configuration file...
Please enter your site passphrase: Your site passphrase
Wrote configuration file: /etc/tripwire/tw.cfg
A clear-text version of the Tripwire configuration file
/etc/tripwire/twcfg.txt
has been preserved for your inspection. It is recommended
that you delete this file manually after you have examined it.
----------------------------------------------
Signing policy file...
Please enter your site passphrase: Your site passphrase
Wrote policy file: /etc/tripwire/tw.pol
A clear-text version of the Tripwire policy file
/etc/tripwire/twpol.txt
has been preserved for your inspection. This implements
a minimal policy, intended only to test essential
Tripwire functionality. You should edit the policy file
to describe your system, and then use twadmin to generate
a new signed copy of the Tripwire policy
Step 2
Once our passphrase keyfiles have been generated, it’s time to run Tripwire in its’ initialization
mode. The initialization mode will create the initial Tripwire database files based on what
information has been provided inside the twpol.txt file. Tripwire must have a database to
compare against, so we first create the file information database. This action will create a file
called “tw.db_[hostname]” in the directory you specified to hold your databases (where
[hostname] will be replaced with your machine hostname).
• To run the Tripwire in initialization mode, use the following command:
[root@deep /]# tripwire --init
Parsing policy file: /etc/tripwire/tw.pol
Generating the database...
*** Processing Unix File System ***
Please enter your local passphrase:
Wrote database file: /var/lib/tripwire/deep.twd
The database was successfully generated.](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-527-2048.jpg)
![Tripwire 2
CHAPTER 3
528
NOTE: Initialization of the database Tripwire uses should be done manually because the key
used to sign the database should be different for each system.
Step 3
Finally, if you have not received any kind of error message, then you can safety remove the
twcfg.txt and twpol.txt files from your system since they are no longer needed and it
would be a security risk to keep these files on your server.
• To remove the files from your system, use the following commands:
[root@deep /]# rm -f /etc/tripwire/twcfg.txt
[root@deep /]# rm -f /etc/tripwire/twpol.txt
NOTE: You have to remove the files from your server ONLY if you are sure that the initialization of
the databases has been completed without any errors. Otherwise you should keep these files and
regenerate a new database once all the errors have been fixed inside the twpol.txt file, since
in many cases errors come from twpol.txt file having some lines referring to files or directories
that do not exist in your system.
Securing Tripwire
It is highly recommended that the database (tw.db_[hostname]) file of Tripwire be moved
someplace (e.g. floppy) where it cannot be modified. This is important because data from
Tripwire is only as trustworthy as its database.
It is also recommended that you make a hardcopy printout of the database contents right away. In
the event that you become suspicious of the integrity of the database, you will be able to
manually compare information against this hardcopy.
Tripwire Administrative Tools
The commands listed below are some of the most used of this software, but many more exist.
Check the Tripwire manual pages for more details.
Running Tripwire in Interactive Checking Mode:
In “Interactive Checking Mode” feature, Tripwire verifies files or directories that have been
added, deleted, or changed from the original database and ask the user whether the database
entry should be updated. This mode is the most convenient way of keeping your database up-to-
date, but it requires that the user be "at the console". If you want to use this mode, then follow the
simple step below.
Once the file information database of Tripwire has been created, we can now run Tripwire in
“Interactive Checking Mode”. This mode will prompt the user for whether or not each changed
entry on the system should be updated to reflect the current state of the file.
• To run Tripwire in Interactive Checking Mode, use the following command:
[root@deep /]# tripwire --check --interactive
Parsing policy file: /etc/tripwire/tw.pol
*** Processing Unix File System ***
Performing integrity check...](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-528-2048.jpg)
![ucspi-tcp 2
CHAPTER 4
536
Pristine source
If you don’t use the RPM package to install this program, it will be difficult for you to locate all the
installed files in the system in the event of an update in the future. To solve the problem, it is a
good idea to make a list of files on the system before you install ucsip-tcp, and one afterwards,
and then compares them using the diff utility to find out what files were placed where.
• Simply run the following command before installing the software:
[root@deep root]# find /* > ucsip-tcp1
• And the following one after you install the software:
[root@deep root]# find /* > ucsip-tcp2
• Then use the following command to get a list of what changed:
[root@deep root]# diff ucsip-tcp1 ucsip-tcp2 > ucsip-tcp-Installed
With this procedure, if any upgrade appears, all you have to do is to read the generated list of
what files were](https://image.slidesharecdn.com/securing-optimizinglinux-150718001925-lva1-app6891/75/Securing-optimizing-linux-the-hacking-solution-536-2048.jpg)
Uploaded byfernando valenzuela hernández
Securing optimizing linux. the hacking solution
This document provides instructions for securing and optimizing a Linux server. It is divided into multiple parts that cover topics such as installation security, system security and optimization, network security, cryptography and authentication, monitoring and integrity, servers, management, DNS, mail, databases and more. The book includes over 1200 pages of content to harden a Linux system against security threats and optimize performance. It discusses securing the kernel, firewall configuration using iptables, installing tools like sudo, ssh and squid, and provides detailed configuration examples.
More Related Content
Similar to Securing optimizing linux. the hacking solution
Securing optimizing linux. the hacking solution
- 1. 1 This book isdedicated to OpenNA staff. Thanks, guys (no-gender)!! --Gerhard Mourani This book is printed on acid-free paper with 85% recycled content, 15% post-consumer waste. Open Network Architecture is commited to using paper with the highest recycled content available consistent with high quality. Copyright © 2002 by Gerhard Mourani and Open Network Architecture, Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as permitted by Canada Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the copyright holders Gerhard Mourani and Open Network Architecture, Inc. 11090 Drouart, Montreal, PQ H3M 2S3, (514) 978-6183, fax (514) 333-0236. Requests to the Publisher for permission should be addressed to the Publishing Manager, at Open Network Architecture, Inc., E-mail: books@openna.com This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold with the understanding that some grammatical mistakes could have occurred but this won’t jeopardize the content or the issue raised herewith. Title: Securing and Optimizing Linux: The Hacking Solution Page Count: 1208 Version: 3.0 Last Revised: 2002-06-26 Publisher: Open Network Architecture, Inc. Editor: Ted Nackad Text Design & Drawings (Graphics): Bruno Mourani Printing History: June 2000: First Publication. Author's: Gerhard Mourani Mail: gmourani@openna.com Website: http://www.openna.com/ National Library Act. R.S., c. N-11, s. 1. Legal Deposit, 2002 Securing and Optimizing Linux: The Hacking Solution / Open Network Architecture, Inc. Published by Open Network Architecture, Inc., 11090 Drouart, Montreal, H3M 2S3, Canada. Includes Index. ISBN 0-9688793-1-4 Printed in Canada
- 2. 2 Overview Part I InstallationSecurity Chapter 1 Introduction Chapter 2 Installation Issues Part II System Security & Optimization Chapter 3 General Security Chapter 4 Pluggable Authentication Modules Chapter 5 General Optimization Chapter 6 Kernel Security & Optimization Chapter 7 Process File System Management Part III Network Security Chapter 8 TCP/IP Network Management Chapter 9 Firewall Basic Concept Chapter 10 GIPTables Firewall Chapter 11 Squid Proxy Server Chapter 12 SquidGuard Filter Chapter 13 FreeS/WAN VPN Part IV Cryptography & Authentication Chapter 14 GnuPG Chapter 15 OpenSSL Chapter 16 OpenSSH Chapter 17 Sudo Part V Monitoring & System Integrity Chapter 18 sXid Chapter 19 LogSentry Chapter 20 HostSentry Chapter 21 PortSentry Chapter 22 Snort Chapter 23 Tripwire Part VI Super-Server Chapter 24 UCSPI-TCP Chapter 25 Xinetd Part VII Management & Limitation Chapter 26 NTP Chapter 27 Quota Part VIII Domain Name System & Dynamic Host Protocol Chapter 28 ISC BIND & DNS Chapter 29 ISC DHCP Part IX Mail Transfer Agent Protocol Chapter 30 Exim Chapter 31 Qmail
- 3. 3 Part X InternetMessage Access Protocol Chapter 32 tpop3d Chapter 33 UW IMAP Chapter 34 Qpopper Part XI Anti-Spam & Anti-Virus Chapter 35 SpamAssassin Chapter 36 Sophos Chapter 37 AMaViS Part XII Database Server Chapter 38 MySQL Chapter 39 PostgreSQL Chapter 40 OpenLDAP Part XIII File Transfer Protocol Chapter 41 ProFTPD Chapter 42 vsFTPD Part XIV Hypertext Transfer Protocol Chapter 43 Apache Chapter 44 PHP Chapter 45 Mod_Perl Part XV NetBios Protocol Chapter 46 Samba Part XVI Backup Chapter 47 Tar & Dump Part XVII Appendixes Appendix A Tweaks, Tips and Administration Tasks Appendix B Port list
- 4. 4 Contents Steps of installation13 Author note 13 Audience 14 These installation instructions assume 15 Obtaining the example configuration files 15 Problem with Securing & Optimizing Linux 15 Acknowledgments 15 Introduction 19 What is Linux? 21 Some good reasons to use Linux 21 Let's dispel some of the fear, uncertainty, and doubt about Linux 21 Why choose pristine source? 22 Compiling software on your system 22 Build & install software on your system 23 Editing files with the vi editor tool 24 Recommended software to include in each type of servers 25 Installation Issues 29 Know your Hardware! 31 Creating the Linux Boot Disk 31 Beginning the installation of Linux 33 Installation Class and Method (Install Options) 34 Partition your system for Linux 35 Disk Partition (Manual Partitioning) 39 Selecting Package Groups 50 Boot Disk Creation 53 How to use RPM Commands 53 Starting and stopping daemon services 56 Software that must be uninstalled after installation of the server 57 Remove unnecessary documentation files 65 Remove unnecessary/empty files and directories 66 Software that must be installed after installation of the server 66 General Security 73 BIOS 75 Unplug your server from the network 75 Security as a policy 76 Choose a right password 76 The root account 77 Set login time out for the root account 77 Shell logging 78 The single-user login mode of Linux 79 Disabling Ctrl-Alt-Delete keyboard shutdown command 79 Limiting the default number of started ttys on the server 80 The LILO and /etc/lilo.conf file 80 The GRUB and /boot/grub/grub.conf file 82 The /etc/services file 84
- 5. 5 The /etc/securetty file85 Special accounts 85 Control mounting a file system 88 Mounting the /usr directory of Linux as read-only 89 Tighten scripts under /etc/init.d 91 Tighten scripts under /etc/cron.daily/ 91 Bits from root-owned programs 91 Don’t let internal machines tell the server what their MAC address is 93 Unusual or hidden files 94 Finding Group and World Writable files and directories 95 Unowned files 96 Finding .rhosts files 96 Physical hard copies of all-important logs 97 Getting some more security by removing manual pages 99 System is compromised! 100 Pluggable Authentication Modules 101 The password length 103 Disabling console program access 105 Disabling all console access 105 The Login access control table 106 Tighten console permissions for privileged users 107 Putting limits on resource 109 Controlling access time to services 111 Blocking; su to root, by one and sundry 112 Using sudo instead of su for logging as super-user 113 General Optimization 116 Static vs. shared libraries 118 The Glibc 2.2 library of Linux 119 Why Linux programs are distributed as source 120 Some misunderstanding in the compiler flags options 121 The gcc specs file 122 Striping all binaries and libraries files 127 Tuning IDE Hard Disk Performance 128 Kernel Security & Optimization 133 Difference between a Modularized Kernel and a Monolithic Kernel 135 Making an emergency boot floppy 138 Preparing the Kernel for the installation 139 Applying the Grsecurity kernel patch 141 Obtaining and Installing Grsecurity 141 Tuning the Kernel 142 Cleaning up the Kernel 143 Configuring the Kernel 145 Compiling the Kernel 190 Installing the Kernel 190 Verifying or upgrading your boot loader 192 Reconfiguring /etc/modules.conf file 194 Rebooting your system to load the new kernel 195 Delete programs, edit files pertaining to modules 195
- 6. 6 Making a newrescue floppy for Modularized Kernel 196 Making a emergency boot floppy disk for Monolithic Kernel 196 Process file system management 199 What is sysctl? 202 /proc/sys/vm: The virtual memory subsystem of Linux 202 /proc/sys/fs: The file system data of Linux 209 /proc/sys/net/ipv4: IPV4 settings of Linux 211 Other possible optimization of the system 219 TCP/IP Network Management 225 TCP/IP security problem overview 228 Installing more than one Ethernet Card per Machine 232 Files-Networking Functionality 233 Testing TCP/IP Networking 237 The last checkup 240 Firewall Basic Concept 241 What is the IANA? 243 The ports numbers 243 What is a Firewall? 245 Packet Filter vs. Application Gateway 245 What is a Network Firewall Security Policy? 247 The Demilitarized Zone 248 Linux IPTables Firewall Packet Filter 249 The Netfilter Architecture 249 GIPTables Firewall 255 Building a kernel with IPTables support 259 Compiling - Optimizing & Installing GIPTables 262 Configuring GIPTables 263 /etc/giptables.conf: The GIPTables Configuration File 263 /etc/rc.d/rc.giptables.blocked: The GIPTables Blocked File 274 /etc/init.d/giptables: The GIPTables Initialization File 275 The GIPTables Firewall Module Files 276 How GIPTables parameters work? 277 Running the type of GIPTables firewall that you need 283 The GIPTables configuration file for a Gateway/Proxy Server 284 GIPTables-Firewall Administrative Tools 302 Squid Proxy Server 305 Compiling - Optimizing & Installing Squid 309 Configuring Squid 313 Running Squid with Users Authentication Support 326 Securing Squid 330 Optimizing Squid 333 Squid Administrative Tools 333 The cachemgr.cgi program utility of Squid 335
- 7. 7 SquidGuard Filter337 Compiling -Optimizing & Installing SquidGuard 340 Configuring SquidGuard 342 Testing SquidGuard 350 Optimizing SquidGuard 351 FreeS/WAN VPN 355 Compiling - Optimizing & Installing FreeS/WAN 360 Configuring FreeS/WAN 363 Configuring RSA private keys secrets 367 Requiring network setup for IPSec 372 Testing the FreeS/WAN installation 374 GnuPG 379 Compiling - Optimizing & Installing GnuPG 382 Using GnuPG under Linux terminal 384 OpenSSL 391 Compiling - Optimizing & Installing OpenSSL 396 Configuring OpenSSL 398 OpenSSL Administrative Tools 404 Securing OpenSSL 409 OpenSSH 411 Compiling - Optimizing & Installing OpenSSH 414 Configuring OpenSSH 417 Running OpenSSH in a chroot jail 427 Creating OpenSSH private & public keys 432 OpenSSH Users Tools 434 Sudo 437 Compiling - Optimizing & Installing Sudo 440 Configuring Sudo 442 A more complex sudoers configuration file 444 Securing Sudo 447 Sudo Users Tools 447 sXid 451 Compiling - Optimizing & Installing sXid 454 Configuring sXid 455 sXid Administrative Tools 457 LogSentry 459
- 8. 8 Compiling - Optimizing& Installing LogSentry 462 Configuring LogSentry 466 HostSentry 467 Compiling - Optimizing & Installing HostSentry 470 Configuring HostSentry 474 PortSentry 481 Compiling - Optimizing & Installing PortSentry 484 Configuring PortSentry 487 Removing hosts that have been blocked by PortSentry 494 Snort 495 Compiling - Optimizing & Installing Snort 499 Configuring Snort 501 Running Snort in a chroot jail 507 Tripwire 511 Compiling - Optimizing & Installing Tripwire 514 Configuring Tripwire 517 Running Tripwire for the first time 526 Securing Tripwire 528 Tripwire Administrative Tools 528 ucspi-tcp 533 Compiling - Optimizing & Installing ucsip-tcp 536 Using ucsip-tcp 538 Xinetd 541 Compiling - Optimizing & Installing Xinetd 544 Configuring Xinetd 546 The /etc/xinetd.d directory 547 NTP 559 Compiling - Optimizing & Installing NTP 564 Configuring NTP 566 Running NTP in Client Mode 566 Running NTP in Server Mode 572 Running NTP in a chroot jail 574 NTP Administrative Tools 578 Quota 581 Build a kernel with Quota support enable 584 Compiling - Optimizing & Installing Quota 584
- 9. 9 Modifying the /etc/fstabfile 586 Creating the aquota.user and aquota.group files 587 Assigning Quota for Users and Groups 587 Quota Administrative Tools 590 ISC BIND & DNS 593 Compiling - Optimizing & Installing ISC BIND & DNS 598 Configuring ISC BIND & DNS 600 Running ISC BIND & DNS as Caching-Only Name Server 601 Running ISC BIND & DNS as Primary Master Name Server 610 Running ISC BIND & DNS as Secondary Slave Name Server 615 Running ISC BIND & DNS in a chroot jail 617 Securing ISC BIND & DNS 621 Optimizing ISC BIND & DNS 638 ISC BIND & DNS Administrative Tools 641 ISC BIND & DNS Users Tools 643 ISC DHCP 645 Building a kernel with ISC DHCP support 649 Compiling - Optimizing & Installing ISC DHCP 650 Configuring ISC DHCP 654 Testing the DHCP server 662 Running ISC DHCP in a chroot jail 664 Securing ISC DHCP 675 Running the DHCP client for Linux 676 Exim 683 Compiling - Optimizing & Installing Exim 688 Configuring Exim 693 Testing Exim 716 Allowing Users to authenticate with Exim before relaying 719 Running Exim with SSL support 722 Running Exim with Virtual Hosts support 729 Running Exim with Maildir support 732 Running Exim with mail quota support 734 Running Exim as a Null Client Mail Server 735 Exim Administrative Tools 738 Qmail 741 Compiling, Optimizing & Installing Qmail 745 Configuring Qmail 751 Testing Qmail 755 Allowing Users to authenticate with Qmail before relaying 756 Running Qmail with SSL support 760 Running Qmail with Virtual Hosts support 765 Running Qmail as a Null Client Mail Server 769 Running Qmail as a Mini-Qmail Mail Server 773 Running qmail-pop3d with SSL support 777
- 10. 10 Qmail Administrative Tools780 Qmail Users Tools 781 tpop3d 785 Compiling - Optimizing & Installing tpop3d 790 Configuring tpop3d 791 Securing tpop3d 795 UW IMAP 797 Compiling - Optimizing & Installing UW IMAP 801 Configuring UW IMAP 805 Enable IMAP or POP services via UCSPI-TCP 807 Enable IMAP or POP services via Xinetd 808 Securing UW IMAP 810 Running UW IMAP with SSL support 811 Qpopper 815 Compiling - Optimizing & Installing Qpopper 819 Configuring Qpopper 821 Securing Qpopper 825 Running Qpopper with SSL support 827 SpamAssassin 835 Compiling - Optimizing & Installing SpamAssassin 839 Configuring SpamAssassin 840 Testing SpamAssassin 842 Running SpamAssassin with Exim 843 Running SpamAssassin with Qmail 844 Sophos 849 Compiling & Installing Sophos 853 Configuring Sophos 854 Testing Sophos 855 AMaViS 857 Verifying & installing all the additional prerequisites to run AMaViS 860 Compiling - Optimizing & Installing AMaViS 872 Running AMaViS with Exim 875 Running AMaViS with Qmail 877 Testing AMaViS 878 MySQL 881 Compiling - Optimizing & Installing MySQL 886 Configuring MySQL 888 Securing MySQL 893
- 11. 11 Optimizing MySQL 894 MySQLAdministrative Tools 899 PostgreSQL 907 Compiling - Optimizing & Installing PostgreSQL 910 Configuring PostgreSQL 913 Running PostgreSQL with SSL support 918 Securing PostgreSQL 924 Optimizing PostgreSQL 928 PostgreSQL Administrative Tools 929 OpenLDAP 935 Compiling - Optimizing & Installing OpenLDAP 940 Configuring OpenLDAP 945 Running OpenLDAP with TLS/SSL support 950 Running OpenLDAP in a chroot jail 954 Securing OpenLDAP 961 Optimizing OpenLDAP 962 OpenLDAP Administrative Tools 963 OpenLDAP Users Tools 967 ProFTPD 971 Compiling - Optimizing & Installing ProFTPD 976 Configuring ProFTPD 980 Creating an account for FTP client to connect to the FTP server 992 Setup an anonymous FTP server 993 Allow anonymous users to upload to the FTP server 997 Running ProFTPD with SSL support 1000 Securing ProFTPD 1005 ProFTPD Administrative Tools 1006 vsFTPd 1009 Compiling - Optimizing & Installing vsFTPd 1014 Configuring vsFTPd 1015 Creating an account for FTP client to connect to the FTP server 1021 Setup an anonymous FTP server 1022 Allow anonymous users to upload to the FTP server 1024 Apache 1029 Compiling - Optimizing & Installing Apache 1034 Configuring Apache 1040 Running Apache with TLS/SSL support 1051 Running Apache in a chroot jail 1055 Running Apache with users authentication support 1063 Caching frequently requested static files 1065 Some statistics about Apache and Linux 1066
- 12. 12 PHP 1069 Compiling -Optimizing & Installing PHP 1073 Configuring PHP 1076 Running PHP in a chroot jail 1084 Running PHP with the PHP Accelerator program 1085 Mod_Perl 1089 Compiling - Optimizing & Installing Mod_Perl 1093 Configuring Mod_Perl 1094 Running Mod_Perl in a chroot jail 1095 Samba 1099 Compiling - Optimizing & Installing Samba 1104 Configuring Samba 1106 Running Samba with TLS/SSL support 1116 Securing Samba 1121 Optimizing Samba 1123 Samba Administrative Tools 1125 Samba Users Tools 1126 Tar & Dump 1129 The tar backup program 1131 Making backups with tar 1132 Automating tasks of backups made with tar 1134 Restoring files with tar 1136 The dump backup program 1138 Making backups with dump 1139 Restoring files with dump 1141 Backing up and restoring over the network 1143 APPENDIX A 1151 APPENDIX B 1157
- 13. Preface 13 Steps of installation Dependingof your level of knowledge in Linux, you can read this book from the beginning through to the end of the chapters that interest you. Each chapter and section of this book appears in a manner that lets you read only the parts of your interest without the need to schedule one day of reading. Too many books on the market take myriad pages to explain something that can be explained in two lines, I’m sure that a lot of you agree with my opinion. This book tries to be different by talking about only the essential and important information that the readers want to know by eliminating all the nonsense. Although you can read this book in the order you want, there is a particular order that you could follow if something seems to be confusing you. The steps shown below are what I recommend: Setup Linux in your computer. Remove all the unnecessary RPM’s packages. Install the necessary RPM’s packages for compilation of software (if needed). Secure the system in general. Optimize the system in general. Reinstall, recompile and customize the Kernel to fit your specific system. Configure firewall script according to which services will be installed in your system. Install OpenSSL to be able to use encryption with the Linux server. Install OpenSSH to be able to make secure remote administration tasks. Install Sudo. Install sXid. Install LogSentry. Install PortSentry. Install Tripwire. Install ICS BIND/DNS. Install Exim or Qmail. Install any software you need after to enable specific services into the server. Author note According to some surveys on the Internet, Linux will be the number one operating system for a server platform in year 2003. Presently it is number two and no one at one time thought that it would be in this second place. Many organizations, companies, universities, governments, and the military, etc, kept quiet about it. Crackers use it as the operating system by excellence to crack computers around the world. Why do so many people use it instead of other well know operating systems? The answer is simple, Linux is free and the most powerful, reliable, and secure operating system in the world, providing it is well configured. Millions of programmers, home users, hackers, developers, etc work to develop on a voluntary basis, different programs related to security, services, and share their work with other people to improve it without expecting anything in return. This is the revolution of the Open Source movement that we see and hear about so often on the Internet and in the media.
- 14. 14 If crackers canuse Linux to penetrate servers, security specialists can use the same means to protect servers (to win a war, you should at least have equivalent weapons to what your enemy may be using). When security holes are encountered, Linux is the one operating system that has a solution and that is not by chance. Now someone may say: with all these beautiful features why is Linux not as popular as other well know operating system? There are many reasons and different answers on the Internet. I would just say that like everything else in life, anything that we are to expect the most of, is more difficult to get than the average and easier to acquire. Linux and *NIX are more difficult to learn than any other operating system. It is only for those who want to know computers in depth and know what they doing. People prefer to use other OS’s, which are easy to operate but hard to understand what is happening in the background since they only have to click on a button without really knowing what their actions imply. Every UNIX operating system like Linux will lead you unconsciously to know exactly what you are doing because if you pursue without understanding what is happening by the decision you made, then nothing will surely work as expected. This is why with Linux; you will know the real meaning of a computer and especially a server environment where every decision warrants an action which will closely impact on the security of your organization and employees. Many Web sites are open to all sorts of "web hacking." According to the Computer Security Institute and the FBI's joint survey, 90% of 643 computer security practitioners from government agencies, private corporations, and universities detected cyber attacks last year. Over $265,589,940 in financial losses was reported by 273 organizations. Many readers of the previous version of this book told me that the book was an easy step by step guide for newbie’s, I am flattered but I prefer to admit that it was targeting for a technical audience and I assumed the reader had some background in Linux, UNIX systems. If this is not true in your case, I highly recommend you to read some good books in network administration related to UNIX and especially to Linux before venturing into this book. Remember talking about security and optimization is a very serious endeavor. It is very important to be attentive and understand every detail in this book and if difficulties arise, try to go back and reread the explanation will save a lot of frustration. Once again, security is not a game and crackers await only one single error from your part to enter your system. A castle has many doors and if just one stays open, will be enough to let intruders into your fortress. You have been warned. Many efforts went into the making of this book, making sure that the results were as accurate as possible. If you find any abnormalities, inconsistent results, errors, omissions or anything else that doesn't look right, please let me know so I can investigate the problem and/or correct the error. Suggestions for future versions are also welcome and appreciated. A web site dedicated to this book is available on the Internet for your convenience. If you any have problem, question, recommendation, etc, please go to the following URL: http://www.openna.com/. We made this site for you. Audience This book is intended for a technical audience and system administrators who manage Linux servers, but it also includes material for home users and others. It discusses how to install and setup a Linux server with all the necessary security and optimization for a high performance Linux specific machine. It can also be applied with some minor changes to other Linux variants without difficulty. Since we speak of optimization and security configuration, we will use a source distribution (tar.gz) program for critical server software like Apache, ISC BIND/DNS, Samba, Squid, OpenSSL etc. Source packages give us fast upgrades; security updates when necessary, and better compilation, customization, and optimization options for specific machines that often aren’t available with RPM packages.
- 15. Preface 15 These installation instructionsassume You have a CD-ROM drive on your computer and the Official Red Hat Linux or OpenNA Linux CD-ROM. Installations were tested on the Official Red Hat Linux version 7.3 and OpenNA Linux. You should familiarize yourself with the hardware on which the operating system will be installed. After examining the hardware, the rest of this document guides you, step-by-step, through the installation process. Obtaining the example configuration files In a true server environment and especially when Graphical User Interface is not installed, we will often use text files, scripts, shell, etc. Throughout this book we will see shell commands, script files, configuration files and many other actions to execute on the terminal of the server. You can enter them manually or use the compressed archive file that I made which contains all configuration examples and paste them directly to your terminal. This seems to be useful in many cases to save time. The example configuration files in this book are available electronically via HTTP from this URL: http://www.openna.com/products/books/securing-optimizing-linux/3rdedition/index.htm • In either case, extract the files into your Linux server from the archive by typing: [root@deep /]# cd /var/tmp [root@deep tmp]# tar xzpf floppy-3.0.tgz If you cannot get the examples from the Internet, please contact the author at this email address: gmourani@openna.com Problem with Securing & Optimizing Linux When you encounter a problem in "Securing & Optimizing Linux" we want to hear about it. Your reports are an important part in making the book more reliable, because even with the utmost care we cannot guarantee that every part of the book will work on every platform under every circumstance. We cannot promise to fix every error right away. If the problem is obvious, critical, or affects a lot of users, chances are that someone will look into it. It could also happen that we tell you to update to a newer version to see if the problem persists there. Or we might decide that the problem cannot be fixed until some major rewriting has been done. If you need help immediately, consider obtaining a commercial support contract or try our Q&A archive from the mailing list for an answer. Below are some important links: OpenNA web site: http://www.openna.com/ Mailing list: http://www.openna.com/support/mailing/ Support: http://www.openna.com/support/ RPM Download: http://www.openna.com/download/ Acknowledgments I would like to thank all the OpenNA staff for their hard works and patience. A special gratitude and many thanks to Colin Henry who made tremendous efforts to make this book grammatically and orthographically sound in a professional manner. Adrian Pascalau for its time and help in the open source community and all Linux users around the world who have participated by providing good comments, ideas, recommendations and suggestions.
- 19. 19 Introduction IN THIS CHAPTER 1.What is Linux? 2. Some good reasons to use Linux 3. Let's dispel some of the fear, uncertainty, and doubt about Linux 4. Why choose Pristine source? 5. Compiling software on your system 6. Build, Install software on your system 7. Editing files with the vi editor tool 8. Recommended software to include in each type of servers
- 21. Introduction 0 CHAPTER 1 21 Introduction Whatis Linux? Linux is an operating system that was first created at the University of Helsinki in Finland by a young student named Linus Torvalds. At this time the student was working on a UNIX system that was running on an expensive platform. Because of his low budget, and his need to work at home, he decided to create a copy of the UNIX system in order to run it on a less expensive platform, such as an IBM PC. He began his work in 1991 when he released version 0.02 and worked steadily until 1994 when version 1.0 of the Linux Kernel was released. The Linux operating system is developed under the GNU General Public License (also known as GNU GPL) and its source code is freely available to everyone who downloads it via the Internet. The CD-ROM version of Linux is also available in many stores, and companies that provide it will charge you for the cost of the media and support. Linux may be used for a wide variety of purposes including networking, software development, and as an end-user platform. Linux is often considered an excellent, low-cost alternative to other more expensive operating systems because you can install it on multiple computers without paying more. Some good reasons to use Linux There are no royalty or licensing fees for using Linux and the source code can be modified to fit your needs. The results can be sold for profit, but the original authors retain copyright and you must provide the source to your modifications. Because it comes with source code to the kernel, it is quite portable. Linux runs on more CPUs and platforms than any other computer operating system. The recent direction of the software and hardware industry is to push consumers to purchase faster computers with more system memory and hard drive storage. Linux systems are not affected by those industries’ orientation because of its capacity to run on any kind of computer, even aging x486-based computers with limited amounts of RAM. Linux is a true multi-tasking operating system similar to its brother, UNIX. It uses sophisticated, state-of-the-art memory management techniques to control all system processes. That means that if a program crashes you can kill it and continue working with confidence. Another benefit is that Linux is practically immunized against all kinds of viruses that we find in other operating systems. To date we have found only two viruses that were effective on Linux systems - well, actually they are Trojan Horses. Let's dispel some of the fear, uncertainty, and doubt about Linux It's a toy operating system Fortune 500 companies, governments, and consumers more and more use Linux as a cost- effective computing solution. It has been used, and is still used, by big companies like IBM, Amtrak, NASA, and others. There's no support Every Linux distribution comes with more than 12,000 pages of documentation. Commercial Linux distributions offer initial support for registered users, and small business and corporate accounts can get 24/7 supports through a number of commercial support companies. As an Open Source operating system, there's no six-month wait for a service release, plus the online Linux community fixes many serious bugs within hours.
- 22. 22 Why choose pristinesource? All the programs in Red Hat and OpenNA distributions of Linux are provided as RPM files. An RPM file, also known, as a “package”, is a way of distributing software so that it can be easily installed, upgraded, queried, and deleted. However, in the Unix world, the defacto-standard for package distribution continues to be by way of so-called “tarballs”. Tarballs are simply compressed files that can be readable and uncompressed with the “tar” utility. Installing from tar is usually significantly more tedious than using RPM. So why would we choose to do so? 1) Unfortunately, it takes a few weeks for developers and helpers to get the latest version of a package converted to RPM’s because many developers first release them as tarballs. 2) When developers and vendors release a new RPM, they include a lot of options that often aren’t necessary. Those organizations and companies don’t know what options you will need and what you will not, so they include the most used to fit the needs of everyone. 3) Often RPMs are not optimized for your specific processors; companies like Red Hat Linux build RPM’s based on a standard PC. This permits their RPM packages to be installed on all sorts of computers since compiling a program for an i386 machine means it will work on all systems. 4) Sometimes you download and install RPM’s, which other people around the world are building and make available for you to use. This can pose conflicts in certain cases depending how this individual built the package, such as errors, security and all the other problems described above. Compiling software on your system A program is something a computer can execute. Originally, somebody wrote the "source code" in a programming language he/she could understand (e.g., C, C++). The program "source code" also makes sense to a compiler that converts the instructions into a binary file suited to whatever processor is wanted (e.g. a 386 or similar). A modern file format for these "executable" programs is ELF. The programmer compiles his source code on the compiler and gets a result of some sort. It's not at all uncommon that early attempts fail to compile, or having compiled, fail to act as expected. Half of programming is tracking down and fixing these problems (debugging). For the beginners there are more aspect and new words relating to the compilation of source code that you must know, these include but are not limited to: Multiple Files (Linking) One-file programs are quite rare. Usually there are a number of files (say *.c, *.cpp, etc) that are each compiled into object files (*.o) and then linked into an executable. The compiler is usually used to perform the linking and calls the 'ld' program behind the scenes. Makefiles Makefiles are intended to aid you in building your program the same way each time. They also often help with increasing the speed of a program. The “make” program uses “dependencies” in the Makefile to decide what parts of the program need to be recompiled. If you change one source file out of fifty you hope to get away with one compile and one link step, instead of starting from scratch.
- 23. Introduction 0 CHAPTER 1 23 Libraries Programscan be linked not only to object files (*.o) but also to libraries that are collections of object files. There are two forms of linking to libraries: static, where the code goes in the executable file, and dynamic, where the code is collected when the program starts to run. Patches It was common for executable files to be given corrections without recompiling them. Now this practice has died out; in modern days, people change a small portion of the source code, putting a change into a file called a “patch”. Where different versions of a program are required, small changes to code can be released this way, saving the trouble of having two large distributions. Errors in Compilation and Linking Errors in compilation and linking are often due to typos, omissions, or misuse of the language. You have to check that the right “includes file” is used for the functions you are calling. Unreferenced symbols are the sign of an incomplete link step. Also check if the necessary development libraries (GLIBC) or tools (GCC, DEV86, MAKE, etc) are installed on your system. Debugging Debugging is a large topic. It usually helps to have statements in the code that inform you of what is happening. To avoid drowning in output you might sometimes get them to print out only the first 3 passes in a loop. Checking that variables have passed correctly between modules often helps. Get familiar with your debugging tools. Build & install software on your system You will see in this book that we use many different compile commands to build and install programs on the server. These commands are UNIX compatible and are used on all variants of *NIX machines to compile and install software. The procedures to compile and install software tarballs on your server are as follows: 1. First of all, you must download the tarball from your trusted software archive site. Usually from the main site of the software you hope to install. 2. After downloading the tarball, change to the /var/tmp directory (note that other paths are possible, at personal discretion) and untar the archive by typing the commands (as root) as in the following example: [root@deep /]# tar xzpf foo.tar.gz The above command will extract all files from the example foo.tar.gz compressed archive and will create a new directory with the name of the software from the path where you executed the command. The “x” option tells tar to extract all files from the archive. The “z” option tells tar that the archive is compressed with gzip utility. The “p” option maintains the original permissions the files had when the archive was created. The “f” option tells tar that the very next argument is the file name.
- 24. 24 Once the tarballhas been decompressed into the appropriate directory, you will almost certainly find a “README” and/or an “INSTALL” file included with the newly decompressed files, with further instructions on how to prepare the software package for use. Likely, you will need to enter commands similar to the following example: ./configure make make install The above commands, ./configure will configure the software to ensure your system has the necessary libraries to successfully compile the package, make will compile all the source files into executable binaries. Finally, make install will install the binaries and any supporting files into the appropriate locations. Other specific commands that you’ll see in this book for compilation and installation procedure will be: make depend strip chown The make depend command will build and make the necessary dependencies for different files. The strip command will discard all symbols from the object files. This means that our binary file will be smaller in size. This will improve the performance of the program, since there will be fewer lines to read by the system when it executes the binary. The chown command will set the correct file owner and group permissions for the binaries. More commands will be explained in the sections concerning program installation. Editing files with the vi editor tool The vi program is a text editor that you can use to edit any text and particularly programs. During installation of software, the user will often have to edit text files, like Makefiles or configuration files. The following are some of the more important keystroke commands to get around in vi. I decided to introduce the vi commands now since it is necessary to use vi throughout this book. Command Result ===================================================================== i --------------------------------- Notifies vi to insert text before the cursor a --------------------------------- Notifies vi to append text after the cursor dd -------------------------------- Notifies vi to delete the current line x --------------------------------- Notifies vi to delete the current character Esc ------------------------------- Notifies vi to end the insert or append mode u --------------------------------- Notifies vi to undo the last command Ctrl+f ---------------------------- Scroll up one page Ctrl+b ---------------------------- Scroll down one page /string --------------------------- Search forward for string :f -------------------------------- Display filename and current line number :q -------------------------------- Quit editor :q! ------------------------------- Quit editor without saving changes :wq ------------------------------- Save changes and exit editor =====================================================================
- 25. Introduction 0 CHAPTER 1 25 Recommendedsoftware to include in each type of servers If you buy binaries, you will not get any equity and ownership of source code. Source code is a very valuable asset and binaries have no value. Buying software may become a thing of the past. You only need to buy good hardware; it is worth spending money on the hardware and gets the software from the Internet. The important point is that it is the computer hardware that is doing the bulk of the work. The hardware is the real workhorse and the software is just driving it. It is for this reason that we believe in working with and using Open source software. Much of the software and services that come with Linux are open source and allow the user to use and modify them in an undiscriminating way according to the General Public License. Linux has quickly become the most practical and friendly used platform for e-business -- and with good reason. Linux offers users stability, functionality and value that rivals any platform in the industry. Millions of users worldwide have chosen Linux for running their applications, from web and email servers to departmental and enterprise vertical application servers. To respond to your needs and to let you know how you can share services between systems I have developed ten different types of servers, which cover the majority of servers' functions and enterprise demands. Often companies try to centralize many services into one server to save money, it is well known and often seen that there are conflicts between the technical departments and purchasing agents of companies about investment and expenditure when it comes to buying new equipment. When we consider security and optimization, it is of the utmost importance not to run too many services on one server, it is highly recommended to distribute tasks and services between multiple systems. The table below shows you which software and services we recommend to for each type of Linux server. The following conventions will explain the interpretations of these tables: Optional Components: components that may be included to improve the features of the server or to fit special requirements. Security Software Required: what we consider as minimum-security software to have installed on the server to improve security. Security Software Recommended: what we recommend for the optimal security of the servers.
- 26. 26 Mail Server WebServer Gateway Server Exim or Qmail (SMTP Server) BIND/DNS (Caching) IPTables Firewall GIPTables ---------- IMAP/POP only for Exim Apache Qmail BIND/DNS (Caching) IPTables Firewall GIPTables BIND/DNS (Caching) Qmail IPTables Firewall GIPTables ---------- Squid SuidGuard Optional Components Optional Components Optional Components Mod_PHP Mod_SSL Mod-Perl DHCP Security Software Required Security Software Required Security Software Required Grsecurity OpenSSL OpenSSH Tripwire Sudo Grsecurity OpenSSL OpenSSH Tripwire Sudo Grsecurity OpenSSL OpenSSH Tripwire Sudo Security Software recommended Security Software recommended Security Software recommended GnuPG sXid Logcheck HostSentry PortSentry GnuPG sXid Logcheck HostSentry PortSentry GnuPG sXid Logcheck HostSentry PortSentry FTP Server Domain Name Server File Sharing Server ProFTPD Qmail BIND/DNS (Caching) IPTables Firewall GIPTables Primary BIND/DNS (Server) Qmail IPTables Firewall GIPTables ---------- Secondary BIND/DNS (Server) Samba Qmail BIND/DNS (Caching) IPTables Firewall GIPTables Optional Components Optional Components Optional Components Anonymous FTP (Server) Security Software Required Security Software Required Security Software Required Grsecurity OpenSSL OpenSSH Tripwire Sudo Grsecurity OpenSSL OpenSSH Tripwire Sudo Grsecurity OpenSSL OpenSSH Tripwire Sudo Security Software recommended Security Software recommended Security Software recommended GnuPG sXid Logcheck HostSentry PortSentry GnuPG sXid Logcheck HostSentry PortSentry GnuPG sXid Logcheck HostSentry PortSentry
- 27. Introduction 0 CHAPTER 1 27 Databaseserver Backup server VPN Server PostgreSQL (Client & Server) Qmail BIND/DNS (Caching) IPTables Firewall GIPTables ---------- MySQL (Client & Server) ---------- OpenLDAP (Client & Servers) Amanda Qmail BIND/DNS (Caching) Dump Utility IPTables Firewall GIPTables FreeS/WAN VPN (Server) Qmail BIND/DNS (Caching) IPTables Firewall GIPTables Optional Components Optional Components Optional Components Security Software Required Security Software Required Security Software Required Grsecurity OpenSSL OpenSSH Tripwire Sudo Grsecurity OpenSSL OpenSSH Tripwire Sudo Grsecurity OpenSSL OpenSSH Tripwire Sudo Security Software recommended Security Software recommended Security Software recommended GnuPG sXid Logcheck HostSentry PortSentry GnuPG sXid Logcheck HostSentry PortSentry GnuPG sXid Logcheck HostSentry PortSentry
- 29. 29 Installation Issues IN THISCHAPTER 1. Know your Hardware! 2. Creating the Linux Boot Disk 3. Beginning the installation of Linux 4. Installation Class and Method (Install Options) 5. Partition your system for Linux 6. Disk Partition (Manual Partitioning) 7. Selecting Package Groups 8. Boot Disk Creation 9. How to use RPM Commands 10. Starting and stopping daemon services 11. Software that must be uninstalled after installation of the server 12. Remove unnecessary documentation files 13. Remove unnecessary/empty files and directories 14. Software that must be installed after installation of the server
- 31. Installation Issues 0 CHAPTER2 31 Linux Installation Abstract This part of the book deals with the basic knowledge required to properly install a Linux OS, in our case this is going to be Red Hat Linux, on your system in the most secure and clean manner available. We have structured this chapter in a manner that follows the original installation of the Red Hat Linux operating system from CD-ROM. Each section below refers to, and will guide you through, the different screens that appear during the setup of your system after booting from the Red Hat boot diskette. We promise that it will be interesting to have the machine you want to install Linux on ready and near you when you follow the steps described below. You will see that through the beginning of the installation of Linux, there are many options, parameters, and hacks that you can set before the system boots up for the first time. Know your Hardware! Understanding the hardware of your computer is essential for a successful installation of Linux. Therefore, you should take a moment and familiarize yourself with your computer hardware. Be prepared to answer the following questions: 1. How many hard drives do you have? 2. What size is each hard drive (eg, 15GB)? 3. If you have more than one hard drive, which is the primary one? 4. What kind of hard drive do you have (eg, IDE ATA/66, SCSI)? 5. How much RAM do you have (eg, 256MB RAM)? 6. Do you have a SCSI adapter? If so, who made it and what model is it? 7. Do you have a RAID system? If so, who made it and what model is it? 8. What type of mouse do you have (eg, PS/2, Microsoft, Logitech)? 9. How many buttons does your mouse have (2/3)? 10. If you have a serial mouse, what COM port is it connected to (eg, COM1)? 11. What is the make and model of your video card? How much video RAM do you have (eg, 8MB)? 12. What kind of monitor do you have (make and model)? 13. Will you be connected to a network? If so, what will be the following: a. Your IP address? b. Your netmask? c. Your gateway address? d. Your domain name server’s IP address? e. Your domain name? f. Your hostname? g. Your types of network(s) card(s) (makes and model)? h. Your number of card(s) (makes and model)? Creating the Linux Boot Disk The first thing to do is to create an installation diskette, also known as a boot disk. If you have purchased the official Red Hat Linux CD-ROM, you will find a floppy disk called “Boot Diskette” in the Red Hat Linux box so you don’t need to create it. Sometimes, you may find that the installation will fail using the standard diskette image that comes with the official Red Hat Linux CD-ROM. If this happens, a revised diskette is required in order for the installation to work properly. In these cases, special images are available via the Red Hat Linux Errata web page to solve the problem (http://www.redhat.com/errata).
- 32. 32 Since this, isa relatively rare occurrence, you will save time if you try to use the standard diskette images first, and then review the Errata only if you experience any problems completing the installation. Below, we will show you two methods to create the installation Boot Disk, the first method is to use an existing Microsoft Windows computer and the second using an existing Linux computer. Making a Diskette under MS-DOS: Before you make the boot disk, insert the Official Red Hat Linux CD-ROM Disk 1 in your computer that runs the Windows operating system. When the program asks for the filename, enter boot.img for the boot disk. To make the floppies under MS-DOS, you need to use these commands (assuming your CD-ROM is drive D: and contain the Official Red Hat Linux CD-ROM). • Open the Command Prompt under Windows: Start | Programs | Command Prompt C:> d: D:> cd dosutils D:dosutils> rawrite Enter disk image source file name: ..imagesboot.img Enter target diskette drive: a: Please insert a formatted diskette into drive A: and press -ENTER- : D:dosutils>exit The rawrite.exe program asks for the filename of the disk image: Enter boot.img and insert a blank floppy into drive A. It will then ask for a disk to write to: Enter a:, and when complete, label the disk “Red Hat boot disk”, for example. Making a Diskette under a Linux-Like OS: To make a diskette under Linux or any other variant of Linux-Like operating system, you must have permission to write to the device representing the floppy drive (known as /dev/fd0H1440 under Linux). This permission is granted when you log in to the system as the super-user “root”. Once you have logged as “root”, insert a blank formatted diskette into the diskette drive of your computer without issuing a mount command on it. Now it’s time to mount the Red Hat Linux CD-ROM on Linux and change to the directory containing the desired image file to create the boot disk. • Insert a blank formatted diskette into the diskette drive Insert the Red Hat Linux CD Part 1 into the CD-ROM drive [root@deep /]# mount /dev/cdrom /mnt/cdrom [root@deep /]# cd /mnt/cdrom/images/ [root@deep images]# dd if=boot.img of=/dev/fd0H1440 bs=1440k 1+0 records in 1+0 records out [root@deep images]# cd / [root@deep /]# umount /mnt/cdrom Don’t forget to label the diskette “Red Hat boot disk”, for example.
- 33. Installation Issues 0 CHAPTER2 33 Beginning the installation of Linux Now that we have made the boot disk, it is time to begin the installation of Linux. Since we’d start the installation directly off the CD-ROM, boot with the boot disk. Insert the boot diskette you create into the drive A: on the computer where you want to install Linux and reboot the computer. At the boot: prompt, press Enter to continue booting and follow the three simple steps below. Step 1 The first step is to choose what language should be used during the installation process. In our example we choose the English language. Once you select the appropriate language, click Next to continue. Step 2 Next, the system allows you to choose your keyboard type, layout type for the keyboard, and the possibility to enable or disable Dead Keys. Once you have made the appropriate selections, click Next to continue.
- 34. 34 Step 3 Finally, wechoose the kind of mouse type we have and if this mouse has two or three buttons. If you have a mouse with just two buttons, you can select the option named “Emulate 3 Buttons” and click both mouse buttons at the same time to act as the middle mouse button. Once we have completed the above three steps, we are ready to begin the installation of Red Hat Linux. Installation Class and Method (Install Options) Red Hat Linux 7.3 includes four different classes, or type of installation. They are: Workstation Server Laptop Custom The first two classes (Workstation and Server) give you the option of simplifying the installation process with a significant loss of configuration flexibility that we don’t want to lose. For this reason we highly recommend you select the “Custom” installation. Only the custom-class installation gives us complete flexibility. During the custom-class installation, it is up to you how disk space should be partitioned. We also have complete control over the different RPM packages that will be installed on the system. The idea is to load the minimum amount of packages, while maintaining maximum efficiency. The less software that resides on the machine, the fewer potential security exploits or holes may appear. From the menu that appears on your screen, select the “Custom” installation class and click Next.
- 35. Installation Issues 0 CHAPTER2 35 Partition your system for Linux Partitioning allows you to divide your hard drive into isolated sections, where each section behaves as its own hard drive. This is a useful security measure and to avoid some possible DoS attacks because we can create separate partition for specific services that we would like to run on our Linux server. See later in this book for more information about which partition strategy to use with security. The system will show you a new screen from where you can choose the tool you would like to use to partition the disks for Linux. From here we have two choices, but before we explain them, it is important to understand partition strategies first.
- 36. 36 We assume thatyou are installing the new Linux server to a new hard drive, with no other existing file system or operating system installed. A good partition strategy is to create a separate partition for each major file system. This enhances security and prevents accidental Denial of Service (DoS) or exploit of SUID programs. Creating multiple partitions offers you the following advantages: Protection against Denial of Service attack. Protection against SUID programs. Faster booting. Easy backup and upgrade management. Ability for better control of mounted file system. Limit each file system’s ability to grow. Improve performance of some program with special setup. WARNING: If a previous file system or operating system exists on the hard drive and computer where you want to install your Linux system, we highly recommend, that you make a backup of your current system before proceeding with the disk partitioning. Partitions Strategy For performance, stability and security reasons you must create something like the following partitions listed below on your computer. We suppose for this partition configuration the fact that you have a SCSI hard drive of 9.1 GB with 256 MB of physical RAM. Of course you will need to adjust the partition sizes and swap space according to your own needs and disk size. Minimal recommended partitions that must be created on your system: This is the minimum number of partitions we recommend creating whatever you want to setup it for, a Web Server, Mail Server, Gateway or something else. /boot 5 MB All Kernel images are kept here. / 256 MB Our root partition. /usr 512 MB Must be large, since many Linux binaries programs are installed here. /home 5700 MB Proportional to the number of users you intend to host. (i.e. 100 MB per users * by the number of users 57 = 5700 MB) /var 256 MB Contains files that change when the system run normally (i.e. Log files). /tmp 329 MB Our temporary files partition (must always reside on its own partition). <Swap> 512 MB Our swap partition. The virtual memory of the Linux operating system. Additional or optional partitions that can be created on your system: Depending on what services the Linux system will be assigned to serve or the specific software requirements, there can be some special partitions you can add to the minimum partitions we recommend. You can create as many partitions as you want to fit you needs. What we show you below are partitions related to programs we describe in the book. /chroot 256 MB If you want to install programs in chroot jail environment (i.e. DNS, Apache). /var/lib 1000 MB Partition to handle SQL or Proxy Database Server files (i.e. MySQL, Squid).
- 37. Installation Issues 0 CHAPTER2 37 All major file systems are on separate partitions As you can see, there are two partitions, which are less common than the others. Let’s explain each of them in more detail: The /chroot partition can be used for DNS Server chrooted, Apache web server chrooted and other chrooted future programs. The chroot() command is a Unix system call that is often used to provide an additional layer of security when untrusted programs are run. The kernel on Unix variants which support chroot() maintains a note of the root directory each process on the system has. Generally this is /, but the chroot() system call can change this. When chroot() is successfully called, the calling process has its idea of the root directory changed to the directory given as the argument to chroot(). The /var/lib partition can be used to handle SQL or Squid Proxy database files on the Linux server. This partition can be useful to limit accidental Denial of Service attack and to improve the performance of the program by tuning the /var/lib file system. Putting /tmp and /home on separate partitions is pretty much mandatory if users have shell access to the server (protection against SUID programs), splitting these off into separate partitions also prevents users from filling up critical file systems (denial of service attack), putting /var, and /usr on separate partitions is also a very good idea. By isolating the /var partition, you protect your root partition from overfilling (Denial of Service attack). In our partition configuration we’ll reserve 256 MB of disk space for chrooted programs like Apache, DNS and other software. This is necessary because Apache DocumentRoot files and other binaries, programs related to it will be installed in this partition if you decide to run Apache web server in a chrooted jail. Note that the size of the Apache chrooted directory on the chrooted partition is proportional to the size of your DocumentRoot files or number of users. NOTE: It is for you to decide how much disk space should be reserved and set for each partition you may need to create on your server. The choice completely depends on you and your computer hardware. If you have a lot of disk space and know that you will need to run many services in chroot jail environment, then you can decide to reserve more space for the chroot jail structure on your system.
- 38. 38 Swap related issues: Swaprelates to virtual RAM on the system. This special device is needed when you run out of physical RAM because you don’t have enough MB of RAM available or your applications required more than what is available on your computer. It is not true that swap space is needed on every system, but to ensure that you do not run out of swap, it is recommended to create a swap partition on the server. The 2.4 kernel of Linux is more aggressive than the 2.2 kernels in its use of swap space and the optimal sizing of swap space remains dependent on the following: 1. The amount of RAM installed. 2. The amount of disk space available for swap. 3. The applications being run. 4. The mix of applications that are run concurrently. No rule-of-thumb can possibly take all these points into account. However, we recommend the following swap sizes: • Single-user systems with less than 128MB physical RAM: 256MB • Single-user systems and low-end servers with more than 128MB physical RAM: two times physical RAM (2xRAM) • Dedicated servers with more than 512MB physical RAM: highly dependent on environment and must be determined on a case-by-case basis) NOTE: Swap is bad and it is recommended that you try to avoid it as much as possible by installing more physical RAM whenever possible. If you see that your system begin to swap memory, then consider buying some more RAM. Remember that swap is bad and your rules are to avoid it as much as possible for optimum performance of your Linux server. Minimum size of partitions for very old hard disk: For information purposes only, this is the minimum size in megabytes, which a Linux installation must have to function properly. The sizes of partitions listed below are really small. This configuration can fit into a very old hard disk of 512MB in size that you might find in old i486 computers. We show you this partition just to get an idea of the minimum requirements. / 35MB /boot 5MB /chroot 10MB /home 100MB /tmp 30MB /usr 232MB /var 25MB WARNING: Trying to compile programs on a 512 MB hard drive, will fail due to the lack of available space. Instead, install RPM’s packages.
- 39. Installation Issues 0 CHAPTER2 39 Disk Partition (Manual Partitioning) Now that we know exactly what partitions we need to create for our new Linux server, it is time to choose the partitioning software we will use to make these partitions. With Red Hat Linux two programs exist to assist you with this step: • Manually partition with Disk druid • Manually partition with fdisk [experts only] Disk Druid is new software used by default in Red Hat Linux to partition your disk drive, this program is easy to use, and allows you to use a graphical interface to create your partitions tables. fdisk was the first partitioning program available on Linux. It is more powerful then Disk Druid and allows you to create your partition table in exactly the way you want it (if you want to put your swap partition near the beginning of your drive, then you will need to use fdisk). Unfortunately, it is also a little more complicated than Disk Druid and many Linux users prefer to use Disk Druid for this reason. Personally, I prefer to create the partitions with the fdisk program and I recommend you use and be familiar with it, because if, in the future you want to add or change some file systems you will need to use fdisk. Partitioning with Disk Druid This section applies only if you chose to use Disk Druid to partition your system. Disk Druid is a program that partitions your hard drive for you. Choose “New” to add a new partition, “Edit” to edit a partition, “Delete” to delete a partition and “Reset” to reset the partitions to the original state. When you add a new partition, a new window appears on your screen and gives you parameters to choose. Mount Point: for where you want to mount your new partition in the filesystem. Filesystem Type: Ext3 for Linux filesystem and Swap for Linux Swap Partition Size (MB): for the size of your new partition in megabytes.
- 40. 40 If you havea SCSI disk, the device name will be /dev/sda and if you have an IDE disk it will be /dev/hda. If you’re looking for high performance and stability, a SCSI disk is highly recommended. Linux refers to disk partitions using a combination of letters and numbers. It uses a naming scheme that is more flexible and conveys more information than the approach used by other operating systems. Here is a summary: First Two Letters – The first two letters of the partition name indicate the type of device on which the partition resides. You’ll normally see either hd (for IDE disks), or sd (for SCSI disks). The Next Letter – This letter indicates which device the partition is on. For example: /dev/hda (the first IDE hard disk) and /dev/hdb (the second IDE disk), etc. Keep this information in mind, it will make things easier to understand when you’re setting up the partitions Linux requires. Now, as an example: To make the partitions listed below on your system (this is the partition we’ll need for our server installation example); the commands below are for Disk Druid: Step 1 Execute all of the following commands with Disk Druid to create the require partitions. New Mount Point: /boot Filesystem Type: ext3 Size (Megs): 24 Ok New Mount Point: / Filesystem Type: ext3 Size (Megs): 256 Ok New Mount Point: /usr Filesystem Type: ext3 Size (Megs): 512 Ok New Mount Point: /home Filesystem Type: ext3 Size (Megs): 4512 Ok New Mount Point: /chroot Filesystem Type: ext3 Size (Megs): 256 Ok
- 41. Installation Issues 0 CHAPTER2 41 New Mount Point: /var Filesystem Type: ext3 Size (Megs): 512 Ok New Mount Point: /var/lib Filesystem Type: ext3 Size (Megs): 1024 Ok New Mount Point: /tmp Filesystem Type: ext3 Size (Megs): 256 Ok New Mount Point: swap Filesystem Type: swap Size (Megs): 1372 Ok Step 2 After you have executed the above commands to create and partition your drive with Disk Druid, press the Next button and continue the installation to choose partitions to format. Partitioning with fdisk This section applies only if you chose to use fdisk to partition your system. The first thing you will want to do is using the p key to check the current partition information. You need to first add your root partition. Use the n key to create a new partition and then select either e or p keys for extended or primary partition. Most likely you will want to create a primary partition. You are asked what partition number should be assigned to it, at which cylinder the partition should start (you will be given a range – just choose the lowest number (1)), and the size of the partition. For example, for a 5MB partition, you would enter +5M for the size when asked. Next, you need to add your extended partition. Use the n key to create a new partition and then select the e key for extended partition. You are asked what partition number should be assigned to it, at which cylinder the partition should start (you will be given a range – just choose the lowest number (2)), and the size of the partition. You would enter the last number for the size when asked (or just press Enter). You will now want to create the swap partition. You need to use the n key for a new partition. Choose logical; tell it where the first cylinder should be (2). Tell fdisk how big you want your swap partition. You then need to change the partition type to Linux swap. Enter the t key to change the type and enter the partition number of your swap partition. Enter the number 82 for the hex code for the Linux swap partition.
- 42. 42 Now that youhave created your Linux boot and Linux swap partition, it is time to add any additional partitions you might need. Use the n key again to create a new partition, and enter all the information just as before. Keep repeating this procedure until all your partitions are created. You can create up to four primary partitions; then you must start putting extended partitions into each primary partition. NOTE: None of the changes you make take effect until you save then and exit fdisk using the w command. You may quit fdisk at any time without saving changes by using the q command. An overview of fdisk The command for help is m To list the current partition table, use p To add a new partition, use n To delete a partition, use d To set or changes the partition type, use t To provide a listing of the different partition types and their ID numbers, use l To saves your information and quits fdisk, use w Now, as an example: To make the partitions listed below on your system (these are the partitions we’ll need for our server installation example); the commands below are for fdisk: Step 1 Execute all of the following commands with fdisk to create the require partitions. Command (m for help): n Command action e extended p primary partition (1-4) p Partition number (1-4): 1 First cylinder (1-1116, default 1): 1 Last cylinder or +size or +sizeM or +sizeK (1-1116, default 1116): +18M Command (m for help): n Command action e extended p primary partition (1-4) e Partition number (1-4): 2 First cylinder (4-1116, default 4): 4 Last cylinder or +size or +sizeM or +sizeK (4-1116, default 1116): 1116 Command (m for help): n Command action l logical (5 or over) p primary partition (1-4) l First cylinder (4-1116, default 4): 4 Last cylinder or +size or +sizeM or +sizeK (4-1116, default 1116): +256M
- 43. Installation Issues 0 CHAPTER2 43 Command (m for help): n Command action l logical (5 or over) p primary partition (1-4) l First cylinder (37-1116, default 37): 37 Last cylinder or +size or +sizeM or +sizeK (37-1116, default 1116): +512M Command (m for help): n Command action l logical (5 or over) p primary partition (1-4) l First cylinder (103-1116, default 103): 103 Last cylinder or +size or +sizeM or +sizeK (103-1116, default 1116): +4512M Command (m for help): n Command action l logical (5 or over) p primary partition (1-4) l First cylinder (679-1116, default 679): 679 Last cylinder or +size or +sizeM or +sizeK (679-1116, default 1116): +256M Command (m for help): n Command action l logical (5 or over) p primary partition (1-4) l First cylinder (712-1116, default 712): 712 Last cylinder or +size or +sizeM or +sizeK (712-1116, default 1116): +512M Command (m for help): n Command action l logical (5 or over) p primary partition (1-4) l First cylinder (778-1116, default 778): 778 Last cylinder or +size or +sizeM or +sizeK (778-1116, default 1116): +1024M Command (m for help): n Command action l logical (5 or over) p primary partition (1-4) l First cylinder (909-1116, default 909): 909 Last cylinder or +size or +sizeM or +sizeK (909-1116, default 1116): +256M Command (m for help): n Command action l logical (5 or over) p primary partition (1-4) l First cylinder (942-1116, default 942): 942 Last cylinder or +size or +sizeM or +sizeK (942-1116, default 1116): 1116 Command (m for help): t Partition number (1-12): 12 Hex code (type L to list codes): 82 Changed system type of partition 12 to 82 (Linux swap)
- 44. 44 Step 2 Now, usethe p command to list the partition that we’ve created, you must see something like the following information on your screen. Command (m for help): p Disk /tmp/sda: 255 heads, 63 sectors, 1116 cylinders Units = cylinders of 16065 * 512 bytes Device Boot /tmp/sda1 /tmp/sda2 /tmp/sda5 /tmp/sda6 /tmp/sda7 /tmp/sda8 /tmp/sda9 /tmp/sda10 /tmp/sda11 /tmp/sda12 Start 1 4 4 37 103 679 712 778 909 942 End 3 1116 36 102 678 711 777 908 941 1116 Blocks 24066 8940172+ 265041 530113+ 4626688+ 265041 530113+ 1052226 265041 1405656 Id 83 5 83 83 83 83 83 83 83 82 System Linux Extended Linux Linux Linux Linux Linux Linux Linux Linux Swap Step 3 If all the partitions look fine and meet your requirements, use the w command to write the table to disk and exit fdisk program: Command (m for help): w The partition table has been altered Step 4 After you have partitioned your drive with fdisk, press Next and continue the installation with Disk Druid to choose the mount point of the directories. Disk Druid contains a list of all disk partitions with file-systems readable by Linux. This gives you the opportunity to assign these partitions to different parts of your Linux system when it boots. Select the partition you wish to assign and press Enter; then enter the mount point for that partition, e.g., /var.
- 45. Installation Issues 0 CHAPTER2 45 Boot Loader Installation On the next screen you will see the Boot Loader Configuration screen. In order to boot your Linux system, you usually need to install a boot loader. With new release of Linux, you can choose to install either GRUB, LILO, or you can choose not to install a boot loader at all. GRUB is the new and recommended method to boot Linux. You can still decide to use LILO, but it’s better to go with GRUB now. From this screen, you will see different configurable options related to GRUB or LILO. The first option is: • Use GRUB as the boot loader This option allows you to use the GRUB software as your boot loader to boot your Linux operating system on the computer. This is the recommended method to use with Linux. GRUB works in the same way as LILO work with many additional security and advanced features that LILO cannot provide you. In our setup, we use this option to boot our Linux server. The second option is: • Use LILO as the boot loader This option allows you to use the LILO software as your boot loader to boot your Linux operating system on the computer. Remember that LILO is now the old method to boot Linux and I recommend you to go with GRUB instead if you want to stay up-to-date with latest technology on the Linux world. In our setup, we don’t choose or use this option. The third option is: • Do not install a boot loader This option allows you to skip installing any type of available boot loader (GRUB or LILO) with Linux. This is useful if you use a boot disk rather than GRUB or LILO to start your operating system. This can greatly improve security in some case since you need to have a bootable Linux floppy with the kernel on it to start the server. But in other hand, you will not be able to restart the server remotely if something happens. In our setup, we don’t use this option. The fourth option is: • Install Boot Loader record on: Master Boot Record (MBR) First sector of boot partition Usually, if Linux is the only operating system on your machine (and this must be the case in a server installation), you should choose the “Master Boot Record (MBR)” option. The MBR is a special area on your hard drive that is automatically loaded by your computer's BIOS, and is the earliest point at which the boot loader can take control of the boot process.
- 46. 46 The fifth optionis: • Force use of LBA32 This option (if checked) allows you to exceed the 1024 cylinder limit for the /boot partition. If you have a system which supports the LBA32 extension for booting operating systems above the 1024 cylinder limit, and you want to place your /boot partition above cylinder 1024, you should select this option but in most case you can live without it and your system will perfectly work. In our setup of the operating system, we don’t use it. The GRUB Password This section applies only if you have selected GRUB as your boot loader. If you are installing GRUB as your boot loader, you should create a password to protect your system. Without a GRUB password, users with access to your system can pass options to the kernel which can compromise your system security. With a GRUB password in place, the password must first be entered in order to select any non-standard boot options.
- 47. Installation Issues 0 CHAPTER2 47 Network Configuration After that, you need to configure your network. If you have multiple Ethernet devices, each device will have its own configuration screen. You will be answered to enter the IP Address, Netmask, Network, Broadcast addresses, and the Gateway, Primary DNS (and if applicable the Secondary DNS and Ternary DNS) addresses. You should know all of the information or you can ask your system administrator to help you get the correct information. Firewall Configuration From this part of the setup installation, we have possibility to configure a Firewall. This is OK for the average end user but NOT for serious Firewall security. This newly added feature uses the old IPCHAINS tool of Linux with the help of a small utility named “lokkit” to set up your firewall. I highly recommend you to deactivate this feature now and see later in this book on how to install and configure IPTables with GIPTable, which is the new Firewall tool to use with Linux and kernel 2.4 generation. GIPTables is simply a Firewall software that can help you to configure IPTables in the most secure and easily way than any other firewall software can provide you. From the next screen that appears, you will see three different security levels available, choose the “No firewall” option and click Next.
- 48. 48 Language Support Selection Withthe internalization, a need for different language support has appeared. From here the installation will ask you to choose the default language that will be used on your Linux system once the installation is complete. If you are only going to use one language on your system, selecting only this language will save significant disk space.
- 49. Installation Issues 0 CHAPTER2 49 Time Zone Selection On the next screen, you will have the opportunity to set your time zone. Once selected click Next. Account Configuration After the clock has been configured, you need to give your system a root password account.
- 50. 50 Authentication Configuration Finally, thelast stage is the authentication configuration. For Authentication Configuration don’t forget to select: Enable MD5 passwords Enable Shadow passwords Enable MD5 passwords - allows a long password to be used (up to 256 characters), instead of the Unix standard eight letters or less. Enable shadow passwords - provides a very secure method of retaining passwords for you. All passwords are stored in a file named shadow, which is readable only by the super-user root. Enable NIS, LDAP, Kerberos and SMB doesn’t need to be selected since we are not configuring these services on this server right know. Selecting Package Groups After your partitions have been configured and selected for formatting and configurations have been set for your specific system, you are ready to select packages for installation. By default, Linux is a powerful operating system that runs many useful services. However, many of these services are unneeded and pose potential security risks. Ideally, each network service should be on a dedicated, single-purpose host. Many Linux operating systems are configured by default to provide a wider set of services and applications than are required to provide a particular network service, so you may need to configure the server to eliminate unneeded services. Offering only essential services on a particular host can enhance your network security in several ways:
- 51. Installation Issues 0 CHAPTER2 51 Other services cannot be used to attack the host and impair or remove desired network services. The host can be configured to better suit the requirements of the particular service. Different services might require different hardware and software configurations, which could lead to needless vulnerabilities or service restrictions. By reducing services, the number of logs and log entries is reduced so detecting unexpected behavior becomes easier. Different individuals may administer different services. By isolating services so each host and service has a single administrator you will minimize the possibility of conflicts between administrators. A proper installation of your Linux server is the first step to a stable, secure system. From the screen menu that appears (Selecting Package Groups), you first have to choose which system components you want to install, in our case; we must DESELECT ALL CHECKED Package Groups on the list. Since we are configuring a Linux server, we don’t need to install a graphical interface (XFree86) on our system (a graphical interface on a server means less processes, less CPU availability, less memory, security risks, and so on), also computers are subject to the treachery of images as well. The image on your computer screen is not a computer file -- it's only an image on a computer screen. Images of files, processes, and network connections are very distant cousins of the actual bits in memory, in network packets, or on disks. Layer upon layer of hardware and software produces the images that you see. When an intruder "owns" a machine, any of those layers could be tampered with. Application software can lie, OS kernels can lie, boot PROMs can lie, and even hard disk drives can lie. Graphical interfaces are usually used on only workstations. Step 1 First of all, it is vital to verify and be SURE to deselect all of the following Package Group: Printing Support Classic X Window System X Window System Laptop Support GNOME KDE Sound and Multimedia Support Network Support Dialup Support Messaging and Web Tools Graphics and Image Manipulation New Server NFS File Server Windows File Server Anonymous FTP Server SQL Database Server Web Server Router / Firewall DNS Name Server Network Managed Workstation Authoring and Publishing Emacs Utilities Legacy Application Support Software Development Kernel Development Windows Compatibility / Interoperability Games and Entertainment Everything To resume, it is very important and I say VERY IMPORTANT to deselect (none is selected) every selected Packages Group before clicking on the Next button for continuing the installation.
- 52. 52 We don’t wantand don’t need to install any additional packages. The default install of this Linux distribution already comes with the most essential programs we need for the base functionality of the operating system. Step 2 At this point, the installation program will check dependencies in packages selected for installation (in our case no packages are selected) and format every partition you selected for formatting in you system. This can take several minutes depending on the speed of your machine. Once all partitions have been formatted, the installation program starts to install Linux to your hard drive.
- 53. Installation Issues 0 CHAPTER2 53 Boot Disk Creation From this section of the installation, we have the possibility to create a boot disk for our newly installed operating system. If you do not want to create a boot disk, you should check the “Skip boot disk creation” checkbox before you click Next. Creating a boot disk must be made if you decide to not install GRUB or LILO on the MBR (the Master Boot Record) or if you are not installing GRUB or LILO at all. How to use RPM Commands This section contains an overview of using RPM for installing, uninstalling, upgrading, querying, listing, and checking RPM packages on your Linux system. You must be familiar with these RPM commands now because we’ll use them often in this book and especially later in this chapter for software that must be uninstalled after installation of the server. Install a RPM package: Note that RPM packages have a file of names like foo-1.0-2.i386.rpm, which include the package name (foo), version (1.0), release (2), and architecture (i386). • To install a RPM package, use the command: [root@deep /]# rpm -ivh foo-1.0-2.i386.rpm foo ################################################## Uninstall a RPM package: Notice that we used the package name “foo”, not the name of the original package file “foo- 1.0-2.i386.rpm”. • To uninstall a RPM package, use the command: [root@deep /]# rpm -e foo
- 54. 54 Upgrade a RPMpackage: With this command, RPM automatically uninstalls the old version of foo package and installs the new one. Always use “rpm -Uvh” command to install packages, since it works fine even when there are no previous versions of the package installed. This is the recommended method of installing package on the system. • To upgrade a RPM package, use the command: [root@deep /]# rpm -Uvh foo-1.0-2.i386.rpm foo ################################################## Force the installation of a RPM package: With this command, RPM will force the installation of the specified package even if some conflict or other kind of problem exists. This command should be used with care and only if you know what you do. In most case, RPM can correctly guest problem and refuse to install. To bypass RPM warning, you can use the RPM command below. • To force the installation of a RPM package, use the command: [root@deep /]# rpm -Uvh --force foo-1.0-2.i386.rpm foo ################################################## Avoid RPM package dependency: With this command, RPM will not take care of package dependency and will install the RPM software on your system. Package dependency is an important concept in the RPM world. Dependency is when some other packages depend of the RPM package you are trying to install. By default, RPM check if all other RPM packages required for the RPM you try to install are present before installing the RPM. If some required packages are not present, RPM will inform you. This is made to avoid problem and be sure that the software you want to install will perfectly work. In some special case, we don’t need to take care of dependency and can use the option below to inform it to skip the dependency check when installing the software. • To avoid RPM package dependency, use the command: [root@deep /]# rpm -Uvh --nodeps foo-1.0-2.i386.rpm foo ################################################## Query a RPM package: This command will print the package name, version, and release number of installed package foo. Use this command to verify that a package is or is not installed on your system. • To query a RPM package, use the command: [root@deep /]# rpm -q foo foo-2.3-8
- 55. Installation Issues 0 CHAPTER2 55 Display RPM package information: This command displays package information; includes name, version, and description of the installed program. Use this command to get information about the installed package. • To display RPM package information, use the command: [root@deep /]# rpm -qi foo Name : foo Relocations: none Version : 2.3 Vendor: OpenNA.com, Inc. Release : 8 Build Date: Thu 24 Aug 2000 11:16:53 AM EDT Install date: Mon 12 Feb 2001 01:17:24 AM EST Build Host: openna.com Group : Applications/Archiving Source RPM: foo-2.3-8.src.rpm Size : 271467 License: distributable Packager : OpenNA.com, Inc. <http://www.openna.com/> Summary : Here will appears summary of the package. Description : Here will appears the description of the package. Display RPM package information before installing the program: This command displays package information; includes name, version, and description of the program without the need to install the program first. Use this command to get information about a package before you install it on your system. • To display package information before installing the program, use the command: [root@deep /]# rpm -qpi foo-2.3-8.i386.rpm Name : foo Relocations: none Version : 2.3 Vendor: OpenNA.com, Inc. Release : 8 Build Date: Thu 24 Aug 2000 11:16:53 AM EDT Install date: Mon 12 Feb 2001 01:17:24 AM EST Build Host: openna.com Group : Applications/Archiving Source RPM: foo-2.3-8.src.rpm Size : 271467 License: distributable Packager : OpenNA.com, Inc. <http://www.openna.com/> Summary : Here will appears summary of the package. Description : Here will appears the description of the package. List files in a installed RPM package: This command will list all files in a installed RPM package. It works only when the package is already installed on your system. • To list files in a installed RPM package, use the command: [root@deep /]# rpm -ql foo /usr/bin/foo /usr/bin/foo1 /usr/sbin/foo2 List files in RPM package that is not already installed: This command will list all files in a RPM package that is not already installed on your system. It is useful when you want to know which components are included in the package before installing it. • To list files in RPM package that is not already installed, use the command: [root@deep /]# rpm -qpl foo /usr/lib/foo /usr/bin/foo1 /usr/sbin/foo2
- 56. 56 Know which filesis part of which RPM package: This command will show you from which RPM package the file comes from. It works only when the package is already installed on your system and it is very useful when you see some files into Linux that you do not know about it and want to get more information about its RPM provenance. • To know which files is part of which RPM package, use the command: [root@deep /]# rpm -qf /etc/passwd setup-2.3.4-1 Check RPM signature package: This command checks the PGP signature of specified package to ensure its integrity and origin. Always use this command first before installing new RPM package on your system. GnuPG or PGP software must be already installed on your system before you can use this command. See the chapter related to GnuPG installation and configuration for more information. • To check a RPM signature package, use the command: [root@deep /]# rpm --checksig foo Examine the md5sum of RPM package: The RPM md5sum is useful to verify that a package has not been corrupted or tampered with. You can use it to be sure that the download of your new RPM package was not corrupted during network transfer. • To examine only the md5sum of the package, use the command: [root@deep /]# rpm --checksig --nogpg foo Starting and stopping daemon services The init program of Linux (also known as process control initialization) is in charge of starting all the normal and authorized processes that need to run at boot time on your system. These may include the APACHE daemons, NETWORK daemons, and anything else that must be running when your machine boots. Each of these processes has a script file under the /etc/init.d directory written to accept an argument, which can be start, stop, restart, etc. As you can imagine, those script are made to simplify the administration of the server and the way we can start or stop services under Linux. Of course, we can use the native way to start all required services under our server, but it is much simple to have some kind of script files that should provide us some easy method to automate and control the procedures. This is why init program and all initialization script files available under the /etc/init.d directory exist. Below are some examples showing you how to execute those scripts by hand. For example: • To start the httpd web server daemon manually under Linux, you’ll type: [root@deep /]# /etc/init.d/httpd start Starting httpd: [OK] • To stop the httpd web server daemon manually under Linux, you’ll type: [root@deep /]# /etc/init.d/httpd stop Shutting down http: [OK]
- 57. Installation Issues 0 CHAPTER2 57 • To restart the httpd web server daemon manually under Linux, you’ll type: [root@deep /]# /etc/init.d/httpd restart Shutting down http: [OK] Starting httpd: [OK] Check inside your /etc/init.d directory for services available and use the commands start | stop | restart to work around. You will see along this book that we often use initialization script file to administration and control the way we start, restart, stop, etc services under Linux. Software that must be uninstalled after installation of the server Red Hat Linux installs other programs on your system by default and doesn’t give you the choice to uninstall them during the install setup. For this reason, you must uninstall the following software after the complete installation of our Linux server. Below is the list of programs and a short description of their purpose. We must uninstall them for increased security and to make more space on our server. For more information and an explanation of their capabilities and uses, please see your Red Hat manual or query the package by making an “rpm -qi foo” command before uninstalling them. The anacron package: The anacron package is similar to the cron package but differ in the way that it does not assume that the system is running continuously and it is a good command scheduler for system which doesn’t run 24 hours a day. In server environment, the system should absolutely run 24/24 hours; therefore we simply don’t need to have this kind of package installed on a server. • To remove the anacron package from your system, use the following commands: [root@deep /]# /etc/init.d/anacron stop [root@deep /]# rpm -e anacron [root@deep /]# rm -rf /var/spool/anacron/ The apmd package: The apmd package or Advanced Power Management Daemon utilities is used on notebook computer. It can watch your notebook's battery and warn all users when the battery is low. As you can imagine, there is no need to have it installed on a server machine. • To remove the apmd package from your system, use the following commands: [root@deep /]# /etc/init.d/apmd stop [root@deep /]# rpm -e apmd The at package: The at package is a utility that will do time-oriented job control by scheduling a command to run later. Unfortunately, it has had a rich history of problems and we can achieve the same functionality with the more secure vixie-cron package. For this reason I recommend you to uninstall it. • To remove the at package from your system, use the following commands: [root@deep /]# /etc/init.d/atd stop [root@deep /]# rpm -e at
- 58. 58 The gpm package: Thegpm package provides mouse support to text-based Linux applications. It’s the software that allows you to cut-and-paste operations using the mouse on your terminal. If most of your entire administration of the server is made via remote connection, you can remove this package to save some processes and memories. We can continue to use cut-and-paste operation via remote connection to the server without problem. The gpm package is only useful if you stay at the console terminal of your server to make administration tasks. • To remove the gpm package from your system, use the following commands: [root@deep /]# /etc/init.d/gpm stop [root@deep /]# rpm -e gpm The dhcpcd package: The dhcpcd package contains the protocol, which allows systems to get their own network configuration information from a DHCP server. If you are going to use DHCP on your network, it is recommended to install the DHCP client included in the pump package, which provides a faster and simpler DHCP client. For more information about DHCP, see its related chapter in this book. • To remove the dhcpcd package from your system, use the following command: [root@deep /]# rpm -e dhcpcd The eject package: The eject package contains an eject program that allows the user to eject removable media (typically CD-ROMs, floppy disks, Iomega Jaz or Zip disks) using software. This is an unneeded program to have installed in a server environment. You should keep it installed only if you’re intended to run a tape backup on your system. • To remove the eject package from your system, use the following command: [root@deep /]# rpm -e eject The hotplug package: The hotplug package is a helper application for loading modules for USB devices. On a server environment, USB devices are not used at all and are required only on Linux workstation. • To remove the hotplug package from your system, use the following command: [root@deep /]# rpm -e hotplug The lokkit package: The lokkit package is a Firewall configuration application for an average end user and it is not designed to configure arbitrary firewalls since it is solely designed to handle typical dialup user and cable modem set-ups. It is not the answer to a complex firewall configuration, and it is not the equal of an expert firewall designer. Therefore remove it from your server and read the chapter related to GIPTables in this book for more information about secure firewall configuration. • To remove the lokkit package from your system, use the following command: [root@deep /]# rpm -e lokkit
- 59. Installation Issues 0 CHAPTER2 59 The ipchains package: The ipchains package is the old tool used with Linux kernel 2.2 for managing Linux kernel packet filtering capabilities and to set up firewalling on the network. A new and more powerful tool called “IPTables” exists and this is the one that we will use later in the book to set up our firewall on Linux. • To remove the ipchains package from your system, use the following command: [root@deep /]# rpm -e ipchains The ksymoops package: The ksymoops package is a small tool used to report kernel oops and error message decoder. This package is useful for developers that work on the Linux kernel and want to debug or for users that want to report bugs with the kernel. The same result can be achieved with the dmesg command of Linux. Therefore, you can remove this package from your secure server. • To remove the ksymoops package from your system, use the following command: [root@deep /]# rpm -e ksymoops The kudzu package: The kudzu package is a hardware-probing tool that runs at system boot time to determine what hardware has been added or removed from the system. Again, in server environment, we don’t upgrade, add or remove hardware every time. Therefore, we can safety remove this small package from our server. • To remove the kudzu package from your system, use the following command: [root@deep /]# rpm -e kudzu The mailcap package: Metamail is a program that uses the mailcap file to determine how it should display non-text or multimedia material. We don’t need to have it installed at all. Remove it. • To remove the mailcap package from your system, use the following command: [root@deep /]# rpm -e mailcap The pciutils package: The pciutils package contains various utilities for inspecting and setting devices connected to the PCI bus. Keep it installed if you want, but I recommend removing it form the server. • To remove the pciutils package from your system, use the following command: [root@deep /]# rpm -e pciutils The raidtools package: The raidtools package includes the tools you need to set up and maintain a software RAID device on a Linux system. You should keep this package only if you have configured your server to run with RAID support. Otherwise, remove it. • To remove the raidtools package from your system, use the following command: [root@deep /]# rpm -e raidtools
- 60. 60 The redhat-logos package: Theredhat-logos package contains files of the Red Hat "Shadow Man" logo and the RPM logo. • To remove the redhat-logos package from your system, use the following command: [root@deep /]# rpm -e redhat-logos The redhat-release package: The redhat-release package contains the Red Hat Linux release files. Please note that if you remove this package, the boot process of the system will generate error messages because it will look for a file called “redhat-release” which will not be available anymore. To solve this problem, we recreate the required file under the appropriated directory and add as content of this file the word “Red Hat Linux”. Of course you can change it for whatever you want. • To remove the redhat-release package from your system, use the command: [root@deep /]# rpm -e redhat-release [root@deep /]# echo Red Hat Linux > /etc/redhat-release The setserial package: The setserial package is a basic system utility for displaying or setting serial port information. • To remove the setserial package from your system, use the command: [root@deep /]# rpm -e setserial The hdparm package: The program hdparm is needed by IDE hard disks but not SCSI hard disks. If you have an IDE disk on your system you must keep this program (hdparm), but if you don’t have an IDE hard disk you can remove it safely from your system. hdparm is a small Linux tool used to optimize your IDE hard drive. SCSI hard drives don’t need to be optimized since they are capable to run at their full speed (80 Mps to 160 Mps) without modification. • To remove the hdparm package from your system, use the following command: [root@deep /]# rpm -e hdparm The mkinitrd package: The program mkinitrd is needed by SCSI or RAID hard disk but not IDE hard disks. If you have a SCSI or RAID disk on your system you must keep this program (mkinitrd), but if you don’t have a SCSI or RAID hard disk you can safely remove it from your system. • To remove the mkinitrd package from your system, use the following command: [root@deep /]# rpm -e --nodeps mkinitrd The xconfig packages: Use the programs kbdconfig, mouseconfig, timeconfig, netconfig, authconfig, ntsysv, and setuptool in order to set your keyboard language and type, your mouse type, your default time zone, your Ethernet devices, your NIS and shadow passwords, your numerous symbolic links in /etc/rc.d directory, and text mode menu utility which allow you to access all of these features.
- 61. Installation Issues 0 CHAPTER2 61 After those configurations have been set during the installation stage of your Linux server it’s rare that you would need to change them again. So, you can uninstall them, and if in the future you need to change your keyboard, mouse, default time, etc again via test mode menu, all you have to do is to install the program with the RPM from your original CD-ROM. • To remove all the above programs from your system, use the following command: [root@deep /]# rpm -e kbdconfig mouseconfig timeconfig netconfig authconfig ntsysv setuptool The newt package: The newt package provides a development library for text mode user interfaces. It’s mainly used by all the above config packages. Since all the config packages are removed from the server, we can safety remove newt from our system. If you have decided to keep the above config packages (kbdconfig, mouseconfig, etc), then you should keep newt, otherwise you should remove it. • To remove the newt package from your system, use the following command: [root@deep /]# rpm -e newt The lilo package: The lilo package provides a boot loader for Linux. Remember that during our Linux installation, we have chosen to go with GRUB instead of LILO as our boot loader for Linux. Therefore, you can safety remove this package from your system since it’s really not used now. • To remove the lilo package from your system, use the following command: [root@deep /]# rpm -e lilo [root@deep /]# rm -f /etc/lilo.conf.anaconda The reiserfs-utils package: The reiserfs-utils package contains a number of utilities for creating, checking, modifying, and correcting any inconsistencies in ReiserFS file-systems. ReiserFS is another file-system like Ext3 for Linux. In our configuration of Linux we use Ext3 as our file-system therefore we don’t need to keep this package installed. Keep this utility package only if you intended to run ReiserFS instead of Ext3 file-system on your Linux server. • To remove the reiserfs-utils package from your system, use the command: [root@deep /]# rpm -e reiserfs-utils The quota package: The program quota is a system administration tools for monitoring and limiting user/group disk usage, per file system. This program must be installed only on servers where the need for monitoring and restricting amount of disk space in user’s directories is require. • To remove the quota package from your system, use the following command: [root@deep /]# rpm -e quota
- 62. 62 The indexhtml package: Theindexhtml package contains the HTML page and graphics for a welcome page shown by your browser under graphical interface installation. These HTML pages are information about Red Hat software. You really don’t need this package under server installation and especially when GUI is not available. Therefore, you can safety remove this package from your system. • To remove the indexhtml package from your system, use the following command: [root@deep /]# rpm -e indexhtml The usbutils package: The usbutils package provides Linux USB utilities for inspecting devices connected to a USB bus on your system. In server environment, we really don’t use any USB devices and can safety remove this package from our server installation. USB will usually be used only in Linux workstation installation where you want to plug printer, camera and any other media of this type. • To remove the usbutils package from your system, use the following command: [root@deep /]# rpm -e usbutils The hwdata package: The hwdata package contains various hardware identification and configuration data mainly used with USB and XFree86. Remember that XFree86 is related to graphical interface and since we don’t use any GUI into our Linux server, we can remove this package from our system. • To remove the hwdata package from your system, use the following command: [root@deep /]# rpm -e hwdata The parted package: The parted package contains various utilities to create, destroy, resize, move and copy hard disk partitions. It is useful when you need to play with your hard disk structure. In most case, we partition and set hard disk organization at the installation of the operating system and don’t need to play or change something once everything are installed and running properly. It’s rare to have to use this package utility on a production server and this is the reason why I recommend you to uninstall it. If you prefer to keep it installed for future possible usage, you are free to do it. • To remove the parted package from your system, use the following command: [root@deep /]# rpm -e parted The hesiod package: The hesiod package is another one that we can uninstall from our Linux server configuration setup. It’s a system which uses existing DNS functionality to provide access to databases of information that change infrequently. In most cases, we don’t need it and you should keep it install only if you think that you will need it for some special configuration of your server. • To remove the hesiod package from your system, use the following command: [root@deep /]# rpm -e hesiod
- 63. Installation Issues 0 CHAPTER2 63 The mt-st package: The mt-st package provides tools for controlling and managing tape drives on your system. You should keep it installed only if you have and want to run a tape backup media on your system. • To remove the mt-st package from your system, use the following command: [root@deep /]# rpm -e mt-st The man-pages package: The man-pages package provides a large collection of additional manual pages (man pages) from the Linux Documentation Project (LDP). By default many manual pages are installed with the operating system and the man-pages package provides additional documentation for those who want to read them on the system. In server environment, I really don’t see the need to have additional manual pages installed since these manual pages can be read online from the Internet or even on another computer running as a development or workstation. • To remove the man-pages package from your system, use the following command: [root@deep /]# rpm -e man-pages The sendmail package: Even if you don’t want to run your system as a full mail server, mailer software is always needed for potential messages sent to the root user by different software services installed on your machine. Sendmail is a Mail Transport Agent (MTA) program that sends mail from one machine to another and it’s the default mail server program installed on Red Hat Linux. Unfortunately, this software has a long history of security problem and for this reason I highly recommend you to not use it on your Linux server. You must uninstall Sendmail and see the part in this book that is related to Mail Transfer Agent configuration and installation for some good alternative like Exim or Qmail. • To remove the sendmail package from your system, use the following commands: [root@deep /]# /etc/init.d/sendmail stop [root@deep /]# rpm -e sendmail The procmail package: Procmail is a mail-processing program, which can be used by Sendmail for all local mail delivery and filtering. This program is required only if you decide to install and use Sendmail on your server and only if Sendmail is installed. Since we’ve decided to not go with Sendmail as our MTA for security reason, we can uninstall procmail from our Linux server. • To remove the procmail package from your system, use the following command: [root@deep /]# rpm -e procmail The openldap package: The OpenLDAP software is a set of protocols for accessing directory services like phone book style information and other kinds of information over the Internet. This useful program is not suitable for everyone and depends of what you want to do with your system. If you want to give it a try, see later in this book under the chapter related to databases for more information. • To remove the OpenLDAP package from your system, use the following command: [root@deep /]# rpm -e --nodeps openldap
- 64. 64 The cyrus-sasl packages: TheCyrus SASL implementation is the Simple Authentication and Security Layer, a method for adding authentication support to connection-based protocols. It is used in conjunction with Cyrus, which is an electronic messaging program like Sendmail. Since Cyrus SASL is made to be used with Sendmail that we have removed previously for security reason, we can safety remove it. • To remove the Cyrus SASL package from your system, use the following command: [root@deep /]# rpm -e --nodeps cyrus-sasl cyrus-sasl-md5 cyrus-sasl-plain The openssl package: OpenSSL is an SSL encryption mechanism which ensures and provides safe and secure transactions of data over networks. This piece of software is one of the most important tools for a Linux server and it is highly recommended that it is installed. Unfortunately, the one that comes with Red Hat Linux is not up to date and not optimized for our specific server. For this reason, we will uninstall it now and see later in this book, under the chapters related to security software, how to install, secure, optimize and use it. • To remove the OpenSSL package from your system, use the following command: [root@deep /]# rpm -e --nodeps openssl [root@deep /]# rm -rf /usr/share/ssl/ The ash package: The ash package is a smaller version of the bourne shell (sh). Since we already use sh, we can uninstall this package from our system. If you use this program in your regular administration task, then keep it installed on your server. In most case, we can remove it. • To remove the ash package from your system, use the following command: [root@deep /]# rpm -e ash The tcsh package: The tcsh package is an enhanced version of csh, another C shell. We already have bash as our default shell program on Linux, and I don’t find any reason to keep another variant installed if we don’t have any program or services that need it to run. Most services under Linux can easily run with our default bash shell program and if you don’t have any program that required tcsh to run, then I recommend you to uninstall it. If in the future, you see that you need to have tcsh installed on your server for some specific program to run, then all you have to do is to install it from your CD-ROM. In most cases, there is no program that needs tsch to run, therefore you can remove it. • To remove the tcsh package from your system, use the following command: [root@deep /]# rpm -e tcsh
- 65. Installation Issues 0 CHAPTER2 65 The specspo package: The specspo package contains the portable object catalogues used to internationalize Red Hat packages. I don’t think that this kind of package is really required on a production server. • To remove the specspo package from your system, use the following command: [root@deep /]# rpm -e specspo The krb5-libs package: The krb5-libs package contains the shared libraries needed by Kerberos 5. Because we’re not using Kerberos, we’ll need to uninstall this package. Kerberos is not secure as you can think and can be cracked easily with some good knowledge of this program. Anyway it is yours to decide if you really need it. • To remove the krb5-libs package from your system, use the following command: [root@deep /]# rpm -e krb5-libs [root@deep /]# rm -rf /usr/kerberos/ The MAKEDEV package: The MAKEDEV package contains the MAKEDEV program, which makes it easier to create and maintain the files in the /dev directory. Program provided by this package is used for creating device files in the /dev directory of your server. In general, we use it under development server when we build new package for our Linux system. On production servers, it’s rare to use it. Therefore we can remove it from our system without any problem. • To remove the MAKEDEV package from your system, use the following command: [root@deep /]# rpm -e MAKEDEV Remove unnecessary documentation files By default the majority of each RPM’s packages installed under Linux come with documentation files related to the software. This documentation contains original files from the program tar archive like README, FAQ, BUG, INSTALL, NEWS, PROJECTS and more. Many of them can be easily retrieved from the website where the program has been downloaded and it makes no sense for them to be kept on your system. I know that hard drives costs have come down considerably recently, but why keep this kind of documentation on a secure server if it unlikely they will not be read more than once. Anyway, have a look inside those files and decide for yourself if you want to remove them or not. • To remove all documentation files from your system, use the following commands: [root@deep /]# cd /usr/share/doc/ [root@deep doc]# rm -rf *
- 66. 66 Remove unnecessary/empty filesand directories There are some files and directories we can remove manually from the file system of Linux to make a clean install. These files and directories are not needed but still exist after our secure installation of Linux and can be removed safely. Some are bugs from the Red Hat installation script and others are created by default even if you don’t use them. • To remove all unnecessary files and directories from your system, use the commands: [root@deep /]# rm -f /etc/exports [root@deep /]# rm -f /etc/printcap [root@deep /]# rm -f /etc/ldap.conf [root@deep /]# rm -f /etc/krb.conf [root@deep /]# rm -f /etc/yp.conf [root@deep /]# rm -f /etc/hosts.allow [root@deep /]# rm -f /etc/hosts.deny [root@deep /]# rm -f /etc/csh.login [root@deep /]# rm -f /etc/csh.cshrc [root@deep /]# rm -f /etc/fstab.REVOKE [root@deep /]# rm -f /etc/pam_smb.conf [root@deep /]# rm -rf /etc/xinetd.d/ [root@deep /]# rm -rf /etc/opt/ [root@deep /]# rm -rf /etc/X11/ [root@deep /]# rm -rf opt/ [root@deep /]# rm -rf /var/opt/ [root@deep /]# rm -rf /var/nis/ [root@deep /]# rm -rf /var/yp/ [root@deep /]# rm -rf /var/lib/games/ [root@deep /]# rm -rf /var/spool/lpd/ [root@deep /]# rm -rf /usr/lib/python1.5/ [root@deep /]# rm -rf /usr/lib/games/ [root@deep /]# rm -rf /usr/X11R6/ [root@deep /]# rm -rf /usr/etc/ [root@deep /]# rm -rf /usr/games/ [root@deep /]# rm -rf /usr/local/ [root@deep /]# rm -rf /usr/dict/ [root@deep /]# rm -f /usr/bin/X11 [root@deep /]# rm -f /usr/lib/X11 NOTE: If in the future you want to install a program which needs some of the files/directories we have removed, then the program will automatically recreate the missing files or directories. Good! Software that must be installed after installation of the server There are certain programs required to be able to compile programs on your server, for this reason you must install the following RPM packages. This part of the installation is very important and requires that you install all the packages described below. These are on your Red Hat Part 1 and Part 2 CD-ROMs under RedHat/RPMS directory and represent the necessary base software needed by Linux to compile and install programs. Please note that if you don’t want to compile software in your server or if you only use RPM’s packages to update programs or if you use a dedicated server to develop, compile or create your own RPM’s packages which will be installed later along your network on the servers, then you DON’T need to install the packages described here.
- 67. Installation Issues 0 CHAPTER2 67 Step 1 First, we mount the CD-ROM drive and move to the RPMS subdirectory of the CD-ROM. • To mount the CD-ROM drive and move to RPM directory, use the following commands: [root@deep /]# mount /dev/cdrom /mnt/cdrom/ had: ATAPI 32X CD-ROM drive, 128kB Cache mount: block device dev/cdrom is write-protected, mounting read-only [root@deep /]# cd /mnt/cdrom/RedHat/RPMS/ These are the packages that we need to be able to compile and install programs on the Linux system. Remember, this is the minimum number of packages that permits you to compile most of the tarballs available for Linux. Other compiler packages exist on the Linux CD-ROM, so verify with the README file that came with the tarballs program you want to install if you receive error messages during compilation of the specific software. The compiler packages: Compiler packages contain programs and languages used to build software on the system. Remember to uninstall the entire following compiler packages after successful installation of all software required on your Linux server. binutils-2.11.93.0.2-11.i386.rpm bison-1.35-1.i386.rpm byacc-1.9-19.i386.rpm cdecl-2.5-22.i386.rpm cpp-2.96-110.i386.rpm cproto-4.6-9.i386.rpm ctags-5.2.2-2.i386.rpm dev86-0.15.5-1.i386.rpm flex-2.5.4a-23.i386.rpm gcc-2.96-110.i386.rpm gcc-c++-2.96-110.i386.rpm glibc-kernheaders-2.4-7.14.i386.rpm m4-1.4.1-7.i386.rpm make-3.79.1-8.i386.rpm patch-2.5.4-12.i386.rpm perl-5.6.1-34.99.6.i386.rpm The development packages: Development packages contain header and other files required during compilation of software. In general, development packages are needed when we want to add some specific functionality to the program that we want to compile. For example if I want to add PAM support to IMAP, I’ll need pam-devel, which contains the required header files for IMAP to compile successfully. As for compiler packages, all development packages must be uninstalled after successful compilation of all the software that you need on your Linux server. Remember to uninstall them since they are not needed for proper functionality of the server, but just to compile the programs. aspell-devel-0.33.7.1-9.i386.rpm db3-devel-3.3.11-6.i386.rpm freetype-devel-2.0.9-2.i386.rpm gdbm-devel-1.8.0-14.i386.rpm gd-devel-1.8.4-4.i386.rpm glibc-devel-2.2.5-34.i386.rpm libjpeg-devel-6b-19.i386.rpm libpng-devel-1.0.12-2.i386.rpm libstdc++-devel-2.96-110.i386.rpm ncurses-devel-5.2-26.i386.rpm pam-devel-0.75-32.i386.rpm pspell-devel-0.12.2-8.i386.rpm zlib-devel-1.1.3-25.7.i386.rpm
- 68. 68 Dependencies packages: Dependencies packagesare other RPM packages needed by the RPM packages that we want to install. This happens because some RPM’s are directly linked with others and depend on each one to function properly. The following packages are required by the above RPM packages and we will install them to satisfy dependencies. After proper compilation and installation of all needed software on the Linux server, we can uninstall them safety (if not needed by special program that we will install). aspell-0.33.7.1-9.i386.rpm freetype-2.0.9-2.i386.rpm gd-1.8.4-4.i386.rpm libjpeg-6b-19.i386.rpm libpng-1.0.12-2.i386.rpm libtool-libs-1.4.2-7.i386.rpm pspell-0.12.2-8.i386.rpm Step 2 It is better to install the software described above together if you don’t want to receive dependencies error messages during the install. Some of the RPMs reside on CD-ROM Part 1 and other on CD-ROM Part2 of Red Hat. For easy installation, I recommend you to copy all of the required packages (compilers and development) to your hard drive and install them from there. • These procedures can be accomplished with the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rpm -Uvh *.rpm Preparing... ########################################### [100%] 1:binutils ########################################### [ 2%] 2:bison ########################################### [ 5%] 3:byacc ########################################### [ 8%] 4:cdecl ########################################### [ 11%] 5:cpp ########################################### [ 13%] 6:cproto ########################################### [ 16%] 7:ctags ########################################### [ 19%] 8:db3-devel ########################################### [ 22%] 9:dev86 ########################################### [ 25%] 10:flex ########################################### [ 27%] 11:freetype ########################################### [ 30%] 12:freetype-devel ########################################### [ 33%] 13:gdbm-devel ########################################### [ 36%] 14:glibc-kernheaders ########################################### [ 38%] 15:glibc-devel ########################################### [ 41%] 16:gcc ########################################### [ 44%] 17:libjpeg ########################################### [ 47%] 18:libjpeg-devel ########################################### [ 50%] 19:libpng ########################################### [ 52%] 20:gd ########################################### [ 55%] 21:gd-devel ########################################### [ 58%] 22:libstdc++-devel ########################################### [ 61%] 23:gcc-c++ ########################################### [ 63%] 24:libtool-libs ########################################### [ 66%] 25:m4 ########################################### [ 69%] 26:make ########################################### [ 72%] 27:pam-devel ########################################### [ 75%] 28:patch ########################################### [ 77%] 29:perl ########################################### [ 80%] 30:ncurses-devel ########################################### [ 83%] 31:pspell ########################################### [ 86%] 32:aspell ########################################### [ 88%] 33:pspell-devel ########################################### [ 91%] 34:aspell-devel ########################################### [ 94%] 35:zlib-devel ########################################### [ 97%] 36:libpng-devel ########################################### [100%]
- 69. Installation Issues 0 CHAPTER2 69 Step 3 After installation and compilation of all programs and services, it’s a good idea to remove all sharp objects (compilers, etc) described above unless they are required by your system. A few reasons are: If a cracker gains access to your server he or she cannot compile or modify binary programs. Also, this will free a lot of space and will help to improve regular scanning of the files on your server for integrity checking. When you run a server, you will give it a special task to accomplish. You will never put all services you want to offer in one machine or you will lose speed (resources available divided by the number of process running on the server). Decrease your security with a lot of services running on the same machine, if a cracker accesses this server, he or she can attack directly all the others available. Having different servers doing different tasks will simplify the administration, and management. You know what task each server is supposed to do, what services should be available, which ports are open to clients access and which one are closed, you know what you are supposed to see in the log files, etc, and give you more control and flexibility on each one (server dedicated for mail, web pages, database, development, backup, etc. For example, one server specialized just for development and testing will mean you will not be compelled to install compiler programs on a server each time you want to compile and install new software on it, and be obliged afterwards to uninstall the compilers, or other sharp objects.
- 73. General Security IN THISCHAPTER 1. BIOS 2. Unplug your server from the network 3. Security as a policy 4. Choose a right password 5. The root account 6. Set login time out for the root account 7. Shell logging 8. The single-user login mode of Linux 9. Disabling Ctrl-Alt-Delete keyboard shutdown command 10. Limiting the default number of started ttys on the server 11. The LILO and /etc/lilo.conf file 12. The GRUB and /boot/grub/grub.conf file 13. The /etc/services file 14. The /etc/securetty file 15. Special accounts 16. Control mounting a file system 17. Mounting the /usr directory of Linux as read-only 18. Tighten scripts under /etc/init.d/ 19. Tighten scripts under /etc/cron.daily 20. Bits from root-owned programs 21. Don’t let internal machines tell the server what their MAC address is 22. Unusual or hidden files 23. Finding Group and World Writable files and directories 24. Unowned files 25. Finding .rhosts files 26. Physical hard copies of all-important logs 27. Getting some more security by removing manual pages 28. System is compromised!
- 75. General Security 0 CHAPTER3 75 Linux General Security Abstract A secure Linux server depends on how the administrator makes it. Once we have eliminated the potential security risk by removing unneeded services, we can start to secure our existing services and software on our server. Within a few hours of installing and configuring your system, you can prevent many attacks before they occur. In this chapter we will discuss some of the more general, basic techniques used to secure your system. The following is a list of features that can be used to help prevent attacks from external and internal sources. BIOS It is recommended to disallow booting from floppy drives and set passwords on BIOS features. You can check your BIOS manual or look at it thoroughly the next time you boot up your system to find out how to do this. Disabling the ability to boot from floppy drives and being able to set a password to access the BIOS features will improve the security of your system. This will block unauthorized people from trying to boot your Linux system with a special boot disk and will protect you from people trying to change BIOS features like allowing boot from floppy drive or booting the server without prompt password. It is important to note that there is a possibility to bypass this security measure if someone has a physical access to your server since they can open the computer and unplug the BIOS battery. This will reset all features to their initial values. Unplug your server from the network It is not wise to apply security changes in your newly installed Linux server if you are online. So it is preferable to deactivate all network interfaces in the system before applying security changes. • To stop specific network devices manually on your system, use the command: [root@deep /]# ifdown eth0 • To start specific network devices manually on your system, use the command: [root@deep /]# ifup eth0 • To shut down all network interfaces, use the following command: [root@deep /]# /etc/init.d/network stop Shutting down interface eth0 [OK] Disabling Ipv4 packet forwarding [OK] • To start all network interfaces, use the following command: [root@deep /]# /etc/init.d/network start Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK]
- 76. 76 Security as apolicy It is important to point out that you cannot implement security if you have not decided what needs to be protected, and from whom. You need a security policy--a list of what you consider allowable and what you do not consider allowable upon which to base any decisions regarding security. The policy should also determine your response to security violations. What you should consider when compiling a security policy will depend entirely on your definition of security. The following questions should provide some general guidelines: How do you classify confidential or sensitive information? Does the system contain confidential or sensitive information? Exactly whom do you want to guard against? Do remote users really need access to your system? Do passwords or encryption provide enough protection? Do you need access to the Internet? How much access do you want to allow to your system from the Internet? What action will you take if you discover a breach in your security? This list is short, and your policy will probably encompass a lot more before it is completed. Any security policy must be based on some degree of paranoia; deciding how much you trust people, both inside and outside your organization. The policy must, however, provide a balance between allowing your users reasonable access to the information they require to do their jobs and totally disallowing access to your information. The point where this line is drawn will determine your policy. Choose a right password The starting point of our Linux general security tour is the password. Many people keep their valuable information and files on a computer, and the only thing preventing others from seeing it is the eight-character string called a password. An unbreakable password, contrary to popular belief, does not exist. Given time and resources all passwords can be guessed either by social engineering or by brute force. Social engineering of server passwords and other access methods are still the easiest and most popular way to gain access to accounts and servers. Often, something as simple as acting as a superior or executive in a company and yelling at the right person at the right time of the day yields terrific results. Running a password cracker on a weekly basis on your system is a good idea. This helps to find and replace passwords that are easily guessed or weak. Also, a password checking mechanism should be present to reject a weak password when choosing an initial password or changing an old one. Character strings that are plain dictionary words, or are all in the same case, or do not contain numbers or special characters should not be accepted as a new password. We recommend the following rules to make passwords effective: They should be at least six characters in length, preferably eight characters including at least one numeral or special character. They must not be trivial; a trivial password is one that is easy to guess and is usually based on the user’s name, family, occupation or some other personal characteristic. They should have an aging period, requiring a new password to be chosen within a specific time frame. They should be revoked and reset after a limited number of concurrent incorrect retries.
- 77. General Security 0 CHAPTER3 77 The root account The "root" account is the most privileged account on a Unix system. The "root" account has no security restrictions imposed upon it. This means the system assumes you know what you are doing, and will do exactly what you request -- no questions asked. Therefore it is easy, with a mistyped command, to wipe out crucial system files. When using this account it is important to be as careful as possible. For security reasons, never log in on your server as "root" unless it is absolutely an instance that necessitates root access. Also, if you are not on your server, never sign in and leave yourself on as "root"--this is VERY, VERY. VERY BAD. Set login time out for the root account Despite the notice to never, if they are not on the server, sign in as “root” and leave it unattended, administrators still stay on as “root” or forget to logout after finishing their work and leave their terminals unattended. Step 1 The answer to solve this problem is to make the bash shell automatically logout after not being used for a period of time. To do that, you must set the special variable of Linux named “TMOUT” to the time in seconds of no input before logout. • Edit your profile file (vi /etc/profile) and add the following line somewhere after the line that read “HISTSIZE=” on this file: HOSTNAME=`/bin/hostname` HISTSIZE=1000 TMOUT=7200 The value we enter for the variable “TMOUT=” is in seconds and represents 2 hours (60 * 60 = 3600 * 2 = 7200 seconds). It is important to note that if you decide to put the above line in your /etc/profile file, then the automatic logout after two hours of inactivity will apply for all users on the system. So, instead, if you prefer to control which users will be automatically logged out and which ones are not, you can set this variable in their individual .bashrc file. Step 2 Once we have added the above line to the profile file, we must add its definition to the export line of the same file as follow. • Edit your profile file (vi /etc/profile) and change the line: export PATH USER LOGNAME MAIL HOSTNAME HISTSIZE INPUTRC To read: export PATH USER LOGNAME MAIL HOSTNAME HISTSIZE TMOUT INPUTRC After this parameter has been set on your system, you must logout and login again (as root) for the change to take effect.
- 78. 78 Shell logging To makeit easy for you to repeat long commands, the bash shell stores up to 500 old commands in the ~/.bash_history file (where “~/” is your home directory). Each user that has an account on the system will have this file .bash_history in their home directory. Reducing the number of old commands the .bash_history files can hold may protect users on the server who enter by mistake their password on the screen in plain text and have their password stored for a long time in the .bash_history file. Step 1 The HISTSIZE line in the /etc/profile file determine the size of old commands the .bash_history file for all users on your system can hold. For all accounts I would highly recommend setting the HISTSIZE in /etc/profile file to a low value such as 10. • Edit the profile file (vi /etc/profile) and change the line: HISTSIZE=1000 To read: HISTSIZE=10 This means, the .bash_history file in each user’s home directory can store 10 old commands and no more. Now, if a cracker tries to see the ~/.bash_history file of users on your server to find some password typed by mistake in plain text, he or she has less chance to find one. Step 2 The administrator should also add into the /etc/profile file the “HISTFILESIZE=0” line, so that each time a user logs out, its .bash_history file will be deleted so crackers will not be able to use .bash_history file of users who are not presently logged into the system. • Edit the profile file (vi /etc/profile) and add the following parameter below the “HISTSIZE=” line: HISTFILESIZE=0 Step 3 Once we have added the above line to the profile file, we must add its definition to the export line of the same file as follow. • Edit your profile file (vi /etc/profile) and change the line: export PATH USER LOGNAME MAIL HOSTNAME HISTSIZE TMOUT INPUTRC To read: export PATH USER LOGNAME MAIL HOSTNAME HISTSIZE HISTFILESIZE TMOUT INPUTRC After this parameter has been set on your system, you must logout and login again (as root) for the change to take effect.
- 79. General Security 0 CHAPTER3 79 The single-user login mode of Linux This part applies for those who use LILO as their boot loader. Linux has a special command (linux single) also known as ‘single-user mode’, which can be entered at the boot prompt during startup of the system. The single-user mode is generally used for system maintenance. You can boot Linux in single-user mode by typing at the LILO boot prompt the command: LILO: linux single This will place the system in Run level 1 where you'll be logged in as the super-user 'root', and where you won't even have to type in a password! Step 1 Requiring no password to boot into root under single-user mode is a bad idea! You can fix this by editing the inittab file (vi /etc/inittab) and change the following line: id:3:initdefault: To read: id:3:initdefault: ~~:S:wait:/sbin/sulogin The addition of the above line will require entering the root password before continuing to boot into single-user mode by making init (8) run the program sulogin (8) before dropping the machine into a root shell for maintenance. Step 2 Now, we have to restart the process control initialization of the server for the changes to take effect. • This can be done with the following command: [root@deep /]# /sbin/init q Disabling Ctrl-Alt-Delete keyboard shutdown command Commenting out the line (with a “#”) listed below in your /etc/inittab file will disable the possibility of using the Control-Alt-Delete command to shutdown your computer. This is pretty important if you don't have the best physical security to the machine. Step 1 • To do this, edit the inittab file (vi /etc/inittab) and change/comment the line: ca::ctrlaltdel:/sbin/shutdown -t3 -r now To read: #ca::ctrlaltdel:/sbin/shutdown -t3 -r now
- 80. 80 Step 2 Now, wehave to restart the process control initialization of the server for the changes to take effect. • This can be done with the following command: [root@deep /]# /sbin/init q Limiting the default number of started ttys on the server Default installed Linux system comes with six virtual consoles in standard run levels. This means that six mingetty processes will always by available at every time on the server. These virtual consoles allow you to login with six different virtual consoles on the system. Step 1 On secure server, we can limit the number to two virtual consoles and save some resources which may be used for other work by the server when required. • To do this, edit the inittab file (vi /etc/inittab) and remove/comment the lines: 1:2345:respawn:/sbin/mingetty tty1 2:2345:respawn:/sbin/mingetty tty2 3:2345:respawn:/sbin/mingetty tty3 4:2345:respawn:/sbin/mingetty tty4 5:2345:respawn:/sbin/mingetty tty5 6:2345:respawn:/sbin/mingetty tty6 To read: 1:2345:respawn:/sbin/mingetty tty1 2:2345:respawn:/sbin/mingetty tty2 #3:2345:respawn:/sbin/mingetty tty3 #4:2345:respawn:/sbin/mingetty tty4 #5:2345:respawn:/sbin/mingetty tty5 #6:2345:respawn:/sbin/mingetty tty6 Step 2 Now, we have to restart the process control initialization of the server for the changes to take effect. • This can be done with the following command: [root@deep /]# /sbin/init q The LILO and /etc/lilo.conf file This part applies for those who use LILO as their boot loader. LILO is a boot loader that can be used to manage the boot process, boot Linux kernel images from floppy disks, hard disks or can even act as a "boot manager" for other operating systems. LILO is very important in the Linux system and for this reason, we must protect it the best we can. The most important configuration file of LILO is the lilo.conf file. It is with this file that we can configure and improve the security of our LILO program and Linux system. Following are three important options that will improve the security of our valuable LILO program.
- 81. General Security 0 CHAPTER3 81 • Adding: timeout=00 This option controls how long (in seconds) LILO waits for user input before booting to the default selection. One of the requirements of C2 security is that this interval be set to 0 unless the system dual boots something else. • Adding: restricted This option asks for a password only, if parameters are specified on the command line (e.g. linux single). The option “restricted” can only be used together with the “password” option. Make sure you use this one on each additional image you may have. • Adding: password=<password> This option asks the user for a password when trying to load the image. Actually the effect of using the password parameter in /etc/lilo.conf will protect the Linux image from booting. This means, it doesn't matter if you load Linux in single mode or if you just do a normal boot. It will always ask you for the password. Now this can have a very bad effect, namely you are not able to reboot Linux remotely any more since it won't come up until you type in the root password at the console. It is for this reason that adding “restricted” with “password” is very important since the option "restricted" relaxes the password protection and a password is required only if parameters are specified at the LILO prompt, (e.g. single). Passwords are always case-sensitive, also make sure the /etc/lilo.conf file is no longer world readable, or any user will be able to read the password. Here is an example of our protected LILO with the lilo.conf file. Step 1 • Edit the lilo.conf file (vi /etc/lilo.conf) and add or change the three options above as show: boot=/dev/sda map=/boot/map install=/boot/boot.b prompt remove this line if you don’t want to pass options at the LILO prompt. timeout=00 change this line to 00 to disable the LILO prompt. linear message=/boot/message remove this line if you don’t want the welcome screen. default=linux restricted add this line to relaxes the password protection. password=<password> add this line and put your password. image=/boot/vmlinuz-2.4.2-2 label=linux initrd=/boot/initrd-2.4.2-2.img read-only root=/dev/sda6 Step 2 Because the configuration file /etc/lilo.conf now contains unencrypted passwords, it should only be readable for the super-user “root”. • To make the /etc/lilo.conf file readable only by the super-user “root”, use the following command: [root@deep /]# chmod 600 /etc/lilo.conf (will be no longer world readable).
- 82. 82 Step 3 Now wemust update our configuration file /etc/lilo.conf for the change to take effect. • To update the /etc/lilo.conf file, use the following command: [root@deep /]# /sbin/lilo -v LILO version 21.4-4, copyright © 1992-1998 Wernerr Almesberger ‘lba32’ extentions copyright © 1999,2000 John Coffman Reading boot sector from /dev/sda had : ATAPI 32X CD-ROM drive, 128kB Cache Merging with /boot/boot.b Mapping message file /boot/message Boot image : /boot/vmlinuz-2.2.16-22 Mapping RAM disk /boot/initrd-2.2.16-22.img Added linux * /boot/boot.0800 exists – no backup copy made. Writing boot sector. Step 4 One more security measure you can take to secure the lilo.conf file is to set it immutable, using the chattr command. • To set the file immutable simply, use the following command: [root@deep /]# chattr +i /etc/lilo.conf And this will prevent any changes (accidental or otherwise) to the lilo.conf file. If you wish to modify the lilo.conf file you will need to unset the immutable flag: • To unset the immutable flag, use the following command: [root@deep /]# chattr -i /etc/lilo.conf WARNING: When you use the password option, then LILO will always ask you for the password, regardless if you pass options at the LILO prompt (e.g. single) or not EXCEPT when you set the "restricted" option in /etc/lilo.conf. The option "restricted" relaxes the password protection and a password is required only if parameters are specified at the LILO prompt, (e.g. single). If you didn't had this option set "restricted", Linux will always ask you for the password and you will not be able to remotely reboot your system, therefore don’t forget to add the option "restricted” with the option "password" into the /etc/lilo.conf file. The GRUB and /boot/grub/grub.conf file This part applies for those who use GRUB as their boot loader. GRUB is another boot loader like LILO but with many more useful feature and power. One of its main advantages compared to LILO is the fact that it provides a small shell interface to manage the operating system. Also, it doesn’t need to be updated each time you recompile a new kernel on your server.
- 83. General Security 0 CHAPTER3 83 GRUB is very important since it is the first software program that runs when the computer starts and we have to secure it as much as possible to avoid any possible problem. In its default installation it’s already well protected and below I explain how its configuration file is made. In regard to LILO, GRUB is really easy to use and configure. Below is a default GRUB configuration file and security I recommend you to apply. The text in bold are the parts of the configuration file that must be customized and adjusted to satisfy our needs. • Edit the grub.conf file (vi /boot/grub/grub.conf) and set your needs. Below is what we recommend you: default=0 timeout=0 splashimage=(hd0,0)/grub/splash.xpm.gz password --md5 $1$oKr0ÝmFo$tPYwkkvQbtqo1erwHj5wb/ title Red Hat Linux (2.4.18-3) root (hd0,0) kernel /vmlinuz-2.4.18-3 ro root=/dev/sda5 initrd /initrd-2.4.18-3.img This tells the grub.conf file to set itself up for this particular configuration with: default=0 The option “default” is used to define the default entry of the configuration file. The number “0” mean that the following parameters are the default entry for the configuration of GRUB. In a server configuration where Linux is the only operating system to boot, the default entry of “0” will be the only one to use and we don’t need to define any additional entry. timeout=0 This option “timeout” is used to define the timeout, in sec seconds to wait, before automatically booting the default entry. As for LILO, one of the requirements of C2 security is that this interval be set to 0 unless the system dual boots something else. One of the disadvantages to set this option to “0” is that you will no longer be able to have access at boot time to the shell interface of the software but this is not really a problem since all we need from the GRUB software is to boot our operating system. splashimage=(hd0,0)/grub/splash.xpm.gz This option “splashimage” is an option added by Red Hat to boot the system with a graphical image. The value is the path of the compressed image to use when booting GRUB. It’s to you to keep this parameter on your system or to remove it. If you want to remove it, just delete the above line with the compressed image from your server. password --md5 $1$bgGCL/$4yF3t0py.IjU0LU.q7YfB1 This option “password” is used to inform GRUB to ask for a password and disallows any interactive control, until you press the key <p> and enter a correct password. The option --md5 tells GRUB that a password in MD5 format is required as a value. If it is omitted, GRUB assumes the specified password is in clear text. When we have installed the operating system, we have already configured GRUB with a password protection. This password is what you see here. If you want to change it, you have to use the “grub-md5-crypt” command to generate a new encrypt password it in MD5 format. • This can be done with the following command: [root@dev /]# grub-md5-crypt Password: $1$bgGCL/$4yF3t0py.IjU0LU.q7YfB1
- 84. 84 Once the abovecommand has been issued, you have to cut and paste the encrypted password to your configuration file. title Red Hat Linux (2.4.18-3) This option “title” is used to define a name to the contents of the rest of the line. It is directly related to the default boot entry. What you enter here is what you will see during boot time. This option is useful when we have more than one OS to start on our computer and allow us to give the name that we want to distinguish them. You are free to enter whatever name you like. root (hd0,0) This option “root” is one of the most important parameter with GRUB and without it nothing will work. It is used to define the current root device to use for booting the operating system. Its definition and configuration is a little bit special as you can see. Here is an explanation of its meaning. The “hd0” parameter represents using the entire disk and the “hd0,0” represents using the partition of the disk (or the boot sector of the partition when installing GRUB). Don’t be confused here because “hd” is valid for IDE and SCSI drives. There is no difference; you always use “hd” even on SCSI drive. kernel /vmlinuz-2.4.18-3 ro root=/dev/sda5 This option “kernel” is used to load the primary boot image (our kernel). The parameter to this option is simply the path where GRUB should find the kernel image from which we want it to boot. The additional lines are to inform it that kernel image is located on the sda5 partition on our server and that we want to load it as read only for security reason. initrd /initrd-2.4.18-3.img This option “initrd” is optional and will appear into your GRUB configuration file only if you run a SCSI computer. For IDE computer, this option is not required and should not be defined inside the configuration file of GRUB. The parameter simply informs GRUB software where our initial ram disk image is located on the server. GRUB reads this initial ram disk and loads it during startup. The /etc/services file The port numbers on which certain "standard" services are offered are defined in the RFC 1700 "Assigned Numbers". The /etc/services file enables server and client programs to convert service names to these numbers (ports). The list is kept on each host and it is stored in the file /etc/services. Only the "root" user is allowed to make modifications to this file. It is rare to edit the /etc/services file since it already contains the more common service names / port numbers. To improve security, we can set the immutable flag on this file to prevent unauthorized deletion or modification. • To immunize the /etc/services file, use the following command: [root@deep /]# chattr +i /etc/services
- 85. General Security 0 CHAPTER3 85 The /etc/securetty file The /etc/securetty file allows you to specify which TTY and VC (virtual console) devices the “root” user is allowed to login on. The /etc/securetty file is read by the login program (usually /bin/login). Its format is a list of the tty and vc devices names allowed, and for all others that are commented out or do not appear in this file, root login is disallowed. Disable any tty and vc devices that you do not need by commenting them out (# at the beginning of the line) or by removing them. • Edit the securetty file (vi /etc/securetty) and comment out or remove the lines: vc/1 #vc/2 #vc/3 #vc/4 #vc/5 #vc/6 #vc/7 #vc/8 #vc/9 #vc/10 #vc/11 tty1 #tty2 #tty3 #tty4 #tty5 #tty6 #tty7 #tty8 #tty9 #tty10 #tty11 Which means root is allowed to login on only tty1 and vc/1. This is my recommendation, allowing “root” to log in on only one tty or vc device and use the su or sudo command to switch to “root” if you need more devices to log in as “root”. Special accounts It is important to DISABLE ALL default vendor accounts that you don’t use on your system (some accounts exist by default even if you have not installed the related services on your server). This should be checked after each upgrade or new software installation. Linux provides these accounts for various system activities, which you may not need if the services are not installed on your server. If you do not need the accounts, remove them. The more accounts you have, the easier it is to access your system. We assume that you are using the shadow password suite on your Linux system. If you are not, you should consider doing so, as it helps to tighten up security somewhat. This is already set if you’ve followed our Linux installation procedure and selected, under the “Authentication Configuration”, the option to “Enable Shadow Passwords” (see the chapter related to the “Installation of your Linux Server” for more information). • To delete user on your system, use the following command: [root@deep /]# userdel username • To delete group on your system, use the following command: [root@deep /]# groupdel username
- 86. 86 Step 1 First wewill remove all default vendor accounts into the /etc/passwd file that are unnecessary for the operation of the secure server configuration that we use in this book. • Type the following commands to delete all default users accounts listed below. [root@deep /]# userdel adm [root@deep /]# userdel lp [root@deep /]# userdel shutdown [root@deep /]# userdel halt [root@deep /]# userdel news [root@deep /]# userdel mailnull [root@deep /]# userdel operator [root@deep /]# userdel games [root@deep /]# userdel gopher [root@deep /]# userdel ftp [root@deep /]# userdel vcsa WARNING: By default, the userdel command will not delete a user’s home directory. If you want the home directories of accounts to be deleted too, then add the -r option to the userdel command. Finally, the -r option must be used only when you have added a new user to the server and want to remove them. It doesn’t need to be used for the removal of the above default user’s accounts since they do not have a home directory. Once the above list of users has been deleted from your Linux system, the /etc/passwd file will look like this: root:x:0:0:root:/root:/bin/bash bin:x:1:1:bin:/bin:/sbin/nologin daemon:x:2:2:daemon:/sbin:/sbin/nologin sync:x:5:0:sync:/sbin:/bin/sync mail:x:8:12:mail:/var/spool/mail:/sbin/nologin uucp:x:10:14:uucp:/var/spool/mail:/sbin/nologin nobody:x:99:99:Nobody:/:/sbin/nologin rpm:x:37:37::/var/lib/rpm:/bin/bash Step 2 After that we have removed all the unnecessary default vendor accounts into the /etc/passwd file from our system, we will remove all default vendor accounts into the /etc/group file. • Type the following commands to delete all default users groups accounts listed below. [root@deep /]# groupdel adm [root@deep /]# groupdel lp [root@deep /]# groupdel news [root@deep /]# groupdel games [root@deep /]# groupdel dip
- 87. General Security 0 CHAPTER3 87 Once the above list of group users has been deleted from your Linux system the /etc/group file will look like this: root:x:0:root bin:x:1:root,bin,daemon daemon:x:2:root,bin,daemon sys:x:3:root,bin tty:x:5: disk:x:6:root mem:x:8: kmem:x:9: wheel:x:10:root mail:x:12:mail uucp:x:14:uucp man:x:15: lock:x:54: nobody:x:99: users:x:100: slocate:x:21: floppy:x:19: utmp:x:22: rpm:x:37: Step 3 Now, you can add all the necessary and allowed users into the system. Below I show you how you should add new user into your Linux server. Adding a new user into your server mean that you have to create the username and assign him/her a password. • To add a new user on your system, use the following command: [root@deep /]# useradd username For example: [root@deep /]# useradd admin • To add or change password for user on your system, use the following command: [root@deep /]# passwd username For example: [root@deep /]# passwd admin The output should look something like this: Changing password for user admin New UNIX password: Retype new UNIX password: passwd: all authentication tokens updated successfully Step 4 The immutable bit can be used to prevent accidentally deleting or overwriting a file that must be protected. It also prevents someone from creating a symbolic link to this file, which has been the source of attacks involving the deletion of /etc/passwd, /etc/shadow, /etc/group or /etc/gshadow files. • To set the immutable bit on the passwords and groups files, use the following commands: [root@deep /]# chattr +i /etc/passwd [root@deep /]# chattr +i /etc/shadow [root@deep /]# chattr +i /etc/group
- 88. 88 [root@deep /]# chattr+i /etc/gshadow WARNING: In the future, if you intend to add or delete users, passwords, user groups, or group files, you must unset the immutable bit on all those files or you will not be able to make and update your changes. Also if you intend to install an RPM program that will automatically add a new user to the different immunized passwd and group files, then you will receive an error message during the install if you have not unset the immutable bit from those files. • To unset the immutable bit on the passwords and groups files, use the commands: [root@deep /]# chattr -i /etc/passwd [root@deep /]# chattr -i /etc/shadow [root@deep /]# chattr -i /etc/group [root@deep /]# chattr -i /etc/gshadow Control mounting a file system You can have more control on mounting file systems like /var/lib, /home or /tmp partitions with some nifty options like noexec, nodev, and nosuid. This can be setup in the /etc/fstab file. The fstab file contains descriptive information about the various file system mount options; each line addresses one file system. Information related to security options in the fstab text file are: defaults Allow everything (quota, read-write, and suid) on this partition. noquota Do not set users quotas on this partition. nosuid Do not set SUID/SGID access on this partition. nodev Do not set character or special devices access on this partition. noexec Do not set execution of any binaries on this partition. quota Allow users quotas on this partition. ro Allow read-only on this partition. rw Allow read-write on this partition. suid Allow SUID/SGID access on this partition. NOTE: For more information on options that you can set in this file (fstab) see the man pages about mount (8). Step 1 • Edit the fstab file (vi /etc/fstab) and change it depending of your needs. For example change: LABEL=/home /home ext3 defaults 1 2 LABEL=/tmp /tmp ext3 defaults 1 2 LABEL=/var/lib /var/lib ext3 defaults 1 2 To read: LABEL=/home /home ext3 defaults,nosuid 1 2 LABEL=/tmp /tmp ext3 defaults,nosuid,noexec 1 2 LABEL=/var/lib /var/lib ext3 defaults,nodev 1 2
- 89. General Security 0 CHAPTER3 89 Meaning, <nosuid>, do not allow set-user-identifier or set-group-identifier bits to take effect, <nodev>, do not interpret character or block special devices on this file system partition, and <noexec>, do not allow execution of any binaries on the mounted file system. Step 2 Once you have made the necessary adjustments to the /etc/fstab file, it is time to inform the Linux system about the modifications. • This can be accomplished with the following commands: [root@deep /]# mount /var/lib -oremount [root@deep /]# mount /home -oremount [root@deep /]# mount /tmp -oremount Each file system that has been modified must be remounted with the command show above. In our example we have modified the /var/lib, /home, and /tmp file system and it is for this reason that we remount these files systems with the above commands. • You can verify if the modifications have been correctly applied to the Linux system with the following command: [root@deep /]# cat /proc/mounts rootfs / rootfs rw 0 0 /dev/root / ext3 rw 0 0 /proc /proc proc rw 0 0 /dev/sda1 /boot ext3 rw 0 0 /dev/sda8 /chroot ext3 rw 0 0 none /dev/pts devpts rw 0 0 /dev/sda7 /home ext3 rw,nosuid 0 0 none /dev/shm tmpfs rw 0 0 /dev/sda11 /tmp ext3 rw,nosuid,noexec 0 0 /dev/sda6 /usr ext3 rw 0 0 /dev/sda9 /var ext3 rw 0 0 /dev/sda10 /var/lib ext3 rw,nodev 0 0 This command will show you all the files systems on your Linux server with parameters applied to them. Mounting the /usr directory of Linux as read-only It is allowable to mount your /usr partition read-only since this is, by definition, static data. Of course, anyone with root access can remount it as writable, but a generic attack script may not know this. On many Linux variants this directory resides in its own partition and the default parameter is to mount it as read-write. We can change this parameter to make it read-only for better security. Please be sure that your /usr directory is on its own partition or the following hack will not work for you.
- 90. 90 Step 1 Mounting the/usr partition as read-only eliminates possible problems that someone may try to change or modify vital files inside it. To mount the /usr file system of Linux as read-only, follow the simple steps below. • Edit the fstab file (vi /etc/fstab) and change the line: LABEL=/usr /usr ext3 defaults 1 2 To read: LABEL=/usr /usr ext3 defaults,ro 1 2 We add the “ro” option to this line to specify to mount this partition as read-only. Step 2 Make the Linux system aware about the modification you have made to the /etc/fstab file. • This can be accomplished with the following command: [root@deep /]# mount /usr -oremount • Then test your results with the following command: [root@deep /]# cat /proc/mounts rootfs / rootfs rw 0 0 /dev/root / ext3 rw 0 0 /proc /proc proc rw 0 0 /dev/sda1 /boot ext3 rw 0 0 /dev/sda8 /chroot ext3 rw 0 0 none /dev/pts devpts rw 0 0 /dev/sda7 /home ext3 rw,nosuid 0 0 none /dev/shm tmpfs rw 0 0 /dev/sda11 /tmp ext3 rw,nosuid,noexec 0 0 /dev/sda6 /usr ext3 ro 0 0 /dev/sda9 /var ext3 rw 0 0 /dev/sda10 /var/lib ext3 rw,nodev 0 0 If you see something like: /dev/sda6 /usr ext3 ro 0 0, congratulations! WARNING: If in the future you want to install some RPM package or program from source code, it is important to reset the modification you have made to the /usr directory to its initial state (read- write) or you will not be able to install new software because the /usr partition is set as read- only. All you have to do if you want to put the /usr partition to its original state is to edit the /etc/fstab file again and remove the “ro” option then remount the /usr file system with the “mount -oremount” command again.
- 91. General Security 0 CHAPTER3 91 Tighten scripts under /etc/init.d Fix the permissions of the script files that are responsible for starting and stopping all your normal processes that need to run at boot time. • To fix the permissions of those files, use the following command: [root@deep /]# chmod 0700 /etc/init.d/* Which means just the super-user “root” is allowed to Read, Write, and Execute scripts files on this directory. I don’t think regular users need to know what’s inside those script files. WARNING: If you install a new program or update a program that use the init system V script located under /etc/init.d/ directory, don’t forget to change or verify the permission of this script file again. Tighten scripts under /etc/cron.daily/ As for the above hack, we can tighten the security of all script files that are responsible for executing scheduled job on our server. Those files have a default permission mode of (0755- rwxr-xr-x), which is too high for what they should accomplish. • To fix the permissions of those files, use the following command: [root@deep /]# chmod 0550 /etc/cron.daily/* The same is true for other cron directories under the /etc directory of your system. If files exist under the other cron directories, then use the above command to change their default permission mode for better security. WARNING: If you install a new program or update a program that provides and install a cron file on your server, don’t forget to change or verify the permission of this script file again. Bits from root-owned programs A regular user will be able to run a program as root if it is set to SUID root. All programs and files on your computer with the ’s’ bits appearing on its mode, have the SUID (-rwsr-xr-x) or SGID (-r-xr-sr-x) bit enabled. Because these programs grant special privileges to the user who is executing them, it is important to remove the 's' bits from root-owned programs that won't absolutely require such privilege. This can be accomplished by executing the command chmod a-s with the name(s) of the SUID/SGID files as its arguments. Such programs include, but aren't limited to: Programs you never use. Programs that you don't want any non-root users to run. Programs you use occasionally, and don't mind having to su (1) to root to run.
- 92. 92 Step 1 We've placedan asterisk (*) next to each program we personally might disable and consider being not absolutely required for the duty work of the server. Remember that your system needs some suid root programs to work properly, so be careful. • To find all files with the ‘s’ bits from root-owned programs, use the command: [root@deep]# find / -type f ( -perm -04000 -o -perm -02000 ) -exec ls -l {} ; *-rwsr-xr-x 1 root root 34296 Mar 27 20:40 /usr/bin/chage *-rwsr-xr-x 1 root root 36100 Mar 27 20:40 /usr/bin/gpasswd -rwxr-sr-x 1 root slocate 25020 Jun 25 2001 /usr/bin/slocate -r-s--x--x 1 root root 15104 Mar 13 20:44 /usr/bin/passwd *-r-xr-sr-x 1 root tty 6920 Mar 14 15:24 /usr/bin/wall *-rws--x--x 1 root root 12072 Apr 1 18:26 /usr/bin/chfn *-rws--x--x 1 root root 11496 Apr 1 18:26 /usr/bin/chsh *-rws--x--x 1 root root 4764 Apr 1 18:26 /usr/bin/newgrp *-rwxr-sr-x 1 root tty 8584 Apr 1 18:26 /usr/bin/write -rwsr-xr-x 1 root root 21080 Apr 15 00:49 /usr/bin/crontab *-rwsr-xr-x 1 root root 32673 Apr 18 17:40 /usr/sbin/ping6 *-rwsr-xr-x 1 root root 13994 Apr 18 17:40 /usr/sbin/traceroute6 -rwxr-sr-x 1 root utmp 6604 Jun 24 2001 /usr/sbin/utempter -rws--x--x 1 root root 22388 Apr 15 18:15 /usr/sbin/userhelper *-rwsr-xr-x 1 root root 17461 Apr 19 12:35 /usr/sbin/usernetctl *-rwsr-xr-x 1 root root 35192 Apr 18 17:40 /bin/ping *-rwsr-xr-x 1 root root 60104 Apr 1 18:26 /bin/mount *-rwsr-xr-x 1 root root 30664 Apr 1 18:26 /bin/umount -rwsr-xr-x 1 root root 19116 Apr 8 12:02 /bin/su -r-sr-xr-x 1 root root 120264 Apr 9 23:24 /sbin/pwdb_chkpwd -r-sr-xr-x 1 root root 16992 Apr 9 23:24 /sbin/unix_chkpwd *-rwxr-sr-x 1 root root 14657 Apr 19 12:35 /sbin/netreport Step 2 • To disable the suid bits on selected programs above, use the following commands: [root@deep /]# chmod a-s /usr/bin/chage [root@deep /]# chmod a-s /usr/bin/gpasswd [root@deep /]# chmod a-s /usr/bin/wall [root@deep /]# chmod a-s /usr/bin/chfn [root@deep /]# chmod a-s /usr/bin/chsh [root@deep /]# chmod a-s /usr/bin/newgrp [root@deep /]# chmod a-s /usr/bin/write [root@deep /]# chmod a-s /usr/sbin/ping6 [root@deep /]# chmod a-s /usr/sbin/traceroute6 [root@deep /]# chmod a-s /usr/sbin/usernetctl [root@deep /]# chmod a-s /bin/ping [root@deep /]# chmod a-s /bin/mount [root@deep /]# chmod a-s /bin/umount [root@deep /]# chmod a-s /sbin/netreport
- 93. General Security 0 CHAPTER3 93 Don’t let internal machines tell the server what their MAC address is To avoid the risk that a user could easily change a computers IP address and appear as someone else to the firewall, you can force the ARP cache entries of Linux using the arp command utility. A special option can be used with the arp utility to avoid letting INTERNAL machines tell the server what their MAC (Media Access Control) address is and the IP address associated with it. Step1 ARP is a small utility, which manipulates the kernel’s ARP (Address Resolution Protocol) cache. Through all possible options associated with this utility, the primary one is clearing an address mapping entry and manually setting up one. In the hope to more secure our server from the INTERNAL, we will manually set MAC address (sometimes called Hardware addresses) of all known computers in our network statically by using static ARP entries. • For each IP address of INTERNAL computers in your network, use the following command to know the MAC address associate with the IP address: [root@deep /]# ifconfig eth0 Link encap:Ethernet HWaddr 00:50:DA:C6:D3:FF inet addr:207.35.78.3 Bcast:207.35.78.32 Mask:255.255.255.224 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1887318 errors:0 dropped:0 overruns:1 frame:0 TX packets:2709329 errors:0 dropped:0 overruns:0 carrier:1 collisions:18685 txqueuelen:100 Interrupt:10 Base address:0xb000 eth1 Link encap:Ethernet HWaddr 00:50:DA:C6:D3:09 inet addr:192.168.1.11 Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:182937 errors:0 dropped:0 overruns:0 frame:0 TX packets:179612 errors:0 dropped:0 overruns:0 carrier:0 collisions:7434 txqueuelen:100 Interrupt:11 Base address:0xa800 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:3924 Metric:1 RX packets:7465 errors:0 dropped:0 overruns:0 frame:0 TX packets:7465 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 The MAC (Media Access Control) address will be the letters and numbers that come after “HWaddr” (the Hardware Address). In the above example our MAC address are: 00:50:DA:C6:D3:FF for the interface eth0 and 00:50:DA:C6:D3:09 for the interface eth1. Step 2 Once we know the MAC (Media Access Control) address associated with IP address, we can add them manually to the ARP entries of the Linux server. • To add manually MAC address to ARP entries, use the following commands: [root@deep /]# arp -s 207.35.78.3 00:50:DA:C6:D3:FF [root@deep /]# arp -s 192.168.1.11 00:50:DA:C6:D3:09
- 94. 94 The “-s” optionmeans to manually create an ARP address mapping entry for host hostname with hardware address set to hw_addr class. You can add you ARP commands to the /etc/rc.local file if you want to keep your configuration if the system reboots. Step 3 • To verify if the modifications have been added to the system, use the following command: [root@deep /]# arp Address Hwtype Hwaddress Flags Mask Iface 207.35.78.3 ether 00:20:78:13:86:92 CM eth1 192.168.1.11 ether 00:E0:18:90:1B:56 CM eth1 WARNING: If you receive error message like: SIOCSARP: Invalid argument, it is because the MAC (Media Access Control) address you want to add is the one of your server. You must add only MAC address of INTERNAL computers in your private network. This hack doesn’t apply to external node on the Internet. You can now be reassured that someone will not change the system's IP address of an INTERNAL system and get through. If they do change the IP address, the server simply won't talk to them. With the new iptables tool of Linux, which replace the old ipchains utility for packet filter administration and firewall setup, MAC addresses can be filtered and configured in the firewall rules too. Unusual or hidden files It is important to look everywhere on the system for unusual or hidden files (files that start with a period and are normally not shown by the “ls” command), as these can be used to hide tools and information (password cracking programs, password files from other systems, etc.). A common technique on UNIX systems is to put a hidden directory or file in a user's account with an unusual name, something like '...' or '.. ' (dot dot space) or '..^G' (dot dot control-G). The find program can be used to look for hidden files. • To look for hidden files, use the following commands: [root@deep /]# find / -name ".. " -print -xdev [root@deep /]# find / -name ".*" -print -xdev | cat –v /etc/skel/.bash_logout /etc/skel/.bash_profile /etc/skel/.bashrc /etc/.pwd.lock /root/.bash_logout /root/.Xresources /root/.bash_profile /root/.bashrc /root/.cshrc /root/.tcshrc /root/.bash_history /usr/lib/perl5/5.6.1/i386-linux/.packlist /home/admin/.bash_logout /home/admin/.bash_profile /home/admin/.bashrc /.autofsck
- 95. General Security 0 CHAPTER3 95 Finding Group and World Writable files and directories Group and world writable files and directories, particularly system files (partitions), can be a security hole if a cracker gains access to your system and modifies them. Additionally, world- writable directories are dangerous, since they allow a cracker to add or delete files as he or she wishes in these directories. In the normal course of operation, several files will be writable, including some from the /dev, /var/mail directories, and all symbolic links on your system. • To locate all group & world-writable files on your system, use the command: [root@deep /]# find / -type f ( -perm -2 -o -perm -20 ) -exec ls -lg {} ; -rw-rw-r-- 1 root utmp 107904 Jun 17 12:04 /var/log/wtmp -rw-rw-r-- 1 root utmp 4608 Jun 17 12:04 /var/run/utmp • To locate all group & world-writable directories on your system, use the command: [root@deep /]# find / -type d ( -perm -2 -o -perm -20 ) -exec ls -ldg {} ; drwxrwxr-x 12 root man 4096 Jun 17 06:50 /var/cache/man/X11R6 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat1 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat2 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat3 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat4 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat5 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat6 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat7 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat8 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/cat9 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/X11R6/catn drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat1 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat2 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat3 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat4 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat5 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat6 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat7 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat8 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/cat9 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/catn drwxrwxr-x 12 root man 4096 Jun 17 06:50 /var/cache/man/local drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat1 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat2 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat3 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat4 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat5 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat6 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat7 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat8 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/cat9 drwxrwxr-x 2 root man 4096 Mar 25 09:17 /var/cache/man/local/catn drwxrwxr-x 3 root lock 4096 Jun 17 06:49 /var/lock drwxrwxr-x 2 root root 4096 Apr 19 12:35 /var/run/netreport drwxrwxr-x 2 root 12 4096 Jun 17 12:30 /var/spool/mail drwxrwxrwt 2 root root 4096 Jun 17 11:29 /var/tmp drwxrwxrwt 2 root root 4096 Jun 17 06:52 /tmp WARNING: A file and directory integrity checker like “Tripwire” software can be used regularly to scan, manage and find modified group or world writable files and directories easily. See later in this book for more information about Tripwire.
- 96. 96 Unowned files Don’t permitany unowned file on your server. Unowned files may also be an indication that an intruder has accessed your system. If you find unowned file or directory on your system, verify its integrity, and if all looks fine, give it an owner name. Some time you may uninstall a program and get an unowned file or directory related to this software; in this case you can remove the file or directory safely. • To locate files on your system that do not have an owner, use the following command: [root@deep /]# find / -nouser -o -nogroup WARNING: It is important to note that files reported under /dev/ directory don’t count. Finding .rhosts files Finding all existing .rhosts files that could exist on your server should be a part of your regular system administration duties, as these files should not be permitted on your system. Remember that a cracker only needs one insecure account to potentially gain access to your entire network. Step 1 If you are doing a new install of Linux (like we did) you should not have any .rhosts files on your system. If the result returns nothing, then you are safe and your system contain no .rhosts files in the /home directory at this time. • You can locate all existing .rhosts files on your system with the following command: [root@deep /]# find /home -name .rhosts Step 2 You can also use a cron job to periodically check for, report the contents of, and delete $HOME/.rhosts files. Also, users should be made aware that you regularly perform this type of audit, as directed by your security policy. • Create the rhosts.cron file (touch /etc/cron.daily/rhosts.cron) and add the following lines inside the script file. #!/bin/sh /usr/bin/find /home -name .rhosts | (cat <<EOF This is an automated report of possible existent ..rhosts files on the server deep.openna.com, generated by the find utility command. New detected ..rhosts. files under the ./home/. directory include: EOF cat ) | /bin/mail -s "Content of .rhosts file audit report" root • Now make this script executable, verify the owner, and change the group to “root”. [root@deep /]# chmod 550 /etc/cron.daily/rhosts.cron [root@deep /]# chown 0.0 /etc/cron.daily/rhosts.cron Each day mail will be sent to “root” with a subject:” Content of .rhosts file audit report” containing potential new .rhosts files.
- 97. General Security 0 CHAPTER3 97 Physical hard copies of all-important logs One of the most important security considerations is the integrity of the different log files under the /var/log directory on your server. If despite each of the security functions put in place on our server, a cracker can gain access to it; our last defence is the log file system, so it is very important to consider a method of being sure of the integrity of our log files. If you have a printer installed on your server, or on a machine on your network, a good idea would be to have actual physical hard copies of all-important logs. This can be easily accomplished by using a continuous feed printer and having the syslog program sending all logs you think are important out to /dev/lp0 (the printer device). Cracker can change the files, programs, etc on your server, but can do nothing when you have a printer that prints a real paper copy of all of your important logs. As an example: For logging of all telnet, mail, boot messages and ssh connections from your server to the printer attached to THIS server, you would want to add the following line to the /etc/syslog.conf file: Step 1 • Edit the syslog.conf file (vi /etc/syslog.conf) and add at the end of this file the following line: authpriv.*;mail.*;local7.*;auth.*;daemon.info /dev/lp0 Step 2 • Now restart your syslog daemon for the change to take effect: [root@deep /]# /etc/init.d/syslog restart Shutting down kernel logger: [OK] Shutting down system logger: [OK] Starting system logger: [OK] Starting kernel logger: [OK] As an example: For logging of all telnet, mail, boot messages and ssh connections from your server to the printer attached to a REMOTE server in your local network, then you would want to add the following line to /etc/syslog.conf file on the REMOTE server. Step 1 • Edit the syslog.conf file (vi /etc/syslog.conf) on the REMOTE server (for example: printer.openna.com) and add at the end of this file the following line: authpriv.*;mail.*;local7.*;auth.*;daemon.info /dev/lp0 If you don’t have a printer in your network, you can also copy all the log files to another machine; simply omit the above first step of adding /dev/lp0 to your syslog.conf file on remote and go directly to the “-r” option second step on remote. Using the feature of copying all the log files to another machine will give you the possibility to control all syslog messages on one host and will tear down administration needs.
- 98. 98 Step 2 Since thedefault configuration of the syslog daemon is to not receive any messages from the network, we must enable on the REMOTE server the facility to receive messages from the network. To enable the facility to receive messages from the network on the REMOTE server, add the following option “-r” to your syslog daemon script file (only on the REMOTE host): • Edit the syslog daemon (vi +24 /etc/rc.d/init.d/syslog) and change: daemon syslogd -m 0 To read: daemon syslogd -r -m 0 Step 3 • Restart your syslog daemon on the remote host for the change to take effect: [root@mail /]# /etc/init.d/syslog restart Shutting down kernel logger: [OK] Shutting down system logger: [OK] Starting system logger: [OK] Starting kernel logger: [OK] Step 4 • Edit the syslog.conf file (vi /etc/syslog.conf) on the LOCAL server, and add at the end of this file the following line: authpriv.*;mail.*;local7.*;auth.*;daemon.info @printer Where (printer) represent the hostname of the REMOTE server. Now if anyone ever hacks your machine and attempts to erase vital system logs, you still have a hard copy of everything. It should then be fairly simple to trace where they came from and deal with it accordingly. Step 5 • Restart your syslog daemon on the LOCAL server for the change to take effect: [root@deep /]# /etc/init.d/syslog restart Shutting down kernel logger: [OK] Shutting down system logger: [OK] Starting system logger: [OK] Starting kernel logger: [OK] WARNING: Never use your Gateway Server as a host to control all syslog messages; this is a very bad idea. More options and strategies exist with the sysklogd program, see the man pages about sysklogd (8), syslog(2), and syslog.conf(5) for more information.
- 99. General Security 0 CHAPTER3 99 Getting some more security by removing manual pages Here we have to think a little bit about manual pages installed on all Linux system. Manual pages also known as man-pages are compressed files located under the /usr/share/man directory on your system. These documentation files are very useful to get quick information on how service, program, commands, and configuration files of specific software work. These files are readable by the man program and depend of other installed software on Linux to work and display the information. On production servers where specific task are assigned and where we only run services to the internal or external, does we really need to have these manual pages and related software installed? Do we will connect to these production servers to read these manual pages? Does this is really important to have them duplicated on all of our different servers? Personally, I don’t think because we can have all of these useful documentation files available on our Linux workstation or development server each time we need to consult them. If you have made attention to what we have done previously to secure our server, you will remember that most of all group and world-writable directories on our system comes from the /var/cache directory which is owned by the man program associated with manual pages. By removing manual pages and related software from our server, we can get some more security and save some not negligible space which could help when we scan our server with integrity tool like Tripwire. This also allow us to remove other software directly related to man program and limit the number of installed component on our production server without scarifying in the functionally of the server. If this is what you want to do, here are the steps to follow. Step 1 First of all, we should remove the man software from our system. The man software is the program we use to read manual pages. By removing this software we eliminate most of all group and world-writable directories from our system. • To remove the man software, use the following command: [root@deep /]# rpm -e man Step 2 Once the above software has been removed, we can continue with groff. Groff is a document formatting system that takes standard text and formatting commands as input and produces formatted output. This software is used by man to format man-pages. • To remove the groff software, use the following command: [root@deep /]# rpm -e groff Step 3 Because we don’t use manual pages anymore on our production servers, we can remove all man-pages that are already installed and available under the /usr/share/man directory. • To remove all preinstalled man-pages from your server, use the following commands: [root@deep /]# cd /usr/share/man/ [root@deep man]# rm -f man*/*.gz
- 100. 100 Step 4 Finally, itis important to note that any future installation and upgrade of RPM packages on the system should be made with the “--excludedocs” option. This RPM option allow us to install or upgrade the RPM package without the need to install the documentation part that may comes with the software. For example, if I want to install or upgrade the bind package, I will use the following RPM command. • To install or upgrade RPM without documentation, use the following command: [root@deep /]# rpm –Uvh --exculdedocs bind-version.i386.rpm System is compromised! If you believe that your system has been compromised, contact CERT ® Coordination Center or your representative in FIRST (Forum of Incident Response and Security Teams). Internet Email: cert@cert.org CERT Hotline: (+1) 412-268-7090 Facsimile: (+1) 412-268-6989 CERT/CC personnel answer 8:00 a.m. – 8:00 p.m. EST (GMT –5)/EDT (GMT –4)) on working days; they are on call for emergencies during other hours and on weekends and holidays.
- 101. Pluggable Authentication Modules INTHIS CHAPTER 1. The password length 2. Disabling console program access 3. Disabling all console access 4. The Login access control table 5. Tighten console permissions for privileged users 6. Putting limits on resource 7. Controlling access time to services 8. Blocking; su to root, by one and sundry 9. Using sudo instead of su for logging as super-user
- 103. Pluggable Authentication Modules0 CHAPTER 4 103 Linux PAM Abstract The Pluggable Authentication Modules (PAM) consists of shared libraries, which enable administrators to choose how applications authenticate users. Basically, PAM enables the separation of authentication schemes from the applications. This is accomplished by providing a library of functions that applications can use for requesting user authentications. ssh, pop, imap, etc. are PAM-aware applications, hence these applications can be changed from providing a password to providing a voice sample or fingerprint by simply changing the PAM modules without having to rewrite any code in these applications. The configuration files of the PAM modules are located in the directory /etc/pam.d and the modules (shared libraries) themselves are located in the directory /lib/security. The /etc/pam.d directory has a collection of named files of its own, e.g. ssh, pop, imap, etc. PAM- aware applications that do not have a configuration file will automatically be pointed to the default configuration file 'other'. In the next section we will set up some recommended minimum-security restrictions using PAM. The password length The minimum acceptable password length by default when you install your Linux system is five. This means that when a new user is given access to the server, his/her password length will be at minimum five mixes of character strings, letter, number, special character etc. This is not enough and must be eight or more. The password length under Linux by the use of its PAM feature is controlled by five arguments: minlen, dcredit, ucredit, lcredit, and ocredit. Step 1 To prevent non-security-minded people or administrators from being able to enter just five characters for the valuable password, edit the rather important /etc/pam.d/system-auth file and enforce the minimum password length. • Edit the system-auth file (vi /etc/pam.d/system-auth) and change the line: password required /lib/security/pam_cracklib.so retry=3 type= To read: password required /lib/security/pam_cracklib.so retry=3 minlen=12 type= After changing the above line, the /etc/pam.d/system-auth file should look like this: #%PAM-1.0 auth required /lib/security/pam_env.so auth sufficient /lib/security/pam_unix.so likeauth nullok auth required /lib/security/pam_deny.so account required /lib/security/pam_unix.so password required /lib/security/pam_cracklib.so retry=3 minlen=12 type= password sufficient /lib/security/pam_unix.so nullok use_authtok md5 shadow password required /lib/security/pam_deny.so session required /lib/security/pam_limits.so session required /lib/security/pam_unix.so
- 104. 104 WARNING: It isimportant to note that when you set the password for a user under ‘root’ account, then these restrictions don't apply!! This is the case on all Unix OS. The super-user ‘root’ can override pretty much everything. Instead, log as the user account from which you apply this restriction and try to change the password. You will see that it works. You need to keep in mind that this module includes a credit mechanism. E.g. if you define minlen=12, then you will get 1 credit for e.g. including a single digit number in your password, or for including a non-alphanumeric character. Getting 1 credit means that the module will accept a password of the length of minlen-credit. When you check the parameters of the cracklib module, you will see that it has some parameters that let you define what a credit is (http://www.kernel.org/pub/linux/libs/pam/Linux-PAM-html/pam.html). For example: minlen The following password was accepted --------- --------------------------------------------------- 14 gjtodgsdf1$ You can see that I got 1 credit for a alphanumeric character and a credit for each non- alphanumeric character. "gjtodgsdf1$" has a length of 11, 1 credit for alpha-numeric, 2 credits for non-alphanumeric character (1 and $) which gives me a credit of 3, hence the password length of 11 was accepted. At any rate, the minimum length is adjusted by the mixture of types of characters used in the password. Using digits (up to the number specified with the "dcredit=" parameter, which defaults to 1) or uppercase letters "ucredit" or lowercase letters "lcredit" or other types of letters "ocredit" will decrease the minimum length by up to four since the default parameter for these arguments is 1 and there is four different arguments that you can add. A password with 9 lowercase letters in it will pass a minimum length set to 10 unless "lcredit=0" is used, because a credit is granted for the use of a lowercase letter. If the mixture includes an uppercase letter, a lowercase letter, and a digit, then a minlength of 8 effectively becomes 5. NOTE: With the new MD5 passwords capability, which is installed by default in all modern Linux operating system, a long password can be used now (up to 256 characters), instead of the Unix standard eight letters or less. If you want to change the password length of 8 characters to example 16 characters, all you have to do is to replace the number 12 by 20 in the “minlen=12” line of the /etc/pam.d/system-auth file.
- 105. Pluggable Authentication Modules0 CHAPTER 4 105 Disabling console program access In a safe environment, where we are sure that console is secured because passwords for BIOS and GRUB or LILO are set and all physical power and reset switches on the system are disabled, it may be advantageous to entirely disable all console-equivalent access to programs like poweroff, reboot, and halt for regular users on your server. • To do this, run the following command: [root@deep /]# rm -f /etc/security/console.apps/<servicename> Where <servicename> is the name of the program to which you wish to disable console- equivalent access. • To disable console program access, use the following commands: [root@deep /]# rm -f /etc/security/console.apps/halt [root@deep /]# rm -f /etc/security/console.apps/poweroff [root@deep /]# rm -f /etc/security/console.apps/reboot This will disable console-equivalent access to programs halt, poweroff, and reboot. Disabling all console access The Linux-PAM library installed by default on your system allows the system administrator to choose how applications authenticate users, such as for console access, program and file access. Step 1 In order to disable all these accesses for the users, you must comment out all lines that refer to pam_console.so in the /etc/pam.d directory. This step is a continuation of the hack “Disabling console program access”. The following script will do the trick automatically for you. • As ‘root’ creates the disabling.sh script file (touch disabling.sh) and add the following lines inside: # !/bin/sh cd /etc/pam.d for i in * ; do sed '/[^#].*pam_console.so/s/^/#/' < $i > foo && mv foo $i done Step 2 Now, we have to make the script executable and run it. • This can be done with the following commands: [root@deep /]# chmod 700 disabling.sh [root@deep /]# ./disabling.sh This will comment out all lines that refer to pam_console.so for all files located under /etc/pam.d directory. Once the script has been executed, you can remove it from your system.
- 106. 106 The Login accesscontrol table On a server environment where authorized and legitimate logins can come from everywhere, it is important to have the possibility to use a security file which allows us to have more control over users who can connect to the server. What we are looking here is to have more control on not allowing some legitimated accounts to login from anywhere. Fortunately, this file exists and is called "access.conf", you can find it under your /etc/security directory. The access.conf file which comes already installed with your native Linux system allow us to control which authorized users can/cannot log in to the server or to the console and from where. Don't forget that users access can come everywhere from remote host or directly from the console of the system. Configuration of the access.conf file of Linux is not complicated to understand. Below I show you how to configure it to be very restrictive and secure. Step 1 By default denying access to every one, is the first step of a reliable security policy. In this way we eliminate the possibility of forgetting someone or to making a mistake. • Edit the access.conf file (vi /etc/security/access.conf) and add the following line at the end of the file. -:ALL EXCEPT root gmourani:ALL This access policy means to disallow console logins as well as remote accounts login to all from anywhere except for user ‘root’ and ‘gmourani’. With this choice of policy, we deny non- networked and remote logins to every user with a shell account on the system from everywhere and allow only the selected users. Take a note that many possibilities exist as for example allowing the same users ‘root’ and ‘gmourani’ to log only to the system from remote host with IP address 207.35.78.2. To enable this policy, all we need to do is to change the above policy to this one: • Edit the access.conf file (vi /etc/security/access.conf) and add the following lines at the end of the file. -:ALL EXCEPT root gmourani:207.35.78.2 -:ALL:LOCAL Here the second policy line means to disallow all local access to the console for every users even for the super-user ‘root’, therefore if you want to log as ‘root’ you need first to log as user ‘gmourani’ from remote host with IP address 207.35.78.2 and su to ‘root’ (this is why I added ‘root’ to the users allowed to connect from remote host 207.35.78.2).
- 107. Pluggable Authentication Modules0 CHAPTER 4 107 Step 2 To be able to use the access.conf feature of Linux, make sure to add the following line to /etc/pam.d/system-auth and sshd if you use this service or it will not work. • Edit the login file (vi /etc/pam.d/system-auth) and add the following line. account required /lib/security/pam_access.so After adding the above line, the /etc/pam.d/system-auth file should look like this: #%PAM-1.0 auth required /lib/security/pam_env.so auth sufficient /lib/security/pam_unix.so likeauth nullok auth required /lib/security/pam_deny.so account required /lib/security/pam_unix.so account required /lib/security/pam_access.so password required /lib/security/pam_cracklib.so retry=3 minlen=12 type= password sufficient /lib/security/pam_unix.so nullok use_authtok md5 shadow password required /lib/security/pam_deny.so session required /lib/security/pam_limits.so session required /lib/security/pam_unix.so NOTE: Please read information about possible configurations of this file inside the access.conf file since your policies will certainly differ from the example that I show you above. Tighten console permissions for privileged users The console.perms security file of Linux, which use the pam_console.so module to operate, is designed to give to privileged users at the physical console (virtual terminals and local xdm- managed X sessions) capabilities that they would not otherwise have, and to take those capabilities away when they are no longer logged in at the console. It provides two main kinds of capabilities: file permissions and authentication. When a user logs in at the console and no other user is currently logged in at the console, the pam_console.so module will change permissions and ownership of files as described in the file /etc/security/console.perms. Please note that privileged users are nothing in common with regular users you may add to the server, they are special users like floppy, cdrom, scanner, etc which in an networking server environment are also considered and treated as users.
- 108. 108 Step 1 The defaultconsole.perms configuration file of Linux is secure enough for regular use of the system where an Xwindow interface is considered to be installed but in a highly secure environment where the Graphical User Interface (GUI) is not installed or where some special devices like sound, jaz, etc have no reason to exist, we can tighten the console.perms security file of Linux to be more secure by eliminating non-existent or unneeded privileged users to have capabilities that they would not otherwise have. • Edit the console.perms file (vi /etc/security/console.perms), and change the default lines inside this file: # file classes -- these are regular expressions <console>=tty[0-9][0-9]* vc/[0-9][0-9]* :[0-9].[0-9] :[0-9] <xconsole>=:[0-9].[0-9] :[0-9] # device classes -- these are shell-style globs <floppy>=/dev/fd[0-1]* /dev/floppy/* /mnt/floppy* <sound>=/dev/dsp* /dev/audio* /dev/midi* /dev/mixer* /dev/sequencer /dev/sound/* /dev/beep <cdrom>=/dev/cdrom* /dev/cdroms/* /dev/cdwriter* /mnt/cdrom* <pilot>=/dev/pilot <jaz>=/mnt/jaz* <zip>=/mnt/pocketzip* /mnt/zip* <ls120>=/dev/ls120 /mnt/ls120* <scanner>=/dev/scanner /dev/usb/scanner* <rio500>=/dev/usb/rio500 <camera>=/mnt/camera* /dev/usb/dc2xx* /dev/usb/mdc800* <memstick>=/mnt/memstick* <flash>=/mnt/flash* <diskonkey>=/mnt/diskonkey* <rem_ide>=/mnt/microdrive* <fb>=/dev/fb /dev/fb[0-9]* /dev/fb/* <kbd>=/dev/kbd <joystick>=/dev/js[0-9]* <v4l>=/dev/video* /dev/radio* /dev/winradio* /dev/vtx* /dev/vbi* /dev/video/* <gpm>=/dev/gpmctl <dri>=/dev/nvidia* /dev/3dfx* <mainboard>=/dev/apm_bios # permission definitions <console> 0660 <floppy> 0660 root.floppy <console> 0600 <sound> 0600 root <console> 0600 <cdrom> 0660 root.disk <console> 0600 <pilot> 0660 root.uucp <console> 0600 <jaz> 0660 root.disk <console> 0600 <zip> 0660 root.disk <console> 0600 <ls120> 0660 root.disk <console> 0600 <scanner> 0600 root <console> 0600 <camera> 0600 root <console> 0600 <memstick> 0600 root <console> 0600 <flash> 0600 root <console> 0600 <diskonkey> 0660 root.disk <console> 0600 <rem_ide> 0660 root.disk <console> 0600 <fb> 0600 root <console> 0600 <kbd> 0600 root <console> 0600 <joystick> 0600 root
- 109. Pluggable Authentication Modules0 CHAPTER 4 109 <console> 0600 <v4l> 0600 root <console> 0700 <gpm> 0700 root <console> 0600 <mainboard> 0600 root <console> 0600 <rio500> 0600 root <xconsole> 0600 /dev/console 0600 root.root <xconsole> 0600 <dri> 0600 root To read : # file classes -- these are regular expressions <console>=tty[0-9][0-9]* vc/[0-9][0-9]* :[0-9].[0-9] :[0-9] # device classes -- these are shell-style globs <floppy>=/dev/fd[0-1]* /dev/floppy/* /mnt/floppy* <cdrom>=/dev/cdrom* /dev/cdroms/* /dev/cdwriter* /mnt/cdrom* <pilot>=/dev/pilot <fb>=/dev/fb /dev/fb[0-9]* /dev/fb/* <kbd>=/dev/kbd <gpm>=/dev/gpmctl <mainboard>=/dev/apm_bios # permission definitions <console> 0660 <floppy> 0660 root.floppy <console> 0600 <cdrom> 0660 root.disk <console> 0600 <pilot> 0660 root.uucp <console> 0600 <fb> 0600 root <console> 0600 <kbd> 0600 root <console> 0700 <gpm> 0700 root <console> 0600 <mainboard> 0600 root Here we removed every privileged user related to the Graphical User Interface and others related to sound, zip drive, jaz drive, scanner, joystick and video media at the physical console on the server. Putting limits on resource The limits.conf file located under the /etc/security directory can be used to control and limit resources for the users on your system. It is important to set resource limits on all your users so they can't perform Denial of Service attacks (number of processes, amount of memory, etc) on the server. These limits will have to be set up for the user when he or she logs in. Step 1 For example, limits for all users on your system might look like this: • Edit the limits.conf file (vi /etc/security/limits.conf) and change the lines: #* soft core 0 #* hard rss 10000 #@student hard nproc 20 To read: * hard core 0 * hard rss 5000 * hard nproc 35
- 110. 110 This says toprohibit the creation of core files “core 0”, restrict the number of processes to 35 “nproc 35”, and restrict memory usage to 5M “rss 5000” for everyone except the super user “root”. All of the above only concerns users who have entered through the login prompt on your system. With this kind of quota, you have more control on the processes, core files, and memory usage that users may have on your system. The asterisk “*” mean: all users that logs in on the server. Putting an asterisk “*” to cover all users can pose problem with daemon users account like “www” for a web server, “mysql” for a SQL database server, etc. If we put an asterisk, then, these users will be affected by the restriction and limitation of processes or memory usage. To solve the problem, we can choose an existing group name in our system and add every regular user to this group. In this manner, the restrictions and limitations will apply to all users who are members of this group name only. A special group account named “users” can be used for this purpose. This is the recommended method on putting limit on user resources. • Edit the limits.conf file (vi /etc/security/limits.conf) and change the lines: #* soft core 0 #* hard rss 10000 #@student hard nproc 20 To read: @users hard core 0 @users hard rss 5000 @users hard nproc 35 If you decide to use a group name like “@users” to control and limit resources for the users on your system, then it is important to not forget to change the GUI (Group User ID) of these users to be “100”. “100” is the numeric value of the user’s ID “users”. • The command to create a new user with group name which is set by default to users is: [root@deep /]# useradd -g100 admin The “-g100” option represents the number of the user’s initial login group and in our case “100” is the group account name “users”. The “admin” parameter is the user name we want to add to the group name “users”. WARNING: Use the same command above for all users on your system you want to be member of the “users” group account. It is also preferable to set this parameter first before adding users to the system.
- 111. Pluggable Authentication Modules0 CHAPTER 4 111 Controlling access time to services As the Linux-PAM system said, running a well-regulated system occasionally involves restricting access to certain services in a selective manner. The time.conf security file, which is provided by the pam_time.so module of Linux, offers some time control for access to services offered by a system. Its actions are determined through the configuration file called time.conf and located under /etc/security directory. Step 1 The time.conf file can be configured to deny access to (individual) users based on their name, the time of day, the day of week, the service they are applying for and their terminal from which they are making their request. • Edit the time.conf file (vi /etc/security/time.conf), and add the following line: login ; tty* & !ttyp* ; !root & gmourani ; !Al0000-2400 The above time control access line means to deny all user access to console-login at all times except for the super-user 'root' and the user 'gmourani'. Take a note that many combinations exist as described in the time.conf file, we can, for example, allow user ‘admin’ to access the console-login any time except at the weekend and on Tuesday from 8AM to 6PM with the following statement. • Edit the time.conf file (vi /etc/security/time.conf), and add the following line: login ; * ; !admin ; !Wd0000-2400 & Tu0800-1800 Step 2 To be able to use the time.conf feature of Linux, make sure to add the following line to /etc/pam.d/system-auth and sshd if you use this service or nothing will work. • Edit the system-auth file (vi /etc/pam.d/system-auth) and add the line. account required /lib/security/pam_time.so After adding the line above, the /etc/pam.d/system-auth file should look like this: #%PAM-1.0 auth required /lib/security/pam_env.so auth sufficient /lib/security/pam_unix.so likeauth nullok auth required /lib/security/pam_deny.so account required /lib/security/pam_unix.so account required /lib/security/pam_access.so account required /lib/security/pam_time.so password required /lib/security/pam_cracklib.so retry=3 minlen=12 type= password sufficient /lib/security/pam_unix.so nullok use_authtok md5 shadow password required /lib/security/pam_deny.so session required /lib/security/pam_limits.so session required /lib/security/pam_unix.so
- 112. 112 Blocking; su toroot, by one and sundry The su (Substitute User) command allows you to become other existing users on the system. For example you can temporarily become ‘root’ and execute commands as the super-user ‘root’. Step 1 If you don’t want anyone to su to root or want to restrict the su command to certain users then uncomment the following line of your su configuration file in the /etc/pam.d directory. We highly recommend that you limit the persons allowed to su to the root account. • Edit the su file (vi /etc/pam.d/su) and uncomment the following line in the file: auth required /lib/security/pam_wheel.so use_uid After this line has been uncommented, the /etc/pam.d/su file should look like this: #%PAM-1.0 auth sufficient /lib/security/pam_rootok.so auth required /lib/security/pam_wheel.so use_uid auth required /lib/security/pam_stack.so service=system-auth account required /lib/security/pam_stack.so service=system-auth password required /lib/security/pam_stack.so service=system-auth session required /lib/security/pam_stack.so service=system-auth session optional /lib/security/pam_xauth.so Which means only those who are members of the “wheel” group can su to root; it also includes logging. Note that the “wheel” group is a special account on your system that can be used for this purpose. You cannot use any group name you want to make this hack. This hack combined with specifying which TTY and VC devices super-user root is allowed to login on will improve your security a lot on the system. Step 2 Now that we have defined the “wheel” group in our /etc/pam.d/su file configuration, it is time to add some users who will be allowed to su to super-user “root” account. • If you want to make, for example, the user “admin” a member of the “wheel” group, and thus be able to su to root, use the following command: [root@deep /]# usermod -G10 admin Which means “G” is a list of supplementary groups, where the user is also a member of. “10” is the numeric value of the user’s ID “wheel”, and “admin” is the user we want to add to the “wheel” group. Use the same command above for all users on your system you want to be able to su to super-user “root” account. NOTE: For Linux users, who use the Xwindow interface, it is important to note that if you can't su in a GNOME terminal, it’s because you’ve used the wrong terminal. (So don't think that this advice doesn't work simply because of a GNOME terminal problem!)
- 113. Pluggable Authentication Modules0 CHAPTER 4 113 Facultative: A special line exists in the su file /etc/pam.d/su which allows you to implicitly trust users in the “wheel” group (for security reasons, I don’t recommend using this option). This mean that all users who are members of the “wheel” group can su to root without the need to enter the super-user “root” password. • To allow users who are members of the “wheel” group to su to root account without the need to enter the “root” password, edit the su file (vi /etc/pam.d/su) and uncomment the following line in the file: auth sufficient /lib/security/pam_wheel.so trust use_uid After this line has been uncommented, the /etc/pam.d/su file should look like this: #%PAM-1.0 auth sufficient /lib/security/pam_rootok.so auth sufficient /lib/security/pam_wheel.so trust use_uid auth required /lib/security/pam_stack.so service=system-auth account required /lib/security/pam_stack.so service=system-auth password required /lib/security/pam_stack.so service=system-auth session required /lib/security/pam_stack.so service=system-auth session optional /lib/security/pam_xauth.so Using sudo instead of su for logging as super-user There is a security tool called “sudo” that we discuss into this book. This security software allow us to archive the same result as using the su command to get root privilege on the server but in a more secure and informative way. With sudo installed in our server, we can get information about who is connected as super-user root as well as many other useful features. Please see the chapter related to this security program into this book for more information about sudo. If you want to use sudo to allow and control which is allowed to logging as super-user root on your server, then you no longer need to use the su command of Linux to archive this task and we can remove the SUID bit on this command to completely disable su and use sudo. This let us remove one more SUID bit on our secure server and have a more complete and powerful security software to control access to super-user root. This is the method I highly recommend you to use instead of the su command of Linux. Step 1 To archive this result, we have to remove the SUID bit of the su command and install the sudo security software as explained further down in this book. This also implies that we don’t need to modify the above su configuration file on our system. To recap, all we need to do is to remove the SUID bit on the su command, and install sudo in our server. • To remove the SUID bit on the su binary, use the following command: [root@deep /]# chmod a-s /bin/su
- 116. General Optimization IN THISCHAPTER 1. Static vs. shared libraries 2. The Glibc 2 library of Linux 3. Why Linux programs are distributed as source 4. Some misunderstanding in the compiler flags options 5. The gcc specs file 6. Striping all binaries and libraries files 7. Tuning IDE Hard Disk Performance
- 118. General Optimization 0 CHAPTER5 118 Linux General Optimization Abstract At this stage of your configuration, you should now have a Linux server optimally configured and secured. Our server contains the most essential package and programs installed to be able to work properly and the most essential general system security configuration. Before we continue and begin to install the services we want to share with our customers, it is important to tune our Linux server to make it runs faster. The tuning we will perform in the following part will be applied to the whole system. It also applies to present as well as future programs, such as services that we will later install. Generally, if you don’t use an x386 Intel processor, Red Hat Linux out of the box is not optimized for your specific CPU architecture (most people now run Linux on a Pentium processor). The sections below will guide you through different steps to optimize your Linux server for your specific processor, memory, and network. Static vs. shared libraries During compilation and build time of a program, the last stage (where all the parts of the program are joined together) is to link the software through the Linux libraries if needed. These libraries, which come in both shared and static formats, contain common system code which are kept in one place and shared between programs. Obviously there are some tasks that many programs will want to do, like opening files, and the codes that perform these functions are provided by the Linux libraries. On many Linux system these libraries files can be found into the /lib, /usr/lib, and /usr/share directories. The default behavior of Linux is to link shared and if it cannot find the shared libraries, then is to link statically. One of the differences between using static or shared libraries are: When using a static library, the linker finds the bits that the program modules need, and directly copies them into the executable output file that it generates. For shared libraries, it leaves a note in the output saying, “When this program is run, it will first have to load this library”. Performance-wise, for most systems, worrying about static vs. dynamic is a moot point. There simply isn’t enough difference to measure. Security-wise there are valid arguments both ways. Static linking is less secure because it locks in the library bugs; unless you rebuild all such programs, your system won’t be properly secured. Static linking is more secure because it avoids library attacks. The choice is yours: run a daemon which will remain vulnerable to library attacks, or run one which remains vulnerable to library bugs. Portability-wise, the only difference is the size of the file you’ll be transferring between systems. To make setup easier, a statically linked daemon is only needed when the libraries are completely unavailable. That is rarely the case. Finally, on a busy system (when performance becomes a true issue), by statically linking you’ll be DEGRADING performance. Being bigger, as more and more statically linked daemons are running, your system begins to swap sooner and since none of the code is shared, swapping will have a larger effect on performance. So, when looking to improve performance, you’ll want to use shared libraries as much as possible. <Gregory A Lundberg>
- 119. General Optimization 0 CHAPTER5 119 If you decide to compile program statically, you will generally need to add the “-static” and/or “--disable-shared” options flag to your compile line during compilation of your software. Be aware that it is not always possible to use and compile statically all programs, this highly depends on how developers are coding and developed the software. To resume: 1. If you want to compile program with shared libraries, you will use something like: CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS ./Configure 2. If you want to compile program with static libraries, you will use something like: CFLAGS="-O2 –static -march=i686 -funroll-loops"; export CFLAGS ./Configure --disable-shared WARNING: On Linux, static libraries have names like libc.a, while shared libraries are called libc.so.x.y.z where x.y.z is some form of version number since it would be quite a pain to recompile programs each time the version number changed so instead programs reference libraries by these shorter names and depend on the dynamic linker to make these shorter names symlinks to the current version. Shared libraries often have links pointing to them. The Glibc 2.2 library of Linux The Glibc 2.2, which replaces the libc4 and libc5 that came before it, is the latest version of the GNU C Library for Linux and it contains standard libraries used by multiple programs on the system as described in the previous section. This particular package contains the most important sets of shared and static libraries, which provides the core functionality for C programs to run and without it, a Linux system would not function. Under Red Hat Linux and most other Linux variant this package comes configured to run under i386 processor for portability reasons and this will pose problems for us if we want to compile programs under Linux because even if we have put in all the optimization flags we need to improve the speed of our server, when the compiler includes static or shared libraries files to our program, these library files will run optimized for an i386 processor. In this case, our program will have some parts of its binaries optimized for an i686 processor (the program itself) and another parts optimized for an i386 processor (the GLIBC libraries). To solve the problem, you have to check inside your vendor CD-ROM media for available GLIBC RPM packages made to run on i686 CPU architecture. All vendors known about this issue and provide alternative GLIBC packages for i686 processors.
- 120. General Optimization 0 CHAPTER5 120 Why Linux programs are distributed as source Linux has been ported to run on a large number of different machines and rather than provide a copy for each machine Linux can run on, it's much simpler just to distribute the source and let the end user compile it. The creators of the distribution have no idea if you're going to be running it on a 386 or on a Pentium III and above so they have to write programs that work on all processors and this is where the problem comes, because all the programs that were installed with your distribution are going to be compiled so they work on the 386 for portability, meaning that they don't use any new feature like MMX which can only be found on newer generation of processors. Fortunately, various compiler options exist to optimize program you want to install under Linux for your specific CPU architecture. This is great for those of us that want to tweak every ounce of performance out of the program, now we get to decide how the program is compiled. If you want some speed out of your programs you've got to know a fair amount about the various option flags you can use to compile. The first thing you want to set is your CPU type, that's done with the “-march=cpu_type” (processor machine architecture) flag, an example would be “-march=i686” or “-march=k6”, this will allow the compiler to select the appropriate optimizations for the processor, but this is only the beginning of what can be done. You can set the “-O” flag anywhere from 1 to 3 to tell the compiler how aggressive to be with the optimizations, “-O3” will produce the fastest programs assuming the compiler didn't optimize an important part of a subroutine out. The next thing you might want to do is check out the “-f” options of the compiler, these are things like “-funroll-loops”, and “-fomit-frame- pointer”. WARNING: Compiling with the “-fomit-frame-pointer” switch option will use the stack for accessing variables. Unfortunately, debugging is almost impossible with this option. Also take special attention to the above optimization number “-O3”; “O” is a capital o and not a 0 (zero). What I recommend you to use for all software that you want to compile on your server is the following optimizations FLAGS: CFLAGS="-O2 -march=i686 -funroll-loops" As you can see, I don’t use the “-O3” and “-fomit-frame-pointer” options because some software have problem to run with these optimization options.
- 121. General Optimization 0 CHAPTER5 121 Some misunderstanding in the compiler flags options At lot of discussions exist in the Linux community about the “-O” option and its level numbers. Some Linux users try to convince that level number up to “-O3” like “-O9” will produce faster program. The “-O9” flag doesn't do anything over “-O3”, if you don't believe me make a small file, call it testO3.c and see: Step 1 • Create the testO3.c file with the following command: [root@deep tmp]# touch testO3.c Step 2 • Run the GCC compiler with “-O3” flag through the testO3.c file with the command: [root@deep tmp]# gcc -O3 -S -fverbose-asm testO3.c Step 3 Look at testO3.s that it made, then run again with “-O9” and compare the output. • Create the testO9.c file with the following command: [root@deep tmp]# touch testO9.c Step 4 • Run the GCC compiler again with “-O9” flag through the testO9.c file with command: [root@deep tmp]# gcc -O9 -S -fverbose-asm testO9.c Step 5 Now if you compare the output you will see no difference between the both files. • To compare the output, use the following command: [root@deep tmp]# diff testO3.s testO9.s > difference WARNING: The “-O3” flag level number is the best and highest optimization flag you can use during optimization of programs under Linux.
- 122. General Optimization 0 CHAPTER5 122 The gcc specs file The /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file of Red Hat Linux is a set of defines that the gcc compiler uses internally to set various aspects of the compile environment. All customizations that you put in this file will apply for the entire variable environment on your system, so putting optimization flags in this file is a good choice. To squeeze the maximum performance from your x86 programs, you can use full optimization when compiling with the “-O3” flag. Many programs contain “-O2” in the Makefile. The “-O3” level number is the highest level of optimization. It will increase the size of what it produces, but it runs faster in most case. You can also use the “-march=cpu_type” switch to optimize the program for the CPU listed to the best of GCC’s ability. However, the resulting code will only be run able on the indicated CPU or higher. Below are the optimization flags that we recommend you to put in your /usr/lib/gcc- lib/i386-redhat-linux/2.96/specs file depending on your CPU architecture. The optimization options apply only when we compile and install a new program in our server. These optimizations don’t play any role in our Linux base system; it just tells our compiler to optimize the new programs that we will install with the optimization flags we have specified in the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file. Adding options listed below depending of your CPU architecture to the gcc 2.96 specs file will save you having to change every CFLAGS in future Makefiles. Step 1 The first thing to do is to verify the compiler version installed on your Linux server. • To verify the compiler version installed on your system, use the command: [root@deep /]# gcc -v Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs gcc version 2.96 20000731 (Red Hat Linux 7.3 2.96-110) Step 2 For CPU i686 or PentiumPro, Pentium II, Pentium III, and Athlon Edit the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file, scroll down a ways... You'll see a section like the following: *cpp_cpu_default: -D__tune_i386__ *cpp_cpu: -Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__ %{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__ %{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__ %{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro - D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__ %{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:- D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:- D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__ }%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:- D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__ }%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}} *cc1_cpu: %{!mcpu*: %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium} %{mpentiumpro:-mcpu=pentiumpro}}
- 123. General Optimization 0 CHAPTER5 123 Change it for the following: *cpp_cpu_default: -D__tune_i686__ *cpp_cpu: -Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__ %{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__ %{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__ %{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro - D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__ %{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:- D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:- D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__ }%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:- D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__ }%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}} *cc1_cpu: %{!mcpu*: -O2 -march=i686 -funroll-loops %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium} %{mpentiumpro:-mcpu=pentiumpro}} WARNING: Make sure that you’re putting –O2 and not -02 (dash zero three). For CPU i586 or Pentium Edit the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file, scroll down a ways... You'll see a section like the following: *cpp_cpu_default: -D__tune_i386__ *cpp_cpu: -Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__ %{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__ %{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__ %{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro - D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__ %{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:- D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:- D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__ }%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:- D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__ }%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}} *cc1_cpu: %{!mcpu*: %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium} %{mpentiumpro:-mcpu=pentiumpro}} Change it for the following: *cpp_cpu_default: -D__tune_i586__ *cpp_cpu: -Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__ %{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__ %{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__ %{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro -
- 124. General Optimization 0 CHAPTER5 124 D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__ %{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:- D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:- D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__ }%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:- D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__ }%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}} *cc1_cpu: %{!mcpu*: -O2 -march=i586 -funroll-loops %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium} %{mpentiumpro:-mcpu=pentiumpro}} WARNING: Make sure that you’re putting –O2 and not -02 (dash zero three). For CPU i486 Edit the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file, scroll down a ways... You'll see a section like the following: *cpp_cpu_default: -D__tune_i386__ *cpp_cpu: -Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__ %{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__ %{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__ %{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro - D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__ %{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:- D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:- D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__ }%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:- D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__ }%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}} *cc1_cpu: %{!mcpu*: %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium} %{mpentiumpro:-mcpu=pentiumpro}} Change it for the following: *cpp_cpu_default: -D__tune_i486__ *cpp_cpu: -Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__ %{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__ %{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__ %{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro - D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__ %{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:- D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:- D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__ }%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:- D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__ }%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}} *cc1_cpu:
- 125. General Optimization 0 CHAPTER5 125 %{!mcpu*: -O2 -march=i486 -funroll-loops %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium} %{mpentiumpro:-mcpu=pentiumpro}} WARNING: Make sure that you’re putting –O2 and not -02 (dash zero three). For CPU AMD K6 or K6-2 Edit the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file, scroll down a ways... You'll see a section like the following: *cpp_cpu_default: -D__tune_i386__ *cpp_cpu: -Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__ %{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__ %{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__ %{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro - D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__ %{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:- D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:- D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__ }%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:- D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__ }%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}} *cc1_cpu: %{!mcpu*: %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium} %{mpentiumpro:-mcpu=pentiumpro}} Change it for the following: *cpp_cpu_default: -D__tune_k6__ *cpp_cpu: -Acpu(i386) -Amachine(i386) %{!ansi:-Di386} -D__i386 -D__i386__ %{march=i386:%{!mcpu*:-D__tune_i386__ }}%{march=i486:-D__i486 -D__i486__ %{!mcpu*:-D__tune_i486__ }}%{march=pentium|march=i586:-D__pentium -D__pentium__ %{!mcpu*:-D__tune_pentium__ }}%{march=pentiumpro|march=i686:-D__pentiumpro - D__pentiumpro__ %{!mcpu*:-D__tune_pentiumpro__ }}%{march=k6:-D__k6 -D__k6__ %{!mcpu*:-D__tune_k6__ }}%{march=athlon:-D__athlon -D__athlon__ %{!mcpu*:- D__tune_athlon__ }}%{m386|mcpu=i386:-D__tune_i386__ }%{m486|mcpu=i486:- D__tune_i486__ }%{mpentium|mcpu=pentium|mcpu=i586:-D__tune_pentium__ }%{mpentiumpro|mcpu=pentiumpro|mcpu=i686:-D__tune_pentiumpro__ }%{mcpu=k6:- D__tune_k6__ }%{mcpu=athlon:-D__tune_athlon__ }%{!march*:%{!mcpu*:%{!m386:%{!m486:%{!mpentium*:%(cpp_cpu_default)}}}}} *cc1_cpu: %{!mcpu*: -O2 -march=k6 -funroll-loops %{m386:-mcpu=i386} %{m486:-mcpu=i486} %{mpentium:-mcpu=pentium} %{mpentiumpro:-mcpu=pentiumpro}} WARNING: Make sure that you’re putting –O2 and not -02 (dash zero three).
- 126. General Optimization 0 CHAPTER5 126 Step3 Once our optimization flags have been applied to the gcc 2.96 specs file, it time to verify if the modification work. • To verify if the optimization work, use the following commands: [root@deep tmp]# touch cpu.c [root@deep tmp]# gcc cpu.c -S -fverbose-asm [root@deep tmp]# less cpu.s What you'll get is a file that contains depending of options you have chose, something like: .file "ccnVPjeW.i" .version "01.01" # GNU C version 2.96 20000731 (Red Hat Linux 7.3 2.96-110) (i386-redhat-linux) compiled by GNU C version 2.96 20000731 (Red Hat Linux 7.3 2.96-110). # options passed: -O2 -march=i686 -funroll-loops -fverbose-asm # options enabled: -fdefer-pop -foptimize-sibling-calls -fcse-follow-jumps # -fcse-skip-blocks -fexpensive-optimizations -fthread-jumps # -fstrength-reduce -funroll-loops -fpeephole -fforce-mem -ffunction-cse # -finline -fkeep-static-consts -fcaller-saves -fpcc-struct-return -fgcse # -frerun-cse-after-loop -frerun-loop-opt -fdelete-null-pointer-checks # -fschedule-insns2 -fsched-interblock -fsched-spec -fbranch-count-reg # -fnew-exceptions -fcommon -fverbose-asm -fgnu-linker -fregmove # -foptimize-register-move -fargument-alias -fstrict-aliasing # -fmerge-constants -fident -fpeephole2 -fmath-errno -m80387 -mhard-float # -mno-soft-float -mieee-fp -mfp-ret-in-387 -march=i686 gcc2_compiled.: .ident "GCC: (GNU) 2.96 20000731 (Red Hat Linux 7.3 2.96-110)" WARNING: In our example we are optimized the specs file for a i686 CPU processor. It is important to note that most of the “-f” options are automatically included when you use “-O2” and don't need to be specified again. The changes that were shown were made so that a command like "gcc" would really be the command "gcc -march=i686" without having to change every single Makefile which can really be a pain. Below is the explanation of the different optimization options we use: • The “-march=cpu_type” optimization flag The “-march=cpu_type” optimization option will set the default CPU to use for the machine type when scheduling instructions. • The “-funroll-loops” optimization flag The “-funroll-loops” optimization option will perform the optimization of loop unrolling and will do it only for loops whose number of iterations can be determined at compile time or run time. • The “-fomit-frame-pointer” optimization flag The “-fomit-frame-pointer” optimization option, one of the most interesting, will allow the program to not keep the frame pointer in a register for functions that don't need one. This avoids the instructions to save, set up and restores frame pointers; it also makes an extra register available in many functions and makes debugging impossible on most machines.
- 127. General Optimization 0 CHAPTER5 127 WARNING: All future optimizations that we will describe in this book refer by default to a Pentium PRO/II/III and higher i686 CPU family. So you must adjust the compilation flags for your specific CPU processor type in the /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs file and during your compilation time. Striping all binaries and libraries files When compiler builds program it add many comments and other stuffs inside the resulting binary and library code which are used to debug application on the system or to make the code more readable by developers. To get the latest bit of optimization, we can remove all comments and other unneeded stuffs since there are only used for debugging purpose. This will make the software runs a little bit faster because it will have fewer codes to read when executing. I don’t know if it’s a good idea to talk about this hack because it’s really dangerous to apply and can make your system unstable or completely unworkable if you don’t take care of what you do. The process of eliminating all unneeded comments and other unneeded stuffs from your binaries and libraries files is made by the use of the strip command of Linux. This command should be used with care and in the good manner or you will certainly have a bad surprise. Bellow, I will explain you how to apply it on your system and on which files or where you should use it. It is very important to know that it’s NOT all binaries and especially libraries files that need to be striped by this method but ONLY some of them. If you apply this hack on your entire system, then something will inevitably break, you have been warned. Finally, you should use this hack on servers where you DON’T compile software. If you compile software on the server where you want to apply this hack, then nothing will work and you will not be able to compile any software on it. Use this hack on server which doesn’t have any compiler packages installed to make compilation. Step 1 The first step in our procedure will be to be sure that the strip command is available on our server; this command comes from the “binutils” RPM package. Therefore, if it is not installed, install it from your CD-ROM. Step 2 Once the strip program is installed on our server, it’s time to strip the required files. With the commands below, we strip all binaries program available under the /bin, /sbin, /usr/bin and /usr/sbin directories of your server. • To strip all binaries program, use the following commands: [root@deep /]# strip /bin/* [root@deep /]# strip /sbin/* [root@deep /]# strip /usr/bin/* [root@deep /]# strip /usr/sbin/* NOTE: When issuing the above commands, you will receive some error messages like “File format not recognized” on your terminal. This is normal because some of the binaries files are symbolic link pointing to other binaries on your system and the strip command generate the warning because it cannot strip symbolic links.
- 128. General Optimization 0 CHAPTER5 128 Step 3 Now, it’s time to strip the libraries files. This is where the action can become dangerous if you don’t take care or abuse the strip command. With the commands below, we strip all libraries files available under the /lib and /usr/lib directory of the system. • To strip all libraries files, use the following command: [root@deep /]# strip -R .comment /usr/lib/*.so.* [root@deep /]# strip -R .comment /lib/*.so.* NOTE: Make attention to the above command, you can see here that I use the “-R” option with the strip command. This option allows us to select a specific name to strip from the target libraries files. With the “.comment” name, we inform the command to remove any lines inside the libraries codes where this name appears. You can see that I don’t use the strip command without any option as I do for the above step related to binaries program. This is very important and you should never use the strip command without the above option to strip libraries files on your system. Tuning IDE Hard Disk Performance Accessing a hard disk can be 50 to 100 times slower than reading data from RAM. File caches using RAM can alleviate this. However, low memory conditions will reduce the amount of memory available for the file-system cache slowing things down. File systems can also become heavily fragmented, slowing down disk accesses. Heavy use of symbolic links on Unix systems can slow down disk accesses too. Default Linux installs are also notorious for setting hard disk default settings which are tuned for compatibility and not for speed. Use the command hdparm to tune your Linux hard disk settings. The hdparm is a tool, which can be used to tune and improve the performance of your IDE hard disk. By default, any IDE drives you have in your Linux system are not optimized. Even if you have an ULTRA DMA system you will not be able to take full advantage of its speed if you are not using the hdparm tool to enable its features. This is because there is many different hard drive makes and models and Linux cannot know every feature of each one. Performance increases have been reported on massive disk I/O operations by setting the IDE drivers to use DMA, 32-bit transfers and multiple sector modes. The kernel seems to use more conservative settings unless told otherwise. The magic command to change the setting of your drive is hdparm. Before going into the optimization of your hard drive, it is important to verify that the hdparm package is installed in your system. If you have followed every step during the installation of Linux on your computer, then this package is not installed. To verify if hdparm package is installed on your system, use the command: [root@deep /]# rpm -q hdparm package hdparm is not installed If the hdparm package seems not to be installed, you’ll need to mount your CD-ROM drive containing the Linux CD-ROM Part 1 and install it.
- 129. General Optimization 0 CHAPTER5 129 • To mount the CD-ROM drive, use the following commands: [root@deep /]# mount /dev/cdrom /mnt/cdrom/ had: ATAPI 32X CD-ROM drive, 128kB Cache mount: block device dev/cdrom is write-protected, mounting read-only • To install the hdparm package on your Linux system, use the following command: [root@deep /]# cd /mnt/cdrom/RedHat/RPMS/ [root@deep RPMS]# rpm -Uvh hdparm-version.i386.rpm hdparm ################################################## • To unmount your CD-ROM drive, use the following command: [root@deep RPMS]# cd /; umount /mnt/cdrom/ Once hdparm package is installed on the system, it is time to go into the optimization of your hard drive. It is important to note that depending on your model and make, there will be some parameters that will apply and other that don’t. It is to your responsibility to know and understand your disk drive before applying any optimization parameters as described below. Finally, and especially for UltraDMA systems, it is vital to verify under your BIOS settings if the parameters related to DMA support on your computer are enabled or you will inevitably break your hard disk. You have been warned. Step 1 The first parameter applies to the majority of all modern drives and models in the market and enables 32-bit I/O over PCI buses. This option is one of the most important and will usually double the speed of your drive. • To enable 32-bit I/O over the PCI buses, use the following command: [root@deep /]# /sbin/hdparm -c3 /dev/hda (or hdb, hdc etc). This will usually, depending on your IDE Disk Drive model, cut the timing buffered disk reads time by two. The hdparm (8) manpage says that you may need to use “-c3” for many chipsets since it works with nearly all 32-bit IDE chipsets. All (E)IDE drives still have only a 16-bit connection over the ribbon cable from the interface card. Step 2 The second parameter applies only on standard DMA disk and will activate the simple DMA feature of the disk. This feature is for old disk drives with DMA capabilities. • To enable DMA, use the following command: [root@deep /]# /sbin/hdparm -d1 /dev/hda (or hdb, hdc etc). This may depend on support for your motherboard chipset being compiled into your kernel. Also, this command will enable DMA support for your hard drive only for interfaces which support DMA, it will cut the timing buffered disk reads time and will improve the performance by two.
- 130. General Optimization 0 CHAPTER5 130 Step 3 Multiword DMA mode 2, also known as ATA2 disk drive is the successor of the simple DMA drive. If you have this kind of hard drive, then you must enable the parameter in your Linux system. • To enable multiword DMA mode 2 transfers, use the following command: [root@deep /]# /sbin/hdparm -d1 -X34 /dev/hda (or hdb, hdc etc). This sets the IDE transfer mode for newer (E)IDE/ATA2 drives. (Check your hardware manual to see if you have it). Step 4 As for DMA mode 2, the UltraDMA mode 2 is an improvement of the DMA technology. If you have this kind of drive in your system, then choose this mode. • To enable UltraDMA mode 2 transfers, use the following command: [root@deep /]# /sbin/hdparm -d1 -X66 /dev/hda (or hdb, hdc etc) See your manual page about hdparm for more information. USE THIS OPTION WITH EXTREME CAUTION! Step 5 The UltraDMA mode 4 is one of the latest entries and one of the most popular at this time; it is also known and referred as ATA/66. I guess that most of you have this kind of drive installed and if it is the case then it is the one that you must choose for sure. • To enable UltraDMA mode4 transfers, use the following command: [root@deep /]# /sbin/hdparm -d1 -X12 -X68 /dev/hda (or hdb, hdc etc) This will enable UltraDMA ATA/66 mode on your drive. See your manual page about hdparm for more information. USE THIS OPTION WITH EXTREME CAUTION! Step 6 Multiple sector mode (aka IDE Block Mode), is a feature of most modern IDE hard drives, permitting the transfer of multiple sectors per I/O interrupt, rather than the usual one sector per interrupt. When this feature is enabled, it typically reduces operating system overhead for disk I/O by 30-50%. On many systems it also provides increased data throughput of anywhere from 5% to 50%. • To set multiple sector mode I/O, use the following command: [root@deep /]# /sbin/hdparm -mXX /dev/hda (or hdb, hdc etc) Where “XX” represent the maximum setting supported by your drive. The “-i” flag can be used to find the maximum setting supported by an installed drive: look for MaxMultSect in the output.
- 131. General Optimization 0 CHAPTER5 131 • To find the maximum setting of your drive, use the following command: [root@deep /]# /sbin/hdparm -i /dev/hda (or hdb, hdc etc) /dev/hda: Model=QUANTUM FIREBALLP LM15, FwRev=A35.0700, SerialNo=883012661990 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs } RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4 BuffType=3(DualPortCache), BuffSize=1900kB, MaxMultSect=16, MultSect=16 DblWordIO=no, OldPIO=2, DMA=yes, OldDMA=2 CurCHS=16383/16/63, CurSects=-66060037, LBA=yes, LBAsects=29336832 tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2 IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4 UDMA modes: mode0 mode1 mode2 mode3 *mode4 Step 7 The get/set sector count is used to improve performance in sequential reads of large files! The default setting is 8 sectors (4KB) and we will double and change it for 16. USE THIS OPTION WITH EXTREME CAUTION! • To improve the get/set sector count for file system read-ahead, use the command: [root@deep /]# /sbin/hdparm -a16 /dev/hda (or hdb, hdc etc) Step 8 The get/set interrupt-unmask flag will greatly improve Linux's responsiveness and eliminates "serial port overrun" errors. USE THIS OPTION WITH EXTREME CAUTION! • To improve and get/set interrupt-unmask flag for the drive, use the command: [root@deep /]# /sbin/hdparm -u1 /dev/hda (or hdb, hdc etc) Step 9 The IDE drive's write-caching feature will improve the performance of the hard disk. USE THIS OPTION WITH EXTREME CAUTION! • To enable the IDE drive's write-caching feature, use the following command: [root@deep /]# /sbin/hdparm -W1 /dev/hda (or hdb, hdc etc) Step 10 These options will allow the drive to retain your settings over a soft reset (as done during the error recovery sequence). It is important to note that not all drives support this feature. • To enables the drive to retain your settings, use the command: [root@deep /]# /sbin/hdparm -K1 -k1 /dev/hda (or hdb, hdc etc)
- 132. General Optimization 0 CHAPTER5 132 Step 11 Once every tuning related to your specific drive have been set, you can test the results and see if you want to keep them or not. • You can test the results of your changes by running hdparm in performance test mode: [root@deep /]# /sbin/hdparm -vtT /dev/hda (or hdb, hdc etc). /dev/hda: multcount = 16 (on) I/O support = 3 (32-bit w/sync) unmaskirq = 1 (on) using_dma = 1 (on) keepsettings = 1 (on) nowerr = 0 (off) readonly = 0 (off) readahead = 16 (on) geometry = 1826/255/63, sectors = 29336832, start = 0 Timing buffer-cache reads: 128 MB in 0.85 seconds = 150.59 MB/sec Timing buffered disk reads: 64 MB in 2.54 seconds = 25.20 MB/sec Once you have a set of hdparm options, you can put the commands in your /etc/rc.local file to run it every time you reboot the machine. When running from /etc/rc.local, you can add the “-q” option for reducing screen clutter. In my case, I will put the following configuration in the end of my rc.local file: /sbin/hdparm -q -c3 -d1 -X12 -X68 -m16 -a16 -u1 -W1 -k1 -K1 /dev/had
- 133. Kernel Security &Optimization IN THIS CHAPTER 1. Difference between a Modularized Kernel and a Monolithic Kernel 2. Making an emergency boot floppy 3. Preparing the Kernel for the installation 4. Applying the Grsecurity kernel patch 5. Obtaining and Installing Grsecurity 6. Tuning the Kernel 7. Cleaning up the Kernel 8. Configuring the Kernel 9. Compiling the Kernel 10. Installing the Kernel 11. Verifying or upgrading your boot loader 12. Reconfiguring /etc/modules.conf file 13. Rebooting your system to load the new kernel 14. Delete programs, edit files pertaining to modules 15. Making a new rescue floppy for Modularized Kernel 16. Making a emergency boot floppy disk for Monolithic Kernel
- 135. Kernel Security &Optimization 0 CHAPTER 6 135 Linux Kernel Abstract Well, our Linux server seems to be getting in shape now! But wait, what is the most important part of our server? Yes, it’s the kernel. The Linux kernel is the core of our operating system, and without it there is no Linux at all. So we must configure the kernel to fit our needs and compile only the features we really need. The new generation of Linux Kernel 2.4 was seemingly written with the server in mind. Many of the old limits, which prevented Linux being adopted in the “enterprise” market, have been lifted. The first thing to do next is to build a kernel that best suits your system. It’s very simple to do but, in any case, refer to the README file in the /usr/src/linux source directory after uncompressing the archive on your system. When configuring your kernel, only compile in code that you need. A few reasons that come to mind are: The Kernel will be faster (less code to run); You will have more memory (Kernel parts are NEVER swapped to the virtual memory); More stable (Ever probed for a non-existent card?); Unnecessary parts can be used by an attacker to gain access to the machine or other machines on the network. Modules are also slower than support compiled directly in the kernel. In our configuration and compilation we will firstly show you how to build a monolithic kernel, which is the recommended method for better performance and security and a modularized kernel for easily portability between different Linux systems. Monolithic kernel means to only answer yes or no to the questions (don’t make anything modular) and omits the steps: make modules and make modules_install. Difference between a Modularized Kernel and a Monolithic Kernel I don't want to go into deeply technical descriptions here. I'll try to stay as simple as I can in my explanation, this will allow us to better understand the differences. Firstly, it is evident that not all computers are identical; someone may have a new computer with the latest processor, a lot of memory, running on a SCSI sub-system with a good motherboard, where others may have an old computer with an older Pentium II processor, 128 MB of memory, on an IDE sub-system with a standard motherboard. These differences push kernel developers to constantly add or update for new drivers and features into the kernel code and are one of the reasons why a Modularized Kernel exists. Without all of those differences, it would be simple to provide a kernel where all the drivers and features are already included, but this is impossible because we all have different computers. Someone may say: “ok we can include all presently available drivers and features into the kernel and it will run on any computer”. This approach poses some problems. Firstly, it will make the kernel binary bigger and slower. Secondly, the Kernel will probe for nonexistent hardware, features and maintenance of other programs that directly depend on the kernel and will become more complicated.
- 136. Kernel Security &Optimization 0 CHAPTER 6 136 A solution was found and this was the Modularized Kernel approach. A technique that allows small pieces of compiled code to be inserted in or removed from the running kernel. In this way the Kernel will only load and run drivers and features that your computer have and will forget about the others. This practice is what all Linux vendors use to provide Linux kernels. They build and link every driver and feature as a module (which keeps the binary kernel smaller) that can be recognized and loaded if, and only if, they are needed by the kernel or the system. Kernel developers provide the ability to build a Modularized kernel, through an option that asks you during kernel configuration if you want to build the available drivers/features as a module. This option appears at the beginning of the Kernel configuration in the following form "Enable loadable module support (CONFIG_MODULES) [Y/n/?]". If you answer "Yes" here, then the compiled Kernel will be a Modularized Kernel and all future questions appearing during kernel configuration will give you the choice to compile the drivers/features into the Kernel code as a module by answering "m" for module, "y" for yes includes the code, or "n" do not include the code. Alternatively, if you answer "No" to the question "Enable loadable module support (CONFIG_MODULES) [Y/n/?]", then the corresponding Kernel will be a Monolithic kernel and all future questions appearing during kernel configuration will let you answer either "y" (yes, include the driver/feature) or "n" (no, do not include the drivers/feature). This allows you to build a Kernel where every driver/feature is compiled into it. To recap, Modularized Kernels allow small pieces of compiled code, which reside under the /lib/modules/2.4.x-x/ kernel directory to be inserted into or removed from the running kernel and a Monolithic Kernel contains the drivers/features into its compiled code. Some people will say that a loadable module is as good as hard-linked code. But what sort of speed difference is seen when using loadable modules instead of hard-linked code? Well, here’s an extract of the kernel mailing list archive: The immediate response from some was "almost nothing," but further consideration has shown this not to be true. There are, in fact, a number of costs associated with loadable modules. The biggest, perhaps, relates to how loadable modules are placed in kernel memory. The code for a module needs to live in a contiguous address space. The kernel sets up that address space with a function called vmalloc, which allocates memory with virtual addresses. In other words, a loadable module is in an address space that is visible to the kernel, but which is separate from where the core kernel code goes. This difference is important. The core kernel address space is a direct map of physical memory; it can be handled very efficiently in the processor's page table. Indeed, on some processors, a single page table entry covers the entire kernel. Space obtained from vmalloc, instead, uses one page table entry per memory page. A greater number of page table entries mean more lookups, and more translation buffer misses. One estimate is that the slowdown can be as much as 5%. Given this problem, why not load modules into the regular kernel memory space? Module code requires a contiguous address space. Since the standard kernel space is a direct map of physical memory, contiguous address spaces must also be contiguous in physical memory. Once the system has been running for a while, finding even two physically contiguous pages can be a challenge; finding enough to load a large module can be almost impossible. Modules also seem to have endemic problems with race conditions - it is possible, for example, for the kernel to attempt to access a newly-loaded module before it is fully initialized. Modules can also, in some situations, be removed while still in use. Such occurrences are obviously quite rare, but they can be catastrophic when they happen. The race conditions can be fixed with enough work, but that may require changing some fundamental kernel interfaces. In general, dealing with loadable modules is not an easy task; as one kernel hacker told us in a private message: "Doing live surgery on the kernel is never going to be pretty."
- 137. Kernel Security &Optimization 0 CHAPTER 6 137 These installation instructions assume Commands are Unix-compatible. The source path is /usr/src. Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Latest Kernel version number is 2.4.18 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages The following is based on information listed by the Linux Kernel Archives as of 2002/06/04. Please check http://www.kernel.org/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: Kernel Homepage: http://www.kernel.org/ Kernel FTP Site: 204.152.189.116 You must be sure to download: linux-2.4.18.tar.gz Prerequisites Depending on whether you want a firewall, user quota support with your system or if you have a SCSI/RAID controller, the Linux Kernel requires that the listed software below be already installed on your system to be able to compile successfully. If this is not the case, you must install them from your Linux CD-ROM or source archive files. Please make sure you have all of these programs installed on your system before proceeding with this chapter. iptables package, is the new secure and more powerful program used by Linux to set up firewalls as well as IP masquerading on your system. Install this package if you want to support Firewalls on your server. quota package, is a system administration tool for monitoring and limiting users' and/or groups' disk usage, per file system. Install this package if you want a tool to control the size of user’s directories on your server. mkinitrd package, creates filesystem images for use as initial ramdisk (initrd) images. These ramdisk images are often used to preload the block device modules (SCSI or RAID) needed to access the root filesystem. Install this package if you have a SCSI or RAID system where the Kernel is compiled as a Modularized Kernel. mkbootdisk package, creates a standalone boot floppy disk for booting the running system. Install this package only if you have a Modularized Kernel installed on your system. This package is not needed for Monolithic Kernel. The dosfstools package includes the mkdosfs and dosfsck utilities, which make and check MS-DOS FAT filesystems on hard drives or on floppies. You only need to install this package on Modularized Kernel. NOTE: For more information on Iptables Netfilter Firewall configuration or quota software, see the related chapters later in this book.
- 138. Kernel Security &Optimization 0 CHAPTER 6 138 Pristine source If you don’t use the RPM package to install the kernel, it will be difficult for you to locate all the files installed onto the system if you want to update your kernel in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install the kernel, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the kernel: [root@deep root]# find /* > Kernel1 • And the following one after you install the kernel: [root@deep root]# find /* > Kernel2 • Then use this command to get a list of what changed: [root@deep root]# diff Kernel1 Kernel2 > Kernel-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new kernel. In our example above, we use the /root directory of the system to store all the generated file lists. Making an emergency boot floppy The first step before going into the configuration and compilation of our new kernel is to create an emergency boot floppy in case something goes wrong during the build of your new Linux Kernel. Here we create the boot floppy for a modularized kernel since the kernel that is presently installed on our system should be a modularized kernel. We’ll see later that a method for creating the boot disk for mololithic kernel exists and is different from what we use here. To create the emergency boot floppy for a modularized kernel, follow these steps. Step1 We have to find the present modularized kernel version used on our system. We need this information to be able to create our emergency boot floppy disk. • To know which kernel version is running on your system, use the command: [root@dev /]# uname -a Linux dev 2.4.18-3 #1 Thu Apr 18 07:37:53 EDT 2002 i686 unknown From the above command, we know now that our kernel version is 2.4.18-3. Therefore we will use this information in the next step to create the boot disk. Step2 Once we know which kernel version we are currently running, we can use the command below to create the boot disk. • Put a floppy in your system and execute the following command as root: [root@deep /]# mkbootdisk --device /dev/fd0H1440 2.4.18-3 Insert a disk in /dev/fd0. Any information on the disk will be lost. Press <Enter> to continue or ^C to abort:
- 139. Kernel Security &Optimization 0 CHAPTER 6 139 NOTE: In this example, the current kernel on our system is version 2.4.18-3 and this is why we use “2.4.18-3” here. If you kernel version is different from what we use here, you just have to change the example version number for the one that you have according to the result returned by the “uname -a” command. Following these guidelines, you will now have a boot floppy with a known working kernel in case of problems with the upgrade. I recommend rebooting the system with the floppy to make sure that the floppy works correctly before continuing. Preparing the Kernel for the installation We have to copy the new kernel tar archive to the appropriate location on our server /usr/src and then we remove the old kernel from our system before installing a new one. Removing the old kernel will not freeze your computer until you try to reboot it before installing the new one because the Linux kernel resides in memory. Step 1 We must copy the archive file of the kernel to the /usr/src directory and move to this directory. • To copy the tar archive of the Linux kernel to the /usr/src directory, use the command: [root@deep /]# cp linux-version.tar.gz /usr/src/ • To move to the /usr/src directory, use the following command: [root@deep /]# cd /usr/src/ Step 2 Depending on how the Linux Kernel has been previously installed on your system, there are two ways to uninstall it, these are shown below. If you already have installed a Linux kernel with a tar archive before These steps are required ONLY if you have previously installed a Linux kernel from a tar archive. If it is a fresh, first install of the kernel, then uninstall the kernel-headers-version. i386.rpm, kernel-version.i386.rpm packages that are on your system. • Move to the /usr/src directory if you are not already in it with the following command: [root@deep /]# cd /usr/src/ • Remove the Linux symbolic link with the following command: [root@deep src]# rm -f linux • Remove the Linux kernel headers directory with the following command: [root@deep src]# rm -rf linux-2.4.x/ • Remove the Linux kernel with the following command: [root@deep src]# rm -f /boot/vmlinuz-2.4.x • Remove the Linux System.map file with the following command: [root@deep src]# rm -f /boot/System.map-2.4.x • Remove the Linux kernel modules directory (if available) with the following command: [root@deep src]# rm -rf /lib/modules/2.4.x/
- 140. Kernel Security &Optimization 0 CHAPTER 6 140 NOTE: Removing the old kernel modules is only required if you have installed a modularized kernel version before. If the modules directory doesn’t exist under the /lib/modules directory, it’s because your old kernel version is not a modularized kernel. If the original kernel’s RPM packages are installed on your system If the original kernel RPM packages are installed on your system, instead of the Linux kernel tar archive, which they would be if you have just finished installing your new Linux system, or have previously used an RPM package to upgrade your system, then use the following command to uninstall the Linux kernel: • Verify which kernel RPM packages are installed on your server with the command: [root@deep src]# rpm -qa | grep kernel kernel-2.4.18-3 • Also, verify if Kernel header package is installed on your server with the command: [root@deep src]# rpm -q glibc-kernheaders glibc-kernheaders-2.4-7.14 The above command shows us that kernel and glibc-kernheaders are the only kernel RPM packages installed on our system. We uninstall them as show below. • To uninstall the linux kernel RPM, use the following command: [root@deep src]# rpm -e --nodeps kernel glibc-kernheaders NOTE: If you receive an error message like: cannot remove /lib/modules/2.4.x directory, directory not empty, then remove the directory manually with command like: rm –rf /lib/modules/2.4.x/ form your system. This directory is related to the old kernel and it is not required for the new kernel we want to install. Step 3 Once we have uninstalled the old kernel and our new kernel tar archive has been copied to the /usr/src directory, we must uncompress it and remove the tar archive (linux-version. tar.gz) from the system to conserve disk space. • To uncompress the kernel, use the following command: [root@deep src]# tar xzpf linux-version.tar.gz • To remove the kernel tar archive from the system, use the following command: [root@deep src]# rm -f linux-version.tar.gz WARNING: If kernel compilation is something new for you, then it is recommended to keep the kernel tar archive (linux-version.tar.gz) until the end of the installation. This way, if you make a mistake during compilation, you have the source available to try again.
- 141. Kernel Security &Optimization 0 CHAPTER 6 141 Applying the Grsecurity kernel patch Grsecurity is a single patch file for the newest stable versions of Linux kernel that is an attempt to greatly improve the security of a Linux system. It mainly accomplishes this by physically patching the Linux kernel to make processes more restricted. Many other projects like Grsecurity exist on the Internet. There are some well know like the RSBAC (www.rsbac.org), LIDS (www.lids.org), SELinux (www.nsa.gov/selinux/), and OpenWall (www.openwall.com/linux/), projects, all of which fulfills only one part of a complete security system. What makes Grsecurity so different and better than these other projects is mainly because Grsecurity provides greatly needed additional security to Linux systems. In other words, it covers all features of the other projects and adds additional security features that the other projects do NOT cover. The types of added Grsecurity security are categorized as: 1. Complete Buffer Overflow Exploitation Protection. 2. File system Race Protection. 3. System Auditing. 4. Change-root Protections. 5. Portscan and OS Fingerprinting Protection. 6. Restricted Users. 7. Configurability Options. 8. Access Control. Grsecurity patch may change from version to version, and some may contain various other security features. It is important to use the Grsecurity patch that corresponds to the Linux Kernel version that you compile on your server. If you compile kernel version 2.4.18, you have to download Grsecurity patch for kernel version 2.4.18, etc. WARNING: When applying the Grsecurity patch to the Kernel, a new “Grsecurity” configuration section will be added at the end of your Linux Kernel configuration allowing you to configure and enable the security features that you want. Obtaining and Installing Grsecurity To obtain the Grsecurity patch suitable for your Linux Kernel, simply follow the "download" link on the Grsecurity home page, and download the file listed there. Grsecurity patch is available directly via http://www.grsecurity.org/. Remember that Grsecurity is specific to one kernel version, and as of this writing that version is 2.4.18. • To apply the Grsecurity patch to the Linux kernel, use the commands: [root@deep /]# cp grsecurity-1.9.4-2.4.18.patch /usr/src/ [root@deep /]# cd /usr/src/linux/ [root@deep linux]# patch -p1 < ../grsecurity-1.9.4-2.4.18.patch [root@deep linux]# cd ../ [root@deep src]# rm -f grsecurity-1.9.4-2.4.18.patch The step of patching your new kernel with Grsecurity patch is completed. Now follow the rest of this Kernel installation guide to build the Linux kernel and reboot your system. Configuration relating to Grsecurity will appears at the end of your Kernel configuration.
- 142. Kernel Security &Optimization 0 CHAPTER 6 142 Tuning the Kernel Ok, the old kernel has been uninstalled from our system; we have copied the new one to its appropriate location, uncompressed it, and added the Grsecurity patch (if wanted). Now, we must tune our new kernel to the maximum of its capabilities. All optimizations shown below are just increases of the default kernel parameters. • Edit the sem.h file (vi +66 /usr/src/linux/include/linux/sem.h) and change the following parameter: #define SEMMNI 128 /* <= IPCMNI max # of semaphore identifiers */ To read: #define SEMMNI 512 /* <= IPCMNI max # of semaphore identifiers */ • Edit the limits.h file (vi /usr/src/linux/include/linux/limits.h) and change the following parameters: #define NR_OPEN 1024 To read: #define NR_OPEN 8192 #define OPEN_MAX 256 /* # open files a process may have */ To read: #define OPEN_MAX 8192 /* # open files a process may have */ • Edit the posix_types.h file (vi +25 /usr/src/linux/include/linux/posix_types.h) and change the following parameter: #define __FD_SETSIZE 1024 To read: #define __FD_SETSIZE 8192 Finally, we must instruct the kernel to fit our specific CPU architecture and optimization flags. Depending on your CPU architecture and optimization flags, this step will improve the performance of the kernel. As an example with a PII 400MHz the BogoMIPS will become 799.54 instead of the default number of 400.00. Take note that it is not because BogoMIPS show you a number of 799.54 for a 400MHz CPU that your processor runs at this speed now. The BogoMIPS result can just be considered as a benchmark since it is a meaningless benchmark measurement.
- 143. Kernel Security &Optimization 0 CHAPTER 6 143 • Edit the Makefile file (vi +20 /usr/src/linux/Makefile) and change the line: HOSTCFLAGS = -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer To read: HOSTCFLAGS = -Wall -Wstrict-prototypes -O2 -march=i686 -funroll-loops - fomit-frame- pointer • Edit the Makefile file (vi +91 /usr/src/linux/Makefile) and change the line: CFLAGS := $(CPPFLAGS) -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strict-aliasing To read: CFLAGS := $(CPPFLAGS) -Wall -Wstrict-prototypes -O2 -march=i686 -funroll- loops -fomit-frame-pointer -fno-strict-aliasing WARNING: In the last example, we optimize the code for an i686 CPU architecture, if you have a different processor, you'll have to adjust the “-march=i686" options for your specific processor. Never compile the Kernel with optimization code number superior to “-O2”, this do nothing more and should produce an unstable kernel in some cases. Therefore use “-O2” at the maximum optimization number but never something superior to it. Cleaning up the Kernel It is important to make sure that your /usr/include/asm, and /usr/include/linux subdirectories are just symlinks to the kernel sources. Step 1 The asm and linux subdirectories are soft links to the real include kernel source header directories needed for our Linux architecture, for example /usr/src/linux/include/asm- i386 for asm. • To symlink the asm, and linux subdirectories to the kernel sources, type the following commands on your terminal: [root@deep src]# cd /usr/include/ [root@deep include]# rm -f asm linux [root@deep include]# ln -s /usr/src/linux/include/asm-i386 asm [root@deep include]# ln -s /usr/src/linux/include/linux linux This is a very important part of the configuration: we remove the asm, and linux directories under /usr/include then create new links that point to the same name directories under the new kernel source version directory. The /usr/include directory contains important header files needed by your kernel and programs to be able to compile on your system.
- 144. Kernel Security &Optimization 0 CHAPTER 6 144 WARNING: If the previously installed kernel in your system was made by RPM packages, then the asm and linux soft links will not exist since the uninstall of kernel-headers RPM package removes them automatically for you. Don’t forget to create them. Step 2 Make sure you have no stale .o files or dependencies lying around. • To be sure that we have no stale .o files or dependencies lying around, type the following commands on your terminal: [root@deep include]# cd /usr/src/linux/ [root@deep linux]# make mrproper WARNING: These two steps simply clean up anything that might have accidentally been left in the source tree by the development team. You should now have the source correctly installed. You can configure the kernel in one of three ways. The first method is to use the make config command. It provides you with a text-based interface for answering all the configuration options. You are prompted for all the options you need to set up your kernel. The second method is to use the make menuconfig command, which provides all the kernel options in an easy-to-use menu. The third is to use the make xconfig command (only available if the graphical interface of Linux is installed on the system), which provides a full graphical interface to all the kernel options. Step 3 For configuration in this guide, we will use the make config command because we have not installed the XFree86 Window Interface on our Linux server or the necessary packages to use the make menuconfig command. • Type the following commands on your terminal to load the kernel configuration: [root@deep /]# cd /usr/src/linux/ (if you are not already in this directory). [root@deep linux]# make config rm -f include/asm ( cd include ; ln -sf asm-i386 asm) /bin/sh scripts/Configure arch/i386/config.in # # Using defaults found in arch/i386/defconfig #
- 145. Kernel Security &Optimization 0 CHAPTER 6 145 Configuring the Kernel As soon as you enter make config at the prompt as described in the previous step, a list of kernel configurable options will be displayed for you to choose to configure the kernel, you must indicate what features and devices drivers you want to include on your system and select how to include support for specific devices. Typically, for each configuration option, you have to respond with one of the following choices: [y] To compile into the kernel and always be loaded. [m] To use a module for that feature and load that segment of code on demand. [n] To skip and excludes the support for that specific device from the kernel. WARNING: It is important to note that an [n] or [y] means the default choice. If a device does not have a modular device driver or you have not compiled the Kernel as a Modularized Kernel, you will not see the [m] option. Some time an [?] option will appear in the choices. This mean that you can get more information about the feature when you type the ? + ENTER key. Choosing the [?] help option opens another terminal describing the option. A new Linux kernel is very specific to our computer hardware since we have to choose the right drivers as well as features that we need to include and compile into the Kernel code. This implies a good understanding and knowledge of your computer hardware. It is simply inconceivable to build a Linux system if you don't know what hardware your computer has, especially if you spend money to buy a computer and then take time to configure it. Therefore we assume that you know all of your hardware and be aware that during the kernel configuration, you will be asked to answer some important questions related to your specific computer hardware. Be prepared to answer the following questions: 1. What type of processor do you have on your computer (i.e. Pentium III, AMD)? 2. How many processor do you have on your computer (i.e. 1, 2, 3)? 3. What kind of hard drive do you have on your computer (i.e. IDE, SCSI) ? 4. How many hard drives do you have on your computer? (i.e. want to make RAID)? 5. How much memories (RAM) do you have on your computer (i.e. 512 MB RAM)? 6. Do you have a network card? If so, who made it and what model is it? 7. Do you have a SCSI adapter? If so, who made it and what model is it? 8. Do you have a RAID system? If so, who made it and what model is it? 9. What type of mouse do you have (eg, PS/2, Microsoft, Logitech)? 10. If you have a serial mouse, what COM port is it connected to (eg, COM1)? 11. What is the make and model of your video card? All of the above questions are very important and if you don't know the answers for all of them, then it is recommended to get the information before going into Linux kernel configuration, compilation and installation. If you have all of the information, then you can read the rest of this guide.
- 146. Kernel Security &Optimization 0 CHAPTER 6 146 Monolithic kernel configuration As we know now, they are two possible different configurations for the kernel. The first is called a monolithic kernel the second is called a modularized kernel. Below we begin by showing you the configuration of a monolithic kernel which is to compile the required code and drivers directly into the kernel by answering the different kernel questions only by yes or no. Don’t forget to only compile code that you need and use. A new kernel is very specific to your computer hardware, in the monolithic kernel configuration part below; we have the following hardware for our example. Of course you must change them to fit whatever components you have in your system. 1 Pentium II 400 MHz (i686) processor 1 SCSI Motherboard 1 SCSI Hard Disk 1 SCSI Controler Adaptec AIC 7xxx 1 CD-ROM ATAPI IDE 1 Floppy Disk 2 Ethernet Cards Intel EtherExpressPro 10/100 1 Mouse PS/2 If you don’t want some of the options listed in the monolithic kernel configuration that I enable by default, answer n (for no) instead of y (for yes) to the related questions. If you want some other options that I disable, then answer y instead of n. Finally, the procedure of building a new kernel is quite long, therefore I recommend you to take your time. Some coffees and cigarettes will surely be welcome during these steps. # Using defaults found in arch/i386/defconfig # * * Code maturity level options * Prompt for development and/or incomplete code/drivers (CONFIG_EXPERIMENTAL) [N/y/?] Press Enter This option relates to features or drivers that are currently considered to be in the alpha-test phase and for a production system, it is highly recommended to answer by N to this question. Just hit [Enter] to approve the default choice, which is NO (Do not prompt for development and/or incomplete code/drivers). * * Loadable module support * Enable loadable module support (CONFIG_MODULES) [Y/n/?] n This option is very important since it asks us if we want to enable loadable module support into the Kernel. Remember that our goal in this part of the guide is to build a Monolithic Linux Kernel where all drivers and features are directly integrated and compiled into the Kernel code, therefore our answer to this question must be No (Do not enable loadable module support). This means that we have to enter n as the answer to this question to change the default value, which is Y for Yes.
- 147. Kernel Security &Optimization 0 CHAPTER 6 147 * * Processor type and features * Processor family (386, 486, 586/K5/5x86/6x86/6x86MX, Pentium-Classic, Pentium-MMX, Pentium-Pro/Celeron/Pentium-II, Pentium-III/Celeron(Coppermine), Pentium-4, K6/K6-II/K6-III, Athlon/Duron/K7, Crusoe, Winchip-C6, Winchip-2, Winchip-2A/Winchip-3, CyrixIII/C3) [Pentium-III/Celeron(Coppermine)] Pentium-4 This option asks you what type of processor you have on your computer and the default choice is in brackets [Pentium-III/Celeron(Coppermine)]. Therefore, if you processor is not a Pentium-III/Celerom(Coppermine) model, you have to enter your processor model here. The available choices are listed in the question. In the above example, I changed the default value [Pentium-III/Celeron(Coppermine)] and chose a Pentium-4 model. Toshiba Laptop support (CONFIG_TOSHIBA) [N/y/?] Press Enter This option asks you if you want to add a driver support for Toshiba Laptop. If you intend to run this kernel on a Toshiba portable, then you have to answer Yes to this question and change the default value which is No. In our example, we keep the default value of No by pressing the [Enter] key. Dell Inspiron 8000 support (CONFIG_I8K) [N/y/?] Press Enter As for the previous option, this one asks you if you want to add a driver support for Dell Inspiron 8000 Laptop. If you intend to run this kernel on a Dell Inspiron 8000, then you have to answer Yes to this question and change the default value which is No. In our example, we keep the default value of No by pressing the [Enter] key again. /dev/cpu/microcode - Intel IA32 CPU microcode support (CONFIG_MICROCODE) [N/y/?] Press Enter This option allows you to update the microcode on Intel processors in the IA32 family, e.g. Pentium Pro, Pentium II, Pentium III, Pentium 4, Xeon etc. If you say Y here and change the default value of N, then you'll need to say Y also to "/dev file system support" in the 'File systems' section below. In most situations, this option is simply not needed and you can safety keep the default setting of N by pressing the [Enter] key. /dev/cpu/*/msr - Model-specific register support (CONFIG_X86_MSR) [N/y/?] y This option allows you to enable a device that gives privileged processes access to the x86 Model-Specific Registers (MSRs). On multi-processor the MSR accesses are directed to a specific CPU. It is a good idea to answer Y to this option to enable it, even if you only have one processor on your system. Therefore answer Y to this question. /dev/cpu/*/cpuid - CPU information support (CONFIG_X86_CPUID) [N/y/?] y This option allows you to enable a device that gives processes access to the x86 CPUID instructions to be executed on a specific processor. As for the previous option, it is a good idea to change the default value of N to Y. High Memory Support (off, 4GB, 64GB) [off] Press Enter This option allows you to be able to use up to 64 Gigabytes of physical memory on x86 systems. Usually, and it’s true for most of us, we have a system that will never run with more than 1 Gigabyte total physical RAM and if this is you case, then answer "off" here (default choice and suitable for most users). In other part if you have a system that has between 1 and 4 Gigabytes physical RAM, then answer "4GB" here. If more than 4 Gigabytes is used then answer "64GB" here. In our example, we will configure the kernel to use a maximum of 1 Gigabyte total physical RAM by pressing the [Enter] key to accept the default choice "off".
- 148. Kernel Security &Optimization 0 CHAPTER 6 148 Math emulation (CONFIG_MATH_EMULATION) [N/y/?] Press Enter This option allows Linux to emulate a math coprocessor (a feature used for floating point operations). In general, only old 486SX and 386 processors do not have a a math coprocessor built in. All modern Pentium I, II, III, IV and later, AMD, Cyrix, etc have a math coprocessor built in and do not require this option to be turned on. If you have a very old 486SX or 386 processor in your computer, then you will have to change the default value of N here to become Y, but in general, everyone will keep the default value of N here since they have a math coprocessor built in their processor chip. MTRR (Memory Type Range Register) support (CONFIG_MTRR) [N/y/?] Press Enter This option on Intel family processors (Pentium Pro, Pentium II and later) allows the Memory Type Range Registers (MTRRs) to be used to control processor access to memory ranges. This is most useful if you have XFree86 graphical interface installed on your computer or have more than 1 processor on your system. Changing the default value of N here to become Y on a Linux system where a graphical interface or multiple processors are present will increase the performance of the system. If you don't use a graphical interface, or do not have more than one processor installed on you system, you must keep the default value of N by pressing the [Enter] key. Finally, it is important to note that this option is valid only if you have Intel family processors (Pentium Pro, Pentium II and later) installed on your computer. It doesn't work with AMD or Cyrix processors. Symmetric multi-processing support (CONFIG_SMP) [Y/n/?] n This option enables support for systems with more than one CPU. If you have a system with only one CPU, like most personal computers, say N. If you have a system with more than one CPU, say Y. In our example, we only have one processor installed on our computer and will change the default value of Y to become N. Local APIC support on uniprocessors (CONFIG_X86_UP_APIC) [N/y/?] (NEW) Press Enter This option appears only if you have answered N to the previous option and will let you enable support for local APIC on system with single processor. Local APIC (Advanced Programmable Interrupt Controller) is an integrated interrupt controller in the CPU that supports CPU-generated self-interrupts (timer, performance counters), and the NMI watchdog, which detects hard lockups on the server. On systems with single processor, you may have to answer Y here, but on systems with multiple processors, you don't have to answer this question since the kernel will automatically enable it. Usually we keep the default choice of N here since the feature is available only on modern Pentium processors like the IV family. Pentium I, II, and III don’t support this option, as well as AMD and Cyrix processors. * * General setup * Networking support (CONFIG_NET) [Y/n/?] Press Enter This option is very important and must be set to Y all the time, which is the default value. It allows your Linux system to support and use networking. We use the default value of Y by pressing the [Enter] key. PCI support (CONFIG_PCI) [Y/n/?] Press Enter This option allows us to enable or disable PCI support on Linux. In most cases, we have to say Y here. PCI's are the white slots on your motherboard where you can add network cards, video cards, etc. Since most of us use a PC and have this kind of slot available, it is important to say Y here by pressing the [Enter] key to use the default choice.
- 149. Kernel Security &Optimization 0 CHAPTER 6 149 PCI access mode (BIOS, Direct, Any) [Any] Press Enter On PCI systems, the BIOS can be used to detect the PCI devices and determine their configuration. However, some old PCI motherboards have BIOS bugs and may crash if this is done. Also, some embedded PCI-based systems don't have any BIOS at all. Linux can also try to detect the PCI hardware directly without using the BIOS. With this option, you can specify how Linux should detect the PCI devices. If you choose "BIOS", the BIOS will be used, if you choose "Direct", the BIOS won't be used, and if you choose "Any", the kernel will try the direct access method and falls back to the BIOS if that doesn't work. If unsure, go with the default, which is "Any" by pressing the [Enter] key. PCI device name database (CONFIG_PCI_NAMES) [Y/n/?] n This option lets you enable a feature that allows the Kernel to contain a database of all known PCI device names via different files under the /proc filesystem and make the information comprehensible to the user or disable this feature and get device ID numbers instead of names. Disabling this feature will save you about 80KB of kernel image size and will make the kernel smaller in size, which is a good thing for increased performance. If you disable this feature, the kernel will still run as usual, but will show you device ID numbers instead of names. Therefore, we will change the default value of Y (get PCI device by names) to become N (get PCI device by ID numbers) and will save 80KB of Kernel image size. EISA support (CONFIG_EISA) [N/y/?] Press Enter This option allows you to enable support for the Extended Industry Standard Architecture (EISA) bus. EISA is now obsolete by the PCI bus and you may enable this option only if you use some older ISA card on your system. Since most of us don't have these types of card, we can safety keep the default value of N here by pressing the [Enter] key. MCA support (CONFIG_MCA) [N/y/?] Press Enter This option allows us to enable MCA support with the kernel. MCA (MicroChannel Architecture) is found in some IBM PS/2 machines and laptops. It is a bus system similar to PCI or ISA. It is rare that we have this kind of bus on our system and we can safety keep the default value of N (do not support MCA on this kernel) by pressing the [Enter] key again. Support for hot-pluggable devices (CONFIG_HOTPLUG) [Y/n/?] n This option is often required only on laptop computer where you have a PCMCIA or PC-cards that you can plug or unplug at any time. If the kernel that you compile is made to run on a standard computer (not laptop, portable), then you have to say N here by changing the default option of Y to become N. System V IPC (CONFIG_SYSVIPC) [Y/n/?] Press Enter This option allows us to enable IPC on Linux. IPC (Inter Process Communication) is a suite of library functions and system calls which let processes (running programs) synchronize and exchange information. It is generally considered to be a good thing, and some programs won't run unless you say Y here. Therefore, we keep the default setting of Y by pressing the [Enter] key. BSD Process Accounting (CONFIG_BSD_PROCESS_ACCT) [N/y/?] Press Enter This option allows us to enable a user level programs, which will be able to instruct the kernel (via a special system call) to write process accounting information to a file. It is not vital or even really necessary to enable this kind of option to get a working Linux kernel. Therefore, we keep the default value of N here by pressing the [Enter] key.
- 150. Kernel Security &Optimization 0 CHAPTER 6 150 Sysctl support (CONFIG_SYSCTL) [Y/n/?] Press Enter This option is very important and especially with Linux kernel 2.4.x generation. It is the feature that allows us to dynamically change certain kernel parameters and variables on the fly through the /proc filesystem with the use of the /etc/sysctl.conf file. We keep the default value of Y by pressing the [Enter] key. Kernel core (/proc/kcore) format (ELF, A.OUT) [ELF] Press Enter This option allows a file under the /proc filesystem, which is called "kcore" to contain the Kernel core image in two different formats, which are ELF or A.OUT. For newer Linux systems, ELF (Executable and Linkable Format) is highly recommended and is the default option that we choose by pressing the [Enter] key. A.OUT (Assembler.OUTput) was the old format method used, but is now really obsolete. Kernel support for a.out binaries (CONFIG_BINFMT_AOUT) [Y/n/?] n This option is very important and allows us to provide support for A.OUT. A.OUT (Assembler.OUTput) is a set of formats for libraries and executables used in the earliest versions of UNIX and, as we said before, it is now really obsolete and was replaced with the ELF format. To be sure none of our programs will use this older executable format, we will change the default value of Y to become N. Kernel support for ELF binaries (CONFIG_BINFMT_ELF) [Y/n/?] Press Enter Here, it is important to answer Y to this question since this binary format is the one used now on modern Linux systems. ELF (Executable and Linkable Format) is a format for libraries and executables used across different architectures and operating systems. Many new executables are distributed solely in ELF format and you definitely want to say Y to this option by pressing the [Enter] key. Kernel support for MISC binaries (CONFIG_BINFMT_MISC) [Y/n/?] Press Enter This option enables wrapper-driven binary formats into the kernel. This feature is required by programs that need an interpreter to run, like Java, Python or Emacs-Lisp. Since we're sure to use one of these types of programs, it is safe to accept the default value of Y by pressing the [Enter] key. Power Management support (CONFIG_PM) [Y/n/?] n This option allows Power Management Support for your computer. With this feature parts of your computer are shut off or put into a power conserving "sleep" mode if they are not being used. In general it is good to enable this option on portable computer (Laptop) only. Note that, even if you say N here, Linux, on the x86 architecture, will issue the hlt instruction if nothing is to be done, thereby sending the processor to sleep and saving power. Therefore our choice will be N here. * * Memory Technology Devices (MTD) * Memory Technology Device (MTD) support (CONFIG_MTD) [N/y/?] Press Enter This option enables MTD (Memory Technology Devices) on your computer. MTD are flash, RAM and similar chips, often used for solid-state file systems on embedded devices. If you have some of these devices in your system, then answer Y to the question. In most cases, the default choice of N is recommended. * * Parallel port support * Parallel port support (CONFIG_PARPORT) [N/y/?] Press Enter If you want to use devices connected to your machine's parallel port like a printer, zip drive, etc, then you need to say Y here otherwise the default value N is recommended.
- 151. Kernel Security &Optimization 0 CHAPTER 6 151 * * Plug and Play configuration * Plug and Play support (CONFIG_PNP) [Y/n/?] n Plug and Play (PnP) is a standard for peripherals which allows those peripherals to be configured by software. If you answer to this question by Y, then Linux will be able to configure your Plug and Play devices. Under Linux, we really don't need to enable PNP support and our choice will be N here. * * Block devices * Normal PC floppy disk support (CONFIG_BLK_DEV_FD) [Y/n/?] Press Enter This option allows us to use the floppy disk drive(s) in our PC under Linux. Since everyone has and usually needs a floppy disk in their computer, the answer to this question will be Y. If you run a Linux server in highly secure environment, you could answer to this question by N since we never use floppy disk on this type of system. XT hard disk support (CONFIG_BLK_DEV_XD) [N/y/?] Press Enter This option allows us to enable the very old 8 bit hard disk controllers used in the IBM XT computer and since it's pretty unlikely that you have one of these then you must answer to this question by N. Therefore we simply press [Enter] because the default answer to this question is N. Compaq SMART2 support (CONFIG_BLK_CPQ_DA) [N/y/?] Press Enter This option allows us to enable support for the Compaq Smart Array controllers. If you use these kinds of boards in your system, then you should say Y here, otherwise the default value of N is recommended. Compaq Smart Array 5xxx support (CONFIG_BLK_CPQ_CISS_DA) [N/y/?] Press Enter This option allows us to enable support for the Compaq Smart Array 5xxx controllers. If you use these kinds of boards in your system, then you should say Y here, otherwise the default value of N is recommended. Mylex DAC960/DAC1100 PCI RAID Controller support (CONFIG_BLK_DEV_DAC960) [N/y/?] Press Enter This option enables support for the Mylex DAC960, AcceleRAID, and eXtremeRAID PCI RAID controllers. If you use these kinds of boards in your system, then you should say Y here, otherwise the default value of N is recommended. Loopback device support (CONFIG_BLK_DEV_LOOP) [N/y/?] Press Enter This option is a little different from other kernel options and if you enable it, you will be able to use a regular file as a block device that let you have access to some special advanced features and possibilities under Linux. The default value for this option will be ok for most of us and only advanced users or user that know why they need these features will enable the "Loopback device support" option. For SCSI systems with a modularized kernel, it is important to say Y here since SCSI drivers will use this device. Network block device support (CONFIG_BLK_DEV_NBD) [N/y/?] Press Enter This option enables another advanced feature under Linux, the possibility for your system to be a client for network block devices. The default value for this option would be ok for most of us and only advanced users or user that know why they need this feature will enable it.
- 152. Kernel Security &Optimization 0 CHAPTER 6 152 RAM disk support (CONFIG_BLK_DEV_RAM) [N/y/?] Press Enter This option will allow you to use a portion of your RAM memory as a block device, so that you can make file systems on it, read and write to it and do all the other things that you can do with normal block devices (such as hard drives). It is usually used to load and store a copy of a minimal root file system off of a floppy into RAM during the initial install of Linux. Again, most normal users won't need the RAM disk functionality and will answer N to this question. For SCSI systems with a modularized kernel, it is important to say Y here since SCSI drivers will use this feature. * * Multi-device support (RAID and LVM) * Multiple devices driver support (RAID and LVM) (CONFIG_MD) [N/y/?] Press Enter This option is required only for RAID and logical volume management (LVM). If you use them, then change the default value of N to become Y. * * Networking options * Packet socket (CONFIG_PACKET) [Y/n/?] Press Enter This option allows you to enable applications, which communicate directly with network devices without an intermediate network protocol implemented in the kernel like the tcpdump program. It is a good idea to enable this feature for most of us. Packet socket: mmapped IO (CONFIG_PACKET_MMAP) [N/y/?] y This option allows packet protocol driver to use an IO (Input/Output) mechanism that results in faster communication. Say Y here. Kernel/User netlink socket (CONFIG_NETLINK) [N/y/?] y This option allows us to enable two-way communication between the kernel and user processes. Say Y here. Routing messages (CONFIG_RTNETLINK) [N/y/?] (NEW) y If we have said Y to the previous option, we must say Y here too or the previous option will not work. Therefore our choice is Y. Netlink device emulation (CONFIG_NETLINK_DEV) [N/y/?] (NEW) y This option is a backward compatibility option, and we have to choose Y for now. Network packet filtering (replaces ipchains) (CONFIG_NETFILTER) [N/y/?] y This option enables support for a packet filter firewall on your system (Netfilter). It is very important to answer to this question by Y if you want to support firewall and IPTables on your computer. If you answer N to this question, then the firewalling features will not be available, even if your have the IPTables software installed on your system. Network packet filtering debugging (CONFIG_NETFILTER_DEBUG) [N/y/?] (NEW) y This option turns on support for debugging the netfilter code. It is a good idea to enable it. Socket Filtering (CONFIG_FILTER) [N/y/?] Press Enter This option allows us to enable Linux Socket Filter, a feature needed by PPP packet filtering in general. Therefore you only need to say Y here if you want to use PPP packet filtering on your system. Since we use a network card to get a connection to the Internet and not PPP (modem link), we keep the default value of N here.
- 153. Kernel Security &Optimization 0 CHAPTER 6 153 Unix domain sockets (CONFIG_UNIX) [Y/n/?] Press Enter This option is very important and must always be set to Y. It allows us to include support for Unix domain sockets; sockets are the standard Unix mechanism for establishing and accessing network connections. It is vital to say Y to this option. TCP/IP networking (CONFIG_INET) [Y/n/?] Press Enter Another very important and vital option under Linux. This option allows us to enable support for TCP/IP networking on the computer. TCP/IP are the protocols used on the Internet and on most local Ethernets. It is highly recommended to say Y here even if you are not connected to the Internet. IP: multicasting (CONFIG_IP_MULTICAST) [Y/n/?] n This option allows us to enable a code that addresses several networked computers at once. You need it if you intend to participate in the MBONE, a high bandwidth network on top of the Internet which carries audio and video broadcasts. For most people, it's safe to say N here. IP: advanced router (CONFIG_IP_ADVANCED_ROUTER) [N/y/?] n This option allows us to configure our Linux system to run mostly as a router. The answer to this question won't directly affect the kernel: answering N will just cause the configuration to skip all the questions about advanced routing. In many cases, we can safety keep the default value of N here. Only users that want to run their Linux system primarily as a router will answer to this question by Y to be presented a list of advanced routing features to enable or reject. If you want to configure your Linux server as a Gateway server, you need to answer Y to this question and all questions related to this option. IP: kernel level autoconfiguration (CONFIG_IP_PNP) [N/y/?] Press Enter This option must be set to Y only for diskless machines requiring network access to boot. For most people, it's safe to say N here. IP: tunneling (CONFIG_NET_IPIP) [N/y/?] Press Enter This option will enable Tunneling on our system. Tunneling is a means of encapsulating data of one protocol type within another protocol and sending it over a channel that understands the encapsulating protocol. This is an advanced feature and only advanced users who know why they need it must answer Y to this question. IP: GRE tunnels over IP (CONFIG_NET_IPGRE) [N/y/?] Press Enter This option will enable another kind of Tunneling feature on our system. This method is known as GRE (Generic Routing Encapsulation). This is an advanced feature and only advanced users that know why they need it must answer Y to this question. IP: TCP Explicit Congestion Notification support (CONFIG_INET_ECN) [N/y/?] Press Enter This option enables Explicit Congestion Notification on your system. ECN allows routers to notify clients about network congestion, resulting in fewer dropped packets and increased network performance. This is a very good feature but, unfortunately, there are many broken firewalls on the Internet, which refuse connections from ECN-enabled machines, and it may be a while before these firewalls are fixed. Until then, to access a site behind such a firewall you will have to disable this option by saying N here.
- 154. Kernel Security &Optimization 0 CHAPTER 6 154 IP: TCP syncookie support (disabled per default) (CONFIG_SYN_COOKIES) [N/y/?] y This option is very important and every one must answer to this question by Y because normal TCP/IP networking is open to an attack known as "SYN flooding". This denial-of-service attack prevents legitimate remote users from being able to connect to your computer during an ongoing attack and requires very little work from the attacker, who can operate from anywhere on the Internet. SYN cookies provide protection against this type of attack. Therefore don't forget to answer Y to this question. * * IP: Netfilter Configuration All questions under the "IP: Netfilter Configuration" section of the Kernel configuration are related to packet filter firewall support and features. We recommend you enable everything. Below, we show you the answer for each question without any explanation on the features. If you need to get more information about the features that you don't understand, you can simply type ? [Enter] at the prompt to get help. * Connection tracking (required for masq/NAT) (CONFIG_IP_NF_CONNTRACK) [N/y/?] (NEW) y FTP protocol support (CONFIG_IP_NF_FTP) [N/y/?] (NEW) y IRC protocol support (CONFIG_IP_NF_IRC) [N/y/?] (NEW) y IP tables support (required for filtering/masq/NAT) (CONFIG_IP_NF_IPTABLES) [N/y/?] (NEW) y limit match support (CONFIG_IP_NF_MATCH_LIMIT) [N/y/?] (NEW) y MAC address match support (CONFIG_IP_NF_MATCH_MAC) [N/y/?] (NEW) y netfilter MARK match support (CONFIG_IP_NF_MATCH_MARK) [N/y/?] (NEW) y Multiple port match support (CONFIG_IP_NF_MATCH_MULTIPORT) [N/y/?] (NEW) y TOS match support (CONFIG_IP_NF_MATCH_TOS) [N/y/?] (NEW) y LENGTH match support (CONFIG_IP_NF_MATCH_LENGTH) [N/y/?] (NEW) y TTL match support (CONFIG_IP_NF_MATCH_TTL) [N/y/?] (NEW) y tcpmss match support (CONFIG_IP_NF_MATCH_TCPMSS) [N/y/?] (NEW) y Connection state match support (CONFIG_IP_NF_MATCH_STATE) [N/y/?] (NEW) y Packet filtering (CONFIG_IP_NF_FILTER) [N/y/?] (NEW) y REJECT target support (CONFIG_IP_NF_TARGET_REJECT) [N/y/?] (NEW) y Full NAT (CONFIG_IP_NF_NAT) [N/y/?] (NEW) y Packet mangling (CONFIG_IP_NF_MANGLE) [N/y/?] (NEW) y TOS target support (CONFIG_IP_NF_TARGET_TOS) [N/y/?] (NEW) y MARK target support (CONFIG_IP_NF_TARGET_MARK) [N/y/?] (NEW) y LOG target support (CONFIG_IP_NF_TARGET_LOG) [N/y/?] (NEW) y TCPMSS target support (CONFIG_IP_NF_TARGET_TCPMSS) [N/y/?] (NEW) y WARNING: If you want to enable IPTables support into the kernel, the iptables program must be installed first or you will receive error messages during kernel compilation. This is because when IPTables support is enabled, the kernel will associate some part of the iptables program with it configuration. Therefore don’t forget to install IPTables before configuring kernel with IPTables support. Finally the same warning is true for quota support into the kernel. * * * The IPX protocol (CONFIG_IPX) [N/y/?] Press Enter This option allows us to enable Novell networking protocol support on Linux. You need it if you want to access Novell NetWare file or print servers using Linux or if you want to configure your Linux system to run as a Novell NetWare file or print server. In most cases, this is not required and we can answer N to this question.
- 155. Kernel Security &Optimization 0 CHAPTER 6 155 Appletalk protocol support (CONFIG_ATALK) [N/y/?] Press Enter This option allows us to enable AppleTalk protocol support on Linux. You need it if you want to access Apple computers using Linux. In most cases, this is not required and we can answer N to this question. DECnet Support (CONFIG_DECNET) [N/y/?] Press Enter This option allows us to enable DECnet networking protocol support on Linux. The DECnet networking protocol was used in many products made by Digital (now Compaq). In most cases, this is not required and we can answer N to this question. 802.1d Ethernet Bridging (CONFIG_BRIDGE) [N/y/?] Press Enter This option will allow your Linux system to act as an Ethernet bridge, which means that the different Ethernet segments it is connected to will appear as one Ethernet to the participants. Several such bridges can work together to create even larger networks of Ethernets using the IEEE 802.1 spanning tree algorithm. In most cases, this is not required and we can answer N to this question. * * QoS and/or fair queueing * QoS and/or fair queueing (CONFIG_NET_SCHED) [N/y/?] Press Enter This option allows us to enable QoS (Quality of Service) support on Linux. When the kernel has several packets to send out over a network device, it has to decide which ones to send first, which ones to delay, and which ones to drop. This is the job of the packet scheduler, and several different algorithms for how to do this "fairly" have been proposed. If we answer N to this question, the standard packet scheduler, which is a FIFO (first come, first served) will be used by default. The standard packet scheduler is enough for most of us and if you are running a router system or are an advanced user who wants to experiment in some new way with TCP/IP networking, then you can say Y to this question and be able to choose from among several alternative algorithms which can then be attached to different network devices. In most cases, we say N to this question. * * Telephony Support * Linux telephony support (CONFIG_PHONE) [N/y/?] Press Enter This option allows us to use a regular phone for voice-over-IP applications. This also means that you have to have a telephony card attached to your computer and you know what to do. Most people will simply answer N to this question. * * ATA/IDE/MFM/RLL support * ATA/IDE/MFM/RLL support (CONFIG_IDE) [Y/n/?] Press Enter This option allows the kernel to manage low cost mass storage units such as ATA/(E)IDE and ATAPI units. The most common cases are IDE hard drives and ATAPI CD-ROM drives and since we all have one of these devices attached to our computer we can safety say Y to this question. The only time that you can answer to this question by N is when you know that your system is pure SCSI. * * IDE, ATA and ATAPI Block devices * Enhanced IDE/MFM/RLL disk/cdrom/tape/floppy support (CONFIG_BLK_DEV_IDE) [Y/n/?] Press Enter This option allows us to enable the new enhanced driver with IDE/MFM/RLL disk/cdrom/tape/floppy drives. If you have one or more IDE drives, it is required to answer Y to this question.
- 156. Kernel Security &Optimization 0 CHAPTER 6 156 Use old disk-only driver on primary interface (CONFIG_BLK_DEV_HD_IDE) [N/y/?] Press Enter This option allows us to enable the old hard disk driver to control IDE/MFM/RLL disk/cdrom/tape/floppy drives. It is highly recommended not to enable it, since we've already used the new enhanced driver option with IDE/MFM/RLL disk/cdrom/tape/floppy drives above. This option may be useful for older systems. Include IDE/ATA-2 DISK support (CONFIG_BLK_DEV_IDEDISK) [Y/n/?] Press Enter This option allows us to enable another enhanced support for MFM/RLL/IDE hard disks. The only time that you can answer N to this question is when you know that your system is pure SCSI. Therefore, we'll enable support for this option by saying Y to the question. Use multi-mode by default (CONFIG_IDEDISK_MULTI_MODE) [Y/n/?] n This option allows us to fix possible error messages that can appear on IDE systems. This error message may look like: hda: set_multmode: status=0x51 { DriveReady SeekComplete Error } hda: set_multmode: error=0x04 { DriveStatusError } If you get this kind of error message on your system, then you have to say Y to this option. We suppose that you have a good IDE disk drive and that this error will never appear for you, in this case, we will change the default value of Y to become N. Include IDE/ATAPI CDROM support (CONFIG_BLK_DEV_IDECD) [Y/n/?] n This option allows us to instruct Linux to identify more than one CD-ROM drive along with other IDE devices, as "hdb" or "hdc", or something similar at boot time. If you have only one CD-ROM drive installed on your system, you can say N to this question even if it uses the ATAPI protocol and save some KB into the kernel. You have to say Y to this option only if you handle more than one CD-ROM drive in your computer. Since most people will usually have only one CD-ROM installed into their computer, we will change the default value of Y to become N. If you have one CD-ROM and another one CD-ROM for burning disk, then you have answer Y to this question. Include IDE/ATAPI TAPE support (CONFIG_BLK_DEV_IDETAPE) [N/y/?] Press Enter This option allows us to enable an IDE tape drive using the ATAPI protocol. If you have an IDE tape drive installed on your system, you must answer Y to this question. Most people will say N here. Include IDE/ATAPI FLOPPY support (CONFIG_BLK_DEV_IDEFLOPPY) [N/y/?] Press Enter This option allows us to enable an IDE floppy drive which uses the ATAPI protocol. If you have this kind of floppy drive installed on your system, you must answer this question Y. Most people will say N here. SCSI emulation support (CONFIG_BLK_DEV_IDESCSI) [N/y/?] Press Enter This option provides SCSI host adapter emulation for IDE ATAPI devices, and will allow us to use a SCSI device driver instead of a native ATAPI driver. If you intend to install and use a CD- RW drive on your computer, then you have to say Y here. Again, most people will say N here.
- 157. Kernel Security &Optimization 0 CHAPTER 6 157 * * IDE chipset support/bugfixes * CMD640 chipset bugfix/support (CONFIG_BLK_DEV_CMD640) [Y/n/?] n This option allows us to include code which tries to automatically detect and correct the CMD640 problems under Linux. The CMD-Technologies CMD640 IDE chip is used on many common 486 and Pentium motherboards, usually in combination with a "Neptune" or "SiS" chipset. Unfortunately, it has a number of rather nasty design flaws that can cause severe data corruption under many common conditions. To know if you need to enable this option for your system to correct this bug, edit the /proc/cpuinfo file and see if the parameter "f00f_bug" is set to no or yes. If the "f00f_bug" value is set to no, then you don't need to enable this option and can say N to the question, otherwise you have to say Y here. RZ1000 chipset bugfix/support (CONFIG_BLK_DEV_RZ1000) [Y/n/?] n This option allows us to include code which tries to automatically detect and correct the RZ1000 problems under Linux. The PC-Technologies RZ1000 IDE chip is used on many common 486 and Pentium motherboards, usually along with the "Neptune" chipset. As for the CMD640 bug above, it also has a number of rather nasty design flaws that can cause severe data corruption under many common conditions. To know if you need to enable this option for your system to correct this bug, edit the /proc/cpuinfo file and see if the parameter "coma_bug" is set to no or yes. If the "coma_bug" value is set to no, then you don't need to enable this option and can say N to the question, otherwise you have to say Y here. Generic PCI IDE chipset support (CONFIG_BLK_DEV_IDEPCI) [Y/n/?] Press Enter This option helps the IDE driver to automatically detect and configure all PCI-based IDE interfaces in your system. If you have PCI systems which use IDE drive(s), then say Y to this question. Most of us have PCI systems which use IDE and have answer Y this question. Sharing PCI IDE interrupts support (CONFIG_IDEPCI_SHARE_IRQ) [Y/n/?] Press Enter This option allows us to enable support for sharing a single IRQ with other cards under ATA/IDE chipsets. In general, everyone has answer Y this question. Generic PCI bus-master DMA support (CONFIG_BLK_DEV_IDEDMA_PCI) [Y/n/?] Press Enter This option allows us to reduce CPU overhead with IDE drive(s) on PCI system capable of bus- master DMA operation. If you have a modern IDE ATA/33/66/100 hard drive, then it is recommended to answer this question Y. Boot off-board chipsets first support (CONFIG_BLK_DEV_OFFBOARD) [N/y/?] Press Enter This option allows us to reverse the device scan order to improve the usability of some boot managers such as lilo when booting from a drive on an off-board controller. In many cases, this option is really not required and you can accept the default value of N here. Use PCI DMA by default when available (CONFIG_IDEDMA_PCI_AUTO) [Y/n/?] Press Enter A very important option to enable on all modern IDE disk drives. This option allows us to use the DMA feature of our disk drive under Linux to improve performance. Most people running a capable DMA drive will answer to this question by Y.
- 158. Kernel Security &Optimization 0 CHAPTER 6 158 All of the following Kernel options are related to the special onboard chipsets that you may have on your motherboard. Therefore, specific drivers are provided for each of them and you have to choose from the list the one that matches you chipset. If you have an Intel onboard chipset, then you can safety choose the default answer of N to all of the questions, since the kernel supports it naturally. Other chipset models must be selected from the list. In many cases, if your chipset is not listed, this means that it is automatically supported by the Kernel. Note that two options have their default answer set to Y (Intel PIIXn chipsets support (CONFIG_BLK_DEV_PIIX) [Y/n/?] and PIIXn Tuning support (CONFIG_PIIX_TUNING) [Y/n/?]). If you have a Pentium II or later processor, you must keep the default value of these two option to Y. AEC62XX chipset support (CONFIG_BLK_DEV_AEC62XX) [N/y/?] Press Enter ALI M15x3 chipset support (CONFIG_BLK_DEV_ALI15X3) [N/y/?] Press Enter CMD64X chipset support (CONFIG_BLK_DEV_CMD64X) [N/y/?] Press Enter CY82C693 chipset support (CONFIG_BLK_DEV_CY82C693) [N/y/?] Press Enter Cyrix CS5530 MediaGX chipset support (CONFIG_BLK_DEV_CS5530) [N/y/?] Press Enter HPT34X chipset support (CONFIG_BLK_DEV_HPT34X) [N/y/?] Press Enter HPT366 chipset support (CONFIG_BLK_DEV_HPT366) [N/y/?] Press Enter Intel PIIXn chipsets support (CONFIG_BLK_DEV_PIIX) [Y/n/?] Press Enter PIIXn Tuning support (CONFIG_PIIX_TUNING) [Y/n/?] Press Enter NS87415 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_NS87415) [N/y/?] Press Enter PROMISE PDC202{46|62|65|67|68} support (CONFIG_BLK_DEV_PDC202XX) [N/y/?] Press Enter ServerWorks OSB4/CSB5 chipsets support (CONFIG_BLK_DEV_SVWKS) [N/y/?] Press Enter SiS5513 chipset support (CONFIG_BLK_DEV_SIS5513) [N/y/?] Press Enter SLC90E66 chipset support (CONFIG_BLK_DEV_SLC90E66) [N/y/?] Press Enter Tekram TRM290 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_TRM290) [N/y/?] Press Enter VIA82CXXX chipset support (CONFIG_BLK_DEV_VIA82CXXX) [N/y/?] Press Enter Other IDE chipset support (CONFIG_IDE_CHIPSETS) [N/y/?] Press Enter IGNORE word93 Validation BITS (CONFIG_IDEDMA_IVB) [N/y/?] Press Enter * * SCSI support * SCSI support (CONFIG_SCSI) [Y/n/?] Press Enter This option allows us to enable SCSI hard disks, SCSI tape drives, SCSI CD-ROM's or any other SCSI devices under Linux. If you have a SCSI like system, you need to answer Y to this question. If you don't have any SCSI devices on your system, you can safety answer N to the question. For users that have a SCSI system, it is very important for you to know the name of your SCSI host adapter (the card inside your computer that "speaks" the SCSI protocol, also called SCSI controller), because you will be asked for it if you enable this option. Once again, if you don't have a SCSI system, simply answer N to this question and skip this section of the Linux Kernel configuration. * * SCSI support type (disk, tape, CD-ROM) * SCSI disk support (CONFIG_BLK_DEV_SD) [Y/n/?] Press Enter This option allows us to enable support for a SCSI hard disk under Linux. If you have enabled the SCSI support feature above because you have a SCSI hard drive on your system, then it's here that you have to specify it by answering Y to the question.
- 159. Kernel Security &Optimization 0 CHAPTER 6 159 Maximum number of SCSI disks that can be loaded as modules (CONFIG_SD_EXTRA_DEVS) [40] Press Enter This option allows us to control the amount of additional space allocated in tables for drivers that are loaded as modules after the kernel is booted. In the event that the SCSI core itself was loaded as a module, this value is the number of additional disks that can be loaded after the first host driver is loaded. Since we're compiling a Monolithic Kernel where no modules are available, this option doesn't concern us and we can safety press the [Enter] key to accept the default value. SCSI tape support (CONFIG_CHR_DEV_ST) [N/y/?] Press Enter This option allows us to enable support for a SCSI tape drive under Linux. If you have enabled the SCSI support feature above because you have a SCSI tape drive on your system, then it's here that you have to specify it by answering Y to the question. For most SCSI users, the answer is N (no, we don't have a SCSI tape drive on this computer). SCSI OnStream SC-x0 tape support (CONFIG_CHR_DEV_OSST) [N/y/?] Press Enter This option allows us to enable support for the OnStream SC-x0 SCSI tape drives under Linux. If you have this kind of SCSI tape drive installed on your computer, then you have to answer to this question by Y. Most SCSI users will simply say N to this question. SCSI CD-ROM support (CONFIG_BLK_DEV_SR) [N/y/?] Press Enter This option allows us to enable support for a SCSI CD-ROM under Linux. If you have enabled the SCSI support feature above because you have a SCSI CD-ROM on your system, then it's here that you have to specify it by answering Y to the question. For most SCSI users, the answer is N (no, we don't have a SCSI CD-ROM on this computer). SCSI generic support (CONFIG_CHR_DEV_SG) [N/y/?] Press Enter This option allows us to enable support to use SCSI scanners, synthesizers or CD-writers or just about anything having "SCSI" in its name other than hard disks, CD-ROMs or tapes. If you have one of these SCSI items installed on your computer, then you have to say Y here as well as for the "SCSI disk support" option above to enable the driver. VERY IMPORTANT NOTE: For users having IDE CD-writers, you have to say Y to this question too, even if your CD-writers are not SCSI CD-writers. Most SCSI users will simply say N to his question. * * Some SCSI devices (e.g. CD jukebox) support multiple LUNs * Enable extra checks in new queueing code (CONFIG_SCSI_DEBUG_QUEUES) [Y/n/?] Press Enter This option turns on a lot of additional consistency checking for the new queuing code on SCSI devices. It is a good idea to enable it by saying Y to the question. Probe all LUNs on each SCSI device (CONFIG_SCSI_MULTI_LUN) [Y/n/?] n This option force the SCSI driver to probe for multiple LUN's (Logical Unit Number) on your system and will certainly affect the performance of the system. Most SCSI users will simply disable this option by saying N to this question to improve performance. If you enable this option, then we assume that you know what you're doing.
- 160. Kernel Security &Optimization 0 CHAPTER 6 160 Verbose SCSI error reporting (kernel size +=12K) (CONFIG_SCSI_CONSTANTS) [Y/n/?] n This option allows any error messages regarding your SCSI hardware to be more understandable, this enlarges your kernel by about 12 KB. If performance is important to you, we highly recommend you to disable this option by answering the question N. SCSI logging facility (CONFIG_SCSI_LOGGING) [N/y/?] Press Enter This option allows us to turns on a logging facility that can be used to debug a number of SCSI related problems. Again, if performance is important to you, we highly recommend you to disable this option by keeping the default value of N here. * * SCSI low-level drivers Below you will be presented a list of available SCSI controllers to choose from, simply select the SCSI controller that is installed on your system and disable all the others. As an example, we will pretend that we have an Adaptec AIC7080 controller and will enable it further down. We chose an Adaptec AIC7080 model for our example; don't forget to change our choice if you have another kind of SCSI controller installed on your system. * 3ware Hardware ATA-RAID support (CONFIG_BLK_DEV_3W_XXXX_RAID) [N/y/?] Press Enter 7000FASST SCSI support (CONFIG_SCSI_7000FASST) [N/y/?] Press Enter ACARD SCSI support (CONFIG_SCSI_ACARD) [N/y/?] Press Enter Adaptec AHA152X/2825 support (CONFIG_SCSI_AHA152X) [N/y/?] Press Enter Adaptec AHA1542 support (CONFIG_SCSI_AHA1542) [N/y/?] Press Enter Adaptec AHA1740 support (CONFIG_SCSI_AHA1740) [N/y/?] Press Enter Adaptec AIC7xxx support (CONFIG_SCSI_AIC7XXX) [N/y/?] y Maximum number of TCQ commands per device (CONFIG_AIC7XXX_CMDS_PER_DEVICE) [253] (NEW) Press Enter Initial bus reset delay in milli-seconds (CONFIG_AIC7XXX_RESET_DELAY_MS) [15000] (NEW) Press Enter Build Adapter Firmware with Kernel Build (CONFIG_AIC7XXX_BUILD_FIRMWARE) [N/y/?] (NEW) Press Enter Adaptec I2O RAID support (CONFIG_SCSI_DPT_I2O) [N/y/?] Press Enter AdvanSys SCSI support (CONFIG_SCSI_ADVANSYS) [N/y/?] Press Enter Always IN2000 SCSI support (CONFIG_SCSI_IN2000) [N/y/?] Press Enter AM53/79C974 PCI SCSI support (CONFIG_SCSI_AM53C974) [N/y/?] Press Enter AMI MegaRAID support (CONFIG_SCSI_MEGARAID) [N/y/?] Press Enter BusLogic SCSI support (CONFIG_SCSI_BUSLOGIC) [N/y/?] Press Enter Compaq Fibre Channel 64-bit/66Mhz HBA support (CONFIG_SCSI_CPQFCTS) [N/y/?] Press Enter DMX3191D SCSI support (CONFIG_SCSI_DMX3191D) [N/y/?] Press Enter DTC3180/3280 SCSI support (CONFIG_SCSI_DTC3280) [N/y/?] Press Enter EATA ISA/EISA/PCI (DPT and generic EATA/DMA-compliant boards) support (CONFIG_SCSI_EATA) [N/y/?] Press Enter EATA-DMA [Obsolete] (DPT, NEC, AT&T, SNI, AST, Olivetti, Alphatronix) support (CONFIG_SCSI_EATA_DMA) [N/y/?] Press Enter EATA-PIO (old DPT PM2001, PM2012A) support (CONFIG_SCSI_EATA_PIO) [N/y/?] Press Enter Future Domain 16xx SCSI/AHA-2920A support (CONFIG_SCSI_FUTURE_DOMAIN) [N/y/?] Press Enter Intel/ICP (former GDT SCSI Disk Array) RAID Controller support (CONFIG_SCSI_GDTH) [N/y/?] Press Enter Generic NCR5380/53c400 SCSI support (CONFIG_SCSI_GENERIC_NCR5380) [N/y/?] Press Enter IBM ServeRAID support (CONFIG_SCSI_IPS) [N/y/?] Press Enter Initio 9100U(W) support (CONFIG_SCSI_INITIO) [N/y/?] Press Enter Initio INI-A100U2W support (CONFIG_SCSI_INIA100) [N/y/?] Press Enter
- 161. Kernel Security &Optimization 0 CHAPTER 6 161 NCR53c406a SCSI support (CONFIG_SCSI_NCR53C406A) [N/y/?] Press Enter NCR53c7,8xx SCSI support (CONFIG_SCSI_NCR53C7xx) [N/y/?] Press Enter SYM53C8XX Version 2 SCSI support (CONFIG_SCSI_SYM53C8XX_2) [N/y/?] Press Enter NCR53C8XX SCSI support (CONFIG_SCSI_NCR53C8XX) [N/y/?] Press Enter SYM53C8XX SCSI support (CONFIG_SCSI_SYM53C8XX) [Y/n/?] n PAS16 SCSI support (CONFIG_SCSI_PAS16) [N/y/?] Press Enter PCI2000 support (CONFIG_SCSI_PCI2000) [N/y/?] Press Enter PCI2220i support (CONFIG_SCSI_PCI2220I) [N/y/?] Press Enter PSI240i support (CONFIG_SCSI_PSI240I) [N/y/?] Press Enter Qlogic FAS SCSI support (CONFIG_SCSI_QLOGIC_FAS) [N/y/?] Press Enter Qlogic ISP SCSI support (CONFIG_SCSI_QLOGIC_ISP) [N/y/?] Press Enter Qlogic ISP FC SCSI support (CONFIG_SCSI_QLOGIC_FC) [N/y/?] Press Enter Qlogic QLA 1280 SCSI support (CONFIG_SCSI_QLOGIC_1280) [N/y/?] Press Enter Seagate ST-02 and Future Domain TMC-8xx SCSI support (CONFIG_SCSI_SEAGATE) [N/y/?] Press Enter Simple 53c710 SCSI support (Compaq, NCR machines) (CONFIG_SCSI_SIM710) [N/y/?] Press Enter Symbios 53c416 SCSI support (CONFIG_SCSI_SYM53C416) [N/y/?] Press Enter Tekram DC390(T) and Am53/79C974 SCSI support (CONFIG_SCSI_DC390T) [N/y/?] Press Enter Trantor T128/T128F/T228 SCSI support (CONFIG_SCSI_T128) [N/y/?] Press Enter UltraStor 14F/34F support (CONFIG_SCSI_U14_34F) [N/y/?] Press Enter UltraStor SCSI support (CONFIG_SCSI_ULTRASTOR) [N/y/?] Press Enter * * Fusion MPT device support * Fusion MPT (base + ScsiHost) drivers (CONFIG_FUSION) [N/y/?] Press Enter * * I2O device support * I2O support (CONFIG_I2O) [N/y/?] Press Enter This option allows us to enable support for Intelligent Input/Output (I2O) architecture. In order for this to work, you need to have an I2O interface adapter card in your computer. If you have this kind of I2O interface adapter card installed on your system, then you can say Y to the question and you will get a choice of interface adapter drivers and OSM's. Most users simply say N here. * * Network device support * Network device support (CONFIG_NETDEVICES) [Y/n/?] Press Enter This option is one of the most important and allows us to enable support and feature for network cards under Linux. Therefore, we have to answer Y to this question. * * ARCnet devices * ARCnet support (CONFIG_ARCNET) [N/y/?] Press Enter This option allows us to enable ARCnet chipset support under Linux. If you have a network card of this type installed on your system, then say Y here, otherwise and for most users, you have to keep the default value of N here. Dummy net driver support (CONFIG_DUMMY) [Y/n/?] n This option is only useful for PPP dial up modem users on Linux. If you don't use your system to make PPP connections, then you can safety answer N to this question. Only users who have a modem and want to use it to establish a connection via PPP or SLIP need to say Y here. Since, in our example, we use a network card to make a network connection, we will answer the question N.
- 162. Kernel Security &Optimization 0 CHAPTER 6 162 Bonding driver support (CONFIG_BONDING) [N/y/?] Press Enter This option allows us to 'bond' multiple Ethernet Channels together. This is called 'Etherchannel' by Cisco, 'Trunking' by Sun, and 'Bonding' in Linux. It is a technique to merge Ethernet segments together for doubling the speed of a connection. In most cases, we can safety choose the default choice of N here. You must have two Ethernet cards installed on your system and on the remote computer where you want to use this technique to be able to use it. Also, this is an advanced feature and only experienced Linux users may need it. Therefore, we will answer to the question by N. EQL (serial line load balancing) support (CONFIG_EQUALIZER) [N/y/?] Press Enter This option allows us to enable the same feature as the previous option, but this time for two modems and two telephone lines. Therefore, we will simply say N to this question. Universal TUN/TAP device driver support (CONFIG_TUN) [N/y/?] Press Enter This option allows us to enable TUN/TAP support under Linux. TUN/TAP provides packet reception and transmission for user space programs. It can be viewed as a simple Point-to-Point or Ethernet device, which instead of receiving packets from a physical media, receives them from user space program and instead of sending packets via physical media, writes them to the user space program. It is rare that we need this feature, if you need it then simply say Y here. Most users will say N here. * * Ethernet (10 or 100Mbit) * Ethernet (10 or 100Mbit) (CONFIG_NET_ETHERNET) [Y/n/?] Press Enter This option is very important and allows us to enable support for Ethernet Network Interface Cards (NIC's) under Linux. Now, everyone has a NIC in their computer and if you want to be able to use your network card, then you have to say Y here. Note that the answer to this question won't directly affect the kernel: saying N will just cause the configuration to skip all the questions about Ethernet network cards. We must say Y here to be able to select the network card that we have in our computer from the list of supported network cards. It is very important to know the name of the network card(s) installed in your system because you will be asked for it. As an example we will pretend that we have an "EtherExpressPro/100" network card in our computer and we will enable support for it. This is an example and don't forget to change our default choice if you have another kind of network card installed in your system. In general, we say Y for the network card that we have and N for all other network cards. Sun Happy Meal 10/100baseT support (CONFIG_HAPPYMEAL) [N/y/?] Press Enter Sun GEM support (CONFIG_SUNGEM) [N/y/?] Press Enter 3COM cards (CONFIG_NET_VENDOR_3COM) [N/y/?] Press Enter AMD LANCE and PCnet (AT1500 and NE2100) support (CONFIG_LANCE) [N/y/?] Press Enter Western Digital/SMC cards (CONFIG_NET_VENDOR_SMC) [N/y/?] Press Enter Racal-Interlan (Micom) NI cards (CONFIG_NET_VENDOR_RACAL) [N/y/?] Press Enter DEPCA, DE10x, DE200, DE201, DE202, DE422 support (CONFIG_DEPCA) [N/y/?] Press Enter HP 10/100VG PCLAN (ISA, EISA, PCI) support (CONFIG_HP100) [N/y/?] Press Enter Other ISA cards (CONFIG_NET_ISA) [N/y/?] Press Enter EISA, VLB, PCI and on board controllers (CONFIG_NET_PCI) [Y/n/?] Press Enter AMD PCnet32 PCI support (CONFIG_PCNET32) [N/y/?] Press Enter Apricot Xen-II on board Ethernet (CONFIG_APRICOT) [N/y/?] Press Enter CS89x0 support (CONFIG_CS89x0) [N/y/?] Press Enter DECchip Tulip (dc21x4x) PCI support (CONFIG_TULIP) [N/y/?] Press Enter Generic DECchip & DIGITAL EtherWORKS PCI/EISA (CONFIG_DE4X5) [N/y/?] Press Enter Digi Intl. RightSwitch SE-X support (CONFIG_DGRS) [N/y/?] Press Enter Davicom DM910x/DM980x support (CONFIG_DM9102) [N/y/?] Press Enter
- 163. Kernel Security &Optimization 0 CHAPTER 6 163 EtherExpressPro/100 support (CONFIG_EEPRO100) [Y/n/?] Press Enter Myson MTD-8xx PCI Ethernet support (CONFIG_FEALNX) [N/y/?] Press Enter National Semiconductor DP8381x series PCI Ethernet support (CONFIG_NATSEMI) [N/y/?] Press Enter PCI NE2000 and clones support (see help) (CONFIG_NE2K_PCI) [N/y/?] Press Enter RealTek RTL-8139 PCI Fast Ethernet Adapter support (CONFIG_8139TOO) [N/y/?] Press Enter SiS 900/7016 PCI Fast Ethernet Adapter support (CONFIG_SIS900) [N/y/?] Press Enter SMC EtherPower II (CONFIG_EPIC100) [N/y/?] Press Enter Sundance Alta support (CONFIG_SUNDANCE) [N/y/?] Press Enter TI ThunderLAN support (CONFIG_TLAN) [N/y/?] Press Enter VIA Rhine support (CONFIG_VIA_RHINE) [N/y/?] Press Enter Winbond W89c840 Ethernet support (CONFIG_WINBOND_840) [N/y/?] Press Enter Pocket and portable adapters (CONFIG_NET_POCKET) [N/y/?] Press Enter * * Ethernet (1000 Mbit) * Alteon AceNIC/3Com 3C985/NetGear GA620 Gigabit support (CONFIG_ACENIC) [N/y/?] Press Enter D-Link DL2000-based Gigabit Ethernet support (CONFIG_DL2K) [N/y/?] Press Enter National Semiconduct DP83820 support (CONFIG_NS83820) [N/y/?] Press Enter Packet Engines Hamachi GNIC-II support (CONFIG_HAMACHI) [N/y/?] Press Enter SysKonnect SK-98xx support (CONFIG_SK98LIN) [N/y/?] Press Enter FDDI driver support (CONFIG_FDDI) [N/y/?] Press Enter This option allows us to enable FDDI card support under Linux. FDDI (Fiber Distributed Data Interface) is a high speed local area network design to runs over copper or fiber. If you are connected to such a network and want a driver for the FDDI card in your computer, say Y here. Most users will simply say N here. PPP (point-to-point protocol) support (CONFIG_PPP) [N/y/?] Press Enter This option allows us to enable PPP support under Linux. PPP (Point to Point Protocol) is the protocol used by modern modems to establish a remote connection with your ISP. If you have a modem card installed on your system to make a remote connection with your ISP, then you need to answer Y to this question. If you don't use PPP to connect on the Internet, then you can safety say N here. In our example, we assume that you use another method, like a network interface, to connect to the Internet and say N here. SLIP (serial line) support (CONFIG_SLIP) [N/y/?] Press Enter This option allows us to enable SLIP or CSLIP support under Linux. These protocols are really old now and have been replaced by the PPP protocol (see the above option). If for any reason you still use them, then say Y here. Most users will answer this question N. * * Wireless LAN (non-hamradio) * Wireless LAN (non-hamradio) (CONFIG_NET_RADIO) [N/y/?] Press Enter This option allows us to enable support for wireless LANs and everything having to do with radio, but not with amateur radio or FM broadcasting. If you need such support on your system, then say Y here, also if you need Wireless Extensions with wireless PCMCIA (PC-) cards on Linux, you need to say Y here too. Most users will simply say N here. * * Token Ring devices * Token Ring driver support (CONFIG_TR) [N/y/?] Press Enter This option allows us to enable support for Token Ring under Linux. Token Ring is IBM's way of communication on a local network; the rest of the world uses Ethernet. If you need Token Ring support on your computer, then say Y here. Most people will select the default choice of N here.
- 164. Kernel Security &Optimization 0 CHAPTER 6 164 Fibre Channel driver support (CONFIG_NET_FC) [N/y/?] Press Enter This option allows us to enable Fibre Channel support under Linux. Fibre Channel is a high speed serial protocol mainly used to connect large storage devices to the computer. If you have a Fibre channel adapter card in your computer, then you can say Y here. Most users will simply say N here. * * Wan interfaces * Wan interfaces support (CONFIG_WAN) [N/y/?] Press Enter This option allows us to enable support for Wan interfaces under Linux. Wide Area Networks (WANs), such as X.25, frame relay and leased lines, are used to interconnect Local Area Networks (LANs) over vast distances. If you have these kinds of cards installed on your system, then you can answer Y to the question. Most users will say N here. * * Amateur Radio support * Amateur Radio support (CONFIG_HAMRADIO) [N/y/?] Press Enter This option allows us to enable Amateur Radio support under Linux. If you want to connect your Linux box to an amateur radio, answer Y here. Note that the answer to this question won't directly affect the kernel: saying N will just cause the configuration to skip all the questions about amateur radio. Most people will say N here. * * IrDA (infrared) support * IrDA subsystem support (CONFIG_IRDA) [N/y/?] Press Enter This option allows us to enable support for the IrDA (TM) protocols under Linux. IrDA (Infrared Data Associations) is a support for wireless infrared communication. Laptops or computers that use infrared and PDA's users will say Y here. Most users will say N here. * * ISDN subsystem * ISDN support (CONFIG_ISDN) [N/y/?] Press Enter This option allows us to enable ISDN support under Linux. ISDN (Integrated Services Digital Networks) is a special type of fully digital telephone service; it's mostly used to connect to your Internet service provider (with SLIP or PPP). If you have this type of card installed on your computer (popular in Europe), then you have to say Y here otherwise, you will certainly keep the default value of N here. * * Old CD-ROM drivers (not SCSI, not IDE) * Support non-SCSI/IDE/ATAPI CDROM drives (CONFIG_CD_NO_IDESCSI) [N/y/?] Press Enter If you have a CD-ROM drive that is neither SCSI nor IDE/ATAPI, say Y here, otherwise N. Note that the answer to this question doesn't directly affect the kernel: saying N will just cause the configuration to skip all the questions about these CD-ROM drives. Most users will say N here. * * Input core support * Input core support (CONFIG_INPUT) [N/y/?] Press Enter This option allows us to enable any of the USB HID (Human Interface Device) options in the USB support section which require Input core support. If you intended to use USB on Linux, say Y here otherwise, you can safety say N here. It is rare that we have to use USB on server systems.
- 165. Kernel Security &Optimization 0 CHAPTER 6 165 * * Character devices * Virtual terminal (CONFIG_VT) [Y/n/?] Press Enter This option is very important and allows us to enable support for terminal devices with display and keyboard devices. On Linux, you need at least one virtual terminal device in order to make use of your keyboard and monitor. Therefore, we have to say Y here. Support for console on virtual terminal (CONFIG_VT_CONSOLE) [Y/n/?] Press Enter This option allows us to enable support for consoles on virtual terminals. Most users will simply keep the default value of Y here. Standard/generic (8250/16550 and compatible UARTs) serial support (CONFIG_SERIAL) [Y/n/?] n This option allows us to use serial mice, modems and similar devices that connect to the standard serial ports under Linux. If you use your Linux system as a workstation with graphical interface installed, then you must answer Y to the question. If you use your Linux system for dedicated Ethernet WWW/FTP servers, then you can say N here. In our example, we assume that your Linux system is dedicated and configured to run as a server only and answer N to the question. Non-standard serial port support (CONFIG_SERIAL_NONSTANDARD) [N/y/?] Press Enter This option allows us to enable support for any non-standard serial boards under Linux. These are usually used for systems that need many serial ports, because they serve many terminals or dial-in connections. If this is not true in your case, then you can safety say N here. Most people will simply say N here. Unix98 PTY support (CONFIG_UNIX98_PTYS) [Y/n/?] Press Enter This option is important in modern Linux system and allows us to enable pseudo terminal (PTY) support under Linux. Everyone must say Y here. Information about pseudo terminal (PTY) can be found in the Kernel documentation. Maximum number of Unix98 PTYs in use (0-2048) (CONFIG_UNIX98_PTY_COUNT) [256] 128 This option allows us to set the maximum number of Unix98 PTY's that can be used at any one time. It is important to note that each additional set of 256 PTY's occupy approximately 8 KB of kernel memory on 32-bit architectures and for security as well as performance reasons, we must keep the number as low as possible. Therefore, we will change the default value of 256 to 128. * * I2C support * I2C support (CONFIG_I2C) [N/y/?] Press Enter This option allows us to enable I2C and SMBus support under Linux. IC2 is a slow serial bus protocol used in many micro controller applications and developed by Philips. If you need this feature, then say Y to the question otherwise say N here. Most people will say N here. Bus Mouse Support (CONFIG_BUSMOUSE) [N/y/?] Press Enter This option allows us to enable bus mouse support under Linux. All modern computer now use a serial mouse. If you still continue to use a bus mouse on your system, then you have to say Y here. Laptop users also need to say Y here. Most users will simply say N here.
- 166. Kernel Security &Optimization 0 CHAPTER 6 166 Mouse Support (not serial and bus mice) (CONFIG_MOUSE) [Y/n/?] Press Enter This option allows us to enable support for no serial or bus mice support under Linux. This is for machines with a mouse, which is neither a serial, nor a bus mouse. Examples are PS/2 mice. If you have a PS/2 mouse, then say Y here, otherwise say N here. Laptop and workstation users also need to say Y here. If you use your Linux system for dedicated Ethernet WWW/FTP servers, then you can say N here and save some space in your Kernel code. PS/2 mouse (aka "auxiliary device") support (CONFIG_PSMOUSE) [Y/n/?] Press Enter This option enables support for a PS/2 mouse on your system. The PS/2 mouse connects to a special mouse port that looks much like the keyboard port (small circular connector with 6 pins). If you have this kind of mouse (most modern computers and laptops have and use it), then say Y here otherwise say N. C&T 82C710 mouse port support (as on TI Travelmate) (CONFIG_82C710_MOUSE) [N/y/?] Press Enter This option allows us to enable support for a certain kind of PS/2 mouse used on the TI Travelmate. If you have this kind of mouse installed on your system, then say Y here. Most users will say N to this question. PC110 digitizer pad support (CONFIG_PC110_PAD) [N/y/?] Press Enter This drives the digitizer pad on the IBM PC110 palmtop. It can turn the digitizer pad into PS/2 mouse emulation with tap gestures or into an absolute pad. Most users will answer this question N. QIC-02 tape support (CONFIG_QIC02_TAPE) [N/y/?] Press Enter This option allows us to enable QIC-02 tape support under Linux. QIC-02 is a non-SCSI tape drive and if you use it, then says Y to the question. Most users will say N here. * * Watchdog Cards * Watchdog Timer Support (CONFIG_WATCHDOG) [N/y/?] Press Enter This option enables Watchdog Timer support under Linux. For details about watchdog, please read Documentation/watchdog.txt in the kernel source. Most users will say N here. Intel i8x0 Random Number Generator support (CONFIG_INTEL_RNG) [N/y/?] Press Enter This option allows us to enable support for a driver that provides kernel-side support for the Random Number Generator hardware found on Intel i8xx-based motherboards. If you have this generator hardware installed on your computer, then you can say Y to the question. Most people will say N here. /dev/nvram support (CONFIG_NVRAM) [N/y/?] Press Enter This option allows us to get read and write access to the 50 bytes of non-volatile memory in the real time clock (RTC). This memory is conventionally called "CMOS RAM" on PCs and may be used to view settings there, or to change them (with some utility). Most users will say N to this question. Enhanced Real Time Clock Support (CONFIG_RTC) [N/y/?] Press Enter This option allows us to get access to the real time clock (or hardware clock) built into the computer. If you run Linux on a multiprocessor machine and said Y to "Symmetric Multi Processing" above, you should say Y here to read and set the RTC in an SMP compatible fashion.
- 167. Kernel Security &Optimization 0 CHAPTER 6 167 Double Talk PC internal speech card support (CONFIG_DTLK) [N/y/?] Press Enter This option allows us to enable support for the DoubleTalk PC under Linux. If you have a speech card installed on your computer, then answer to this question by Y. Most users will say N here. Siemens R3964 line discipline (CONFIG_R3964) [N/y/?] Press Enter This option allows synchronous communication with devices using the Siemens R3964 packet protocol. Unless you are dealing with special hardware like PLC's, you are unlikely to need this. Most users will simply say N here. Applicom intelligent fieldbus card support (CONFIG_APPLICOM) [N/y/?] Press Enter This option provides the kernel-side support for the intelligent fieldbus cards made by Applicom International. Unless you are dealing with such a card, you are unlikely to need this. Most users will simply say N here. * * Ftape, the floppy tape device driver * Ftape (QIC-80/Travan) support (CONFIG_FTAPE) [N/y/?] Press Enter This option allows us to enable support for some well know tape drives under Linux. If you enable this option, then you'll be asked to select your make and model from the available list. If you don't use tape drive on your computer, then you can say N to this option. /dev/agpgart (AGP Support) (CONFIG_AGP) [Y/n/?] n This option is only pertinent if you use XFree86 on your computer and want to use the GLX or DRI features for better performance. Enable this option only if you use graphical interface on your computer. If you system is a server, then you really don't need this option. In our example, we are configuring the kernel for a server purpose, therefore, we will say N here. Direct Rendering Manager (XFree86 4.1.0 and higher DRI support) (CONFIG_DRM) [Y/n/?] n This option is directly related to the use of XFree86 and graphical interface on your computer. It allows us to enable Kernel-level support for the Direct Rendering Infrastructure (DRI) for better performance of the system. If your computer doesn't have a graphical interface installed, and is configured as a server, then you don't need to enable this option. If you say Y here because you have and use a graphical interface, then you need to select the module that's right for your graphics card from the available list. In our example, we are configuring the kernel for a server purpose, therefore, we will say N here. ACP Modem (Mwave) support (CONFIG_MWAVE) [N/y/?] Press Enter This option allows us to enable ACP modem (Mwave) support under Linux. ACP is a WinModem composed of a kernel driver and a user level application. Together these components support direct attachment to public switched telephone networks (PSTNs) and support selected countries. Most user will simply say N here. * * Multimedia devices * Video For Linux (CONFIG_VIDEO_DEV) [N/y/?] Press Enter This option allows us to enable support for audio/video capture and overlay devices and FM radio cards under Linux. If you use graphical interface on your system, you need to say Y here. If your system is configured as a server and you don't have graphical interface installed on it, then you can safety say N here. In our example, we are configuring the kernel for a server purpose, therefore, we will say N here.
- 168. Kernel Security &Optimization 0 CHAPTER 6 168 * * File systems * Quota support (CONFIG_QUOTA) [N/y/?] y This option is important to allow us to set per user/group limits for disk usage (also called disk quotas) under Linux. It is sometimes required on server environments where such a purpose is required. If you use Linux as a workstation with graphical interface, it is not useful to enable it. Only enable this option on server environment when required. Kernel automounter support (CONFIG_AUTOFS_FS) [N/y/?] Press Enter This option allows us to automatically mount remote file systems on demand. A newer version of the automounter with more features is now available; therefore we really don't need to enable this option. Say N here. Kernel automounter version 4 support (also supports v3) (CONFIG_AUTOFS4_FS) [Y/n/?] n This option allows us to enable support for the new automounter tool. If you are not a part of a fairly large, distributed network or don't have a laptop which needs to dynamically reconfigure to the local network, you probably do not need an automounter, and can say N here. Ext3 journalling file system support (EXPERIMENTAL) (CONFIG_EXT3_FS) [N/y/?] y This option is very important and allows us to enable support for the new EXT3 journaling file system support under Linux. EXT3 is the journaling version of the Second extended file system (often called ext3), the de facto standard Linux file system (method to organize files on a storage device) for hard disks. Most users will say Y here to be able to get advantage of the new EXT3 journaling file system. JBD (ext3) debugging support (CONFIG_JBD_DEBUG) [N/y/?] y This option allows us to enable debugging output while the system is running EXT3 under Linux, in order to help track down any problems you are having. It is a good idea to enable it, therefore answer Y to the question. DOS FAT fs support (CONFIG_FAT_FS) [N/y/?] Press Enter If you want to use one of the FAT-based file systems (the MS-DOS, VFAT (Windows 95) and UMSDOS (used to run Linux on top of an ordinary DOS partition) file systems, then you must say Y here. In many case, we really don't need this kind of feature and can safety say N here. Recommended choice is N and especially for a linux server system. Compressed ROM file system support (CONFIG_CRAMFS) [N/y/?] Press Enter This option allows us to enable CramFs support under Linux. CramFs (Compressed ROM File System) is designed to be a simple, small, and compressed file system for ROM based embedded systems. Most users will simply say N here. Virtual memory file system support (former shm fs) (CONFIG_TMPFS) [Y/n/?] Press Enter This option allows us to enable support for Tmpfs under Linux. Tmpfs is a file system which keeps all files in virtual memory. It is a good idea to enable it on your computer. Simple RAM-based file system support (CONFIG_RAMFS) [N/y/?] Press Enter This option allows us to enable support for Ramfs under Linux. Ramfs is a file system which keeps all files in RAM. It allows read and write access. Most users will say N here. ISO 9660 CDROM file system support (CONFIG_ISO9660_FS) [Y/n/?] Press Enter This option is important and allows Linux to access and read your CD-ROM. Since everyone has and uses a CD-ROM, it is vital to say Y to this question.
- 169. Kernel Security &Optimization 0 CHAPTER 6 169 Microsoft Joliet CDROM extensions (CONFIG_JOLIET) [N/y/?] Press Enter This option allows us to be able to read Joliet CD-ROM's under Linux. Joliet is a Microsoft extension for the ISO 9660 CD-ROM file system, which allows for long filenames in unicode format. If you think that you'll need to access some files written in this format under Linux with your CD-ROM, then you need to say Y here. For workstations, you may need to say Y here but for Linux servers, you have to say N here. Transparent decompression extension (CONFIG_ZISOFS) [N/y/?] Press Enter This is a Linux-specific extension to RockRidge which lets you store data in compressed form on a CD-ROM and have it transparently decompressed when the CD-ROM is accessed. Again, for workstations you may need it but for servers, you don't need it. Say N. Minix fs support (CONFIG_MINIX_FS) [N/y/?] Press Enter This option allows us to enable Minix fs support with Linux. The minix file system (method to organize files on a hard disk partition or a floppy disk) was the original file system for Linux, but has been superseded by the second extended file system (ext2fs). Simply say N here. FreeVxFS file system support (VERITAS VxFS(TM) compatible) (CONFIG_VXFS_FS) [N/y/?] Press Enter FreeVxFS is a file system driver that support the VERITAS VxFS(TM) file system format. VERITAS VxFS(TM) is the standard file system of SCO UnixWare (and possibly others) and optionally available for Sunsoft Solaris, HP-UX and many other operating systems. If you want to support such file system on your computer, then say Y here otherwise say N. NTFS file system support (read only) (CONFIG_NTFS_FS) [N/y/?] Press Enter NTFS is the file system of Microsoft Windows NT. Say Y if you want to get read access to files on NTFS partitions of your hard drives otherwise say N here. Most users will say N. OS/2 HPFS file system support (CONFIG_HPFS_FS) [N/y/?] Press Enter OS/2 is IBM's operating system for PC's, the same as Warp, and HPFS is the file system used for organizing files on OS/2 hard disk partitions. Say Y if you want to be able to read files from and write files to an OS/2 HPFS partition on your hard drive. Most users will say N here. /proc file system support (CONFIG_PROC_FS) [Y/n/?] Press Enter This is a virtual file system providing information about the status of the system. Several programs depend on this, so everyone should say Y here. /dev/pts file system for Unix98 PTYs (CONFIG_DEVPTS_FS) [Y/n/?] Press Enter This option allows us to get a virtual file system which can be mounted on /dev/pts with "mount -t devpts". This, together with the pseudo terminal master multiplexer /dev/ptmx, is used for pseudo terminal support as described in The Open Group's Unix98 standard. Again, everyone should say Y here. ROM file system support (CONFIG_ROMFS_FS) [N/y/?] Press Enter This is a very small read-only file system mainly intended for initial ram disks of installation disks, but it could be used for other read-only media as well. Most people will simply say N to this question. If you want to run SCSI system on modularized kernel, you should say Y here. Second extended fs support (CONFIG_EXT2_FS) [Y/n/?] Press Enter This is the de facto standard Linux file system (method to organize files on a storage device) for hard disks and you must say Y to this question.
- 170. Kernel Security &Optimization 0 CHAPTER 6 170 System V/Xenix/V7/Coherent file system support (CONFIG_SYSV_FS) [N/y/?] Press Enter SCO, Xenix and Coherent are commercial Unix systems for Intel machines, and Version 7 was used on the DEC PDP-11. Saying Y here would allow you to read from their floppies and hard disk partitions. Most people will say N to this question. UDF file system support (read only) (CONFIG_UDF_FS) [N/y/?] Press Enter This is the new file system used on some CD-ROMs and DVD's. Say Y if you intend to mount DVD discs or CD-RW's written in packet mode, or if written to by other UDF utilities, such as DirectCD. Only enable this option if you have some such need. UFS file system support (read only) (CONFIG_UFS_FS) [N/y/?] Press Enter BSD and derivative versions of Unix (such as SunOS, FreeBSD, NetBSD, OpenBSD and NeXTstep) use a file system called UFS. Some System V Unixes can create and mount hard disk partitions and diskettes using this file system as well. Saying Y here will allow you to read from these partitions. Most users will say N here. * * Network File Systems * Coda file system support (advanced network fs) (CONFIG_CODA_FS) [N/y/?] Press Enter Coda is an advanced network file system, similar to NFS in that it enables you to mount file systems of a remote server and access them with regular Unix commands as if they were sitting on your hard disk. Enable this option only if you need it otherwise disable it. NFS file system support (CONFIG_NFS_FS) [Y/n/?] n If you are connected to some other (usually local) Unix computer (using SLIP, PLIP, PPP or Ethernet) and want to mount files residing on that computer (the NFS server) using the Network File Sharing protocol, say Y here. NFS server support (CONFIG_NFSD) [Y/n/?] n If you want your Linux box to act as an NFS *server*, so that other computers on your local network which support NFS can access certain directories on your box transparently, you have two options: you can use the self-contained user space program nfsd, in which case you should say N here, or you can say Y and use the kernel based NFS server. The advantage of the kernel based solution is that it is faster. Finally, if you don't want to support NFS server at all, simply say N here. SMB file system support (to mount Windows shares etc.) (CONFIG_SMB_FS) [N/y/?] Press Enter SMB (Server Message Block) is the protocol Windows for Workgroups (WfW), Windows 95/98, Windows NT, 2000, XP and OS/2 Lan Manager use to share files and printers over local networks. Saying Y here allows you to mount their file systems (often called "shares" in this context) and access them just like any other Unix directory. Enable this option only if you need it. In most cases the answer to this question will be N even if you install Samba on your system. NCP file system support (to mount NetWare volumes) (CONFIG_NCP_FS) [N/y/?] Press Enter NCP (NetWare Core Protocol) is a protocol that runs over IPX and is used by Novell NetWare clients to talk to file servers. It is to IPX what NFS is to TCP/IP, if that helps. Saying Y here allows you to mount NetWare file server volumes and to access them just like any other Unix directory. Enable this option only if you need it. In most cases the answer to this question will be N. You do not have to say Y here if you want your Linux box to act as a file *server* for Novell NetWare clients.
- 171. Kernel Security &Optimization 0 CHAPTER 6 171 * * Partition Types * Advanced partition selection (CONFIG_PARTITION_ADVANCED) [N/y/?] Press Enter This option allows us to enable the use of hard disks under Linux which were partitioned under an operating system running on a different architecture than the Linux system. Note that the answer to this question won't directly affect the kernel: saying N will just cause the configuration to skip all the questions about foreign partitioning schemes. * * Console drivers * VGA text console (CONFIG_VGA_CONSOLE) [Y/n/?] Press Enter This option allows us to use Linux in text mode through a display that complies with the generic VGA standard. Virtually everyone wants that. Everyone should say Y here. Video mode selection support (CONFIG_VIDEO_SELECT) [N/y/?] Press Enter This option allows us to enable support for text mode selection on kernel startup. In general we don't need to enable this option. Say N here and read the file Documentation/svga.txt for more information about the Video mode selection support if you are curious. * * Sound * Sound card support (CONFIG_SOUND) [Y/n/?] n This option allows us to enable sound support under Linux. If you have a sound card in your computer, then say Y here and select from the available list of sound card the one that you have. If you run Linux as a workstation, you may need to say Y here, if you run Linux as a server, you really don't need to enable this option and can safety say N here. * * USB support * Support for USB (CONFIG_USB) [Y/n/?] n This option allows us to enable USB support under Linux. If your computer has a USB port and you want to use USB devices, then you have to say Y here. For servers you really don't need to say Y here and can safety say N. * * Kernel hacking * Kernel debugging (CONFIG_DEBUG_KERNEL) [N/y/?] Press Enter You have to say Y here only if you are developing drivers or trying to debug and identify kernel problems. Most users will simply say N here. * * Grsecurity * Grsecurity (CONFIG_GRKERNSEC) [N/y/?] y This option allows us to enable Grsecurity support under Linux. If you say Y to this option, you will be able to configure many Grsecurity features that will enhance the security of your Linux system in many ways. This option is available ONLY if you have patched your Linux kernel with the Grsecurity patch as discussed previously in this chapter. For best security of your Linux server say Y here. Security level (Low, Medium, High, Customized) [Customized] Press Enter This Grsecurity option allows us to choose between three predefined Grsecurity configurations or to customize the Grsecurity configuration as we want. For better control of what the patch does and what we may or may not need, we chose to have full control about what should or shouldn’t be enable with Grsecurity by pressing the [Enter] key to accept the default choice of [Customized], which let us see all available security features and chose only the one we need for our server and kernel security setup.
- 172. Kernel Security &Optimization 0 CHAPTER 6 172 * * Buffer Overflow Protection * Openwall non-executable stack (CONFIG_GRKERNSEC_STACK) [N/y/?] y This Grsecurity option allows us to enable the non-executable stack protection on the system. If you say Y here, your system will not allow execution of code on the stack, making buffer overflow exploitation more difficult. It’s a good idea to say Y here. Gcc trampoline support (CONFIG_GRKERNSEC_STACK_GCC) [N/y/?] Press Enter This Grsecurity option allows us to support trampoline code along with the above stack protection. Trampolining is an action to use the ability of the stack to contain executable code, which can improve program efficiency in a few cases. Since few programs and some version of GCC use and need 'trampolining', it is preferable to NOT enable this option to avoid break of some program on the system. Say N here by pressing the [Enter] key. Read-only kernel memory (CONFIG_GRKERNSEC_KMEM) [N/y/?] y This Grsecurity option allows us to enable the read-only kernel memory protection on the system. If you say Y here, root will not be able to modify the contents of kernel memory. It’s a good idea to say Y here. If you are building a Monolithic Kernel, then the ability of an attacker to insert foreign code into a running kernel is completely removed. Yes, another good idea to build a Monolithic Kernel instead of a Modularized Kernel. Say Y here. * * Access Control Lists * Grsecurity ACL system (CONFIG_GRKERNSEC_ACL) [N/y/?] y This Grsecurity option allows us to enable the Access Control List system (ACL) for Grsecurity. It’s a good idea to say Y here. ACL allows us to better control what program, files, etc on the system are allowed to do. We use it to apply a security policy that will work for the entire system. You can install and run Grsecurity without ACL but it is recommended for optimum security to enable this feature and use it. Once properly implemented, it will become impossible for a cracker to access and damage your Linux server. Personally, with Grsecurity ACL, I don’t know how someone could break into a Linux system. Say Y here. ACL Debugging Messages (CONFIG_GR_DEBUG) [N/y/?] y This Grsecurity option allows the Grsecurity ACL system to print debugging messages as an aid to finding problems in your ACL sets. It’s really a good idea to say Y here since it can become very difficult to debug problems with ACL if this option is not enable. Say Y here. Extra ACL Debugging Messages (CONFIG_GR_SUPERDEBUG) [N/y/?] Press Enter This Grsecurity option allows us to enable additional debugging messages that can help in finding problems in ACL sets or to gain a better understanding of the internal workings of the ACL system. In many cases, it is not necessary to enable this additional debugging messages feature for ACL and we will say N here. Denied capability logging (CONFIG_GRKERNSEC_ACL_CAPLOG) [N/y/?] y This Grsecurity option allows us to enable the denied capability logging support protection with ACL. If you say Y here, logs will be produced when a root-owned process does not have a needed capability raised in his set. This can help to debug the ACL of Grsecurity again. It’s a good idea to say Y here.
- 173. Kernel Security &Optimization 0 CHAPTER 6 173 Path to gradm (CONFIG_GRADM_PATH) [/sbin/gradm] Press Enter This Grsecurity option allows us to specify the path of the gradm binary installed on the system. gradm is the binary program used to manage Grsecurity ACL on the server. You have to download and install it to be able to use ACL on your server. Please read the next chapter in this book to get information about gradm and how to setup Grsecurity ACL on your system. The default path for gradm as shown above is correct and we press the [Enter] key to accept the default value for the path. Maximum tries before password lockout (CONFIG_GR_MAXTRIES) [3] 2 Once the gradm program is installed on your server to control and manage Grsecurity ACL, you will have to create a password for gradm to work. This option allows us to specify the number of times a user can attempt to authorize them selves with the Grsecurity ACL system. Here we change the default value of 3 to become 2, meaning that users are allowed to authorize themselves with the Grsecurity ACL system twice. Time to wait after max password tries, in seconds (CONFIG_GR_TIMEOUT) [30] Press Enter This option specifies the time the user must wait after attempting to authorize to the ACL system with the maximum number of invalid passwords. Just press [Enter] to accept the default entry. * * Filesystem Protections * Proc restrictions (CONFIG_GRKERNSEC_PROC) [N/y/?] y This Grsecurity option allows us to enable the proc restrictions protection on the system. If you say Y here, the permissions of the /proc file system will be altered to enhance system security and privacy. It’s a very good idea to say Y here. Restrict to user only (CONFIG_GRKERNSEC_PROC_USER) [N/y/?] y This Grsecurity option allows us to enable restrict to user only protection on the system. If you say Y here, non-root users will only be able to view their own processes, restricts them from viewing network-related information, viewing kernel symbol and module information. It’s a very good idea to say Y here. Additional restrictions (CONFIG_GRKERNSEC_PROC_ADD) [N/y/?] y This Grsecurity option allows us to enable additional restrictions protection on the system. If you say Y here, additional restrictions will be placed on the /proc file system that keep normal users from viewing cpu and device information. Again, it’s a good idea to say Y here. Linking restrictions (CONFIG_GRKERNSEC_LINK) [N/y/?] y This Grsecurity option allows us to enable the linking restrictions protection on the system. If you say Y here, /tmp race exploits will be prevented, since users will no longer be able to follow symlinks owned by other users in world-writeable +t directories, unless the owner of the symlink is the owner of the directory. Users will also not be able to hard link to files they do not own. It’s really a good idea to say Y here. FIFO restrictions (CONFIG_GRKERNSEC_FIFO) [N/y/?] y This Grsecurity option allows us to enable FIFO restrictions protection on the system. If you say Y here, users will not be able to write to FIFOs they don't own in world-writeable +t directories, unless the owner of the FIFO is the same owner of the directory it's held in. It’s a good idea to say Y here.
- 174. Kernel Security &Optimization 0 CHAPTER 6 174 Secure file descriptors (CONFIG_GRKERNSEC_FD) [N/y/?] y This Grsecurity option allows us to enable secure file descriptors protection on the system. If you say Y here, binaries will be protected from data spoofing attacks. It’s a very good idea to say Y here. Chroot jail restrictions (CONFIG_GRKERNSEC_CHROOT) [N/y/?] y This Grsecurity option allows us to enable chroot jail restrictions protection on the system. If you say Y here, you will be able to choose several options that will make breaking out of a chrooted jail much more difficult. It’s a very good idea to say Y here. Restricted signals (CONFIG_GRKERNSEC_CHROOT_SIG) [N/y/?] y This Grsecurity option allows us to enable the restricted signals protection on the system. If you say Y here, processes inside a chroot will not be able to send signals outside of the chroot. It’s a good idea to say Y here. Deny mounts (CONFIG_GRKERNSEC_CHROOT_MOUNT) [N/y/?] y This Grsecurity option allows us to enable deny mounts protection on the system. If you say Y here, processes inside a chroot will not be able to mount or remount file systems. It’s a good idea to say Y here. Deny double-chroots (CONFIG_GRKERNSEC_CHROOT_DOUBLE) [N/y/?] y This Grsecurity option allows us to enable the deny double-chroot protection on the system. If you say Y here, processes inside a chroot will not be able to chroot again. This is a widely used method of breaking out of a chroot jail and should not be allowed. It’s a good idea to say Y here. Enforce chdir("/") on all chroots (CONFIG_GRKERNSEC_CHROOT_CHDIR) [N/y/?] y This Grsecurity option allows us to enable the enforce chdir("/") on all chroots protection on the system. If you say Y here, the current working directory of all newly-chrooted applications will be set to the root directory of the chroot. It’s a good idea to say Y here. Deny (f)chmod +s (CONFIG_GRKERNSEC_CHROOT_CHMOD) [N/y/?] y This Grsecurity option allows us to enable the deny (f)chmod +s protection on the system. If you say Y here, processes inside a chroot will not be able to chmod or fchmod files to make them have suid or sgid bits. It’s a really good idea to say Y here. Deny mknod (CONFIG_GRKERNSEC_CHROOT_MKNOD) [N/y/?] y This Grsecurity option allows us to enable the deny mknod protection on the system. If you say Y here, processes inside a chroot will not be allowed to mknod (create device on the system). It’s a good idea to say Y here, unless you run into software incompatibilities. Deny ptraces (CONFIG_GRKERNSEC_CHROOT_PTRACE) [N/y/?] y This Grsecurity option allows us to enable the deny ptraces protection on the system. If you say Y here, processes inside a chroot will not be able to ptrace other processes. It’s a good idea to say Y here. Restrict priority changes (CONFIG_GRKERNSEC_CHROOT_NICE) [N/y/?] y This Grsecurity option allows us to enable the restrict priority changes protection on the system. If you say Y here, processes inside a chroot will not be able to raise the priority of processes in the chroot, or alter the priority of processes outside the chroot. It’s a good idea to say Y here.
- 175. Kernel Security &Optimization 0 CHAPTER 6 175 Capability restrictions within chroot (CONFIG_GRKERNSEC_CHROOT_CAPS) [N/y/?] Press Enter This Grsecurity option allows us to enable the capability restrictions within chroot protection on the system. If you say Y here, the capabilities on all root processes within a chroot jail will be lowered to stop module insertion, raw i/o, system and net admin tasks, rebooting the system, modifying immutable files, and changing the system time. This option can break some applications on the server and we disable it by answering to the question by N. Secure keymap loading (CONFIG_GRKERNSEC_KBMAP) [N/y/?] y This Grsecurity option allows us to enable the secure keymap loading protection on the system. If you say Y here, KDSKBENT and KDSKBSENT ioctl calls being called by unprivileged users will be denied. This means that unprivileged users with access to the console will NOT be able to modify keyboard bindings. It’s a good idea to say Y here. * * Kernel Auditing * Single group for auditing (CONFIG_GRKERNSEC_AUDIT_GROUP) [N/y/?] Press Enter This Grsecurity option allows us to enable single group for auditing protection on the system. If you say Y here, the exec, chdir, (un)mount, and ipc logging features of Grsecurity will only operate on a group you specify. By default Grsecurity produces a large amount of logs from the entire system on a production server; we don’t really need the entire auditing feature provided by Grsecurity even on specified group. Therefore we simply say N to this question and enable later in this Grsecurity kernel configuration the auditing log that we really need for production servers. Exec logging (CONFIG_GRKERNSEC_EXECLOG) [N/y/?] Press Enter This Grsecurity option allows us to enable the exec logging protection on the system. If you say Y here, execution of any program on the server will be logged to syslog. This option when enabled will produce a LOT of logs, especially on an active system. Therefore, we don’t recommend you to enable this security option. If you are those people that like to spend all their time reading log files, you could enable this option but be aware that it will take you a day to read all the logs generated by this option on a production server. Be reasonable, and say N here. Log execs within chroot (CONFIG_GRKERNSEC_CHROOT_EXECLOG) [N/y/?] y This Grsecurity option allows us to enable the log execs within chroot protection on the system. If you say Y here, all executions of any program inside a chroot jail environment will be logged to syslog. Contrary to the previous option that logs execution of any program on the system, this option ONLY logs the execution of program inside a chroot jail. In general, services running in chroot jail environment are limited, meaning that your log file will not become too BIG to read; therefore we can say Y here to enable this security option on the server. Chdir logging (CONFIG_GRKERNSEC_AUDIT_CHDIR) [N/y/?] Press Enter This Grsecurity option allows us to enable the chdir logging protection on the system. If you say Y here, all chdir() calls will be logged. Chdir() calls is when you navigate via your console on your server by using the ‘cd’ command. Imagine how many times you use the ‘cd’ command on your server when you performing some administration tasks. I recommend you to NOT enable this option if you want to avoid some BIG log files to read again. Say N here. (Un)Mount logging (CONFIG_GRKERNSEC_AUDIT_MOUNT) [N/y/?] Press Enter This Grsecurity option allows us to enable the (Un)Mount logging protection on the system. If you say Y here, all mounts and unmounts calls will be logged to syslog. This means that each time you mount or unmount drive on your server; the action will be logged to syslog. You can enable this security option if you like, but personally, I prefer to disable it. Say N here.
- 176. Kernel Security &Optimization 0 CHAPTER 6 176 IPC logging (CONFIG_GRKERNSEC_AUDIT_IPC) [N/y/?] y This Grsecurity option allows us to enable the IPC logging protection on the system. If you say Y here, creation and removal of message queues, semaphores, and shared memory will be logged. It’s a good idea to say Y here. Ptrace logging (CONFIG_GRKERNSEC_AUDIT_PTRACE) [N/y/?] Press Enter This Grsecurity option allows us to enable the ptrace logging protection on the system. If you say Y here, all successful ptraces will be logged. Ptraces are special operations performed when programs like strace or gdb are run. In general we never install programs like strace and gdb on a production server. These programs are only required on development server to debug software. If you don’t have these kinds of programs installed on your server, you can safety say N here, otherwise say Y to this security option. Signal logging (CONFIG_GRKERNSEC_SIGNAL) [N/y/?] y This Grsecurity option allows us to enable the signal logging protection on the system. If you say Y here, certain important signals will be logged, such as SIGSEGV that will inform you of when an error in a program occurred, which in some cases could mean a possible exploit attempt. It’s a good idea to say Y here. Fork failure logging (CONFIG_GRKERNSEC_FORKFAIL) [N/y/?] y This Grsecurity option allows us to enable the fork failure logging protection on the system. If you say Y here, all failed fork() attempts will be logged. This could suggest a fork bomb, or someone attempting to overstep their process limit. It’s a good idea to say Y here. Set*id logging (CONFIG_GRKERNSEC_SUID) [N/y/?] Press Enter This Grsecurity option allows us to enable the set*id logging protection on the system. If you say Y here, all set*id() calls will be logged. Enabling this security option could produce a lot of logs on an active system that run some services that use set*id() calls to operate. Mailman, Exim, Sendmail are known to software that uses many set*id() calls. Also, we already have other security programs doing the same job like sXid, therefore we don’t really need to enable this option. It is your’s to decide if you really need it or not. Personally, I disable this option by saying N here and use sXid to achieve the same result. Log set*ids to root (CONFIG_GRKERNSEC_SUID_ROOT) [N/y/?] y This Grsecurity option allows us to enable the log set*ids to root protection on the system. If you say Y here, only set*id() calls where a user is changing to the GID or UID of the root user will be logged. Such information could be useful when detecting a possible intrusion attempt. This option will produce smaller logs than logging all calls; therefore it’s a good idea to say Y here. Time change logging (CONFIG_GRKERNSEC_TIME) [N/y/?] y This Grsecurity option allows us to enable the time change logging protection on the system. If you say Y here, any changes of the system clock will be logged. It’s a good idea to say Y here. * * Executable Protections * Exec process limiting (CONFIG_GRKERNSEC_EXECVE) [N/y/?] y This Grsecurity option allows us to enable the exec process limiting protection on the system. If you say Y here, users with a resource limit on processes will have the value checked during execve() calls (execution of program). It’s really a good idea to say Y here.
- 177. Kernel Security &Optimization 0 CHAPTER 6 177 Dmesg(8) restriction (CONFIG_GRKERNSEC_DMESG) [N/y/?] y This Grsecurity option allows us to enable the dmesg restriction protection on the system. If you say Y here, non-root users will not be able to use dmesg(8) to view up to the last 4kb of messages in the kernel's log buffer. Again, it’s really a good idea to say Y here. Randomized PIDs (CONFIG_GRKERNSEC_RANDPID) [N/y/?] y This Grsecurity option allows us to enable the randomized PIDs protection on the system. If you say Y here, all PIDs created on the system will be pseudo-randomly generated. This is extremely effective to disallow an attacker from guessing pids of daemons, etc. It’s a good idea to say Y here. Altered default IPC permissions (CONFIG_GRKERNSEC_IPC) [N/y/?] Press Enter This Grsecurity option allows us to enable the altered default IPC permissions protection on the system. If you say Y here, the default permissions for IPC objects will be set based on the file system umask of the user creating the object. This is a good security feature but unfortunately, it is known to break software like Apache. Therefore we say N here. Limit uid/gid changes to root (CONFIG_GRKERNSEC_TTYROOT) [N/y/?] y This Grsecurity option allows us to enable the limit UID/GID changes to root protection on the system. If you say Y here, you will be able choose from three options that will allow you to restrict access to the root account by console type. Therefore we say Y here. Deny physical consoles (tty) (CONFIG_GRKERNSEC_TTYROOT_PHYS) [N/y/?] Press Enter This Grsecurity option allows us to enable the deny physical consoles (tty) protection on the system. If you say Y here, access to root from physical consoles will be denied. This is only recommended for rare cases where you will never need to be physically at the machine. For most of us, this is not the case and we have to say N here. Deny serial consoles (ttyS) (CONFIG_GRKERNSEC_TTYROOT_SERIAL) [N/y/?] y This Grsecurity option allows us to enable deny serial consoles (ttyS) protection on the system. If you say Y here, access to root from serial consoles will be denied. Most people can say Y here, since most don't use serial devices for their console access. Deny pseudo consoles (pty) (CONFIG_GRKERNSEC_TTYROOT_PSEUDO) [N/y/?] Press Enter This Grsecurity option allows us to enable the deny pseudo consoles (pty) protection on the system. If you say Y here, access to root from pseudo consoles will be denied. Pseudo consoles include consoles from telnet, ssh, or any other kind of interactive shell initiated from the network. In general, most of us use at least SSH to make a remote connection. Therefore we have to say N here, if we want to continue to use SSH for secure remote connection. Fork-bomb protection (CONFIG_GRKERNSEC_FORKBOMB) [N/y/?] y This Grsecurity option allows us to enable fork-bomb protection on the system. If you say Y here, you will be able to configure a group to add to users on your system that you want to be unable to fork-bomb the system. It’s a good idea to say Y here and chose in the next security option the GID to run this protection with. GID for restricted users (CONFIG_GRKERNSEC_FORKBOMB_GID) [1006] Press Enter This Grsecurity option allows us to enable the GID for restricted users’ protection on the system. Here we have to enter a GID number as the value, the default value is 1006 and we can keep it by pressing the [Enter] key. It is important to note that this GID should be added to any user for which the feature should be activated. See the next chapter of this book for more information about the procedures to follow. At this time you only need to accept the default value.
- 178. Kernel Security &Optimization 0 CHAPTER 6 178 Forks allowed per second (CONFIG_GRKERNSEC_FORKBOMB_SEC) [40] Press Enter This Grsecurity option allows us to enable the forks allowed per second protection on the system. This option is a continuation of the above forks options, here we have to enter the maximum number of forks allowed per second, and the default setting should be fine for most users. Maximum processes allowed (CONFIG_GRKERNSEC_FORKBOMB_MAX) [20] 35 This Grsecurity option allows us to enable the maximum processes allowed on the system. Here we have to enter the maximum number of processes users in the fork-bomb protected group can run. We change the default value of 20 to become 35. 35 is the number you have set into the /etc/security/limit.conf file. Please see what is set into your limit.conf file and report it here. In general, 35 is a good value to go with. Trusted path execution (CONFIG_GRKERNSEC_TPE) [N/y/?] y This Grsecurity option allows us to enable trusted path execution protection on the system. If you say Y here, you will be able to choose a GID to add to the supplementary groups of users you want to mark as "untrusted." These users will not be able to execute any files that are not in root-owned directories writeable only by root. It’s a good idea to say Y here. Glibc protection (CONFIG_GRKERNSEC_TPE_GLIBC) [N/y/?] y This Grsecurity option allows us to enable the glibc protection on the system. If you say Y here, all non-root users executing any files while glibc specific environment variables such as LD_PRELOAD are set, will have their environment cleared of these variables, since they could be used to evade the trusted path execution protection. It also protects against evasion through executing the dynamic linker to run a rogue binary. It is highly recommended you say Y here. Partially restrict non-root users (CONFIG_GRKERNSEC_TPE_ALL) [N/y/?] y This Grsecurity option allows us to enable the partially restrict non-root users protection on the system. If you say Y here, All non-root users other than the ones in the group specified in the main TPE option will only be allowed to execute files in directories they own that are not group or world-writeable, or in directories owned by root and writeable only by root. It’s a good idea to say Y here. GID for untrusted users: (CONFIG_GRKERNSEC_TPE_GID) [1005] Press Enter This Grsecurity option allows us to enable the GID for untrusted user’s protection on the system. Here we have to enter a GID number as the value, the default value is 1005 and we can keep it by pressing the [Enter] key. It is important to note that this GID should be added to any user for which the feature should be activated. See the next chapter of this book for more information about the procedures to follow. At this time you only need to accept the default value. Restricted ptrace (CONFIG_GRKERNSEC_PTRACE) [N/y/?] y This Grsecurity option allows us to enable the restricted ptrace protection on the system. If you say Y here, no one but root will be able to ptrace processes. Tracing syscalls inside the kernel will also be disabled. It’s a good idea to say Y here. Allow ptrace for group (CONFIG_GRKERNSEC_PTRACE_GROUP) [N/y/?] Press Enter This Grsecurity option allows us to enable the allow ptrace for group protection on the system. If you say Y here, you will be able to choose a GID of users will be able to ptrace. Remember that ptraces are special operations performed when programs like strace or gdb are run for debugging purpose. Since these kind of program should in general be run on development server and by a root user only, we can safety say N here.
- 179. Kernel Security &Optimization 0 CHAPTER 6 179 * * Network Protections * Randomized IP IDs (CONFIG_GRKERNSEC_RANDID) [N/y/?] y This Grsecurity option allows us to enable the allow randomized IP IDs protection on the system. If you say Y here, the entire ID field on all outgoing packets will be randomized. This hinders OS fingerprinters and keeps your machine from being used as a bounce for an untraceable portscan. It’s a good idea to say Y here. Randomized TCP source ports (CONFIG_GRKERNSEC_RANDSRC) [N/y/?] y This Grsecurity option allows us to enable the randomized TCP source ports protection on the system. If you say Y here, situations where a source port is generated on the fly for the TCP protocol (ie. with connect() ) will be altered so that the source port is generated at random, instead of a simple incrementing algorithm. It’s a good idea to say Y here. Randomized RPC XIDs (CONFIG_GRKERNSEC_RANDRPC) [N/y/?] y This Grsecurity option allows us to enable the randomized RPC XIDs protection on the system. If you say Y here, the method of determining XIDs for RPC requests will be randomized, instead of using linux's default behavior of simply incrementing the XID. This allows us to have a more secure RPC connection on the system. It’s a good idea to say Y here. Altered Ping IDs (CONFIG_GRKERNSEC_RANDPING) [N/y/?] y This Grsecurity option allows us to enable the altered Ping IDs protection on the system. If you say Y here, the way Linux handles echo replies will be changed so that the reply uses an ID equal to the ID of the echo request. This will help in confusing OS detection. It’s a good idea to say Y here. Randomized TTL (CONFIG_GRKERNSEC_RANDTTL) [N/y/?] y This Grsecurity option allows us to enable the randomized TTL protection on the system. If you say Y here, your TTL (time to live) for packets will be set at random, with a base of the sysctl ttl default, to further confuse OS detection. It’s a good idea to say Y here. Socket restrictions (CONFIG_GRKERNSEC_SOCKET) [N/y/?] y This Grsecurity option allows us to enable the socket restrictions protection on the system. If you say Y here, you will be able to choose from three options related to socket protection on the server. From these three available security options, you’ll have to choose the one that best fits your requirements. Therefore, it’s a good idea to say Y here since we have to choose one of the three available security options for our needs. Deny any sockets to group (CONFIG_GRKERNSEC_SOCKET_ALL) [N/y/?] y This Grsecurity option allows us to enable the socket restrictions protection on the system. If you say Y here, you will be able to choose a GID of whose users will be unable to connect to other hosts from your machine or run server applications from your machine. This is the security option that we’ll choose. Say Y here. GID to deny all sockets for: (CONFIG_GRKERNSEC_SOCKET_ALL_GID) [1004] Press Enter This Grsecurity option allows us to enable the GID to deny all sockets protection on the system. Here we have to enter a GID number as the value, the default value is 1004 and we can keep it by pressing the [Enter] key. It is important to note that this GID should be added to any user for which the feature should be activated. See the next chapter of this book for more information about the procedures to follow. At this time you only need to accept the default value.
- 180. Kernel Security &Optimization 0 CHAPTER 6 180 Deny client sockets to group (CONFIG_GRKERNSEC_SOCKET_CLIENT) [N/y/?] Press Enter This Grsecurity option allows us to enable the deny client sockets to group protection on the system. If you say Y here, you will be able to choose a GID of whose users will be unable to connect to other hosts from your machine, but will be able to run servers. We have already chosen the above option that enables socket protection on both ways (users cannot connect to other hosts or run server applications from our machine), therefore we don’t need to say Y here. Say N here. Deny server sockets to group (CONFIG_GRKERNSEC_SOCKET_SERVER) [N/y/?] Press Enter This Grsecurity option allows us to enable the deny server sockets to group protection on the system. If you say Y here, you will be able to choose a GID of whose users will be unable to run server applications from your machine. As for the above option, we already have chosen the first option that enable socket protection on both way (users cannot connect to other hosts or run server applications from our machine), therefore we don’t need to say Y here. Say N here. * * Sysctl support * Sysctl support (CONFIG_GRKERNSEC_SYSCTL) [N/y/?] Press Enter This Grsecurity option allows us to enable Grsecurity sysctl support protection on the system. If you say Y here, you will be able to change the options that Grsecurity runs with at bootup, without having to recompile your kernel. You can echo values to files in /proc/sys/kernel/grsecurity to enable (1) or disable (0) various features. Enabling this option will reduce the effectiveness of the added security of the Grsecurity patch, therefore we say N here. * * Miscellaneous Features * Seconds in between log messages(minimum) (CONFIG_GRKERNSEC_FLOODTIME) [30] Press Enter This Grsecurity option allows us to enable the seconds in between log messages protection on the system. This option allows you to enforce the number of seconds between Grsecurity log messages. The default should be suitable for most people. Just press the [Enter] key here to accept the default value. BSD-style coredumps (CONFIG_GRKERNSEC_COREDUMP) [N/y/?] y This Grsecurity option allows us to enable the BSD-style coredumps protection on the system. If you say Y here, Linux will use a style similar to BSD for coredumps, core.processname. Not a security feature, just a useful one. Say Y here. *** End of Linux kernel configuration. *** Check the top-level Makefile for additional configuration. *** Next, you must run 'make dep'.
- 181. Kernel Security &Optimization 0 CHAPTER 6 181 Modularized kernel configuration Building kernel with modules (Modularized kernel) has some advantages. It allows easy portability between different Linux systems, since you can choose and build different parts of the kernel as a module and load that segment of code on demand. Below we show you the configuration of Modularized kernel, which is to compile some needed codes and drivers as a module into the kernel by answering the different questions using y, n or m. As for the previous Monolithic kernel configuration, don’t forget to only compile code that you need and use. A new kernel is very specific to your computer hardware, in the Modularized kernel configuration part below; we assume the following hardware for our example. Of course you must change them to fit your system components. 1 Pentium-III 667 MHz (i686) processor 1 Motherboard Asus P3V4X Pro 133Mhz EIDE 1 Hard Disk Ultra ATA/100 EIDE 1 Chipset Apollo Pro133A 1 CD-ROM ATAPI IDE 1 Floppy Disk 2 Ethernet Cards 3COM 3c597 PCI 10/100 1 Mouse PS/2 If you don’t want some options listed in the Modularized kernel configuration that I enable by default, answer n (for no) instead of y (for yes) or m (for modularized if possible) to the related questions. If you want some other options that I disable, then answer y or m instead of n. rm -f include/asm ( cd include ; ln -sf asm-i386 asm) /bin/sh scripts/Configure arch/i386/config.in # # Using defaults found in arch/i386/defconfig # * * Code maturity level options * Prompt for development and/or incomplete code/drivers (CONFIG_EXPERIMENTAL) [N/y/?] Press Enter * * Loadable module support * Enable loadable module support (CONFIG_MODULES) [Y/n/?] Press Enter Set version information on all module symbols (CONFIG_MODVERSIONS) [Y/n/?] n Kernel module loader (CONFIG_KMOD) [Y/n/?] Press Enter * * Processor type and features * Processor family (386, 486, 586/K5/5x86/6x86/6x86MX, Pentium-Classic, Pentium-MMX, Pentium-Pro/Celeron/Pentium-II, Pentium-III/Celeron(Coppermine), Pentium-4, K6/K6-II/K6- III, Athlon/Duron/K7, Elan, Crusoe, Winchip-C6, Winchip-2, Winchip-2A/Winchip-3, CyrixIII/C3) [Pentium-III/Celeron(Coppermine)] Press Enter Toshiba Laptop support (CONFIG_TOSHIBA) [N/y/m/?] Press Enter Dell laptop support (CONFIG_I8K) [N/y/m/?] Press Enter /dev/cpu/microcode - Intel IA32 CPU microcode support (CONFIG_MICROCODE) [N/y/m/?] m /dev/cpu/*/msr - Model-specific register support (CONFIG_X86_MSR) [N/y/m/?] m /dev/cpu/*/cpuid - CPU information support (CONFIG_X86_CPUID) [N/y/m/?] m High Memory Support (off, 4GB, 64GB) [off] Press Enter Math emulation (CONFIG_MATH_EMULATION) [N/y/?] Press Enter MTRR (Memory Type Range Register) support (CONFIG_MTRR) [N/y/?] Press Enter Symmetric multi-processing support (CONFIG_SMP) [Y/n/?] n Local APIC support on uniprocessors (CONFIG_X86_UP_APIC) [N/y/?] (NEW) y IO-APIC support on uniprocessors (CONFIG_X86_UP_IOAPIC) [N/y/?] (NEW) y * * General setup
- 182. Kernel Security &Optimization 0 CHAPTER 6 182 * Networking support (CONFIG_NET) [Y/n/?] Press Enter PCI support (CONFIG_PCI) [Y/n/?] Press Enter PCI access mode (BIOS, Direct, Any) [Any] Press Enter PCI device name database (CONFIG_PCI_NAMES) [Y/n/?] n EISA support (CONFIG_EISA) [N/y/?] Press Enter MCA support (CONFIG_MCA) [N/y/?] Press Enter Support for hot-pluggable devices (CONFIG_HOTPLUG) [Y/n/?] n System V IPC (CONFIG_SYSVIPC) [Y/n/?] Press Enter BSD Process Accounting (CONFIG_BSD_PROCESS_ACCT) [N/y/?] Press Enter Sysctl support (CONFIG_SYSCTL) [Y/n/?] Press Enter Kernel core (/proc/kcore) format (ELF, A.OUT) [ELF] Press Enter Kernel support for a.out binaries (CONFIG_BINFMT_AOUT) [Y/m/n/?] m Kernel support for ELF binaries (CONFIG_BINFMT_ELF) [Y/m/n/?] Press Enter Kernel support for MISC binaries (CONFIG_BINFMT_MISC) [Y/m/n/?] m Power Management support (CONFIG_PM) [Y/n/?] n * * Memory Technology Devices (MTD) * Memory Technology Device (MTD) support (CONFIG_MTD) [N/y/m/?] Press Enter * * Parallel port support * Parallel port support (CONFIG_PARPORT) [N/y/m/?] Press Enter * * Plug and Play configuration * Plug and Play support (CONFIG_PNP) [Y/m/n/?] n * * Block devices * Normal PC floppy disk support (CONFIG_BLK_DEV_FD) [Y/m/n/?] Press Enter XT hard disk support (CONFIG_BLK_DEV_XD) [N/y/m/?] Press Enter Compaq SMART2 support (CONFIG_BLK_CPQ_DA) [N/y/m/?] Press Enter Compaq Smart Array 5xxx support (CONFIG_BLK_CPQ_CISS_DA) [N/y/m/?] Press Enter Mylex DAC960/DAC1100 PCI RAID Controller support (CONFIG_BLK_DEV_DAC960) [N/y/m/?] Press Enter Loopback device support (CONFIG_BLK_DEV_LOOP) [N/y/m/?] Press Enter Network block device support (CONFIG_BLK_DEV_NBD) [N/y/m/?] Press Enter RAM disk support (CONFIG_BLK_DEV_RAM) [N/y/m/?] Press Enter * * Multi-device support (RAID and LVM) * Multiple devices driver support (RAID and LVM) (CONFIG_MD) [N/y/?] Press Enter * * Networking options * Packet socket (CONFIG_PACKET) [Y/m/n/?] Press Enter Packet socket: mmapped IO (CONFIG_PACKET_MMAP) [N/y/?] y Netlink device emulation (CONFIG_NETLINK_DEV) [N/y/m/?] (NEW) m Network packet filtering (replaces ipchains) (CONFIG_NETFILTER) [N/y/?] y Network packet filtering debugging (CONFIG_NETFILTER_DEBUG) [N/y/?] (NEW) y Socket Filtering (CONFIG_FILTER) [N/y/?] Press Enter Unix domain sockets (CONFIG_UNIX) [Y/m/n/?] Press Enter TCP/IP networking (CONFIG_INET) [Y/n/?] Press Enter IP: multicasting (CONFIG_IP_MULTICAST) [Y/n/?] n IP: advanced router (CONFIG_IP_ADVANCED_ROUTER) [N/y/?] Press Enter IP: kernel level autoconfiguration (CONFIG_IP_PNP) [N/y/?] Press Enter IP: tunneling (CONFIG_NET_IPIP) [N/y/m/?] Press Enter IP: GRE tunnels over IP (CONFIG_NET_IPGRE) [N/y/m/?] Press Enter IP: TCP Explicit Congestion Notification support (CONFIG_INET_ECN) [N/y/?] Press Enter IP: TCP syncookie support (disabled default) (CONFIG_SYN_COOKIES) [N/y/?] y * * IP: Netfilter Configuration * Connection tracking (required for masq/NAT) (CONFIG_IP_NF_CONNTRACK) [N/y/m/?] (NEW) m FTP protocol support (CONFIG_IP_NF_FTP) [N/m/?] (NEW) m IRC protocol support (CONFIG_IP_NF_IRC) [N/m/?] (NEW) m IP tables support (required for filtering/masq/NAT) (CONFIG_IP_NF_IPTABLES) [N/y/m/?] (NEW) m limit match support (CONFIG_IP_NF_MATCH_LIMIT) [N/m/?] (NEW) m
- 183. Kernel Security &Optimization 0 CHAPTER 6 183 MAC address match support (CONFIG_IP_NF_MATCH_MAC) [N/m/?] (NEW) m netfilter MARK match support (CONFIG_IP_NF_MATCH_MARK) [N/m/?] (NEW) m Multiple port match support (CONFIG_IP_NF_MATCH_MULTIPORT) [N/m/?] (NEW) m TOS match support (CONFIG_IP_NF_MATCH_TOS) [N/m/?] (NEW) m AH/ESP match support (CONFIG_IP_NF_MATCH_AH_ESP) [N/m/?] (NEW) m LENGTH match support (CONFIG_IP_NF_MATCH_LENGTH) [N/m/?] (NEW) m TTL match support (CONFIG_IP_NF_MATCH_TTL) [N/m/?] (NEW) m tcpmss match support (CONFIG_IP_NF_MATCH_TCPMSS) [N/m/?] (NEW) m Connection state match support (CONFIG_IP_NF_MATCH_STATE) [N/m/?] (NEW) m Packet filtering (CONFIG_IP_NF_FILTER) [N/m/?] (NEW) m REJECT target support (CONFIG_IP_NF_TARGET_REJECT) [N/m/?] (NEW) m Full NAT (CONFIG_IP_NF_NAT) [N/m/?] (NEW) m MASQUERADE target support (CONFIG_IP_NF_TARGET_MASQUERADE) [N/m/?] (NEW) m REDIRECT target support (CONFIG_IP_NF_TARGET_REDIRECT) [N/m/?] (NEW) m Packet mangling (CONFIG_IP_NF_MANGLE) [N/m/?] (NEW) m TOS target support (CONFIG_IP_NF_TARGET_TOS) [N/m/?] (NEW) m MARK target support (CONFIG_IP_NF_TARGET_MARK) [N/m/?] (NEW) m LOG target support (CONFIG_IP_NF_TARGET_LOG) [N/m/?] (NEW) m TCPMSS target support (CONFIG_IP_NF_TARGET_TCPMSS) [N/m/?] (NEW) m ipchains (2.2-style) support (CONFIG_IP_NF_COMPAT_IPCHAINS) [N/y/m/?] (NEW) Press Enter ipfwadm (2.0-style) support (CONFIG_IP_NF_COMPAT_IPFWADM) [N/y/m/?] (NEW) Press Enter * * * The IPX protocol (CONFIG_IPX) [N/y/m/?] Press Enter Appletalk protocol support (CONFIG_ATALK) [N/y/m/?] Press Enter DECnet Support (CONFIG_DECNET) [N/y/m/?] Press Enter 802.1d Ethernet Bridging (CONFIG_BRIDGE) [N/y/m/?] Press Enter * * QoS and/or fair queueing * QoS and/or fair queueing (CONFIG_NET_SCHED) [N/y/?] Press Enter * * Telephony Support * Linux telephony support (CONFIG_PHONE) [N/y/m/?] Press Enter * * ATA/IDE/MFM/RLL support * ATA/IDE/MFM/RLL support (CONFIG_IDE) [Y/m/n/?] Press Enter * * IDE, ATA and ATAPI Block devices * Enhanced IDE/MFM/RLL disk/cdrom/tape/floppy support (CONFIG_BLK_DEV_IDE) [Y/m/n/?] Press Enter * * Please see Documentation/ide.txt for help/info on IDE drives * Use old disk-only driver on primary interface (CONFIG_BLK_DEV_HD_IDE) [N/y/?] Press Enter Include IDE/ATA-2 DISK support (CONFIG_BLK_DEV_IDEDISK) [Y/m/n/?] Press Enter Use multi-mode by default (CONFIG_IDEDISK_MULTI_MODE) [Y/n/?] n Include IDE/ATAPI CDROM support (CONFIG_BLK_DEV_IDECD) [Y/m/n/?] Press Enter Include IDE/ATAPI TAPE support (CONFIG_BLK_DEV_IDETAPE) [N/y/m/?] Press Enter Include IDE/ATAPI FLOPPY support (CONFIG_BLK_DEV_IDEFLOPPY) [N/y/m/?] Press Enter SCSI emulation support (CONFIG_BLK_DEV_IDESCSI) [N/y/m/?] Press Enter * * IDE chipset support/bugfixes * CMD640 chipset bugfix/support (CONFIG_BLK_DEV_CMD640) [Y/n/?] n RZ1000 chipset bugfix/support (CONFIG_BLK_DEV_RZ1000) [Y/n/?] n Generic PCI IDE chipset support (CONFIG_BLK_DEV_IDEPCI) [Y/n/?] Press Enter Sharing PCI IDE interrupts support (CONFIG_IDEPCI_SHARE_IRQ) [Y/n/?] Press Enter Generic PCI bus-master DMA support (CONFIG_BLK_DEV_IDEDMA_PCI) [Y/n/?] Press Enter Boot off-board chipsets first support (CONFIG_BLK_DEV_OFFBOARD) [N/y/?] Press Enter Use PCI DMA by default when available (CONFIG_IDEDMA_PCI_AUTO) [Y/n/?] Press Enter AEC62XX chipset support (CONFIG_BLK_DEV_AEC62XX) [N/y/?] Press Enter ALI M15x3 chipset support (CONFIG_BLK_DEV_ALI15X3) [N/y/?] Press Enter AMD Viper support (CONFIG_BLK_DEV_AMD74XX) [N/y/?] Press Enter CMD64X chipset support (CONFIG_BLK_DEV_CMD64X) [N/y/?] Press Enter CY82C693 chipset support (CONFIG_BLK_DEV_CY82C693) [N/y/?] Press Enter
- 184. Kernel Security &Optimization 0 CHAPTER 6 184 Cyrix CS5530 MediaGX chipset support (CONFIG_BLK_DEV_CS5530) [N/y/?] Press Enter HPT34X chipset support (CONFIG_BLK_DEV_HPT34X) [N/y/?] Press Enter HPT366 chipset support (CONFIG_BLK_DEV_HPT366) [N/y/?] Press Enter Intel PIIXn chipsets support (CONFIG_BLK_DEV_PIIX) [Y/n/?] Press Enter PIIXn Tuning support (CONFIG_PIIX_TUNING) [Y/n/?] Press Enter NS87415 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_NS87415) [N/y/?] Press Enter PROMISE PDC202{46|62|65|67|68} support (CONFIG_BLK_DEV_PDC202XX) [N/y/?] Press Enter ServerWorks OSB4/CSB5 chipsets support (CONFIG_BLK_DEV_SVWKS) [N/y/?] Press Enter SiS5513 chipset support (CONFIG_BLK_DEV_SIS5513) [N/y/?] Press Enter SLC90E66 chipset support (CONFIG_BLK_DEV_SLC90E66) [N/y/?] Press Enter Tekram TRM290 chipset support (EXPERIMENTAL) (CONFIG_BLK_DEV_TRM290) [N/y/?] Press Enter VIA82CXXX chipset support (CONFIG_BLK_DEV_VIA82CXXX) [N/y/?] y Other IDE chipset support (CONFIG_IDE_CHIPSETS) [N/y/?] Press Enter IGNORE word93 Validation BITS (CONFIG_IDEDMA_IVB) [N/y/?] Press Enter * * SCSI support * SCSI support (CONFIG_SCSI) [Y/m/n/?] n * * Fusion MPT device support * * * I2O device support * I2O support (CONFIG_I2O) [N/y/m/?] Press Enter * * Network device support * Network device support (CONFIG_NETDEVICES) [Y/n/?] Press Enter * * ARCnet devices * ARCnet support (CONFIG_ARCNET) [N/y/m/?] Press Enter Dummy net driver support (CONFIG_DUMMY) [M/n/y/?] Press Enter Bonding driver support (CONFIG_BONDING) [N/y/m/?] Press Enter EQL (serial line load balancing) support (CONFIG_EQUALIZER) [N/y/m/?] Press Enter Universal TUN/TAP device driver support (CONFIG_TUN) [N/y/m/?] Press Enter * * Ethernet (10 or 100Mbit) * Ethernet (10 or 100Mbit) (CONFIG_NET_ETHERNET) [Y/n/?] Press Enter Sun Happy Meal 10/100baseT support (CONFIG_HAPPYMEAL) [N/y/m/?] Press Enter Sun GEM support (CONFIG_SUNGEM) [N/y/m/?] Press Enter 3COM cards (CONFIG_NET_VENDOR_3COM) [N/y/?] y 3c501 "EtherLink" support (CONFIG_EL1) [N/y/m/?] (NEW) Press Enter 3c503 "EtherLink II" support (CONFIG_EL2) [N/y/m/?] (NEW) Press Enter 3c505 "EtherLink Plus" support (CONFIG_ELPLUS) [N/y/m/?] (NEW) Press Enter 3c509/3c529 (MCA)/3c579 "EtherLink III" support (CONFIG_EL3) [N/y/m/?] (NEW) Press Enter 3c515 ISA "Fast EtherLink" (CONFIG_3C515) [N/y/m/?] (NEW) Press Enter 3c590/3c900 series (592/595/597) "Vortex/Boomerang" support (CONFIG_VORTEX) [N/y/m/?] (NEW) y AMD LANCE and PCnet (AT1500 and NE2100) support (CONFIG_LANCE) [N/y/m/?] Press Enter Western Digital/SMC cards (CONFIG_NET_VENDOR_SMC) [N/y/?] Press Enter Racal-Interlan (Micom) NI cards (CONFIG_NET_VENDOR_RACAL) [N/y/?] Press Enter DEPCA, DE10x, DE200, DE201, DE202, DE422 support (CONFIG_DEPCA) [N/y/m/?] Press Enter HP 10/100VG PCLAN (ISA, EISA, PCI) support (CONFIG_HP100) [N/y/m/?] Press Enter Other ISA cards (CONFIG_NET_ISA) [N/y/?] Press Enter EISA, VLB, PCI and on board controllers (CONFIG_NET_PCI) [Y/n/?] n Pocket and portable adapters (CONFIG_NET_POCKET) [N/y/?] Press Enter * * Ethernet (1000 Mbit) * Alteon AceNIC/3Com 3C985/NetGear GA620 Gigabit support (CONFIG_ACENIC) [N/y/m/?] Press Enter D-Link DL2000-based Gigabit Ethernet support (CONFIG_DL2K) [N/y/m/?] Press Enter National Semiconduct DP83820 support (CONFIG_NS83820) [N/y/m/?] Press Enter Packet Engines Hamachi GNIC-II support (CONFIG_HAMACHI) [N/y/m/?] Press Enter SysKonnect SK-98xx support (CONFIG_SK98LIN) [N/y/m/?] Press Enter FDDI driver support (CONFIG_FDDI) [N/y/?] Press Enter
- 185. Kernel Security &Optimization 0 CHAPTER 6 185 PPP (point-to-point protocol) support (CONFIG_PPP) [N/y/m/?] Press Enter SLIP (serial line) support (CONFIG_SLIP) [N/y/m/?] Press Enter * * Wireless LAN (non-hamradio) * Wireless LAN (non-hamradio) (CONFIG_NET_RADIO) [N/y/?] Press Enter * * Token Ring devices * Token Ring driver support (CONFIG_TR) [N/y/?] Press Enter Fibre Channel driver support (CONFIG_NET_FC) [N/y/?] Press Enter * * Wan interfaces * Wan interfaces support (CONFIG_WAN) [N/y/?] Press Enter * * Amateur Radio support * Amateur Radio support (CONFIG_HAMRADIO) [N/y/?] Press Enter * * IrDA (infrared) support * IrDA subsystem support (CONFIG_IRDA) [N/y/m/?] Press Enter * * ISDN subsystem * ISDN support (CONFIG_ISDN) [N/y/m/?] Press Enter * * Old CD-ROM drivers (not SCSI, not IDE) * Support non-SCSI/IDE/ATAPI CDROM drives (CONFIG_CD_NO_IDESCSI) [N/y/?] Press Enter * * Input core support * Input core support (CONFIG_INPUT) [N/y/m/?] Press Enter * * Character devices * Virtual terminal (CONFIG_VT) [Y/n/?] Press Enter Support for console on virtual terminal (CONFIG_VT_CONSOLE) [Y/n/?] Press Enter Standard/generic (8250/16550 and compatible UARTs) serial support (CONFIG_SERIAL) [Y/m/n/?] Press Enter Support for console on serial port (CONFIG_SERIAL_CONSOLE) [N/y/?] Press Enter Extended dumb serial driver options (CONFIG_SERIAL_EXTENDED) [N/y/?] Press Enter Non-standard serial port support (CONFIG_SERIAL_NONSTANDARD) [N/y/?] Press Enter Unix98 PTY support (CONFIG_UNIX98_PTYS) [Y/n/?] Press Enter Maximum number of Unix98 PTYs in use (0-2048) (CONFIG_UNIX98_PTY_COUNT) [256] 128 * * I2C support * I2C support (CONFIG_I2C) [N/y/m/?] Press Enter * * Mice * Bus Mouse Support (CONFIG_BUSMOUSE) [N/y/m/?] Press Enter Mouse Support (not serial and bus mice) (CONFIG_MOUSE) [Y/m/n/?] n * * Joysticks * * * Input core support is needed for gameports * * * Input core support is needed for joysticks * QIC-02 tape support (CONFIG_QIC02_TAPE) [N/y/m/?] Press Enter * * Watchdog Cards * Watchdog Timer Support (CONFIG_WATCHDOG) [N/y/?] Press Enter Intel i8x0 Random Number Generator support (CONFIG_INTEL_RNG) [N/y/m/?] Press Enter
- 186. Kernel Security &Optimization 0 CHAPTER 6 186 /dev/nvram support (CONFIG_NVRAM) [N/y/m/?] Press Enter Enhanced Real Time Clock Support (CONFIG_RTC) [N/y/m/?] Press Enter Double Talk PC internal speech card support (CONFIG_DTLK) [N/y/m/?] Press Enter Siemens R3964 line discipline (CONFIG_R3964) [N/y/m/?] Press Enter Applicom intelligent fieldbus card support (CONFIG_APPLICOM) [N/y/m/?] Press Enter * * Ftape, the floppy tape device driver * Ftape (QIC-80/Travan) support (CONFIG_FTAPE) [N/y/m/?] Press Enter /dev/agpgart (AGP Support) (CONFIG_AGP) [Y/m/n/?] n Direct Rendering Manager (XFree86 DRI support) (CONFIG_DRM) [Y/n/?] n ACP Modem (Mwave) support (CONFIG_MWAVE) [N/y/m/?] Press Enter * * Multimedia devices * Video For Linux (CONFIG_VIDEO_DEV) [N/y/m/?] Press Enter * * File systems * Quota support (CONFIG_QUOTA) [N/y/?] y Kernel automounter support (CONFIG_AUTOFS_FS) [N/y/m/?] Press Enter Kernel automounter version 4 support (also supports v3) (CONFIG_AUTOFS4_FS) [Y/m/n/?] n Reiserfs support (CONFIG_REISERFS_FS) [N/y/m/?] Press Enter Ext3 journalling file system support (EXPERIMENTAL) (CONFIG_EXT3_FS) [N/y/m/?] y JBD (ext3) debugging support (CONFIG_JBD_DEBUG) [N/y/?] y DOS FAT fs support (CONFIG_FAT_FS) [N/y/m/?] m MSDOS fs support (CONFIG_MSDOS_FS) [N/y/m/?] m VFAT (Windows-95) fs support (CONFIG_VFAT_FS) [N/y/m/?] m Compressed ROM file system support (CONFIG_CRAMFS) [N/y/m/?] Press Enter Virtual memory file system support (former shm fs) (CONFIG_TMPFS) [Y/n/?] Press Enter Simple RAM-based file system support (CONFIG_RAMFS) [N/y/m/?] Press Enter ISO 9660 CDROM file system support (CONFIG_ISO9660_FS) [Y/m/n/?] m Microsoft Joliet CDROM extensions (CONFIG_JOLIET) [N/y/?] y Transparent decompression extension (CONFIG_ZISOFS) [N/y/?] Press Enter Minix fs support (CONFIG_MINIX_FS) [N/y/m/?] Press Enter FreeVxFS file system support (VERITAS VxFS(TM) compatible) (CONFIG_VXFS_FS) [N/y/m/?] Press Enter NTFS file system support (read only) (CONFIG_NTFS_FS) [N/y/m/?] Press Enter OS/2 HPFS file system support (CONFIG_HPFS_FS) [N/y/m/?] Press Enter /proc file system support (CONFIG_PROC_FS) [Y/n/?] Press Enter /dev/pts file system for Unix98 PTYs (CONFIG_DEVPTS_FS) [Y/n/?] Press Enter ROM file system support (CONFIG_ROMFS_FS) [N/y/m/?] Press Enter Second extended fs support (CONFIG_EXT2_FS) [Y/m/n/?] Press Enter System V/Xenix/V7/Coherent file system support (CONFIG_SYSV_FS) [N/y/m/?] Press Enter UDF file system support (read only) (CONFIG_UDF_FS) [N/y/m/?] Press Enter UFS file system support (read only) (CONFIG_UFS_FS) [N/y/m/?] Press Enter * * Network File Systems * Coda file system support (advanced network fs) (CONFIG_CODA_FS) [N/y/m/?] Press Enter NFS file system support (CONFIG_NFS_FS) [Y/m/n/?] n NFS server support (CONFIG_NFSD) [Y/m/n/?] n SMB file system support (to mount Windows shares etc.) (CONFIG_SMB_FS) [N/y/m/?] Press Enter NCP file system support (to mount NetWare volumes) (CONFIG_NCP_FS) [N/y/m/?] Press Enter * * Partition Types * Advanced partition selection (CONFIG_PARTITION_ADVANCED) [N/y/?] Press Enter * * Native Language Support * Default NLS Option (CONFIG_NLS_DEFAULT) [iso8859-1] (NEW) Press Enter Codepage 437 (United States, Canada) (CONFIG_NLS_CODEPAGE_437) [N/y/m/?] (NEW) Press Enter Codepage 737 (Greek) (CONFIG_NLS_CODEPAGE_737) [N/y/m/?] (NEW) Press Enter Codepage 775 (Baltic Rim) (CONFIG_NLS_CODEPAGE_775) [N/y/m/?] (NEW) Press Enter Codepage 850 (Europe) (CONFIG_NLS_CODEPAGE_850) [N/y/m/?] (NEW) Press Enter Codepage 852 (Central/Eastern Europe) (CONFIG_NLS_CODEPAGE_852) [N/y/m/?] (NEW) Press Enter Codepage 855 (Cyrillic) (CONFIG_NLS_CODEPAGE_855) [N/y/m/?] (NEW) Press Enter
- 187. Kernel Security &Optimization 0 CHAPTER 6 187 Codepage 857 (Turkish) (CONFIG_NLS_CODEPAGE_857) [N/y/m/?] (NEW) Press Enter Codepage 860 (Portuguese) (CONFIG_NLS_CODEPAGE_860) [N/y/m/?] (NEW) Press Enter Codepage 861 (Icelandic) (CONFIG_NLS_CODEPAGE_861) [N/y/m/?] (NEW) Press Enter Codepage 862 (Hebrew) (CONFIG_NLS_CODEPAGE_862) [N/y/m/?] (NEW) Press Enter Codepage 863 (Canadian French) (CONFIG_NLS_CODEPAGE_863) [N/y/m/?] (NEW) Press Enter Codepage 864 (Arabic) (CONFIG_NLS_CODEPAGE_864) [N/y/m/?] (NEW) Press Enter Codepage 865 (Norwegian, Danish) (CONFIG_NLS_CODEPAGE_865) [N/y/m/?] (NEW) Press Enter Codepage 866 (Cyrillic/Russian) (CONFIG_NLS_CODEPAGE_866) [N/y/m/?] (NEW) Press Enter Codepage 869 (Greek) (CONFIG_NLS_CODEPAGE_869) [N/y/m/?] (NEW) Press Enter Simplified Chinese charset (CP936, GB2312) (CONFIG_NLS_CODEPAGE_936) [N/y/m/?] (NEW) Press Enter Traditional Chinese charset (Big5) (CONFIG_NLS_CODEPAGE_950) [N/y/m/?] (NEW) Press Enter Japanese charsets (Shift-JIS, EUC-JP) (CONFIG_NLS_CODEPAGE_932) [N/y/m/?] (NEW) Press Enter Korean charset (CP949, EUC-KR) (CONFIG_NLS_CODEPAGE_949) [N/y/m/?] (NEW) Press Enter Thai charset (CP874, TIS-620) (CONFIG_NLS_CODEPAGE_874) [N/y/m/?] (NEW) Press Enter Hebrew charsets (ISO-8859-8, CP1255) (CONFIG_NLS_ISO8859_8) [N/y/m/?] (NEW) Press Enter Windows CP1250 (Slavic/Central European Languages) (CONFIG_NLS_CODEPAGE_1250) [N/y/m/?] (NEW) Press Enter Windows CP1251 (Bulgarian, Belarusian) (CONFIG_NLS_CODEPAGE_1251) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-1 (Latin 1; Western European Languages) (CONFIG_NLS_ISO8859_1) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-2 (Latin 2; Slavic/Central European Languages) (CONFIG_NLS_ISO8859_2) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-3 (Latin 3; Esperanto, Galician, Maltese, Turkish) (CONFIG_NLS_ISO8859_3) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-4 (Latin 4; old Baltic charset) (CONFIG_NLS_ISO8859_4) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-5 (Cyrillic) (CONFIG_NLS_ISO8859_5) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-6 (Arabic) (CONFIG_NLS_ISO8859_6) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-7 (Modern Greek) (CONFIG_NLS_ISO8859_7) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-9 (Latin 5; Turkish) (CONFIG_NLS_ISO8859_9) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-13 (Latin 7; Baltic) (CONFIG_NLS_ISO8859_13) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-14 (Latin 8; Celtic) (CONFIG_NLS_ISO8859_14) [N/y/m/?] (NEW) Press Enter NLS ISO 8859-15 (Latin 9; Western European Languages with Euro) (CONFIG_NLS_ISO8859_15) [N/y/m/?] (NEW) Press Enter NLS KOI8-R (Russian) (CONFIG_NLS_KOI8_R) [N/y/m/?] (NEW) Press Enter NLS KOI8-U/RU (Ukrainian, Belarusian) (CONFIG_NLS_KOI8_U) [N/y/m/?] (NEW) Press Enter NLS UTF8 (CONFIG_NLS_UTF8) [N/y/m/?] (NEW) Press Enter * * Console drivers * VGA text console (CONFIG_VGA_CONSOLE) [Y/n/?] Press Enter Video mode selection support (CONFIG_VIDEO_SELECT) [N/y/?] Press Enter * * Sound * Sound card support (CONFIG_SOUND) [Y/m/n/?] n * * USB support * Support for USB (CONFIG_USB) [Y/m/n/?] n * * USB Controllers * * * USB Device Class drivers * * * SCSI support is needed for USB Storage * * * USB Human Interface Devices (HID) * * * Input core support is needed for USB HID * * * USB Imaging devices *
- 188. Kernel Security &Optimization 0 CHAPTER 6 188 * * USB Multimedia devices * * * Video4Linux support is needed for USB Multimedia device support * * * USB Network adaptors * * * USB port drivers * * * USB Serial Converter support * * * USB Miscellaneous drivers * * * Kernel hacking * Kernel debugging (CONFIG_DEBUG_KERNEL) [N/y/?] Press Enter * * Grsecurity * Grsecurity (CONFIG_GRKERNSEC) [N/y/?] y Security level (Low, Medium, High, Customized) [Customized] Press Enter * * Buffer Overflow Protection * Openwall non-executable stack (CONFIG_GRKERNSEC_STACK) [N/y/?] y Gcc trampoline support (CONFIG_GRKERNSEC_STACK_GCC) [N/y/?] Press Enter Read-only kernel memory (CONFIG_GRKERNSEC_KMEM) [N/y/?] y * * Access Control Lists * Grsecurity ACL system (CONFIG_GRKERNSEC_ACL) [N/y/?] y ACL Debugging Messages (CONFIG_GR_DEBUG) [N/y/?] y Extra ACL Debugging Messages (CONFIG_GR_SUPERDEBUG) [N/y/?] Press Enter Denied capability logging (CONFIG_GRKERNSEC_ACL_CAPLOG) [N/y/?] y Path to gradm (CONFIG_GRADM_PATH) [/sbin/gradm] Press Enter Maximum tries before password lockout (CONFIG_GR_MAXTRIES) [3] 2 Time to wait after max password tries, in seconds (CONFIG_GR_TIMEOUT) [30] Press Enter * * Filesystem Protections * Proc restrictions (CONFIG_GRKERNSEC_PROC) [N/y/?] y Restrict to user only (CONFIG_GRKERNSEC_PROC_USER) [N/y/?] y Additional restrictions (CONFIG_GRKERNSEC_PROC_ADD) [N/y/?] y Linking restrictions (CONFIG_GRKERNSEC_LINK) [N/y/?] y FIFO restrictions (CONFIG_GRKERNSEC_FIFO) [N/y/?] y Secure file descriptors (CONFIG_GRKERNSEC_FD) [N/y/?] y Chroot jail restrictions (CONFIG_GRKERNSEC_CHROOT) [N/y/?] y Restricted signals (CONFIG_GRKERNSEC_CHROOT_SIG) [N/y/?] y Deny mounts (CONFIG_GRKERNSEC_CHROOT_MOUNT) [N/y/?] y Deny double-chroots (CONFIG_GRKERNSEC_CHROOT_DOUBLE) [N/y/?] y Enforce chdir("/") on all chroots (CONFIG_GRKERNSEC_CHROOT_CHDIR) [N/y/?] y Deny (f)chmod +s (CONFIG_GRKERNSEC_CHROOT_CHMOD) [N/y/?] y Deny mknod (CONFIG_GRKERNSEC_CHROOT_MKNOD) [N/y/?] y Deny ptraces (CONFIG_GRKERNSEC_CHROOT_PTRACE) [N/y/?] y Restrict priority changes (CONFIG_GRKERNSEC_CHROOT_NICE) [N/y/?] y Capability restrictions within chroot (CONFIG_GRKERNSEC_CHROOT_CAPS) [N/y/?] Press Enter Secure keymap loading (CONFIG_GRKERNSEC_KBMAP) [N/y/?] y * * Kernel Auditing * Single group for auditing (CONFIG_GRKERNSEC_AUDIT_GROUP) [N/y/?] Press Enter Exec logging (CONFIG_GRKERNSEC_EXECLOG) [N/y/?] Press Enter Log execs within chroot (CONFIG_GRKERNSEC_CHROOT_EXECLOG) [N/y/?] y Chdir logging (CONFIG_GRKERNSEC_AUDIT_CHDIR) [N/y/?] Press Enter
- 189. Kernel Security &Optimization 0 CHAPTER 6 189 (Un)Mount logging (CONFIG_GRKERNSEC_AUDIT_MOUNT) [N/y/?] Press Enter IPC logging (CONFIG_GRKERNSEC_AUDIT_IPC) [N/y/?] y Ptrace logging (CONFIG_GRKERNSEC_AUDIT_PTRACE) [N/y/?] Press Enter Signal logging (CONFIG_GRKERNSEC_SIGNAL) [N/y/?] y Fork failure logging (CONFIG_GRKERNSEC_FORKFAIL) [N/y/?] y Set*id logging (CONFIG_GRKERNSEC_SUID) [N/y/?] Press Enter Log set*ids to root (CONFIG_GRKERNSEC_SUID_ROOT) [N/y/?] y Time change logging (CONFIG_GRKERNSEC_TIME) [N/y/?] y * * Executable Protections * Exec process limiting (CONFIG_GRKERNSEC_EXECVE) [N/y/?] y Dmesg(8) restriction (CONFIG_GRKERNSEC_DMESG) [N/y/?] y Randomized PIDs (CONFIG_GRKERNSEC_RANDPID) [N/y/?] y Altered default IPC permissions (CONFIG_GRKERNSEC_IPC) [N/y/?] Press Enter Limit uid/gid changes to root (CONFIG_GRKERNSEC_TTYROOT) [N/y/?] y Deny physical consoles (tty) (CONFIG_GRKERNSEC_TTYROOT_PHYS) [N/y/?] Press Enter Deny serial consoles (ttyS) (CONFIG_GRKERNSEC_TTYROOT_SERIAL) [N/y/?] y Deny pseudo consoles (pty) (CONFIG_GRKERNSEC_TTYROOT_PSEUDO) [N/y/?] Press Enter Fork-bomb protection (CONFIG_GRKERNSEC_FORKBOMB) [N/y/?] y GID for restricted users (CONFIG_GRKERNSEC_FORKBOMB_GID) [1006] Press Enter Forks allowed per second (CONFIG_GRKERNSEC_FORKBOMB_SEC) [40] Press Enter Maximum processes allowed (CONFIG_GRKERNSEC_FORKBOMB_MAX) [20] 35 Trusted path execution (CONFIG_GRKERNSEC_TPE) [N/y/?] y Glibc protection (CONFIG_GRKERNSEC_TPE_GLIBC) [N/y/?] y Partially restrict non-root users (CONFIG_GRKERNSEC_TPE_ALL) [N/y/?] y GID for untrusted users: (CONFIG_GRKERNSEC_TPE_GID) [1005] Press Enter Restricted ptrace (CONFIG_GRKERNSEC_PTRACE) [N/y/?] y Allow ptrace for group (CONFIG_GRKERNSEC_PTRACE_GROUP) [N/y/?] Press Enter * * Network Protections * Randomized IP IDs (CONFIG_GRKERNSEC_RANDID) [N/y/?] y Randomized TCP source ports (CONFIG_GRKERNSEC_RANDSRC) [N/y/?] y Randomized RPC XIDs (CONFIG_GRKERNSEC_RANDRPC) [N/y/?] y Altered Ping IDs (CONFIG_GRKERNSEC_RANDPING) [N/y/?] y Randomized TTL (CONFIG_GRKERNSEC_RANDTTL) [N/y/?] y Socket restrictions (CONFIG_GRKERNSEC_SOCKET) [N/y/?] y Deny any sockets to group (CONFIG_GRKERNSEC_SOCKET_ALL) [N/y/?] y GID to deny all sockets for: (CONFIG_GRKERNSEC_SOCKET_ALL_GID) [1004] Press Enter Deny client sockets to group (CONFIG_GRKERNSEC_SOCKET_CLIENT) [N/y/?] Press Enter Deny server sockets to group (CONFIG_GRKERNSEC_SOCKET_SERVER) [N/y/?] Press Enter * * Sysctl support * Sysctl support (CONFIG_GRKERNSEC_SYSCTL) [N/y/?] Press Enter * * Miscellaneous Features * Seconds in between log messages(minimum) (CONFIG_GRKERNSEC_FLOODTIME) [30] Press Enter BSD-style coredumps (CONFIG_GRKERNSEC_COREDUMP) [N/y/?] y *** End of Linux kernel configuration. *** Check the top-level Makefile for additional configuration. *** Next, you must run 'make dep'.
- 190. Kernel Security &Optimization 0 CHAPTER 6 190 Compiling the Kernel This section applies to both Monolithic and Modularized kernels. Once the kernel configuration has been completed, return to the /usr/src/linux directory (if you are not already in it) and compile the new kernel. You do so by using the following command: • To compile the Kernel, use the following command: [root@deep linux]# make dep; make clean; make bzImage This line contains three commands in one. The first one, make dep, actually takes your configuration and builds the corresponding dependency tree. This process determines what gets compiled and what doesn’t. The next step, make clean, erases all previous traces of a compilation so as to avoid any mistakes in which the wrong version of a feature gets tied into the kernel. Finally, make bzImage does the full compilation of the kernel. After the process is complete, the kernel is compressed and ready to be installed on your system. Before we can install the new kernel, we must know if we need to compile the corresponding modules. This is required ONLY if you said yes to “Enable loadable module support (CONFIG_MODULES)” and have compiled some options in the kernel configuration above as a module (See Modularized kernel configuration). In this case, you must execute the following commands: • To compile the corresponding modules for your kernel, use the following commands: [root@deep linux]# make modules [root@deep linux]# make modules_install WARNING: The make modules and make modules_install commands are required ONLY if you say yes to “Enable loadable module support (CONFIG_MODULES)” in your kernel configurations (See Modularized kernel configuration) because you want to build a modularized kernel. Installing the Kernel This section applies to both the Monolithic and Modularized kernel. Ok, the kernel has been configured, compiled and is now ready to be installed your system. Below are the steps required to install all the necessary kernel components in your system. Step 1 Copy the file /usr/src/linux/arch/i386/boot/bzImage from the kernel source tree to the /boot directory, and give it an appropriate new name. • To copy the bzImage file to the /boot directory, use the following commands: [root@deep /]# cd /usr/src/linux/ (if you are not already in it) [root@deep linux]# cp arch/i386/boot/bzImage /boot/vmlinuz-2.4.18
- 191. Kernel Security &Optimization 0 CHAPTER 6 191 Step 2 A new System.map file is generated when you compile a kernel, and is a list of all the addresses in that kernel and their corresponding symbols. Every time that you create a new kernel, such a file System.map is created and saved in /usr/src/linux. It's a text file, which is read by a few programs to do address <-> symbol translation, and which you need if you ever get an Oops. Certain commands, like klog, ps, and lsof, use the System.map file to get the name of kernel symbols. Without it some commands like lsof will complain that they can't find a System.map file to match the currently booted kernel. Copy the file /usr/src/linux/System.map from the kernel source tree to the /boot directory, and give it an appropriate new name. • To copy the System.map file to the /boot directory, use the following commands: [root@deep /]# cd /usr/src/linux/ (if you are not already in it) [root@deep linux]# cp System.map /boot/System.map-2.4.18 Step 3 Move into the /boot directory and rebuild the links vmlinuz and System.map. • To rebuild the vmlinuz and System.map files, use the following commands: [root@deep linux]# cd /boot/ [root@deep /boot]# ln -fs vmlinuz-2.4.18 vmlinuz [root@deep /boot]# ln -fs System.map-2.4.18 System.map We must rebuild the links of vmlinuz and System.map to point them to the new installed kernel version. Without the new links LILO or GRUB program will look, by default, for the old version of your Linux kernel. Step 4 Remove obsolete and unnecessary files under the /boot directory to increase disk space: • To remove obsolete and unnecessary files under the /boot directory, use commands: [root@deep /]# cd /boot/ (if you are not already in it) [root@deep /boot]# rm -f module-info [root@deep /boot]# rm -f initrd-2.4.x.img The module-info is a link, which points to the old modules directory of your original kernel. Since we have installed a brand new kernel, we don’t need to keep this broken link. The initrd-2.4.x.img is a file that contains an initial RAM disk image that serves as a system before the disk is available. This file is only available and is installed by the Linux initial setup installation if your system has a SCSI adapter present and only if your system has a SCSI adapter. If we use and have a SCSI system, the required driver now will have been incorporated into our new Linux kernel since we have build it by answering Y to the question related to our SCSI model during the configuration of the kernel, so we can safely remove this file (initrd- 2.4.x.img).
- 192. Kernel Security &Optimization 0 CHAPTER 6 192 Step 5 Create a new Linux kernel directory that will handle all the header files related to Linux kernel for future compilation of other programs on your system. Recall, we had created two symlinks under the /usr/include directory that pointed to the Linux kernel header files to be able to compile it without receiving errors and also be able to compile future programs. The /usr/include directory is where all the header files for your Linux system are kept for reference and dependencies when you compile and install new programs. The asm, and linux links are used when programs need to know some functions which are compile-time specific to the kernel installed on your system. Programs call other headers as well in the /usr/include directory when they must know specific information, dependencies, etc of your system. • To create a new Linux kernel directory to handle all header files, use the commands: [root@deep /]# mkdir -p /usr/src/linux-2.4.18/include [root@deep /]# cd /usr/src/linux/ [root@deep linux]# cp -r include/asm-i386 ../linux-2.4.18/include/ [root@deep linux]# cp -r include/linux ../linux-2.4.18/include/ [root@deep linux]# cd ../ [root@deep src]# rm -rf /usr/src/linux [root@deep src]# cd /usr/src/ (to be sure that we are into the src directory) [root@deep src]# ln -s /usr/src/linux-2.4.18 linux First we create a new directory named “linux-2.4.18” based on the version of the kernel we have installed for easy interpretation, then we copy directories asm-i386, and linux from /usr/src/linux/include to our new location /usr/src/linux-2.4.18/include. After we remove the entire source directory where we compiled the new kernel, we create a new symbolic link named “linux” under /usr/src that points to our new /usr/src/linux- 2.4.18 directory. With these steps, future compiled programs will know where to look for headers related to the kernel on your server. NOTE: This step will allow us to gain space on our hard drive and will reduce the security risks. The Linux kernel source directory handles a lot files and is about 94M in size when uncompressed. With the procedure described above, our Linux kernel directory began approximately 4M in size so we save 90MB for the same functionalities. Verifying or upgrading your boot loader Once the new kernel image has been installed on your server, we have to inform our boot loader about it. This is done inside the configuration file of your boot loader software. Below I show you how to do it depending if you use GRUB or LILO as your boot loader. LILO: This step applies only if you use LILO as your boot loader on the system. If you use GRUB as your boot loader instead of LILO (highly recommended), then you can skip this section and go directly to the next one.
- 193. Kernel Security &Optimization 0 CHAPTER 6 193 Step 1 You need to edit the lilo.conf file to make your new kernel one of the boot time options: • Edit the lilo.conf file (vi /etc/lilo.conf) and make the appropriate change on the line that reads “image=/boot/vmlinuz-x.x.x”. [root@deep /]# vi /etc/lilo.conf boot=/dev/sda map=/boot/map install=/boot/boot.b timeout=00 default=linux restricted password=somepasswd image=/boot/vmlinuz label=linux read-only root=/dev/sda6 Step 2 Once the necessary modifications have been made in the /etc/lilo.conf file as shown above, we update our lilo.conf file for the change to take effect. • This can be done with the following command: [root@deep /]# /sbin/lilo Added linux * GRUB: This step applies only if you use GRUB as your boot loader on the system. In most case, GRUB does not need to be updated when you install a new kernel on your computer but here we have to verify inside our grub.conf file if all default setting are still available and configured because when we uninstall Red Hat kernel RPM package, the software automatically remove some needed parameters from the GRUB configuration file. Step 1 Edit your GRUB configuration file and be sure that everything is correct and look like the following. Your setting should differ from the example below. • Edit the grub.conf file (vi /etc/grub.conf) and check your setting. [root@deep /]# vi /etc/grub.conf default 0 timeout 00 title Red Hat Linux kernel (hd0,0)/vmlinuz ro root=/dev/sda5
- 194. Kernel Security &Optimization 0 CHAPTER 6 194 Reconfiguring /etc/modules.conf file This section applies only if you chose to install a Modularized Kernel on your system. The /etc/modules.conf file represents the (optional) configuration file for loading kernel modules on your system. It is used to modify the behavior of modprobe and depmod programs. This file consists of a set of lines with different parameters. It is important after each upgrade of a modularized kernel to verify if all the information and parameters contained inside it are valid and correct. All the contents of the /etc/modules.conf file apply only for systems where the kernel has been configured with modules (modularized kernel). So if you have recompiled your new kernel with some new options as modules or if you have removed some modules from it, it is important to update or remove the modules.conf file to reflect the changes and eliminate possible error message during booting. As an example, the following is the content of the modules.conf file on my system. Linux has added these parameters automatically, depending of the system hardware during the primary install stage of the operating system. alias scsi_hostadapter aic7xxx alias eth0 eepro100 alias eth1 eepro100 alias parport_lowlevel parport_pc alias usb-controller uhci One important use of the modules.conf file is the possibility of using the “alias” directive to give alias names to modules and link object files to a module. After recompilation of the kernel, and depending on how we have answered the different kernel questions during kernel configuration, it may be possible that we need to make some adjustments to the default parameters, especially if we have answered yes during kernel configuration to some devices available in our system, like network cards and SCSI adapters. If the configuration file /etc/modules.conf is missing, or if any directive is not overridden, the default will be to look under /lib/modules directory containing modules compiled for the current release of the kernel. Therefore, we can remove the /etc/modules.conf file from the system and let the modprobe and depmod programs manage all existing modules for us. To summarize, you can: 1) Keep the modules.conf file; only kernel options which you have answered m during kernel configuration time (of course only if these modules did exist into modules.conf). Any kernel options where you have answered yes or no will not appear into the modules.conf file. 2) Or remove the /etc/modules.conf file from your system and let modprobe and depmod programs manage all existing modules for you. On a server environment, I prefer to use this choice.
- 195. Kernel Security &Optimization 0 CHAPTER 6 195 Rebooting your system to load the new kernel Whether you have installed a new Monolithic Kernel where codes and drivers are compiled into the kernel and are always loaded or a Modularized Kernel where some segment of codes are compiled into the kernel as a module and loaded on demand, it is time to Reboot your system and test your results. • To reboot your Linux system, use the following command: [root@deep /]# reboot When the system is rebooted and you are logged in, verify the new version of your kernel with the following command: • To verify the version of your new kernel, use the following command: [root@deep /]# uname -a Linux dev 2.4.18-grsec-1.9.4 #1 Wed Jun 19 15:14:55 EDT 2002 i686 unknown Congratulations! Delete programs, edit files pertaining to modules This section applies only if you chose to install a Monolithic Kernel on your system. By default when you install Linux for the first time (like we did), the kernel is built as a Modularized Kernel. This means that each device or function we need exists as a module and is controlled by the Kernel Daemon program named kmod. kmod automatically loads some modules and functions into memory as they are needed, and unloads them when they’re no longer being used. kmod and other module management programs included in the modutils RPM package use the modules.conf file located in the /etc directory to know for example which Ethernet card you have, if your Ethernet card requires special configuration and so on. If we don’t use any modules in our newly compiled kernel because we have compiled the kernel as a Monolithic Kernel and ONLY in this case, we can remove the modules.conf file and completely uninstall the modutils RPM package. • To remove the modules.conf file, use the following command: [root@deep /]# rm -f /etc/modules.conf • To uninstall the modutils package, use the following command: [root@deep /]# rpm -e --nodeps modutils WARNING: Once again, the above is required only if you said no to “Enable loadable module support (CONFIG_MODULES)” in your kernel configuration because you have decided to build a Monolithic Kernel.
- 196. Kernel Security &Optimization 0 CHAPTER 6 196 Making a new rescue floppy for Modularized Kernel This section applies only if you chose to install a Modularized Kernel on your system. Immediately after you successfully start your system and log in as root, you should create a new emergency boot floppy disk. The procedure to achieve it is the same as shown at the beginning of this chapter related to Linux Kernel. Please go back to the beginning of this chapter and follow the procedures to recreate a new emergency boot floppy disk suitable for the new install Linux kernel on your system. Don’t forget to test the boot disk to be sure that it works. The mkbootdisk program runs only on Modularized Kernel. So you can’t use it on a Monolithic Kernel; instead create an emergency boot floppy disk for Monolithic kernel as shown below. Making a emergency boot floppy disk for Monolithic Kernel This section applies only if you chose to install a Monolithic Kernel in your system. Because it is possible to create a rescue floppy only on modularized kernel, we must find another way to boot our Linux system for a monolithic kernel if the Linux kernel on the hard disk is damaged. This is possible with a Linux emergency boot floppy disk. You should create it immediately after you successfully start your system and log in as root. To create the emergency boot floppy, follow these steps: Step 1 Insert a floppy disk and format it with the following command: [root@deep /]# fdformat /dev/fd0H1440 Double-sided, 80 tracks, 18 sec/track. Total capacity 1440 kB. Formatting ... done Verifying ... done Step 2 Copy the actual file “vmlinuz” from the /boot directory to the floppy disk: [root@deep /]# cp /boot/vmlinuz /dev/fd0H1440 cp: overwrite `/dev/fd0H1440'? y NOTE: The vmlinuz file is a symbolic link that points to the real Linux kernel. Step 3 Determine the kernel’s root device with the following command: [root@deep /]# rdev /dev/sda6 / NOTE: The kernel’s root device is the disk partition where the root file system is located. In this example, the root device is /dev/sda6; the device name should be different on your system.
- 197. Kernel Security &Optimization 0 CHAPTER 6 197 Step 4 Set the kernel’s root device with the following command: [root@deep /]# rdev /dev/fd0H1440 /dev/sda6 NOTE: To set the kernel’s root device, use the device reported by the “rdev” command utility in the previous step. Step 5 Mark the root device as read-only with the following command: [root@deep /]# rdev -R /dev/fd0H1440 1 NOTE: This causes Linux to initially mount the root file system as read-only. By setting the root device as read-only, you avoid several warnings and error messages. Step 6 Now put the boot floppy in the drive A: and reboot your system with the following command to be sure that your new boot disk is working: [root@something /]# reboot Following these guidelines, you will now have a boot floppy with a known working kernel in case of problems with the upgrade. I recommend rebooting the system with the floppy to make sure that the floppy works correctly. Step 7 Because the mkbootdisk and dosfstools program are required only when you have a Modularized kernel installed in your Linux system, we can remove the unneeded mkbootdisk and dosfstools packages from the system when we have a Monolithic kernel installed on our server. • To uninstall the mkbootdisk and dosfstools utility, use the following command: [root@deep /]# rpm –e mkbootdisk dosfstools
- 199. Process file systemmanagement IN THIS CHAPTER 1. What is sysctl? 2. /proc/sys/vm: The virtual memory subsystem of Linux 3. /proc/sys/fs: The file system data of Linux 4. /proc/sys/net/ipv4: IPV4 settings of Linux 5. Other possible optimization of the system
- 201. Process file systemmanagement 0 CHAPTER 7 201 Linux /proc Abstract The /proc (the process file system), also known as a pseudo-filesystem, is used as an interface to kernel data structures. It doesn’t exist, neither the /proc directory nor its subdirectories or its files actually exist. Most of the files in this special directory are read-only and cannot be changed, but some kernel variables can be changed. It is these files that we will talk about in this chapter of the book. It is important to note that the /proc filesystem is structured in a hierarchy. Most of the entries in the /proc directory are a decimal number, corresponding to a process-ID running on the system. These entries are themselves subdirectories and access to process state that is provided by additional files contained within each subdirectory. Have you ever thought about where all the processes running in the background of your system are handled and managed by the kernel? The answer is the /proc filesystem directory of Linux. But the /proc filesystem doesn’t handle only process ID of the system; it is also responsible for providing and managing all access to the state of each information on the system. This information is comprised of CPU, devices, IDE, SCSI, interrupts, io-ports, memories, modules, partitions, PCI information and much more. Just take a quick look inside your /proc filesystem directory to get an idea of the available features controlled by the kernel through the /proc filesystem. We can read the contents of this information to get an idea of what processor, PCI, network cards, kernel version, partitions, etc that we have on our system. As we said before, not all features available in the /proc filesystem are customizable, most are managed by the kernel and cannot be changed. Most are well controlled by the kernel and should not require any modifications since the kernel does a good job with them. Some can, and need to be, changed and customized to better fit your system resources, and increase security. It is those customizable features related to performance and security of the Linux system under the /proc filesystem that we will explain and customize in this chapter. This is possible with the /etc/sysctl.conf file which contains values that change the default parameters of customizable features in the /proc filesystem. To recap, systcl.conf is the configuration file that talks to sysctl(8) which is an interface that allows you to make changes to a running Linux system. We use systcl.conf to talk to the kernel and say for example: hey, I need more power on the virtual memory, please change your value to this value. Throughout this chapter, we’ll often use it to customize our /proc filesystem on Linux to better utilize resources, power and security of our particular machine. Remember that everyone have a different computer with different hardware, setting and this is why changing some default customizable values in the /proc directory could make the difference on security and speed. In this chapter, we will talk about customized parameters available under the /proc/sys directory since most of all changeable parameters are located under this directory. We will talk about virtual memory, file system, TCP/IP stack security and performance.
- 202. Process file systemmanagement 0 CHAPTER 7 202 What is sysctl? sysctl is an interface that allows you to make changes to a running Linux system. It serves two functions: to read and to modify system settings. • To view all readable variables, use the following command: [root@deep /]# sysctl -a • To read a particular variable, for example, fs.file-max, use the following command: [root@deep /]# sysctl fs.file-max fs.file-max = 8192 • To set a particular variable for fs.file-max, use the following command: [root@deep /]# sysctl -w fs.file-max=5536 fs.file-max = 16384 Settings of sysctl variables are usually strings, numbers, or booleans (a boolean being 1 for yes or a 0 for no). If you set and change variable manually with the sysctl command as show above, your changes will not resists on the next reboot of the system. For this reason, we will use and show you further down in this chapter how to make your changes permanent even on possible reboot of the server by using the /etc/sysctl.conf file. /proc/sys/vm: The virtual memory subsystem of Linux All parameters described in this chapter reside under the /proc/sys/vm directory of the server and can be used to tune the operation of the virtual memory (VM) subsystem of the Linux kernel. Be very careful when attempting this. You can optimize your system, but you can also cause it to crash. Since every system is different, you'll probably want some control over this piece of the system. Finally, these are advanced setting and if you don’t understand them, then don’t try to play in this area or try to use all the examples below in your system. Remember that all systems are different and require different settings and customizations. The majority of the following hacks will work fine on a server with >= at 512MB of RAM or a minimum of 256MB of RAM. Below this amount of memory, nothing is guaranteed and the default setting will just be fine for you. Next I’ll show you parameters that can be optimized. All suggestions I make in this section are valid for all kinds of servers. The only difference depends on the amount of RAM your machine has and this is where the settings will change. The above figure shows a snapshot of /proc/sys/vm directory on an OpenNA Linux & Red Hat Linux system running kernel version 2.4. Please note that this picture may look different on your system.
- 203. Process file systemmanagement 0 CHAPTER 7 203 The bdflush parameters: The bdflush file is closely related to the operation of the virtual memory (VM) subsystem of the Linux kernel and has a little influence on disk usage. This file /proc/sys/vm/bdflush controls the operation of the bdflush kernel daemon. We generally tune this file to improve file system performance. By changing some values from the defaults shown below, the system seems more responsive; e.g. it waits a little more to write to disk and thus avoids some disk access contention. The bdflush parameters currently contain 9 integer values, of which 4 are actually used by the kernel. Only first, fifth, sixth and the seventh parameters are used by the kernel for bdflush setup and all the other parameters are not used and their values are set to ‘0’. Parameter 1 (nfract): The bdflush parameter 1 governs the maximum number of dirty buffers in the buffer cache. Dirty means that the contents of the buffer still have to be written to disk (as opposed to a clean buffer, which can just be forgotten about). Setting this to a high value means that Linux can delay disk writes for a long time, but it also means that it will have to do a lot of I/O (Input/Output) at once when memory becomes short. A low value will spread out disk I/O more evenly at the cost of more frequent I/O operations. The default value is 40%, the minimum is 0%, and the maximum is 100%. We improve the default value here. Parameter 2 (dummy1): This parameter is unused by the system so we don’t need to change the default ones. Parameter 3 (dummy2): This parameter is unused by the system so we don’t need to change the default ones. Parameter 4 (dummy3): This parameter is unused by the system so we don’t need to change the default ones. Parameter 5 (interval): The bdflush parameter 5 specifies the minimum rate at which kupdate will wake and flush. The value is expressed in jiffies (clockticks), the number of jiffies per second is normally 100. Thus, x*HZ is x seconds. The default value is 5 seconds, the minimum is 0 seconds, and the maximum is 600 seconds. We keep the default value here. Parameter 6 (age_buffer): The bdflush parameter 6 governs the maximum time Linux waits before writing out a dirty buffer to disk. The value is in jiffies. The default value is 30 seconds, the minimum is 1 second, and the maximum 6,000 seconds. We keep the default value here. Parameter 7 (nfract_sync): The bdflush parameter 7 governs the percentage of buffer cache that is dirty before bdflush activates synchronously. This can be viewed as the hard limit before bdflush forces buffers to disk. The default is 60%, the minimum is 0%, and the maximum is 100%. We improve the default value here. Parameter 8 (dummy4): This parameter is unused by the system so we don’t need to change the default ones. Parameter 9 (dummy5): This parameter is unused by the system so we don’t need to change the default ones.
- 204. Process file systemmanagement 0 CHAPTER 7 204 The default kernel setup for the bdflush parameters is: "40 64 64 256 500 3000 60 0 0" The default setup for the bdflush parameters under OpenNA Linux is: "60 64 64 256 500 3000 80 0 0" The default setup for the bdflush parameters under Red Hat Linux is: "30 64 64 256 500 3000 60 0 0" Step 1 To change the values of bdflush, type the following command on your terminal: • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following line: # Improve file system performance vm.bdflush = 60 64 64 256 500 3000 80 0 0 Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command in your terminal screen: [root@deep /]# sysctl -w vm.bdflush="60 64 64 256 500 3000 80 0 0" The kswapd parameter: The kswapd file is related to the kernel swapout daemon. This file /proc/sys/vm/kswapd frees memory on the system when it gets fragmented or full. Its task is to keep the memory management system operating efficiently. Since every system is different, you'll probably want some control over this piece of the system. There are three parameters to tune in this file and two of them (tries_base and swap_cluster) have the largest influence on system performance. The kswapd file can be used to tune the operation of the virtual memory (VM) subsystem of the Linux kernel. Parameter 1 (tries_base): The kswapd parameter 1 specifies the maximum number of pages kswapd tries to free in one round. Usually this number will be divided by 4 or 8, so it isn't as big as it looks. Increase this number to cause swap to be released faster, and increase overall swap throughput. The default value is 512 pages. We keep the default value here.
- 205. Process file systemmanagement 0 CHAPTER 7 205 Parameter 2 (tries_min): The kswapd parameter 2 specifies the minimum number of pages kswapd tries to free a least each time it is called. Basically it's just there to make sure that kswapd frees some pages even when it's being called with minimum priority. The default value is 32 pages. We keep the default value here. Parameter 3 (swap_cluster): The kswapd parameter 3 specifies the number of pages kswapd writes in one iteration. You want this large to increase performance so that kswapd does its I/O in large chunks and the disk doesn't have to seek often, but you don't want it to be too large since that would flood the request queue. The default value is 8 pages. We improve the default value here. The default kernel setup for the kswapd parameters is: "512 32 8" The default setup for the kswapd parameters under OpenNA Linux is: "512 32 32" The default setup for the kswapd parameters under Red Hat Linux is: "512 32 8" Step 1 To change the values of kswapd, type the following command on your terminal: • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Increase swap bandwidth system performance vm.kswapd = 512 32 32 Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w vm.kswapd=”512 32 32”
- 206. Process file systemmanagement 0 CHAPTER 7 206 The overcommit_memory parameter: The overcommit_memory parameter is simply a flag that enables memory overcommitment. Memory overcommitment is a procedure to check that a process has enough memory to allocate a new virtual mapping. When this flag is 0, the kernel checks before each malloc() to see if there's enough memory left. If the flag is 1, the system pretends there's always enough memory and don't make the check on the system. This feature can be very useful ONLY on big servers with a lot of pysical memories available (>= 2GB) because there are a lot of programs that malloc() huge amounts of memory "just-in-case" and don't use much of it. The default kernel setup for the overcommit_memory parameter is: "0" The default setup for the overcommit_memory parameter under OpenNA Linux is: "0" The default setup for the overcommit_memory parameter under Red Hat Linux is: "0" Step 1 To change the value of overcommit_memory, type the following command on your terminal: • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Enables/Disables memory overcommitment vm.overcommit_memory = 0 Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] WARNING: Only change the default value of 0 to become 1 on systems with more than 2GB of RAM. Recall that on small systems the value must be set to 0 (overcommit_memory=0). There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w overcommit_memory=0
- 207. Process file systemmanagement 0 CHAPTER 7 207 The page-cluster parameter: The Linux virtual memory subsystem avoids excessive disk seeks by reading multiple pages on a page fault. The number of pages it reads is highly dependent on the amount of memory in your machine. The number of pages the kernel reads in at once is equal to 2 ^ page-cluster. Values above 2 ^ 5 don't make much sense for swap because we only cluster swap data in 32-page groups. The page-cluster parameter is used to tune the operation of the virtual memory (VM) subsystem of the Linux kernel. The default kernel setup for the kswapd parameter is: "3" The default setup for the kswapd parameter under OpenNA Linux is: "5" The default setup for the kswapd parameter under Red Hat Linux is: "4" Step 1 To change the value of page-cluster, type the following command on your terminal: • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Increase number of pages kernel reads in at once vm.page-cluster = 5 Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w vm.page-cluster=5
- 208. Process file systemmanagement 0 CHAPTER 7 208 The pagetable_cache parameter: The kernel keeps a number of page tables in a per-processor cache (this helps a lot on SMP systems). The cache size for each processor will be between the low and the high value. On SMP systems it is used so that the system can do fast pagetable allocations without having to acquire the kernel memory lock. For large systems, the settings are probably OK. For normal systems they won't hurt a bit. For small systems (<16MB RAM) and on a low-memory, single CPU system it might be advantageous to set both values to 0 so you don't waste the memory. The default kernel setup for the kswapd parameters is: "25 50" The default setup for the kswapd parameters under OpenNA Linux is: "25 50" The default setup for the kswapd parameters under Red Hat Linux is: "25 50" Step 1 To change the values of pagetable_cache, type the following command on your terminal: • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Improve number of page tables keeps in a per-processor cache vm.pagetable_cache = 25 50 Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] WARNING: Only change these values on systems with multiple processors (SMP) or on small systems (single processor) with less than 16MB of RAM. Recall that on small systems the both values must be set to 0 (vm.pagetable_cache = 0 0). There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w vm.pagetable_cache=”25 50”
- 209. Process file systemmanagement 0 CHAPTER 7 209 /proc/sys/fs: The file system data of Linux All parameters described later in this chapter reside under the /proc/sys/fs directory of the server and can be used to tune and monitor miscellaneous things in the operation of the Linux kernel. Be very careful when attempting this. You can optimize your system, but you can also cause it to crash. Since every system is different, you'll probably want some control over these pieces of the system. Finally, these are advanced settings and if you don’t understand them, then don’t play in this area or try to use all the examples below in your system. Remember that all systems are different and required different setting and customization. Below I show you only parameters that can be optimized for the system. All suggestions I enumerate in this section are valid for every kind of servers. The only difference depends on the amount of MB of RAM your machines have and this is where settings will change. The above figure shows a snapshot of /proc/sys/fs directory on a OpenNA Linux & Red Hat Linux system running kernel version 2.4. Please note that this picture may look different on your system. The file-max & file-nr parameters: The file-max and file-nr files work together on Linux, we use the file-max parameter to sets the maximum number of file-handles that the Linux kernel will allocate and the file-nr file to get information about the number of allocated file handles, the number of used file handles and the maximum number of file handles presently on the system. A large-scale production server may easily require many thousands of file-handles, depending on the kind and number of services running concurrently on the server. On busy servers where many services are running concurrently, we generally tune this file (file-max) to improve the number of open files available on the system. It is important to note that you need to increase the limit of open files available on your server ONLY when you get lots of error messages about running out of file handles. If you don’t receive this kind of error message, you really DON’T need to increase the default value. • To know the number of allocated file handles, the number of used file handles and the maximum number of file handles on your system, use the following command: [root@deep /]# cat /proc/sys/fs/file-nr 405 137 8192
- 210. Process file systemmanagement 0 CHAPTER 7 210 The first value (405) in our result is the number of allocated file handles, the second value (137) is the number of used file handles, and the last value (8192) is the maximum number of file handles. When the allocated file handles (405) come close to the maximum (8192), but the number of actually used ones (137) is far less, you've encountered a peak in your usage of file handles and you don't need to increase the maximum. The default kernel setup is suitable for most of us. The default kernel setup for the file-max parameter is: "8192" The default setup for the file-max parameter under OpenNA Linux is: "8192" The default setup for the file-max parameter under Red Hat Linux is: "8192" Step 1 To adjust the value of file-max, type the following command on your terminal: • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Increase limit of file-handles fs.file-max = 8192 Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w fs.file-max=8192
- 211. Process file systemmanagement 0 CHAPTER 7 211 /proc/sys/net/ipv4: IPV4 settings of Linux All parameters described below reside under the /proc/sys/net/ipv4 directory of the server and can be used to control the behavior of the IPv4 subsystem of the Linux kernel. Below I show you only the parameters, which can be used for the network security of the system. The above figure shows a snapshot of /proc/sys/net/ipv4 directory on a OpenNA Linux & Red Hat Linux system running kernel version 2.4. Please note that this picture may look different on your system.
- 212. Process file systemmanagement 0 CHAPTER 7 212 Prevent your system responding to ping request: Preventing your system for responding to ping requests can make a big improvement in your network security since no one can ping your server and receive an answer. The TCP/IP protocol suite has a number of weaknesses that allows an attacker to leverage techniques in the form of covert channels to surreptitiously pass data in otherwise benign packets. Step 1 Preventing your server from responding to ping requests can help to minimize this problem. Not responding to pings would at least keep most "crackers" out because they would never know it's there. ICMP blocking can hurt the performance of long-duration TCP connections, and this is due to the fact that MTU discovery relies on ICMP packets to work. • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Enable/Disable ignoring ping request net.ipv4.icmp_echo_ignore_all = 1 When this key is on (1), the computer will ignore all ICMP packets. It is not recommended to turn on this key, except in special situations when you receive an ICMP packet based attack. In the above parameter, we enable this option. Step 2 Once the configuration has been set, you must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w net.ipv4.icmp_echo_ignore_all=1
- 213. Process file systemmanagement 0 CHAPTER 7 213 Refuse responding to broadcasts request: As for the ping request, it’s also important to disable broadcast requests. When a packet is sent to an IP broadcast address (i.e. 192.168.1.255) from a machine on the local network, that packet is delivered to all machines on that network. Then all the machines on a network respond to this ICMP echo request and the result can be severe network congestion or outages (Denial-of-Service attacks). See the RFC 2644 for more information. Step 1 To disable broadcast requests, type the following command on your terminal. • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Enable/Disable ignoring broadcasts request net.ipv4.icmp_echo_ignore_broadcasts = 1 When this key is on (1), the server will never answer to "ping" if its destination address is multicast or broadcast. It is good to turn on (1) this key to avoid your server becoming an involuntary partner of a DoS attack. In the above parameter, we enable this option. Step 2 Once the configuration has been set, you must restart your network for the change to take effect. The command to restart the network is the following: • To restart all networks devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w net.ipv4.icmp_echo_ignore_broadcasts=1
- 214. Process file systemmanagement 0 CHAPTER 7 214 Routing Protocols: Routing and routing protocols can create several problems. IP source routing, where an IP packet contains details of the path to its intended destination, is dangerous because according to RFC 1122 the destination host must respond along the same path. If an attacker was able to send a source routed packet into your network, then he would be able to intercept the replies and fool your host into thinking it is communicating with a trusted host. Step 1 I strongly recommend that you disable IP source routing on all network interfaces on the system to protect your server from this hole. If IP source routing is set to off (0), the server will not accept source-routed frames. Remember that Source-routed frames have an embedded, explicit description of the route between source and destination. Normally, IP packet routing is based solely on destination's address. • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Enable/Disable IP source routing net.ipv4.conf.all.accept_source_route = 0 net.ipv4.conf.default.accept_source_route = 0 Source-routed packets are a powerful concept but were never used, and can bring security problems because they allow a non-blind, REMOTE spoof. It is a very good idea to turn off (0) these keys. In the above parameters, we disable these options. Step 2 Once configurations have been set, you must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w net.ipv4.conf.all.accept_source_route=1 [root@deep /]# sysctl -w net.ipv4.conf.default.accept_source_route=1
- 215. Process file systemmanagement 0 CHAPTER 7 215 Enable TCP SYN Cookie Protection: A "SYN Attack" is a Denial of Service (DoS) attack that consumes all the resources on your machine, forcing you to reboot. Denials of Service attacks (attacks which incapacitate a server due to high traffic volume or ones those tie-up system resources enough that the server cannot respond to a legitimate connection request from a remote system) are easily achievable from internal resources or external connections via extranets and Internet. Enabling TCP SYN Cookie Protection will help to eliminate the problem. Below is a simple explanation of the concept. Every TCP/IP connection begins with a 3-way handshake: 1. Client sends a packet (packet 1) to server with SYN bit on, and waits; 2. Server sends a confirmation packet (packet 2) to client, and waits; 3. Client sends a third packet (packet 3) that consolidates the connection. Once the 3-way handshake is done, the server keeps data of packet 1 in a queue to compare it with packet 3 and establish the connection. This queue is limited in size and a quite high timeout. The SYN-flood attack exploits this fact and sends a lot of type-1 packets with random IP source addresses; the phase 3 answers never arrive; and once the queue is full, the server cannot receive more connections, be they legitimate or forged. The SYN cookie "trick" is to embed a code in the header of phase 2 packets, so server DOES NOT NEED TO KEEP any information about the client. If the phase 3 packet arrives someday, the server will calculate the port and client initial sequence number based solely on that packet - and will be able to establish the connection. Step 1 Since this embedded codification reduces randomness of the server initial sequence number, and thus can increase the "chance" of IP spoof family attacks, SYN cookies are used only in emergency situations, that is, when the half-open connections queue is full. • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Enable/Disable TCP SYN Cookie Protection net.ipv4.tcp_syncookies = 1 If this key is on (1), the kernel will send "SYN cookies" ONLY and ONLY when the half-open connections queue becomes full. This will mitigate the effects of SYN-flood DoS attacks. In the above parameter, we enable this option. Step 2 Once the configuration has been set, you must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] WARNING: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w net.ipv4.tcp_syncookies=1
- 216. Process file systemmanagement 0 CHAPTER 7 216 Disable ICMP Redirect Acceptance: When hosts use a non-optimal or defunct route to a particular destination, an ICMP redirect packet is used by routers to inform the hosts what the correct route should be. If an attacker is able to forge ICMP redirect packets, he or she can alter the routing tables on the host and possibly subvert the security of the host by causing traffic to flow via a path you didn't intend. Step 1 A legitimate ICMP REDIRECT packet is a message from a router that says "router X is better than me to reach network Y". Therefore, in complex networks, it will be highly recommended to keep these keys activated. On simple networks, it’s strongly recommended to disable ICMP Redirect Acceptance into all available interfaces on the server to protect it from this hole. • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Enable/Disable ICMP Redirect Acceptance net.ipv4.conf.all.accept_redirects = 0 net.ipv4.conf.default.accept_redirects = 0 When these keys are off (0), the kernel does not honor ICMP_REDIRECT packets, thus avoiding a whole family of attacks based on forging of this type of packet. In the above parameters, we disable these options. Step 2 Once the configurations have been set, you must restart your network for the change to take effect. The command to restart the network is the following: • To restart all networks devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w net.ipv4.conf.all.accept_redirects=0 [root@deep /]# sysctl -w net.ipv4.conf.default.accept_redirects=0
- 217. Process file systemmanagement 0 CHAPTER 7 217 Enable bad error message Protection: This option will alert you about all bad error messages on your network. Step 1 • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following line: # Enable/Disable bad error message Protection net.ipv4.icmp_ignore_bogus_error_responses = 1 Step 2 Once configuration has been set, you must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w net.ipv4.icmp_ignore_bogus_error_responses=1 Enable IP spoofing protection: The spoofing protection prevents your network from being the source of spoofed (i.e. forged) communications that are often used in DoS Attacks. Step 1 • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Enable/Disable IP spoofing protection net.ipv4.conf.all.rp_filter = 2 net.ipv4.conf.default.rp_filter = 2 These keys control IP Spoof detection and can have the following values: 0=absolutely no checking (the default) 1=simple checking (only obvious spoofs) 2=strong checking (positive source verification) The recommended standard is level 2, which can bring "problems in complex (non loop free) networks". In the above parameters, we enable the “strong checking” option.
- 218. Process file systemmanagement 0 CHAPTER 7 218 Step 2 Once the configurations have been made, you must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] NOTE: This parameter will prevent spoofing attacks against your internal networks but your external addresses can still be spoofed. There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w net.ipv4.conf.all.rp_filter=2 [root@deep /]# sysctl -w net.ipv4.conf.default.rp_filter=2 Enable Log Spoofed, Source Routed and Redirect Packets: This change will log all Spoofed Packets, Source Routed Packets, and Redirect Packets to your log files. Step 1 • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Enable/Disable Log Spoofed, Source Routed, Redirect Packets net.ipv4.conf.all.log_martians = 1 net.ipv4.conf.default.log_martians = 1 When this key is on (1), the kernel will log any "impossible" packets, where the IP source address spoofing is obvious. Example: packet with source address equal to 127.0.0.1 coming from an Ethernet interface. In the above parameter, we enable this option. Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] NOTE: There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w net.ipv4.conf.all.log_martians=1 [root@deep /]# sysctl -w net.ipv4.conf.default.log_martians=1
- 219. Process file systemmanagement 0 CHAPTER 7 219 Other possible optimization of the system All information described next relates to tuning we can make on the system. Be very careful when attempting this. You can optimize your system, but you can also cause it to crash. The shared memory limit parameters: Shared memory is used for inter-process communication, and to store resources that are shared between multiple processes such as cached data and code. If insufficient shared memory is allocated any attempt to access resources that use shared memory such as database connections or executable code will perform poorly. Under UNIX world, shared memory is referred to as "System V IPC". Almost all modern operating systems provide these features, but not all of them have them turned on or sufficiently sized by default, especially systems with BSD heritage. With Linux the default shared memory limit (both SHMMAX and SHMALL) is 32 MB in 2.4 kernels, but fortunately can be changed into the /proc file system. On system with small MB of RAM (>= 128MB), the default setting of 32MB for shared memory on the system is enough and should not be changed. On system with lot of RAM, we can readjust the default setting to better fit our machine and server performance. Below, I show you an example, to allow 128MB of shared memory on the system. The new value you enter for the shared memory should be four time less than what your total MB of RAM is. For example if you have 512MB of RAM installed on your computer, then you can set the default shared memory to 128MB as we do here. If you have less than what I use in this example, you have to adjust the value to fit your needs. Step 1 To change the default values of shared memory, type the following commands on your terminal: • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Improve shared memory size kernel.shmall = 134217728 kernel.shmmax = 134217728 Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following commands into your terminal screen: [root@deep /]# sysctl -w kernel.shmall="134217728" [root@deep /]# sysctl -w kernel.shmmax="134217728"
- 220. Process file systemmanagement 0 CHAPTER 7 220 Tuning the default and maximum window size parameters: The following hack allow us to raise network limits on Linux by requesting that the kernel provide a larger send buffer for the server's connections to its clients. This is possible by tuning the "default send window" and "maximum send window" parameters. Default parameter under Linux is 64kb, this is correct for regular use of the OS but if you run your system as a server, it is recommended to change the default parameter to a sufficiently-high value like 2000kb. 64kb is equal to 65536 (64 * 1024 = 65536), to set the values to 2000kb, we should enter new values of 2048000 (2000 * 1024 = 2048000). Step 1 To change the default values of default and maximum window size, type the following commands on your terminal: • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Improve default and maximum window size net.core.wmem_max = 2048000 net.core.wmem_default = 2048000 Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] NOTE: There is another way to update the entry without restarting the network by using the following commands into your terminal screen: [root@deep /]# sysctl -w net.core.wmem_max="2048000" [root@deep /]# sysctl -w net.core.wmem_default="2048000" The ulimit parameter: The ulimit command provides control over the resources available to the shell and to processes started by it, on systems that allow such control. We can use it to change the default resource set by the system to users or super user “root” but actually, we use it for changing resources for the super user “root” only because resources for normal users are controlled and managed through the /etc/security/limit.conf file. It is not all default resources available through the ulimit parameter that need to be changed but only those which can improve the performance of the server in a high load environment. Most default values for the super user “root” are acceptable and should be kept unchanged. Linux itself has a “Maximum Processes” per user limit. This feature allows us to control the number of processes an existing user or the “root” super user on the server may be authorized to have. To increase performance on highly loaded servers, we can safely set the limit of processes for the super-user “root” to be unlimited. This is what we will do in the following steps.
- 221. Process file systemmanagement 0 CHAPTER 7 221 One question remains, how can we change the default resources for a specific user on the system? Each new user has a hidden file called “.bashrc” available in their home directory. It is into this file that we can change the default value of resources for the specific user. In the example below, we do it for the super user “root” but the procedure is the same for any users on the system. Just edit their corresponding “.bashrc” file and make the change. Step 1 • Edit the .bashrc file for super user “root” (vi /root/.bashrc) and add the line: ulimit -u unlimited NOTE: You must exit and re-login from your terminal for the change to take effect. Step 2 To verify that you are ready to go, make sure that when you type as root the command ulimit -a on your terminal, it shows "unlimited" next to max user processes. [root@deep /]# ulimit -a core file size (blocks, -c) 0 data seg size (kbytes, -d) unlimited file size (blocks, -f) unlimited max locked memory (kbytes, -l) unlimited max memory size (kbytes, -m) unlimited open files (-n) 1024 pipe size (512 bytes, -p) 8 stack size (kbytes, -s) 8192 cpu time (seconds, -t) unlimited max user processes (-u) unlimited virtual memory (kbytes, -v) unlimited NOTE: You may also do ulimit -u unlimited at the command prompt instead of adding it to the /root/.bashrc file but the value will not survive to a reboot. The atime attribute: Linux records information about when files were created and last modified as well as when it was last accessed. There is a cost associated with recording the last access time. The ext2 file system of Linux has an attribute that allows the super-user to mark individual files such that their last access time is not recorded. This may lead to significant performance improvements on often accessed, frequently changing files such as the contents of News Server, Web Server, Proxy Server, Database Server among other directories. • To set the attribute to a file, use: [root@deep /]# chattr +A filename For a specific file For a whole directory tree, do something like: [root@deep /root]# chattr -R +A /var/spool For a News and Mail Server directory [root@deep /root]# chattr -R +A /home/httpd/html For a Web Server directory [root@deep /root]# chattr -R +A /var/lib/mysql For a SQL Database directory
- 222. Process file systemmanagement 0 CHAPTER 7 222 The noatime attribute: Linux has a special mount option for file systems called noatime that can be added to each line that addresses one file system in the /etc/fstab file. If a file system has been mounted with this option, reading accesses to the file system will no longer result in an update to the atime information associated with the file like we have explained previously. The importance of the noatime setting is that it eliminates the need by the system to make writes to the file system for files, which are simply being read. Since writes can be somewhat expensive, this can result in measurable performance gains. Note that the write time information to a file will continue to be updated anytime the file is written to. In our example below, we will set the noatime option to our /chroot file system. Step 1 • Edit the fstab file (vi /etc/fstab) and add in the line that refers to the /chroot file system, the noatime option after the defaults option as show below: LABEL=/chroot /chroot ext3 defaults,noatime 1 2 Step 2 Once you have made the necessary adjustments to the /etc/fstab file, it is time to inform the system about the modification. • This can be accomplished with the following commands: [root@deep /]# mount /chroot -oremount Each file system that has been modified must be remounted with the command shown above. In our example we have modified the /chroot file system and it is for this reason that we remount this file system with the above command. Step 3 • You can verify if the modifications have been correctly applied to the Linux system with the following command: [root@deep /]# cat /proc/mounts /dev/root / ext3 rw 0 0 /proc /proc proc rw 0 0 /dev/sda1 /boot ext3 rw 0 0 /dev/sda10 /cache ext3 rw,nodev 0 0 /dev/sda9 /chroot ext3 rw,noatime 0 0 /dev/sda8 /home ext3 rw,nosuid 0 0 /dev/sda13 /tmp ext3 rw,noexec,nosuid 0 0 /dev/sda7 /usr ext3 rw 0 0 /dev/sda11 /var ext3 rw 0 0 /dev/sda12 /var/lib ext3 rw 0 0 none /dev/pts devpts rw 0 0 This command will show you all file systems on your server with parameters applied to them. If you see something like: /dev/sda9 /chroot ext3 rw,noatime 0 0 Congratulations!
- 225. TCP/IP Network Management INTHIS CHAPTER 1. TCP/IP security problem overview 2. Installing more than one Ethernet Card per Machine 3. Files-Networking Functionality 4. Testing TCP/IP Networking 5. The last checkup
- 227. TCP/IP Network Management0 CHAPTER 8 227 Linux TCP/IP Abstract This chapter has been inserted here because it is preferable not to be connected to the network if the parts "Installation-Related Reference" and "Security and Optimization-Related Reference" of the book have not been completed. It is not wise to apply new security configurations to your system if you are online. Also, don’t forget that the firewall, which represents 50% of networking security, is still not configured on the Linux server. Finally it is very important and I say VERY IMPORTANT that you check all configuration files related to Linux networking to be sure that everything is configured correctly. Please follow all recommendations and steps in this chapter before continuing reading this book. This will allow us to be sure that if something goes wrong in the other chapters, it will be not related to your networking configurations. • To stop specific network device manually on your system, use the following command: [root@deep /]# ifdown eth0 • To start specific network device manually on your system, use the following command: [root@deep /]# ifup eth0 • To stop all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network stop Shutting down interface eth0 [OK] Disabling IPv4 packet forwarding [OK] • To start all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network start Enabling IPv4 packet forwarding [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Until now, we have not played with the networking capabilities of Linux. Linux is one of the best operating systems in the world for networking features. Most Internet sites around the world already know this, and have used it for some time. Understanding your network hardware and all the files related to it is very important if you want to have a full control of what happens on your server. Good knowledge of primary networking commands is vital. Network management covers a wide variety of topics. In general, it includes gathering statistical data and monitoring the status of parts of your network, and taking action as necessary to deal with failures and other changes. The most primitive technique for network monitoring is periodic "pinging" of critical hosts. More sophisticated network monitoring requires the ability to get specific status and statistical information from a range of devices on the network. These should include various sorts of datagram counts, as well as counts of errors of different kinds. For these reasons, in this chapter we will try to answer fundamental questions about networking devices, files related to network functionality, and essential networking commands.
- 228. TCP/IP Network Management0 CHAPTER 8 228 TCP/IP security problem overview It is assumed that the reader is familiar with the basic operation of the TCP/IP protocol suite, which includes IP and TCP header field functions and initial connection negotiation. For the uninitiated, a brief description of TCP/IP connection negotiation is given below. The user is strongly encouraged however to research other published literature on the subject. The IP Packets: The term packet refers to an Internet Protocol (IP) network message. It's the name given to a single, discrete message or piece of information that is sent across an Ethernet network. Structurally, a packet contains an information header and a message body containing the data being transferred. The body of the IP packet- it's data- is all or a piece (a fragment) of a higher- level protocol message. The IP mechanism: Linux supports three IP message types: ICMP, UDP, and TCP. An ICMP (Internet Control Message Protocol) packet is a network-level, IP control and status message. ICMP messages contains information about the communication between the two end-point computers. A UDP (User Datagram Protocol) IP packet carries data between two network-based programs, without any guarantees regarding successful delivery or packet delivery ordering. Sending a UDP packet is akin to sending a postcard to another program. A TCP (Transmission Control Protocol) IP packet carries data between two network-based programs, as well, but the packet header contains additional state information for maintaining an ongoing, reliable connection. Sending a TCP packet is akin to carrying on a phone conversation with another process. Most Internet network services use the TCP communication protocol rather than the UDP communication protocol. In other words, most Internet services are based on the idea of an ongoing connection with two-way communication between a client program and a server program. The IP packet headers: All IP packet headers contain the source and destination IP addresses and the type of IP protocol message (ICMP, UDP or TCP) this packet contains. Beyond this, a packet header contains slightly different fields depending on the protocol type. ICMP packets contain a type field identifying the control or status message, along with a second code field for defining the message more specifically. UDP and TCP packets contain source and destination service port numbers. TCP packets contain additional information about the state of the connection and unique identifiers for each packet. The TCP/IP Security Problem: The TCP/IP protocol suite has a number of weaknesses that allows an attacker to leverage techniques in the form of covert channels to surreptitiously pass data in otherwise benign packets. This section attempts to illustrate these weaknesses in theoretical examples.
- 229. TCP/IP Network Management0 CHAPTER 8 229 Application: A covert channel is described as: "any communication channel that can be exploited by a process to transfer information in a manner that violates the systems security policy. Essentially, it is a method of communication that is not part of an actual computer system design, but can be used to transfer information to users or system processes that normally would not be allowed access to the information. In the case of TCP/IP, there are a number of methods available whereby covert channels can be established and data can be surreptitiously passed between hosts. These methods can be used in a variety of areas such as the following: Bypassing packet filters, network sniffers, and "dirty word" search engines. Encapsulating encrypted or non-encrypted information within otherwise normal packets of information for secret transmission through networks that prohibit such activity "TCP/IP Steganography". Concealing locations of transmitted data by "bouncing" forged packets with encapsulated information off innocuous Internet sites. It is important to realize that TCP is a "connection oriented" or "reliable" protocol. Simply put, TCP has certain features that ensure data arrives at the remote host in a usually intact manner. The basic operation of this relies in the initial TCP "three way handshake" which is described in the three steps below. Step 1 Send a synchronize (SYN) packet and Initial Sequence Number (ISN) Host A wishes to establish a connection to Host B. Host A sends a solitary packet to Host B with the synchronize bit (SYN) set announcing the new connection and an Initial Sequence Number (ISN) which will allow tracking of packets sent between hosts: Host A ------ SYN(ISN) ------> Host B Step 2 Allow remote host to respond with an acknowledgment (ACK) Host B responds to the request by sending a packet with the synchronize bit set (SYN) and ACK (acknowledgment) bit set in the packet back to the calling host. This packet contains not only the responding clients' own sequence number, but the Initial Sequence Number plus one (ISN+1) to indicate the remote packet was correctly received as part of the acknowledgment and is awaiting the next transmission: Host A <------ SYN(ISN+1)/ACK ------ Host B Step 3 Complete the negotiation by sending a final acknowledgment to the remote host. At this point Host A sends back a final ACK packet and sequence number to indicate successful reception and the connection is complete and data can now flow: Host A ------ ACK ------> Host B
- 230. TCP/IP Network Management0 CHAPTER 8 230 The entire connection process happens in a matter of milliseconds and both sides independently acknowledge each packet from this point. This handshake method ensures a "reliable" connection between hosts and is why TCP is considered a "connection oriented" protocol. It should be noted that only TCP packets exhibit this negotiation process. This is not so with UDP packets which are considered "unreliable" and do not attempt to correct errors nor negotiate a connection before sending to a remote host. Encoding Information in a TCP/IP Header: The TCP/IP header contains a number of areas where information can be stored and sent to a remote host in a covert manner. Take the following diagrams, which are textual representations of the IP and TCP headers respectively: IP Header (Numbers represent bits of data from 0 to 32 and the relative position of the fields in the datagram) TCP Header (Numbers represent bits of data from 0 to 32 and the relative position of the fields in the datagram)
- 231. TCP/IP Network Management0 CHAPTER 8 231 Within each header there are multitudes of areas that are not used for normal transmission or are "optional" fields to be set as needed by the sender of the datagrams. An analysis of the areas of a typical IP header that are either unused or optional reveals many possibilities where data can be stored and transmitted. The basis of the exploitation relies in encoding ASCII values of the range 0-255. Using this method it is possible to pass data between hosts in packets that appear to be initial connection requests, established data streams, or other intermediate steps. These packets can contain no actual data, or can contain data designed to look innocent. These packets can also contain forged source and destination IP addresses as well as forged source and destination ports. This can be useful for tunneling information past some types of packet filters. Additionally, forged packets can be used to initiate an anonymous TCP/IP "bounced packet network" whereby packets between systems can be relayed off legitimate sites to thwart tracking by sniffers and other network monitoring devices. Implementations of Security Solutions: The following protocols and systems are commonly used to solve and provide various degrees of security services in a computer network. • IP filtering • Network Address Translation (NAT) • IP Security Architecture (IPSec) • SOCKS • Secure Sockets Layer (SSL) • Application proxies • Firewalls • Kerberos and other authentication systems (AAA servers) • Secure Electronic Transactions (SET) This graph illustrates where those security solutions fit within the TCP/IP layers: Security Solutions in the TCP/IP Layers
- 232. TCP/IP Network Management0 CHAPTER 8 232 Installing more than one Ethernet Card per Machine You might use Linux as a gateway between two Ethernet networks. In that case, you might have two Ethernet cards on your server. To eliminate problems at boot time, the Linux kernel doesn’t detect multiple cards automatically. If you happen to have two or more cards, you should specify the parameters of the cards in the lilo.conf file for a Monolithic kernel or in the modules.conf file for a Modularized kernel. The following are problems you may encounter with your network cards. Problem 1 If the driver(s) of the card(s) is/are being used as a loadable module (Modularized kernel), in the case of PCI drivers, the module will typically detect all of the installed cards automatically. For ISA cards, you need to supply the I/O base address of the card so the module knows where to look. This information is stored in the file /etc/modules.conf. As an example, consider we have two ISA 3c509 cards, one at I/O 0x300 and one at I/O 0x320. • For ISA cards, edit the modules.conf file (vi /etc/modules.conf) and add: alias eth0 3c509 alias eth1 3c509 options 3c509 io=0x300,0x320 This says that the 3c509 driver should be loaded for either eth0 or eth1 (alias eth0, eth1) and it should be loaded with the options io=0x300,0x320 so that the drivers knows where to look for the cards. Note that 0x is important – things like 300h as commonly used in the DOS world won’t work. For PCI cards, you typically only need the alias lines to correlate the ethN interfaces with the appropriate driver name, since the I/O base of a PCI card can be safely detected. • For PCI cards, edit the modules.conf file (vi /etc/modules.conf) and add: alias eth0 3c509 alias eth1 3c509 Problem 2 If the drivers(s) of the card(s) is/are compiled into the kernel (Monolithic kernel), the PCI probes will find all related cards automatically. ISA cards will also find all related cards automatically, but in some circumstance ISA cards still need to do the following. This information is stored in the file /etc/lilo.conf. The method is to pass boot-time arguments to the kernel, which is usually done by LILO. • For ISA cards, edit the lilo.conf file (vi /etc/lilo.conf) and add: append=”ether=0,0,eth1” In this case eth0 and eth1 will be assigned in the order that the cards are found at boot. Remember that this is required only in some circumstance for ISA cards, PCI cards will be found automatically. NOTE: First test your ISA cards without the boot-time arguments in the lilo.conf file, and if this fails, use the boot-time arguments.
- 233. TCP/IP Network Management0 CHAPTER 8 233 Files-Networking Functionality In Linux, the TCP/IP network is configured through several text files. You may have to edit them to make the network work. It’s very important to know the configuration files related to TCP/IP networking, so that you can edit and configure the files if necessary. Remember that our server doesn’t have an Xwindow interface (GUI) to configure files via a graphical interface. Even if you use a GUI in your daily activities it is important to know how to configure the network configuration files in text mode. The following sections describe all the basic TCP/IP configuration files under Linux. The /etc/sysconfig/network-scripts/ifcfg-ethN files: The configuration files for each network device you may have or want to add on your system are located in the /etc/sysconfig/network-scripts directory with Red Hat Linux, and are named ifcfg-eth0 for the first interface and ifcfg-eth1 for the second, etc. It is recommended to verify if all the parameters in this file are correct. Following is a sample /etc/sysconfig/network-scripts/ifcfg-eth0 file: DEVICE=eth0 BOOTPROTO=static BROADCAST=208.164.186.255 IPADDR=208.164.186.1 NETMASK=255.255.255.0 NETWORK=208.164.186.0 ONBOOT=yes USERCTL=no If you want to modify your network address manually, or add a new one on a new interface, edit this file (ifcfg-ethN), or create a new one and make the appropriate changes. DEVICE=devicename, where devicename is the name of the physical network device. BOOTPROTO=proto, where proto is one of the following: • static - The default option of Linux (static IP address) sould be used. • none - No boot-time protocol should be used. • bootp - The bootp (now pump) protocol should be used. • dhcp - The dhcp protocol should be used. BROADCAST=broadcast, where broadcast is the broadcast IP address. IPADDR=ipaddr, where ipaddr is the IP address. NETMASK=netmask, where netmask is the netmask IP value. NETWORK=network, where network is the network IP address. ONBOOT=answer, where answer is yes or no (Does the interface will be active or inactive at boot time). USERCTL=answer, where answer is one of the following: • yes (Non-root users are allowed to control this device). • no (Only the super-user root is allowed to control this device).
- 234. TCP/IP Network Management0 CHAPTER 8 234 The /etc/resolv.conf file: This file /etc/resolv.conf is another text file, used by the resolver—a library that determines the IP address for a host name. It is recommended to verify if all parameters included in this file are correct. Following is a sample /etc/resolv.conf file: domain openna.com search ns1.openna.com ns2.openna.com openna.com nameserver 208.164.186.1 nameserver 208.164.186.2 nameserver 127.0.0.1 NOTE: Name servers are queried in the order they appear in the file (primary, secondary). The /etc/host.conf file: This file /etc/host.conf specifies how names are resolved. Linux uses a resolver library to obtain the IP address corresponding to a host name. It is recommended to verify that all parameters included in this file are correct. Following is a sample /etc/host.conf file: # Lookup names via /etc/hosts first then fall back to DNS resolver. order hosts,bind # We have machines with multiple addresses. multi on The order option indicates the order of services. The sample entry specifies that the resolver library should first consult the /etc/hosts file of Linux to resolve a name and then check the name server (DNS). The multi option determines whether a host in the /etc/hosts file can have multiple IP addresses (multiple interface ethN). Hosts that have more than one IP address are said to be multihomed, because the presence of multiple IP addresses implies that the host has several network interfaces. The /etc/sysconfig/network file: The /etc/sysconfig/network file is used to specify information about the desired network configuration on your server. It is recommended that you verify all the parameters included in this file are correct. Following is a sample /etc/sysconfig/network file: NETWORKING=yes HOSTNAME=deep GATEWAY=207.35.78.1 GATEWAYDEV=eth0
- 235. TCP/IP Network Management0 CHAPTER 8 235 The following values may be used: NETWORKING=answer, where answer is yes or no (Configure networking or not configure networking). HOSTNAME=hostname, where hostname is the hostname of your server. GATEWAY=gwip, where gwip is the IP address of the remote network gateway (if available). GATEWAYDEV=gwdev, where gwdev is the device name (eth#) you use to access the remote gateway. The /etc/sysctl.conf file: With the new version of Red Hat Linux, all kernel parameters available under the /proc/sys/ subdirectory of Linux can be configured at runtime. You can use the new /etc/sysctl.conf file to modify and set kernel parameters at runtime. The sysctl.conf file is read and loaded each time the system reboots or when you restart your network. All settings are now stored in the /etc/sysctl.conf file. All modifications to /proc/sys should be made through /etc/sysctl.conf, because they are better for control, and are executed before rc.local or any other "users" scripts. Below, we’ll focus only on the kernel option for IPv4 forwarding support. See later in this chapter the TCP/IP security parameters related to the sysctl.conf file. To enable IPv4 forwarding on your Linux system, use the following command: Step 1 • Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following line: # Enable packet forwarding (required only for Gateway, VPN, Proxy, PPP) net.ipv4.ip_forward = 1 Step 2 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] WARNING: You must enable packet forwarding only on a machine that serves as a Gateway Server, VPN Server, Proxy Server or with PPP connection. Forwarding allows packets that are destined for another network interface (if you have another one) to pass through the network. There is another way to update the entry without restarting the network by using the following command into your terminal screen: [root@deep /]# sysctl -w net.ipv4.ip_forward=1
- 236. TCP/IP Network Management0 CHAPTER 8 236 The /etc/hosts file: As your machine gets started, it will need to know the mapping of some hostnames to IP addresses before DNS can be referenced. This mapping is kept in the /etc/hosts file. In the absence of a name server, any network program on your system consults this file to determine the IP address that corresponds to a host name. Following is a sample /etc/hosts file: IP Address Hostname Alias 127.0.0.1 localhost.localdomain localhost 208.164.186.1 deep.openna.com deep 208.164.186.2 mail.openna.com mail 208.164.186.3 web.openna.com web The leftmost column is the IP address to be resolved. The next column is that host’s name. Any subsequent columns are the aliases for that host. In the second line, for example, the IP address 208.164.186.1 if for the host deep.openna.com. Another name for deep.openna.com is deep. WARNING: Some people have reporting that a badly formed line in the /etc/hosts file may result to a "Segmentation fault (core dumped)" with the syslogd daemon, therefore I recommend you to double check your entry under this file and be sure that its respond to the example as shown above. The “Alias” part of the line is important if you want to be able to use the FQDN (Fully Qualified Domain Name) of the system reported by the hostname -f command. After you are finished adding and configuring your networking files, don’t forget to restart your network for the changes to take effect. • To restart your network, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] WARNING: Time out problems for telnet or ftp connection are often caused by the server trying to resolve the client IP address to a DNS name. Either DNS isn’t configured properly on your server or the client machines aren’t known to the DNS server. If you intend to run telnet or ftp services on your server, and aren’t using DNS, don’t forget to add the client machine name and IP in your /etc/hosts file on the server or you can expect to wait several minutes for the DNS lookup to time out, before you get a login prompt.
- 237. TCP/IP Network Management0 CHAPTER 8 237 Testing TCP/IP Networking Once we have applied TCP/IP security and optimization parameters to our server and checked or configured all files related to network functionality, we can run some tests to verify that everything works as expected. It is important to note that at this stage every test must be successful and not have any errors. It is to your responsibility to know and understand networking architecture and basic TCP/IP protocols before testing any parts of your networking configuration and topology. Step 1 To begin, we can use the ifconfig utility to display all the network interfaces on the server. • To display all the interfaces you have on your server, use the command: [root@deep /]# ifconfig The output should look something like this: eth0 Link encap:Ethernet HWaddr 00:E0:18:90:1B:56 inet addr:208.164.186.2 Bcast:208.164.186.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1295 errors:0 dropped:0 overruns:0 frame:0 TX packets:1163 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 Interrupt:11 Base address:0xa800 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:3924 Metric:1 RX packets:139 errors:0 dropped:0 overruns:0 frame:0 TX packets:139 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 NOTE: If the ifconfig tool is invoked without any parameters, it displays all interfaces you configured. An option of “-a” shows the inactive one as well. Step 2 If all network interfaces on the server look as you expect, then it is time to verify that you can reach your hosts. Choose a host from your internal network, for instance 192.168.1.1 • To verify that you can reach your internal hosts, use the command: [root@deep /]# ping 192.168.1.1 The output should look something like this: PING 192.168.1.1 (192.168.1.1) from 192.168.1.1 : 56 data bytes 64 bytes from 192.168.1.1: icmp_seq=0 ttl=128 time=1.0 ms 64 bytes from 192.168.1.1: icmp_seq=1 ttl=128 time=1.0 ms 64 bytes from 192.168.1.1: icmp_seq=2 ttl=128 time=1.0 ms 64 bytes from 192.168.1.1: icmp_seq=3 ttl=128 time=1.0 ms --- 192.168.1.1 ping statistics --- 4 packets transmitted, 4 packets received, 0% packet loss round-trip min/avg/max = 1.0/1.0/1.0 ms
- 238. TCP/IP Network Management0 CHAPTER 8 238 WARNING: Do not try to ping a host in which you have applied the previous TCP/IP security settings to prevent your system to respond to ping request. Instead try to ping another host without this feature enable. Also if you don’t receive an answer from the internal host you try to ping, verify if your hubs, routers, network cards, and network topology are correct. If you are able to ping your internal host, congratulations! Now we must ping an external network, for instance 216.148.218.195 • To verify that you can reach the external network, use the command: [root@deep /]# ping 216.148.218.195 The output should look something like this: PING 216.148.218.195 (216.148.218.195) from 216.148.218.195 :56 data byte 64 bytes from 216.148.218.195: icmp_seq=0 ttl=128 time=1.0 ms 64 bytes from 216.148.218.195: icmp_seq=1 ttl=128 time=1.0 ms 64 bytes from 216.148.218.195: icmp_seq=2 ttl=128 time=1.0 ms 64 bytes from 216.148.218.195: icmp_seq=3 ttl=128 time=1.0 ms --- 216.148.218.195 ping statistics --- 4 packets transmitted, 4 packets received, 0% packet loss round-trip min/avg/max = 1.0/1.0/1.0 ms Step 3 You should now display the routing information with the command route to see if the hosts have the correct routing entries. • To display the routing information, use the command: [root@deep /]# route -n The output should look something like this: Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 208.164.186.2 0.0.0.0 255.255.255.255UH 0 0 0 eth0 208.164.186.0 208.164.186.2 255.255.255.0 UG 0 0 0 eth0 208.164.186.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
- 239. Step 4 Another usefuloption is “netstat -vat”, which shows all active and listen TCP connections. • To shows all active and listen TCP connections, use the command: [root@deep /]# netstat -vat The output may look something similar to this example depending if the related services are running. Be aware that your results will almost certainly vary from the ones shown below: Active Internet connections (servers and established) Proto Recv-Q Send-Q Local Address Foreign Address State tcp 0 0 deep.openna.co:domain *:* LISTEN tcp 0 0 localhost:domain *:* LISTEN tcp 0 0 deep.openna.com:ssh gate.openna.com:1682ESTABLISHED tcp 0 0 *:webcache *:* LISTEN tcp 0 0 deep.openar:netbios-ssn *:* LISTEN tcp 0 0 localhost:netbios-ssn *:* LISTEN tcp 0 0 localhost:1032 localhost:1033 ESTABLISHED tcp 0 0 localhost:1033 localhost:1032 ESTABLISHED tcp 0 0 localhost:1030 localhost:1031 ESTABLISHED tcp 0 0 localhost:1031 localhost:1030 ESTABLISHED tcp 0 0 localhost:1028 localhost:1029 ESTABLISHED tcp 0 0 localhost:1029 localhost:1028 ESTABLISHED tcp 0 0 localhost:1026 localhost:1027 ESTABLISHED tcp 0 0 localhost:1027 localhost:1026 ESTABLISHED tcp 0 0 localhost:1024 localhost:1025 ESTABLISHED tcp 0 0 localhost:1025 localhost:1024 ESTABLISHED tcp 0 0 deep.openna.com:www *:* LISTEN tcp 0 0 deep.openna.com:https *:* LISTEN tcp 0 0 *:389 *:* LISTEN tcp 0 0 *:ssh *:* LISTEN Step 5 Sometimes machines on your network will discard your IP packets and finding the offending Gateway responsilbe can be difficult. Fortunately the tracepath utility attempts to trace the route an IP packet would follow to some Internet host. Choose an Internet host, for instance 64.81.28.146 • To print the route packets take to network host, use the command: [root@deep /]# tracepath 64.81.28.146 The output should look something like this: 1?: [LOCALHOST] pmtu 1500 1?: 207.35.78.1 2?: 10.70.1.1 3?: 206.47.228.178 4?: 206.108.97.149 5?: 206.108.103.214 6?: 206.108.103.228 7?: 208.51.134.9 8?: 208.48.234.189 9?: 206.132.41.78 asymm 10 10?: 204.246.213.226 asymm 13 11?: 206.253.192.217 asymm 13 12?: 206.253.195.218 asymm 14 13: 64.81.28.146 asymm 15 139ms reached Resume: pmtu 1500 hops 13 back 15
- 240. TCP/IP Network Management0 CHAPTER 8 240 Step 6 Finally, we will use the hostname command of Linux to show if our systems host name is correct. • To display and print the current host name of your server, use the command: [root@deep /]# hostname deep The hostname command without any options will print the current host name of our system, in this example “deep”. Now, it’s important to verify if the Fully Qualified Domain Name (FQDN) of our server is reported correctly. • To display and print the FQDN of your server, use the command: [root@deep /]# hostname -f deep.openna.com The last checkup If you can answer, “Yes” to each of the questions below, then your network is working and you can continue . Parameters inside ifcfg-ethN files are corrects The /etc/resolv.conf file contain your primary and secondary Domain Name Server All parameters included in the /etc/host.conf file are corrects All parameters included in the /etc/sysconfig/network file are corrects The /etc/hosts file contain the mapping of your hostnames to IP addresses All network interfaces on the server have the right parameter You can reach the internal and external hosts Your hosts have the correct routing entry The status of the interfaces has been checked and looks fine You are able to print the route packets take to network host
- 241. Firewall Basic Concept INTHIS CHAPTER 1. What is the IANA? 2. The ports numbers 3. What is a Firewall? 4. Packet Filter vs. Application Gateway 5. What is a Network Firewall Security Policy? 6. The Demilitarized Zone 7. Linux IPTables Firewall Packet Filter 8. The Netfilter Architecture
- 243. Firewall Basic Concept0 CHAPTER 9 243 Linux Firewall Abstract Before going into the installation, configuration and use of firewall software with Linux, we have to explain a little bit about what a firewall is, how it works and how this effects into your network and servers. A firewall is the first line of defense for your system, it is the first place where network connections and contacts will appear on the server before any server services or programs are started for them. What is the IANA? The Internet Assigned Numbers Authority (IANA) is the central coordinator for the assignment of unique parameter values for Internet protocols. The IANA is chartered by the Internet Society (ISOC), and the Federal Network Council (FNC), to act as the clearinghouse to assigning and coordinating the use of the numerous Internet protocol parameters. The Internet protocol suite, as defined by the Internet Engineering Task Force (IETF) and its steering group (the IESG), contains numerous parameters, such as internet addresses, domain names, autonomous system numbers (used in some routing protocols), protocol numbers, port numbers, management information base object identifiers, including private enterprise numbers, and many others. The common use of Internet protocols by the Internet community requires that the particular values used in these parameter fields be assigned UNIQUELY. It is the task of the IANA to make these unique assignments, as requested, and to maintain a registry of the currently assigned values. As an example, imagine that you have developed a new networking program that runs as a daemon on the server and it requires a port number. It is up to the IANA to register, manage and maintain a unique port number dedicated for and associated with your program. This way, anyone that wants to use your program, will know which unique port number is associated with it. The ports numbers In order for a computer to connect to multiple Internet services at the same time, the concept of a 'port' was introduced. Each computer has 65535 ports available. If your web browser initiates a connection to www.openna.com (port 80 by default) for example, it will pick the first available port ( lets say 10232) and use it to send the connection request to www.openna.com. Openna.com's web server will reply to port 10232 on your PC. This way, your PC knows that this reply is in response to the request sent to www.openna.com earlier. All open ports should have a service or daemon running on them. A service or daemon is simply the software running on these ports, which provides a service to the users who connect to it. If no service or daemon is running on the port, then there is no reason to have the port open on the server and you should close it. The port numbers are divided into three ranges: the Well Known Ports, the Registered Ports, and the Dynamic and/or Private Ports. There are two types of ports, using two different protocols: TCP and UDP. Although they are different protocols, they can have the same port number. The Well Known Ports are those from 0 through 1023, the Registered Ports are those from 1024 through 49151 and the Dynamic and/or Private Ports are those from 49152 through 65535.
- 244. Firewall Basic Concept0 CHAPTER 9 244 The Well Known Ports are assigned by the IANA and on most systems can only be used by system (or root) processes or by programs executed by privileged users (our daemons running in the background). Ports are used by the TCP protocol [RFC793] to name the ends of logical connections, which carry long-term conversations. For the purpose of providing services to unknown callers, a service contact port is defined. The contact port is sometimes called the "well- known port". Wherever possible, the same port assignments are also used by UDP protocol [RFC768]. For many years Well Known Ports were in the range 0-255. Recently, the range for the Well Known Ports has been expanded from 0 to 1023 to respond to the exponential growth of the Internet. The Registered Ports are also listed by the IANA and, on most systems, can be used by ordinary user processes or programs executed by ordinary users. The IANA registers the use of these ports as a convenience to the community. Again, wherever possible, these same port assignments are used with UDP [RFC768]. The Registered Ports are in the range 1024-49151. Finally, the Dynamic and/or Private Ports are those ranging from 49152 through to 65535. 1 2 3 [1] The Well Known Ports represent 2% of all available ports [0-1023]. [2] The Registered Ports represent 73% of all available ports [1024-49151]. [3] The Dynamic and/or Private Ports represent 25% of all available ports [49152-65535]. Ports Numbers Graphical Representation
- 245. Firewall Basic Concept0 CHAPTER 9 245 What is a Firewall? As we said before, if a service or daemon program is not running on its assigned port, then there is no reason to have the related port open on the server. There are some simple reasons why this is bad: 1. We can run a Trojan program on the open port. 2. We can use the open port to access another server. A firewall (software or hardware) will take care of this. It will close all ports that we don’t use on the server. Firewalls can control, manage and supervise all legitimate open ports where services or daemons are running. To recap, an Internet firewall is a software program or hardware device used to protect a server or private network from the ever-raging fire on the Internet. The best practice is to run a firewall on each server, even if you have a router or a big firewall in front of your other servers on the network. This allows us to close any open ports that we don’t use, and to better control what goes in and out of our servers and add another level of security to our network. Packet Filter vs. Application Gateway During the past ten-years, different firewall technologies have been developed to respond to different server’s requirements on the Internet. From this research two distinct categories of firewall software have emerged. The first category of firewall is known as Packet Filtering and the second as an Application Gateway. It is important to make the distinction between the two categories before continuing. This will allow us to understand the technology used in each one and to get a better idea about where we need to use them on our servers and network. Each one has its advantages and this is one of the reasons why both categories still exist. It is up to us to use the right firewall software depending of the kind of services that we want to offer in order to protect our Linux servers and network. Packet Filtering Packet Filtering is the type of firewall that’s built into the Linux kernel (as a kernel module, or compiled in). A filtering firewall works at the network level. Data is only allowed to leave the system if the firewall rules allow it. As packets arrive they are filtered by their type, source address, destination address, and port information contained in each packet header. Most of the time, packet filtering is accomplished by using a router that can forward packets according to filtering rules. When a packet arrives at the packet-filtering router, the router extracts certain information from the packet header and makes decisions according to the filter rules as to whether the packet will be allowed to pass through or be discarded. The following information can be extracted from the packet header: Source IP address Destination IP address TCP/UDP source port TCP/UDP destination port ICMP message type Encapsulated protocol information (TCP, UDP, ICMP or IP tunnel) Because very little data is analyzed and logged, filtering firewalls take less CPU power and create less latency in your network. Two generations of Packet Filtering Firewall software have been made available to the public.
- 246. Firewall Basic Concept0 CHAPTER 9 246 The first generation was called "static", because the method of connecting between the internal and external networks must be left open at all times. Static Packet filtering Firewall (first generation) is well known under Linux as the IPCHAINS firewall software used in the Linux Kernel version 2.2.x. The main disadvantage of this type of firewall is the fact that ports must be left open at all times to allow desired traffic, another important disadvantage is that it allows a direct connection to internal hosts by external clients, and finally it offers no user authentication. To address some of the problems of the first generation of Packet filtering Firewalls, a second generation of Packet Filtering software was developed. The second generation is known as Dynamic Packet Filters or Stateful Packet Filtering also known under Linux as IPTables firewall software, used in Linux Kernel version 2.4.x. The stateful packet filter keeps track of the state and context information of a session. Once a series of packets has passed through the "door" to it’s destination, the firewall closes the door. This solves the problem of having ports open at all times. Another improvement compared to the first generation is the limitation of spoofing attacks. Dynamic Packet Filters is not perfect and external systems are still able to make an IP connection with an internal host and user authentication still not supported. Application Gateways An Application Gateway, also known as proxy software and well known under Linux as “Squid” software, is a firewall system in which processes that provide services maintain complete TCP connection states and sequencing. At this time two generations of Application Gateway Firewall software have been made available to the public. The first generation was simply called an "Application Gateway". With this type of firewall software, all connections to the internal network go through the firewall for verification and approval, based on the set-up policies that you have entered in the configuration file of the Application Gateway Firewall. Contrary to a Packet Filtering Firewall, an Application Gateway Firewall looks in detail at the communication stream before allowing the traffic to pass into the network to its final destination by analyzing application commands inside the payload portion of data packets. Whereas stateful packet filters systems do not. Another important advantage of an Application Gateway is the fact that it does not allow any direct connections between internal and external hosts and it also supports user-level authentication, two points where packet filter lose again. But, Application Gateway Firewall software is not perfect and has some bad points too. The first is that it is slower than packet filtering, it requires that the internal client (i.e. the workstation) to knows about them and it also does not support every type of connection. To address some of the problems encountered in the first generation of this type of firewall software, a second generation has been developed. It’s called the Transparent Application Gateway and one of its main advantage compared to its predecessor is that client workstations do not either have to be aware of the firewall nor run special software to communicate with the external network. This fixes the problem of having the internal client (i.e. the workstation) know about them. Even with all these improvement, some disadvantages still exist. Transparent Application Gateways are slower than packet filters, they consume more system resources and do not support every type of connection.
- 247. Firewall Basic Concept0 CHAPTER 9 247 From the above analysis (Packet Filter vs. Application Gateway), we can summarize the main advantages and disadvantages of each firewall category as follows: Packet Filter advantages are: Application Gateway advantages are: Good for traffic management. Low Overhead / High Throughput. Supports almost any service. Do not allow any direct connections between internal and external hosts. Support user-level authentication. Packet Filter disadvantages are: Application Gateway disadvantages are: Offers no user authentication. Allows direct IP connections to internal hosts by external clients. Slower than packet filters. Do not support every possible type of connection. Therefore we can, with confidence, recommend Packet Filter Firewall software for all servers in the DMZ zone (The Demilitarized Zone) under a Unix environment. For Windows systems, the approach is not recommended, implementations and strategies are different due to the insecure nature of the operating system and it’s programs. Unix systems and their programs have many features to compensate some of the disadvantages of Packet Filter Firewalls and this is the reason why this type of firewall does not pose any problems for Unix systems located in the DMZ like web, mail, ftp, lists, virtual, dns, database, and backup servers. An Application Gateway Firewall is recommended only for a Gateway Server (a machine that makes a bridge between your private internal network and the Internet). Also a Packet Filter Firewall is recommended for Gateway servers and this means that you have to install an Application Gateway Firewall and a Packet Filter Firewall on a Gateway Server. Yes, both are recommended for a secure communication between your private internal hosts and the Internet. Using just one type of firewall on a Gateway Server is not enough. Finally, I will say that installing an Application Gateway Firewall on web, mail, ftp, lists, virtual, dns, database, and backup servers is a waste of time. You only need this kind of firewall software on a Gateway Server. What is a Network Firewall Security Policy? A network firewall security policy defines those services that will be explicitly allowed or denied, how these services will be used and the exceptions to these rules. An organization's overall security policy must be determined according to a security and business-needs analysis. Since a firewall relates to network security alone, a firewall has little value unless the overall security policy is properly defined. Every rule in the firewall security policy should be implemented on a firewall. Generally, a firewall uses one of the following methods. Everything not specifically permitted is denied This approach blocks all traffic between two networks except for those services and applications that are permitted. Therefore, each required service or application should be implemented on an individual basis. No service or application that could possibly be a potential hole on the firewall should be permitted. This is the most secure method of firewalling, denying services and applications unless explicitly allowed by the administrator. On the other hand, from the users point of view, it might be more restrictive and less convenient. This is the method we will use in our Firewall configuration files in this book.
- 248. Firewall Basic Concept0 CHAPTER 9 248 Everything not specifically denied is permitted This approach allows all traffic between two networks except for those services and applications that are denied. Therefore, each untrusted or potentially harmful service or application should be denied individually. Although this is flexible and convenient for the users, it could potentially cause some serious security problems. This method is really not recommended. The Demilitarized Zone A demilitarized zone (DMZ) refers to a part of the network that is neither part of the internal network nor directly part of the Internet. Typically, this is the area between your Internet access router and your bastion host (internal network), though it can be between any two policy-enforcing components of your architecture. A DMZ minimizes the exposure of hosts on your external LAN by allowing only recognized and managed services on those hosts to be accessible by hosts on the Internet. This kind of firewall architecture will be the one we will use throughout this book for all networking services and firewall implementations we want to install. A demilitarized zone (DMZ) is the most commonly used method in firewall security. All web, mail, ftp, lists, virtual, dns, database, and backup servers must be located in this zone. The gateway server also needs to be located in this zone since it makes the bridge between the private internal zone and the Internet. Hub A Server Server Server Server Hub B INTERNET INTRANET
- 249. Firewall Basic Concept0 CHAPTER 9 249 The boxes between Hub A and B are in the 'DMZ'. Hub A only routes traffic between the Internet and the DMZ. Hub B only routes traffic between the DMZ and the Intranet. The theory is that all traffic between the Intranet and the Internet has to pass through a machine in the DMZ. The machine in the DMZ can be used to authenticate, record, and control all traffic via a Packet Filter Firewall or an Application Gateway Firewall software. Linux IPTables Firewall Packet Filter The new Linux kernel, like the two previous kernels, supports a new mechanism for building firewalls, called network packet filtering (netfilter). This new mechanism, which is controlled by a tool named IPTables, is more sophisticated than the previous IPCHAINS and more secure. This easy to configure new mechanism is also the first stateful firewall on a Linux operating system. Stateful firewalling represents a major technological jump in the intelligence of a firewall and allows administrators, for example, to block/detect many stealth scans that were undetected on previous generations of Linux firewalls. It also blocks most of the DoS attacks by rating limiting user-defined packet types, since it keeps each connection passing through it in memory. This new technology means that if a foreign packet tries to enter the network by claiming to be part of an existing connection, IPTables can consult its list of connections, which it keeps in memory, and if it finds that the packet doesn't match any of these, it will drop the packet which will defeat the scan in many cases! I would say that 50% of security on a network depends on a good firewall, and everyone should now be running at least IPTables on a Linux server to reach this level of security. The Netfilter Architecture Netfilter is used to forward, redirect, masquerade and filter packets coming into or out of our network. It is the framework of IPTables and without it IPTables will never work. Currently, four major subsystems exist on top of netfilter but only three are really important to make IPTables work. These three subsystems are what we regularly use to construct and build the rules and chains the firewall will use based on our policies. These subsystems are: The ‘iptables’ packet classification system. The connection-tracking system. The NAT system.
- 250. Firewall Basic Concept0 CHAPTER 9 250 The ‘iptables’ packet classification system: The IPTables packet classification system provides the "filter table" used by the packet filtering system to filter traffic, protocols and IP packets on the server. It is one of the most important subsystem components of IPTables. It works at the network level. Data is only allowed to leave the system if the firewall rules allow it. As packets arrive they are filtered by their type, source address, destination address, and port information contained in each packet. This is possible with the IPTables filter commands, which allow us to build rules to accomplish it. When we use and mix the IPTables filter commands together in the same line (because we know what each command means) we create a rule that the IPTables software understands and applies. Many commands exist and it is not to our intention to list all of them here and explain each of them. We'll only show you the most important ones and their meanings. If you need more detailed information about each IPTables command and how to use them, please read a good firewall book or see the Netfilter web page. After reading this brief introductory chapter about IPTables, you should be able to understand the most important commands, e.g. how a rule is defined, as well as all the subsystem mechanisms of IPTables. This is all we need for the next chapter, where we’ll install and configure the firewall software to interact with IPTables. Now lets explain the most important parts of IPTables NetFilter. The IPTables rules Rules are used in IPTables to define what we want to do with the IP packets and protocols coming in to or out of our machine. 1. Each rule should be defined on one line for the firewall to separate rules. 2. Each new rule should begin with the word "iptables" which refers to the IPTables binary program that will be run. The IPTables chains Three built-in chains (INPUT, OUTPUT, and FORWARD) exist by default with IPTables. These are used to decide if the rule should be applied for INPUT packets, OUTPUT packets or FORWARDed packets. There are several ways to manipulate rules inside a chain. 1. We can append a new rule to a chain (-A). 2. Define which protocol to use (-p) to the chain. 3. Specifying the source (-s) and destination (-d) IP addresses to the chain. 4. Specifying the source (--sport) and destination (--dport) port range specification. 5. On which interface (-i for incoming packet on the interface and -o for outgoing packet on the interface) to match the rule, and so on. The IPTables example for rules and chains Really, we need to show an example to clarify the above. Imagine that we want to block any HTTP packets that come in or out of our server on the external network interface. Here are the rules to accomplish it.
- 251. Firewall Basic Concept0 CHAPTER 9 251 The first rule (the complete line), instructs IPTables to add to its INPUT chain (-A INPUT) a new definition that will drop all packets (-j DROP) entering in the eth0 interface (-i eth0) using the TCP protocol (-p tcp) coming from anywhere (-s 0.0.0.0) on source port between 1024 & 65535 (--sport 1024:65535) to destination IP address 207.35.78.2 (-d 207.35.78.2) on the destination port 80 (--dport 80) for the HTTP service. /sbin/iptables -A INPUT –i eth0 -p tcp -s 0.0.0.0 --sport 1024:65535 -d 207.35.78.2 --dport 80 -j DROP The second rule, instructs IPTables to add to its OUTPUT chain (-A OUTPUT) a new definition that will drop all packets (-j DROP) going out on the eth0 interface (-i eth0) using the TCP protocol (-p tcp) coming from IP address 207.35.78.2 (-s 207.35.78.2) on source port 80 (--sport 80) to anywhere (-d 0.0.0.0) on destination ports between 1024 & 65535 (-- dport 1024:65535) for the HTTP service. /sbin/iptables -A OUTPUT –o eth0 -p tcp -s 207.35.78.2 --sport 80 -d 0.0.0.0 -- dport 1024:65535 -j DROP In the above example, we have defined two new rules. The first rule is for incoming connections with the INPUT chain, and the second rule for outgoing connections with the OUTPUT chain. The connection-tracking system: It is with its "Connection Tracking" feature IPTables can recognize instructions to allow 'NEW' connections, 'RELATED' connections, etc. The connection-tracking subsystem is one of the pieces that makes IPTables more intelligent than its predecessor IPCHAINS. Connection Tracking keeps track of the relationships of packets by maintaining state information about a connection using memory tables. As mentioned previously, firewalls that do this are known as stateful. It is used when you add and declare "state" options like 'NEW', 'ESTABLISHED', 'RELATED', and 'INVALID' into your IPTables rules. This feature becomes enabled when you define the "--state" option in your rules. The "state" feature gives you the opportunity to decide how incoming or outgoing connections should be analyzed and treated. To achieve this, the IPTables "state" feature provide us four possibilities. 1. NEW Allow an incoming or outgoing packet, which creates a new connection. 2. ESTABLISHED Allow an incoming or outgoing packet, which belongs to an existing connection. 3. RELATED Allow an incoming or outgoing packet, which is related to, but no part of, an existing connection. 4. INVALID Allow an incoming or outgoing packet, which could not be identified for some reason. By using the above options with IPTables (highly recommended) we can fine tune our firewall and control much more tightly how packets should be treated before coming into or going out of our server.
- 252. Firewall Basic Concept0 CHAPTER 9 252 The IPTables example for connection-tracking Below are examples on how to use these options with IPTables. We use the previous rules for our example and then we add the connection-tracking feature to it. One other difference with the previous example is that instead of denying traffic we allow it to pass. The first rule, instructs IPTables to add to its INPUT chain (-A INPUT) a new definition that will accept all packets, (-j ACCEPT) which may create new connections or they might belong to an existing connection (-m state --state NEW,ESTABLISHED), to enter in the eth0 interface (-i eth0) with the TCP protocol (-p tcp) coming in from anywhere (-s 0.0.0.0) on source ports between 1024 & 65535 (--sport 1024:65535) to destination IP address 207.35.78.2 (-d 207.35.78.2) on destination port 80 (--dport 80) for the HTTP service. /sbin/iptables -A INPUT –i eth0 -p tcp -s 0.0.0.0 --sport 1024:65535 -d 207.35.78.2 --dport 80 -m state --state NEW,ESTABLISHED -j ACCEPT The second rule, instructs IPTables to add to its OUTPUT chain (-A OUTPUT) a new definition that will accept all packets (-j ACCEPT) which belong to an existing connection (-m state -- state ESTABLISHED) to go out on the eth0 interface (-i eth0) using the TCP protocol (-p tcp) coming from IP address 207.35.78.2 (-s 207.35.78.2) on source port 80 (--sport 80) to anywhere (-d 0.0.0.0) on destination ports between 1024 & 65535 (--dport 1024:65535) for HTTP service. /sbin/iptables -A OUTPUT –o eth0 -p tcp -s 207.35.78.2 --sport 80 -d 0.0.0.0 -- dport 1024:65535 -m state --state ESTABLISHED -j ACCEPT In the above example, we have been using two connection-tracking options to build the rules. For incoming connections, we use the “NEW” and “ESTABLISHED” options to inform IPTables to accept packets which create a new connection, and packets which belong to an existing connection. For outgoing connections, we only use “ESTABLISHED” to inform IPTables to accept packets, which belong to an existing connection. The INVALID state should never be used, since its means that the packet is associated with no known connection. The RELATED state is used in some cases, for example, FTP data transfer or ICPM errors and means that the packet is starting a new connection, but is associated with an existing connection.
- 253. Firewall Basic Concept0 CHAPTER 9 253 The NAT system: NAT (Network Address Translation) transparently makes one set of IP addresses appear to be another set to the external world. It is used when you want to map an entire network onto one IP address, or when you want to forward certain connections to specific servers with private addresses or when you want to use load balancing to distribute load over several systems. NAT can be divided into two different types: Source NAT (SNAT) and Destination NAT (DNAT). 1. Source NAT is when you alter the source address of the first packet (i.e. you are changing where the connection is coming from). Source NAT is always done post-routing, just before the packet goes out onto the wire. Masquerading is a specialized form of SNAT, because you change the source address of the first packet. 2. Destination NAT is when you alter the destination address of the first packet (i.e. you are changing where the connection is going to). Destination NAT is always done pre-routing, when the packet first comes off the wire. Port forwarding, load sharing, and transparent proxying are all forms of DNAT, because you want people to be able to get to the boxes behind the one with the ‘real’ IP address. The IPTables example for NAT Below are examples on how to use these tables with IPTables. The table of NAT rules contains two lists called 'chains'. The two chains are PREROUTING (for Destination NAT, as packets first come in), and POSTROUTING (for Source NAT, as packets leave). For all the NAT operations that you want to do in your firewall script file, you will have to use the ‘- t nat' option to enable the NAT table feature of IPTables, since without this option, the NAT table will not work. If you simply want to tell your Gateway Server that all packets coming from your internal network should be made to look like they are coming from the external interface (eth0) or from your dialup box (ppp0) then you would use the following rules: /sbin/iptables -t nat -A POSTROUTING -o eth0 -j MASQUERADE This says the following to the system: In the NAT table (-t nat), append a rule (-A) after routing (POSTROUTING) for all packets going out eth0 (-o eth0), MASQUERADE the connection (-j MASQUERADE). Now if you want to do port forwarding, meaning for example, that you want TCP packets coming into your external interface, which is directly connected to the Internet on IP address 207.35.78.2 port 8080, to have their destination mapped to your internal interface on IP address 192.168.1.1 on port 80, then you would use the following rules to achieve it. /sbin/iptables -A PREROUTING -t nat -p tcp -d 1.2.3.4 --dport 8080 -j DNAT --to 192.168.1.1:80 This says : Append a pre-routing rule (-A PREROUTING) to the NAT table (-t nat) so that TCP packets (-p tcp) going to 207.35.78.2 (-d 207.35.78.2) on port 8080 (--dport 8080) have their destination mapped (-j DNAT) to 192.168.1.1 on port 80 (--to 192.168.1.1:80).
- 254. Firewall Basic Concept0 CHAPTER 9 254 As we can see, there are many options, parameters, and tables and it is very easy to make a mistake even if we are familiar with Firewall NetFilter technologies like IPTables. We can easily forget some important rule or even open some dangerous ports in error. Building a complete set of rules and chains suitable for all possible types of servers and workstations is a long task and it becomes evident that some predefined firewall rules are required to help us. Conclusion As a change to the previous books where we provided predefined firewall rules to include in your firewall script, we will use a different approach in this new edition of Securing & Optimizing Linux. Two mains reasons justify this change. Firstly, Adrian Pascalau <apascalau@openna.com> has developed a new, very powerful and easy to use firewall software, based on my initial firewall work, which includes support for all possible requirements and needs of an IPTables firewall set-up and secondly because the previous predefined rules were not relevant to the needs of all users. If we had to cover every possible configuration, it would take a complete book in itself, therefore the best solution was to start a new piece in parallel and write a new firewall program based on IPTables that handles and covers, as much as possible, the needs of all Linux users. This has been done and I would like to thanks Adrian for his invaluable help, hard work and expertise in this area. GIPTables- Firewall is the result of this vision and it is the firewall program that we’ll use and explain in the next chapter.
- 255. GIPTables Firewall IN THISCHAPTER 1. Building a kernel with IPTables support 2. Compiling - Optimizing & Installing GIPTables 3. Configuring GIPTables 4. /etc/giptables.conf: The GIPTables Configuration File 5. /etc/rc.d/rc.giptables.blocked: The GIPTables Blocked File 6. /etc/init.d/giptables: The GIPTables Initialization File 7. The GIPTables Firewall Module Files 8. How GIPTables parameters work? 9. Running the type of GIPTables firewall that you need 10. The GIPTables configuration file for a Gateway/Proxy Server 11. GIPTables Administrative Tools
- 257. GIPTables Firewall 1 CHAPTER0 257 Linux GIPTables Abstract GIPTables Firewall is a free set of shell scripts that helps you generate Net filter/IPTables rules for Linux 2.4.x and newer kernels. It is very easy to configure and at present, designed to run on hosts with one or two network cards. It doesn’t require that you to install any additional components to make it work with your Linux system. All you need to set-up a very secure firewall for your Linux machines is IPTables and GIPTables. GIPTables can be used very easily with a host that has only one network card, and this host can be a server or a workstation. It assumes that if your host has two network cards, then the host should be a Gateway Server that connects your INTERNAL private network to the EXTERNAL world (the Internet). Access from your internal network to the external world is automatically controlled and filtered by the SNAT feature of IPTables and GIPTables. This is well known in the Linux world as MASQUERADING. The DNAT feature of IPTables and GIPTables automatically controls access from the Internet to your internal servers where the software will forwards specified incoming connections to your internal server. GIPTables-Firewall has many advantage compared to its competitors. It’s easy to install and configure. It does not require you to install any additional component to make it work. It only needs IPTables to run. It’s uses NAT & MASQ for sharing Internet access when you don't have enough IP. It’s uses the stateful packet filtering (connection tracking) feature of IPTables. It’s automatically does all kinds of network address translation. It’s uses rate-limited connection and logging capability. It provides good protection against all kind of TCP SYN-flooding Denial of Service attacks. It provides good prootections against IP spoofing. It provides TCP packets heath check. It runs on any type of Linux system. It has a flexible and extensible infrastructure. It’s easy to adjust and modify for your needs. It's small and does not use a lot of memory. It merges cleanly with all native Linux programs. It’s well written and very powerful. It covers all needs in a highly secure server environment. GIPTables-Firewall is simply the best firewall software to use with IPTables. It comes with a myriad ready to use of predefined rules. To be protected all we need to do is to answer in its configuration file ‘Yes’ or ‘No’ to the questions. Nothing more than that is required from your part to make it work.
- 258.
- 259. GIPTables Firewall 1 CHAPTER0 259 These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation require using the super-user account “root”. Whether kernel recompilation may be required: Yes Latest GIPTables version number is 1.1 We have only tested GIPTables on OpenNA Linux and Red Hat Linux, but the procedures given in this chapter are likely to work on all Linux platforms. Packages The following is based on information as listed by GIPTables-Firewall as of 2002/06/09. Please regularly check at http://www.giptables.org/ for the latest status. We chose to install from source file because it provides us the opportunity to fine tune the installation. Source code is available from: GIPTables-Firewall Homepage: www.giptables.org You must be sure to download: giptables-1.1.tar.gz Prerequisites Linux GIPTables requires that the listed software below is already installed on your system to be able to run and work successfully. If this is not the case, you must install them from your Linux CD-ROM or source archive file. Please make sure you have all of these programs installed on your machine before you proceed with this chapter. Kernel 2.4 is required to set up GIPTables in your system. iptables package, is the new secure and more powerful program used by Linux to set up GIPTables in your system. Building a kernel with IPTables support The first thing you need to do is to ensure that your kernel has been built with the NetFilter infrastructure compiled in it: NetFilter is a general framework inside the Linux kernel, which other things (such as the iptables module) can plug into. This means you need kernel 2.4.x and answer “y”, “n” or “m” to the following questions depending of the kernel type you have configured. For a Monolithic Kernel, you would answer the questions “y” and your happier running a Modularized Kernel, you would answer the questions “m”. It is important to understand that if IPTables is not enabled in your Kernel, NONE of the information contained in this chapter will work. If your Kernel is one that comes directly from your Linux vendor or is unmodified, then there is a good chance that your kernel is already built to handle IPTables, therefore you wouldn’t have to recompile it and/or go through the setup steps below.
- 260. GIPTables Firewall 1 CHAPTER0 260 Here are the required kernel setups for all type of servers except for a Gateway/Proxy: * Networking options * Packet socket (CONFIG_PACKET) Answer Y here Packet socket: mmapped IO (CONFIG_PACKET_MMAP) Answer Y here Netlink device emulation (CONFIG_NETLINK_DEV) Answer Y here Network packet filtering (replaces ipchains) (CONFIG_NETFILTER) Answer Y here Network packet filtering debugging (CONFIG_NETFILTER_DEBUG) Answer Y here Socket Filtering (CONFIG_FILTER) Answer N here Unix domain sockets (CONFIG_UNIX) Answer Y here TCP/IP networking (CONFIG_INET) Answer Y here IP: multicasting (CONFIG_IP_MULTICAST) Answer N here IP: advanced router (CONFIG_IP_ADVANCED_ROUTER) Answer N here IP: kernel level autoconfiguration (CONFIG_IP_PNP) Answer N here IP: tunneling (CONFIG_NET_IPIP) Answer N here IP: GRE tunnels over IP (CONFIG_NET_IPGRE) Answer N here IP: TCP Explicit Congestion Notification support (CONFIG_INET_ECN) Answer N here IP: TCP syncookie support (disabled per default) (CONFIG_SYN_COOKIES) Answer Y here * * IP: Netfilter Configuration * Connection tracking (required for masq/NAT) (CONFIG_IP_NF_CONNTRACK) Answer Y here FTP protocol support (CONFIG_IP_NF_FTP) Answer Y here IRC protocol support (CONFIG_IP_NF_IRC) Answer N here IP tables support (required for filtering/masq/NAT) (CONFIG_IP_NF_IPTABLES) Answer Y here limit match support (CONFIG_IP_NF_MATCH_LIMIT) Answer Y here MAC address match support (CONFIG_IP_NF_MATCH_MAC) Answer Y here netfilter MARK match support (CONFIG_IP_NF_MATCH_MARK) Answer Y here Multiple port match support (CONFIG_IP_NF_MATCH_MULTIPORT) Answer Y here TOS match support (CONFIG_IP_NF_MATCH_TOS) Answer Y here LENGTH match support (CONFIG_IP_NF_MATCH_LENGTH) Answer Y here TTL match support (CONFIG_IP_NF_MATCH_TTL) Answer Y here tcpmss match support (CONFIG_IP_NF_MATCH_TCPMSS) Answer Y here Connection state match support (CONFIG_IP_NF_MATCH_STATE) Answer Y here Packet filtering (CONFIG_IP_NF_FILTER) Answer Y here REJECT target support (CONFIG_IP_NF_TARGET_REJECT) Answer Y here Full NAT (CONFIG_IP_NF_NAT) Answer N here Packet mangling (CONFIG_IP_NF_MANGLE) Answer Y here TOS target support (CONFIG_IP_NF_TARGET_TOS) Answer Y here MARK target support (CONFIG_IP_NF_TARGET_MARK) Answer Y here LOG target support (CONFIG_IP_NF_TARGET_LOG) Answer Y here TCPMSS target support (CONFIG_IP_NF_TARGET_TCPMSS) Answer Y here Here are the required kernel setups for a Gateway/Proxy server: * Networking options * Packet socket (CONFIG_PACKET) Answer Y here Packet socket: mmapped IO (CONFIG_PACKET_MMAP) Answer Y here Netlink device emulation (CONFIG_NETLINK_DEV) Answer Y here Network packet filtering (replaces ipchains) (CONFIG_NETFILTER) Answer Y here Network packet filtering debugging (CONFIG_NETFILTER_DEBUG) Answer Y here Socket Filtering (CONFIG_FILTER) Answer Y here Unix domain sockets (CONFIG_UNIX) Answer Y here TCP/IP networking (CONFIG_INET) Answer Y here IP: multicasting (CONFIG_IP_MULTICAST) Answer Y here IP: advanced router (CONFIG_IP_ADVANCED_ROUTER) Answer Y here IP: policy routing (CONFIG_IP_MULTIPLE_TABLES) Answer Y here
- 261. GIPTables Firewall 1 CHAPTER0 261 IP: use netfilter MARK value as routing key (CONFIG_IP_ROUTE_FWMARK) Answer Y here IP: fast network address translation (CONFIG_IP_ROUTE_NAT) Answer Y here IP: equal cost multipath (CONFIG_IP_ROUTE_MULTIPATH) Answer Y here IP: use TOS value as routing key (CONFIG_IP_ROUTE_TOS) Answer Y here IP: verbose route monitoring (CONFIG_IP_ROUTE_VERBOSE) Answer Y here IP: large routing tables (CONFIG_IP_ROUTE_LARGE_TABLES) Answer Y here IP: kernel level autoconfiguration (CONFIG_IP_PNP) Answer N here IP: tunneling (CONFIG_NET_IPIP) Answer Y here IP: GRE tunnels over IP (CONFIG_NET_IPGRE) Answer Y here IP: TCP Explicit Congestion Notification support (CONFIG_INET_ECN) Answer N here IP: TCP syncookie support (disabled per default) (CONFIG_SYN_COOKIES) Answer Y here * * IP: Netfilter Configuration * Connection tracking (required for masq/NAT) (CONFIG_IP_NF_CONNTRACK) Answer Y here FTP protocol support (CONFIG_IP_NF_FTP) Answer Y here IRC protocol support (CONFIG_IP_NF_IRC) Answer Y here IP tables support (required for filtering/masq/NAT) (CONFIG_IP_NF_IPTABLES) Answer Y here limit match support (CONFIG_IP_NF_MATCH_LIMIT) Answer Y here MAC address match support (CONFIG_IP_NF_MATCH_MAC) Answer Y here netfilter MARK match support (CONFIG_IP_NF_MATCH_MARK) Answer Y here Multiple port match support (CONFIG_IP_NF_MATCH_MULTIPORT) Answer Y here TOS match support (CONFIG_IP_NF_MATCH_TOS) Answer Y here LENGTH match support (CONFIG_IP_NF_MATCH_LENGTH) Answer Y here TTL match support (CONFIG_IP_NF_MATCH_TTL) Answer Y here tcpmss match support (CONFIG_IP_NF_MATCH_TCPMSS) Answer Y here Connection state match support (CONFIG_IP_NF_MATCH_STATE) Answer Y here Packet filtering (CONFIG_IP_NF_FILTER) Answer Y here REJECT target support (CONFIG_IP_NF_TARGET_REJECT) Answer Y here Full NAT (CONFIG_IP_NF_NAT) Answer Y here MASQUERADE target support (CONFIG_IP_NF_TARGET_MASQUERADE) Answer Y here REDIRECT target support (CONFIG_IP_NF_TARGET_REDIRECT) Answer Y here Packet mangling (CONFIG_IP_NF_MANGLE) Answer Y here TOS target support (CONFIG_IP_NF_TARGET_TOS) Answer Y here MARK target support (CONFIG_IP_NF_TARGET_MARK) Answer Y here LOG target support (CONFIG_IP_NF_TARGET_LOG) Answer Y here TCPMSS target support (CONFIG_IP_NF_TARGET_TCPMSS) Answer Y here ipchains (2.2-style) support (CONFIG_IP_NF_COMPAT_IPCHAINS) Answer N here ipfwadm (2.0-style) support (CONFIG_IP_NF_COMPAT_IPFWADM) Answer N here WARNING: If you have followed the Linux Kernel chapter and have recompiled your Kernel, all the required options for IPTables firewall support, as shown above, are already set. Remember, all servers should be configured to block unused ports, even if they are not a firewall server.
- 262. GIPTables Firewall 1 CHAPTER0 262 Pristine source As we don’t use the RPM package to install the program, it would be difficult for us to locate all the files installed on the system if in the future we want to upgrade. To solve this problem, it is a good idea to make a list of files on the system before you install GIPTables, and then one afterwards, we can then compare them using the diff utility to find out what files were installed and where they were placed. • Simply run the following command before installing the software: [root@deep root]# find /* > GIPTables1 • And the following one after you install the software: [root@deep root]# find /* > GIPTables2 • Then use the following command to get a list of what changed: [root@deep root]# diff GIPTables1 GIPTables2 > GIPTables-Installed With this procedure, if any future upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In the above example, we use the /root directory of the system to store the generated list of files. Compiling - Optimizing & Installing GIPTables To install the GIPTables software on your system, just download the latest version of the software from http://www.giptables.org/ site, and then as user ‘root’ expand the archive under your /var/tmp directory. • To accomplish this use the following commands: [root@deep /]# cp giptables-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf giptables-version.tar.gz Next, move into the newly created GIPTables source directory and perform the following steps to install the software for your system. • To move into the newly created GIPTables source directory use the command: [root@deep tmp]# cd giptables-1.1/ • To install GIPTables enter the following command: [root@deep giptables-1.1]# ./install.sh The “install.sh” script file will simply install any GIPTables components on your system to the right location. Once the installation of GIPTables has been completed, we can free up some disk space by deleting both the program tar archive and the related source directory since they are no longer needed. • To delete GIPTables and its related source directory, use the commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf giptables-version/ [root@deep tmp]# rm -f giptables-version.tar.gz
- 263. GIPTables Firewall 1 CHAPTER0 263 Configuring GIPTables After GIPTables has been installed successfully on your system, your next step is to modify its configuration file to suit your needs. GIPTables is not software that needs to be compiled to work on your system but just to be configured. As you can imagine there are many possible configurations for a firewall design. Some may need to configure it to run for a Web Server, others may need to configure it to run a Mail Server, DNS Server, Virtual Server, Gateway Server, etc, or some others simply want to have the possibility to configure it for a specific requirement. This is one of the advantages of GIPTables. It comes with many pre built configuration files which are suitable for many different types of server. The GIPTables configuration files are very flexible, easy to understand and setup. All you have to do is to answer questions, which refer to a specific firewall option either ‘yes’ to enable or ‘no’ to disable the service. All pre built configuration files are located under the /lib/giptables/conf directory. Please, look in this directory for any existing configuration file relating to the version of the GIPTables software that you have. At the time writing, the following pre configured GIPTables configuration files are available. giptables.conf.dns1 Default configuration file for a Master DNS Server. giptables.conf.dns2 Default configuration file for a Slave DNS Server. giptables.conf.ftpserver Default configuration file for a FTP Server. giptables.conf.gateway Default configuration file for a Gateway Server. giptables.conf.mailserver Default configuration file for a Mail Server. giptables.conf.ppp Default configuration file for a dialup connection. giptables.conf.README Contains all possible configuration parameters. giptables.conf.virtual Default configuration file for a Virtual Server. giptables.conf.webserver Default configuration file for a Web Server. giptables.conf.workstation Default configuration file for a Workstation. /etc/giptables.conf (The GIPTables Configuration File) /etc/rc.d/rc.giptables.blocked (The GIPTables Blocked File) /etc/init.d/giptables (The GIPTables Initialization File) /etc/giptables.conf: The GIPTables Configuration File The /etc/giptables.conf file is the main configuration file for GIPTables. Though there are many options in this file, to get GIPTables up and running you should only need to change a small number of values. But wait a minute, the giptables.conf file does not exist in /etc directory. Why? Remember that many possible firewall configurations exist and depending on both your requirements and the server type that you expect to protect, configurations may differ. GIPTables has some default example configuration files available under the /lib/giptables/conf directory that should suit your needs. You have to pick the one that is suitable for your server type and then create a symbolic link, as “giptables.conf” in the /etc directory that points to it. This is why the giptables.conf file doesn’t exist in the /etc directory, it’s purely a link.
- 264. GIPTables Firewall 1 CHAPTER0 264 Step1 First of all, choose one of the default configuration files that can be found in the /lib/giptables/conf directory. Find one that mostly meets your needs, make a backup copy of it and open it with any kind of text editor to configure it. In our example, we will configure our GIPTables firewall software for a Gateway/Proxy Server with two network interfaces since it is the most complicated configuration we are likely to encounter. • These procedures can be accomplished with the following commands: [root@deep /]# cd /lib/giptables/conf/ [root@deep conf]# cp giptables.conf.gateway giptables.conf.mybox [root@deep conf]# ln -sf /lib/giptables/conf/giptables.conf.mybox /etc/giptables.conf In the above steps, we make a copy of our original “giptables.conf.gateway” file and create a symbolic link pointing to the copy. NOTE: It is a good idea not to directly modify an example configuration file, because if it gets damage, then you have to install the entire package again in order to get it back. Step2 Now our giptables.conf file that is a symbolic link pointing to the original configuration file for a Gateway set-up exists. It is time to edit it and provide or change some minimal values to make it work for our system. In the GIPTables configuration file below, we’ll ONLY explain how to configure and set parameters that are the same for all types of GIPTables firewall configuration. Parts that differ are associated with different available GIPTables modules that must be loaded by the firewall configuration file to enable different services. All available modules with GIPTables firewall are explained later in this document. • Edit the giptables.conf file (vi /etc/giptables.conf) and configure it to fit in with your system and networking setup. Text in bold is what you should change to make the firewall work with your server: The Debug Definition: The first configurable option that is the same for all types of GIPTables firewall configuration is “DEBUG” and by default it is set to “off”. This option is useful only for debugging purposes. # ---------------------------------------------------------------------------- # DEBUG # DEBUG="off" If you set this option to "on", the firewall will display all IPTables rules relating to the GIPTables configuration file that you use to the screen, nothing will go to the kernel. The displayed set of rules will be commented so that you will not end up with lots of rules on the screen that you do not understand. This way you can see only what the firewall is generating, and also you will be able to better understand which rule is for what.
- 265. GIPTables Firewall 1 CHAPTER0 265 When this option is set to "off" (the default setting), the firewall will send all generated rules to the kernel, nothing will be displayed on your screen and the firewall will run on your system. Therefore is you want to run GIPTables on your system, you must be sure that this option is set to ‘off’. NOTE: When the “DEBUG” option is set to “on”, it is possible to redirect the output of the firewall rules to a file, and use this file as learning example of how to set up IPTables rules for different kind of services. This is possible with the following command: [root@deep tmp]# /etc/init.d/giptables start > output-result.txt The Monolithic Kernel Definition: The second configurable option that is the same for all types of GIPTables configuration is the “MONOLITIC_KERNEL” option and it is set to “no” by default. This option exists because the Linux kernel can be compiled so that it contains all the required IPTables driver code directly in it. This way we have a MONOLITHIC KERNEL and we would need to answer by “yes” to the option. # ---------------------------------------------------------------------------- # Some definitions for easy maintenance # Edit these to suit your system # MONOLITIC_KERNEL="no" If you set this option to ‘yes’, then GIPTables will be informed that all native IPTables modules are directly compiled into the kernel. It is important to say ‘yes’ here only if you have a Monolithic Linux Kernel installed on your computer otherwise say ‘no’. Then the firewall will look for and load all the IPTables modules that are required, depending on your configuration file. NOTE: If you compile your kernel as Monolithic, you should know what IPTables modules you need to compile directly into the kernel, since the firewall will not try to load them. If you missed some modules, you will inevitably get errors, or the firewall might not work as expected. The best solution for a Monolithic Kernel set-up is to compile all native iptables modules into the kernel. Also, don’t forget to set MONOLITIC_KERNEL="yes" in the firewall configuration file. The External Network Interface Definition: The next configurable option that is the same for all type of GIPTables firewall configuration is one of the most important settings in the configuration file. # Interface 0: This is our external network interface # It is directly connected to Internet INTERFACE0="eth0" INTERFACE0_IPADDR="x.x.x.x" ANY_IPADDR="0/0"
- 266. GIPTables Firewall 1 CHAPTER0 266 The above definitions set up the parameters associated with our network interface. The first parameter (INTERFACE0="eth0") defines our external interface (the one directly connected to the Internet). By convention, we set it as ‘eth0’, but this is not mandatory and you can change it for whatever your external network interface is. The second parameter (INTERFACE0_IPADDR="x.x.x.x") defines the external IP address associated with the ‘eth0’ interface that your ISP or administrator assigned to you. Remember that every one will have a different IP address, therefore the IP value should be the one assigned to you. The third parameter (ANY_IPADDR="0/0") defines the IP address of any machine. The value of “0/0” means any machines from anywhere. This should NOT be changed, since we use this parameter when we want to talk to any machine out there. WARNING: This warning apply only for a DHCP server configuration. 1) If you get your external IP address from your ISP’s DHCP server, then set the value associated with the “INTERFACE0_IPADDR” parameter To: INTERFACE0_IPADDR=`/lib/giptables/if_ipaddr $INTERFACE0'. 2) Because the firewall is configured to be loaded before any network is initialized, we have to edit /etc/init.d/giptables file and replace the second line that reads: # chkconfig: 2345 08 92 To read: # chkconfig: 2345 11 92 Which will configure our firewall to start up after the network is initialized, and after we received our dynamic IP address from the DHCP server. The Internal Network Interface Definition: The next definition is very important for a ONLY Gateway Server set-up since it allows us to define our second network interface on the system. It is simply NOT required and does not apply on servers or workstations with only one network interface. # Interface 1: This is our internal network interface # It is directly connected to our internal Network 1 INTERFACE1="eth1" INTERFACE1_IPADDR="192.168.1.254" NETWORK1="192.168.1.0/24" The above definitions set up parameters associated with our second network interface (if any). As we can see, the first parameter (INTERFACE1="eth1") defines, in this case, our internal interface name (the one directly connected to our internal private network).
- 267. GIPTables Firewall 1 CHAPTER0 267 The second parameter (INTERFACE1_IPADDR="192.168.1.254") defines the internal IP address associated with the ‘eth1’ interface. Don’t forget to change the IP address if your IP address is different. Finally, the third and new parameter (NETWORK1="192.168.1.0/24") defines our internal subnet. Note that we define it with the IP range to cover every node in our private internal network. As usual, you have to change the example IP address range for the one that you use. NOTE: If you do not have an internal network, then your machine is a Workstation or a Server with only one network interface. In this case just comment out those three options or only the INTERFACE1 option, and the firewall will totally ignore all other options that refer to the internal interface and network. If this is true in your case, then you will have to use another GIPTables example configuration file instead of the giptables.conf.gateway configuration file, which is only suitable for a Gateway Server. The Name Servers Definition: The Name Servers definition is where we define IP addresses for our Primary and Secondary Domain Name Servers. The entries can be the IP addresses that your ISP gave you or the one that your administrator gave you for your network. # Your name servers ip address ISP_PRIMARY_DNS_SERVER="a.a.a.a" ISP_SECONDARY_DNS_SERVER="b.b.b.b" The SYSLOG Server Definition: The SYSLOG Server definition is mandatory. You only need to define it if you have one central log server configured to receive all syslog messages on your network. In general, and if you use this feature, you will have one server on your network configured to receive all log messages and all other servers configured to send their syslog message to the central log server. The value that should be entered into the SYSLOG Server Definition is the IP address of the central log server. # SYSLOG server ip address SYSLOG_SERVER="c.c.c.c" The loopback Interface Definition: The next definition in our GIPTables configuration relates to the loopback interface of Linux and you don’t need to modify it at all. # Loopback interface LOOPBACK_INTERFACE="lo" # Loopback interface
- 268. GIPTables Firewall 1 CHAPTER0 268 The Ports Declarations Definition: The same it true for the definition of privileged and unprivileged ports numbers on our system. The privileged ports numbers are used by daemon services that we run on the server and the unprivileged ports number by clients to establish a connection to the ports on the server. The ports declaration is important for GIPTables to distinguish which ports important services are allowed to run on and on which ports client are allowed to connect. This is a security feature. # Port declarations do not change them PRIV_PORTS="0:1023" UNPRIV_PORTS="1024:65535" The Custom Rules Definition: Most services and rules used on production servers are already included with GIPTables through the different modules files. There can be situations where we need to add additional rules to the firewall; this is possible with the Custom Rules Definition. If you answer “yes” to the definition below, GIPTables will let you add you own custom IPTables rules through the file “rc.giptables.custom” located under the /etc/rc.d directory and the rules will then be added to the firewall. # Loading custom firewall rules from /etc/rc.d/rc.giptables.custom # LOAD_CUSTOM_RULES="yes" If LOAD_CUSTOM_RULES="no", then the Custom Rules Definition is disable. The Logging Definition: This section configures the logging of dropped packets and sends the logging information to a log file of our choice for later investigation. As you will see, for each interface and our internal network we have separate logging options. If you do not have an internal interface, then you can either just ignore, comment out or delete those options that refer to internal interface and internal network (Interface1 and Network1). # ---------------------------------------------------------------------------- # Logging # We log & drop all the packets that are not expected. In order to avoid # our logs being flooded, we rate limit the logging. # Interface 0 log dropped packets INTERFACE0_LOG_DROPPED_PACKETS="yes" INTERFACE0_LOG_LIMIT="5/m" INTERFACE0_LOG_LIMIT_BURST="7" # Interface 1 log dropped packets INTERFACE1_LOG_DROPPED_PACKETS="yes" INTERFACE1_LOG_LIMIT="7/m" INTERFACE1_LOG_LIMIT_BURST="9" # Network 1 log forwarded dropped packets NETWORK1_LOG_DROPPED_PACKETS="yes" NETWORK1_LOG_LIMIT="9/m" NETWORK1_LOG_LIMIT_BURST="11"
- 269. GIPTables Firewall 1 CHAPTER0 269 Our default firewall policy is to DROP everything, and ACCEPT only wanted packets. In an ideal network environment, we do not need to drop a single packet, but when we want to protect our machine or our internal network from the garbage that is out there on the Internet then we really need to consider dropping unwanted packets. What we actually drop are weird packets, incoming connections for services that we do not want to give to the external world, and so on. When those unwanted packets are coming in, we log them just to see when and from where those packets are coming in. Now, there might be a situation when somebody out there will send to us only packets that we don’t want, and because we are logging everything that we drop; soon our logs will fill our disk space. To avoid this, we impose a rate limit to the logging, so that at any time, only the value entered into the LOG_LIMIT parameter will be logged with a burst of the value entered into the LOG_LIMIT_BURST parameter. The LOG_LIMIT module option specifies the maximum average number of matches to allow per second, minute, hour or day by using /second or /s, /minute or /m, /hour or /h and /day or /d. The LOG_LIMIT_BURST module option specifies the exact number of packets to log picked up from the value defined in the LOG_LIMIT module option. Ok, I’m pretty sure that this seems a little confusing. Therefore, if we take the above INTERFACE0 example, the definitions mean that, the first time this rule is reached, the packet will be logged; in fact, since the default burst is 7 (INTERFACE0_LOG_LIMIT_BURST="7"), the first seven packets will be logged. After this, it will be five minutes (INTERFACE0_LOG_LIMIT="5/m") before a packet will be logged from this rule, regardless of how many packets reach it. The log information is sent to the /var/log/messages file. There are different strings that can be used to interpret the /var/log/messages file in order to find different types of dropped packet information: giptables-drop-src-ipaddr: The packet was dropped based on the source IP address. giptables-drop-dst-ipaddr: The packet was dropped based on the destination IP address. giptables-new-no-syn: The TCP packet was dropped because it was a NEW one without SYN flag set. giptables-fragments: The packet was dropped because it was a fragment. giptables-malformed-xmas: The TCP packet was dropped because it looks like a malformed XMAS packet. giptables-malformed-null: The TCP packet was dropped because it looks like a malformed NULL packet.
- 270. GIPTables Firewall 1 CHAPTER0 270 The Network Ghouls Definition: There might be situations when we would like to DROP connections to and from one or more IP addresses. This can be done using the Network Ghouls section and by changing the default value of ‘no’ to ‘yes’. # ---------------------------------------------------------------------------- # Network Ghouls # Refuse any connection from problem sites # NETWORK_GHOULS="no" To enable the Network Ghouls definition, we have to answer ‘yes’ to the first parameter (NETWORK_GHOULS="yes"). If (NETWORK_GHOULS="no"), this section is ignored by the firewall, and it doesn't matter how many IP addresses are added. NOTE: The list of IP addresses that will be blocked from having any kind of access to your server on all interfaces should be defined into the /etc/rc.d/rc.giptables.blocked file when the NETWORK_GHOULS parameter is set to “yes”. The Syn-flood Protection Definition: To protect your machine from SYN-flooding Denial of Service (DoS) attacks, the SYN_FLOOD_PROTECTION parameter should be set to ‘yes’. This allows us to limit the number of incoming TCP connections, and at anytime have a well-defined number of allowed TCP connections on the system. # ---------------------------------------------------------------------------- # Syn-flood protection # Limit the number of incoming tcp connections # SYN_FLOOD_PROTECTION="yes" # Interface 0 incoming syn-flood protection INTERFACE0_IN_SYN_FLOOD_PROTECTION="yes" INTERFACE0_IN_TCP_CONN_LIMIT="1/s" INTERFACE0_IN_TCP_CONN_LIMIT_BURST="3" # Interface 1 incoming syn-flood protection INTERFACE1_IN_SYN_FLOOD_PROTECTION="yes" INTERFACE1_IN_TCP_CONN_LIMIT="3/s" INTERFACE1_IN_TCP_CONN_LIMIT_BURST="5" # Network 1 forwarded incoming syn-flood protection NETWORK1_IN_SYN_FLOOD_PROTECTION="yes" NETWORK1_IN_TCP_CONN_LIMIT="5/s" NETWORK1_IN_TCP_CONN_LIMIT_BURST="7" The TCP_CONN_LIMIT option specifies the maximum average number of new TCP packets that starts a new connection to be accepted per second, minute, hour or day by using /second or /s, /minute or /m, /hour or /h and /day or /d.
- 271. GIPTables Firewall 1 CHAPTER0 271 In our example, we have two interface definitions (INTERFACE0 & INTERFACE1) and one network definition (NETWORK1). The network definition refers to our internal network and the SYN- flood protection feature is enabled on each one. If you don’t have an internal interface, then just ignore the options that refer to internal interface and network (Interface1 and Network1). If SYN_FLOOD_PROTECTION="no", then the entire SYN-flood protections section are ignored. The Sanity Check Definition: The SANITY_CHECK definition allows us to check the sanity (health) of packets that are coming in. If the (SANITY_CHECK) option is set to ‘yes’, Sanity Check protection with your firewall will be enabled. # ---------------------------------------------------------------------------- # Sanity check # SANITY_CHECK="yes" # Make sure NEW incoming TCP connections are SYN packets INTERFACE0_IN_DROP_NEW_WITHOUT_SYN="yes" INTERFACE1_IN_DROP_NEW_WITHOUT_SYN="yes" NETWORK1_IN_DROP_NEW_WITHOUT_SYN="yes" # Drop all incoming fragments INTERFACE0_IN_DROP_ALL_FRAGMENTS="yes" INTERFACE1_IN_DROP_ALL_FRAGMENTS="yes" NETWORK1_IN_DROP_ALL_FRAGMENTS="yes" # Drop all incoming malformed XMAS packets INTERFACE0_IN_DROP_XMAS_PACKETS="yes" INTERFACE1_IN_DROP_XMAS_PACKETS="yes" NETWORK1_IN_DROP_XMAS_PACKETS="yes" # Drop all incoming malformed NULL packets INTERFACE0_IN_DROP_NULL_PACKETS="yes" INTERFACE1_IN_DROP_NULL_PACKETS="yes" NETWORK1_IN_DROP_NULL_PACKETS="yes" There are 4 different kinds of sanity checks used in this version of GIPTables Firewall and each one has a specific function to accomplish, which are. A) Make sure that NEW incoming TCP connections are SYN packets. This will log and drop any new packet that does not have SYN flag set. B) Drop all incoming fragments. This will log and drop any fragment. Fragments can be overlapped, and the subsequent interpretation of such fragments is very OS-dependent. In our protection, we are not going to trust any fragments, thus we log them just to see if we get any, and drop them too. C) Drop all incoming malformed XMAS packets. A typical XMAS scan will most likely show all flags from TCP packet header set. We log and drop all XMAS packets.
- 272. GIPTables Firewall 1 CHAPTER0 272 D) Drop all incoming malformed NULL packets. A NULL packet has no flags set in the TCP header, so it does not do anything and we don’t need it. Those NULL packets are usually used for port scans; therefore we should safely drop all of them. You can set the sanity check protection based on interface or network. If you don’t have an internal interface, then just ignore, comment out or delete the options that refer to internal interface and network (Interface1 and Network1). If SANITY_CHECK="no", then the entire sanity check section is ignored. The Spoofing & Bad Addresses Definition: All IP packet headers contain the source and destination IP addresses and the type of IP protocol message (ICMP, UDP or TCP) the packet contains. The only means of identification under the Internet Protocol (IP) is the source address in the IP packet header. This is a problem that opens the door to source address spoofing, where the sender may replace its address with either a nonexistent address, or the address of some other site. Also, there are at least seven sets of source addresses you should always refuse on your external interface. These are incoming packets claiming to be from: Your external IP address Class A private IP addresses Class B private IP addresses Class C private IP addresses Class D multicast addresses Class E reserved addresses The loopback interface With the exception of your own IP address, blocking outgoing packets containing these source addresses also protects you from possible configuration errors on your part. In this section we log and drop all incoming packets with source IP addresses that we do not expect or want. There are some important one that really need to be monitored and controlled as shown below: # ---------------------------------------------------------------------------- # Spoofing and bad addresses # REFUSE_SPOOFING="yes" There is no way for a packet that come in from the Internet on our external interface to have its source IP address the same with our external IP address. If this happen, then packets are spoofed; therefore we log and drop them. A) We log and drop all incoming packets claiming to be from the IP addresses of our interfaces. In a Gateway firewall configuration, we have two network interfaces, and two IP addresses associated with them. Therefore, we should protect both interfaces as follow.
- 273. GIPTables Firewall 1 CHAPTER0 273 # Refuse incoming packets claiming to be from the IP addresses of our interfaces REFUSE_SPOOFING_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_IN_REFUSE_SPOOFING[0]="yes" INTERFACE1_IN_REFUSE_SPOOFING[0]="no" NETWORK1_IN_REFUSE_SPOOFING[0]="yes" REFUSE_SPOOFING_IPADDR[1]=$INTERFACE1_IPADDR INTERFACE0_IN_REFUSE_SPOOFING[1]="no" INTERFACE1_IN_REFUSE_SPOOFING[1]="yes" NETWORK1_IN_REFUSE_SPOOFING[1]="no" B) We log and drop all incoming packets claiming to be from the broadcast source address range. We accept broadcast source packets only in one situation: when we have a DHCP Server, and this, because a DHCP Client will request its IP address by sending out and DHCP discovery packet that has source IP address "0.0.0.0" and destination IP address "255.255.255.255". In this situation, the Gateway Server is also a DHCP Server, so we will accept by default those broadcast source packets only on the internal interface. # Refuse incoming packets claiming to be from broadcast-src address range REFUSE_SPOOFING_IPADDR[2]="0.0.0.0/8" INTERFACE0_IN_REFUSE_SPOOFING[2]="yes" INTERFACE1_IN_REFUSE_SPOOFING[2]="no" NETWORK1_IN_REFUSE_SPOOFING[2]="yes" C) We log and drop all incoming packets claiming to be from the reserved loopback IP address range. This is so obvious. We should never have incoming packets with source IP address from the loopback address range. We can refuse them safely on all our interfaces. # Refuse incoming packets claiming to be from reserved loopback address range REFUSE_SPOOFING_IPADDR[3]="127.0.0.0/8" INTERFACE0_IN_REFUSE_SPOOFING[3]="yes" INTERFACE1_IN_REFUSE_SPOOFING[3]="yes" NETWORK1_IN_REFUSE_SPOOFING[3]="yes" E) We log and drop all incoming packets claiming to be from the well-known private networks: A, B, C. We can safely refuse all packets claiming to be from those private networks on all of our interfaces, and internal network. # Refuse incoming packets claiming to be from class A private network REFUSE_SPOOFING_IPADDR[4]="10.0.0.0/8" INTERFACE0_IN_REFUSE_SPOOFING[4]="yes" INTERFACE1_IN_REFUSE_SPOOFING[4]="yes" NETWORK1_IN_REFUSE_SPOOFING[4]="yes"
- 274. GIPTables Firewall 1 CHAPTER0 274 # Refuse incoming packets claiming to be from class B private network REFUSE_SPOOFING_IPADDR[5]="172.16.0.0/12" INTERFACE0_IN_REFUSE_SPOOFING[5]="yes" INTERFACE1_IN_REFUSE_SPOOFING[5]="yes" NETWORK1_IN_REFUSE_SPOOFING[5]="yes" # Refuse incoming packets claiming to be from class C private network REFUSE_SPOOFING_IPADDR[6]="192.168.0.0/16" INTERFACE0_IN_REFUSE_SPOOFING[6]="yes" INTERFACE1_IN_REFUSE_SPOOFING[6]="no" NETWORK1_IN_REFUSE_SPOOFING[6]="yes" WARNING: There is only one exception in which case we do not refuse incoming packets on our internal interface claiming to be from our internal private network. This appears only for a Gateway Server when your internal network is from class C. You should not refuse incoming packets on internal interface from your internal network. F) We log and drop all incoming packets claiming to be from class D, E, and unallocated IP addresses. These are classes that are not currently used or that are unallocated. There is no reason for an incoming packet to have a source IP address from one of those classes. # Refuse incoming packets claiming to be from class D, E, and unallocated REFUSE_SPOOFING_IPADDR[7]="224.0.0.0/3" INTERFACE0_IN_REFUSE_SPOOFING[7]="yes" INTERFACE1_IN_REFUSE_SPOOFING[7]="yes" NETWORK1_IN_REFUSE_SPOOFING[7]="yes" The above Spoofing and bad address protection assume that you have two network interfaces installed on your system. This configuration is suitable for a Gateway Server. If you only have one network interface on your server, then you can ignore, comment out or remove those options that refer to internal interface and network (Interface1 and Network1). If REFUSE_SPOOFING="no" then the entire spoofing protection section is ignored. The above configuration closes our discussion about parameters that are the same for all types of GIPTables firewall configurations. Once you have configured all of the customized values in this part of the GIPTables configuration file, suitable for your type of system, you are ready to start the software. /etc/rc.d/rc.giptables.blocked: The GIPTables Blocked File Sometimes you’ll know an address that you would like to block from having any access at all to your server. Instead of entering the entire iptables line per IP address for those jerks on the internet, you can write them into the rc.giptables.blocked file, that will take the IP addresses, strip out any comments and run the resulting list through an iptables routine. The net effect is the /etc/giptables.conf file increases no more than needed, especially when one might have a large number of IP addresses to deny.
- 275. GIPTables Firewall 1 CHAPTER0 275 Step 1 Edit the rc.giptables.blocked file (vi /etc/rc.d/rc.giptables.blocked) and add all the IP addresses that you want blocked from having any access to your server. For example, I’ve put the following IP addresses in this file: # ---------------------------------------------------------------------------- # GIPTables Firewall v0.1-fox # Copyright (C) 2001, 2002 Adrian Pascalau <apascalau@openna.com> # rc.giptables.blocked file # # ---------------------------------------------------------------------------- # This file is part of GIPTables Firewall # # GIPTables Firewall is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or # (at your option) any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 204.254.45.9 # Cracker site with priority 01. 187.231.11.5 # Spam site with priority 07. #214.34.144.4 # Temporally reactivated, please verify with log file. # ---------------------------------------------------------------------------- # End of file Here we can see how this file can be useful. Now we can add the bad IP address, with some comments if necessary to remember why we’ve added the IP address, into the /etc/rc.d/rc.giptables.blocked file and restart GIPTables for the changes to take effect. /etc/init.d/giptables: The GIPTables Initialization File The /etc/init.d/giptables script file is responsible for automatically starting and stopping the GIPTables Firewall. It can take several parameters like ‘start’, ‘stop’, ‘restart’ and ‘panic’. The ‘start’ parameter will actually start the firewall, read the configuration file, clear any pre-defined rules and chains from the kernel, set DROP as the default policy which will deny everything by default and then generate the IPTables rules according to your GIPTables configuration file. The ‘stop’ parameter will stop the firewall, clear any pre-defined rules and chains from the kernel, and set ACCEPT as the default policy for all IPTables default chains. The ‘restart’ option is really just ‘start’ as this firewall isn't a daemon and ‘start’ clears any pre-defined rules anyway. This is really only here to make those who expect it happy. The ‘panic’ option should be used when you want to cut any connections to and from your machine. It will clear any pre-defined rules and chains from the kernel, set default policy as DROP for all IPTables default chains and let through only the packets destined for the loopback interface.
- 276. GIPTables Firewall 1 CHAPTER0 276 • To start GIPTables on your system, use the following command: [root@deep /]# /etc/init.d/giptables start The GIPTables Firewall Module Files Once you have chosen the right GIPTables configuration file suitable for the type of server that you want to protect and all the parameters, which are the same for all type of GIPTables firewall have been configured, GIPTables should have enough information to run properly on your system. Really, your work is done and your firewall is well protected against attacks. Everyone has a different set-up for his firewall design and sometimes we need to implement a new service and then open and control the port associated with this service on our server. GIPTables allows us to add, modify, delete, and customize any existing or expected services in a simple manner through its modules feature. With GIPTables, each service like DNS, FTP, HTTPD, etc have their own modules. Those modules are loaded only when defined in the giptables.conf file, so that if there are no options related to FTP for example, the FTP module will not be loaded. You can specify on which interface or network the module will work, and what kind of requests (incoming or outgoing) can go thought that interface or network. All GIPTables modules are located under the /lib/giptables/modules directory and it’s in these module files that we handle all rules relating to the specific service. When we configure, customize and enable service parameters in the giptables.conf file, the parameter in question get its information about IPTables rules that must be used through the modules files available under the /lib/giptables/modules directory. If the parameter of the specific service that you want to enable is not defined into the GIPTables configuration file, then this service will not load its IPTables rules from its modules file and will not run with your GIPTables Firewall software. If you look in the /lib/giptables/modules directory, you’ll find the following modules for the services that can be enabled with GIPTables Firewall. giptables-ANY The ANY module, which refer to ANY services giptables-AUTH The AUTH module, which refer to AUTH services giptables-DHCP The DHCP module, which refer to DHCP services giptables-DNS The DNS module, which refer to DNS services giptables-FINGER The FINGER module, which refer to FINGER services giptables-FTP The FTP module, which refer to FTP services giptables-HTTP The HTTP module, which refer to HTTP services giptables-HTTPS The HTTPS module, which refer to HTTPS services giptables-ICMP The ICMP module, which refer to ICMP services giptables-IMAP The IMAP module, which refer to IMAP services giptables-IMAPS The IMAPS module, which refer to IMAPS services giptables-LDAP The LDAP module, which refer to LDAP services giptables-LDAPS The LDAPS module, which refer to LDAPS services giptables-MYSQL The MYSQL module, which refer to MYSQL services giptables-NETBIOS The NetBIOS module, which refer to NetBIOS services giptables-NNTP The NNTP module, which refer to NNTP services giptables-NNTPS The NNTPS module, which refer to NNTPS services giptables-NTP The NTP module, which refer to NTP services giptables-ORACLE The ORACLE module, which refer to ORACLE services giptables-POP3 The POP3 module, which refer to POP3 services giptables-POP3S The POP3S module, which refer to POP3S services giptables-POSTGRES The POSTGRES module, which refer to POSTGRES services
- 277. GIPTables Firewall 1 CHAPTER0 277 giptables-SMTP The SMTP module, which refer to SMTP services giptables-SMTPS The SMTPS module, which refer to SMTPS services giptables-SQUID The SQUID module, which refer to SQUID services giptables-SSH The SSH module, which refer to SSH services giptables-SYSLOG The SYSLOG module, which refer to SYSLOG services giptables-TELNET The TELNET module, which refer to TELNET services giptables-TELNETS The TELNETS module, which refer to TELNETS services giptables-TRACEROUTE The TRACEROUTE module, which refer to TRACEROUTE services giptables-WEBCACHE The WEBCACHE module, which refer to WEBCACHE services giptables-WHOIS The WHOIS module, which refer to WHOIS services How GIPTables parameters work? As we’ve shown, GIPTables modules are ONLY loaded when we define and enable their parameters into the GIPTables configuration file. Therefore, if we want to add to our existing configuration a new service that doesn’t exist, we have to define, enable and configure the service with the right parameters. The best way to get an idea about the implementation is to include a new service into our existing GIPTables configuration file. In our next example, we will add the MySQL service to our Gateway Server GIPTables Firewall. We’ll go through the steps that you need to do to add the MySQL service to your GIPTables Firewall. Note that all of the following steps will be the same for any additional services that you might want to add to your existing GIPTables configuration file. Step1 The first step will be to enable the MySQL service module into the GIPTables configuration file. We do this by adding the following lines into the file. Text in bold is what should be added to enable the example MySQL service. • Edit the giptables.conf file (vi /etc/giptables.conf) and add the line. ACCEPT_MYSQL="yes" The above line informs the software to enable the MySQL module service for the MySQL database on any network interfaces or network present on the system and for any requests (incoming or outgoing). Step2 Once the MySQL module service has been enabled, we need to add the right parameters lines specific to the MySQL service to the GIPTables configuration file. Remember that GIPTables is a flexible program that lets us control traffic on external interface, internal interface, and internal network for incoming and outgoing traffic. For a Gateway Server, all options are required but for a server with one network interface, we only need to control traffic on the external interface for incoming and outgoing packets. NOTE: It is important to note that each GIPTables parameter has the same definition and only parts, which relate to services that we want to define change.
- 278. GIPTables Firewall 1 CHAPTER0 278 Enabling outgoing client requests In the example below, we define and enable MySQL outgoing client requests for a Gateway Server. The difference about parameters with other type of servers is that we need to define additional network interface (INTERFACE1) and network (NETWORK1) for a Gateway Server set- up. All text in bold should be configured to define and enable the MySQL service. # ---------------------------------------------------------------------------- # MYSQL outgoing client request # # Interface 0 MYSQL outgoing client request INTERFACE0_MYSQL_CLIENT="yes" INTERFACE0_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_MYSQL_OUT_DST_IPADDR[0]=$ANY_IPADDR In the above example, we first enable MySQL outgoing client request on the external interface (INTERFACE0_MYSQL_CLIENT="yes"). Next, we instruct the system that the parameters apply to interface 0 (INTERFACE0) for the MySQL service (MYSQL) for outgoing requests (OUT) with the source IP address (SRC_IPADDR) coming from our external interface IP address ($INTERFACE0_IPADDR). Which means, packets having our external interface IP address, as a source IP address will be able to go out and/or start a new connection. Finally, we inform the system that the parameters also apply to interface 0 (INTERFACE0) for the MySQL service (MYSQL) for outgoing requests (OUT) with the destination IP address (DST_IPADDR) going to anywhere ($ANY_IPADDR). And this means, packets having our external interface, as the destination IP address will be able to go out and/or start a new connection. Using the connection tracking capability of IPTables, the related MySQL incoming packets are automatically allowed back in by the firewall. In this case, our machine can be a MySQL client that is allowed to access any MySQL server on the Internet. If we want to restrict access to only one external MySQL server, the parameters should be configured like in the example below: # Interface 0 MYSQL outgoing client request INTERFACE0_MYSQL_CLIENT="yes" INTERFACE0_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_MYSQL_OUT_DST_IPADDR[0]="x.x.x.x" In this case, "x.x.x.x" is the IP address of the external MySQL server that we want to access. For a second MySQL server, another set of parameters should be added, like in the example below:
- 279. GIPTables Firewall 1 CHAPTER0 279 # Interface 0 MYSQL outgoing client request INTERFACE0_MYSQL_CLIENT="yes" INTERFACE0_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_MYSQL_OUT_DST_IPADDR[0]="x.x.x.x" INTERFACE0_MYSQL_OUT_SRC_IPADDR[1]=$INTERFACE0_IPADDR INTERFACE0_MYSQL_OUT_DST_IPADDR[1]="y.y.y.y" "x.x.x.x" is the IP address of the first external MySQL server that we want to access and "y.y.y.y" is the IP address of the second external MySQL server that we want to access. Please note that the index of parameters has been increased, so that the first set of parameters have the index 0, and the second set of parameters have the index 1. NOTE: This rule is the same for all GIPTables Firewall parameters that have an index. If you would like to add a second set of parameters, just copy/paste them, make the required changes and do not forget to increase the index. On a Gateway Server or machines with two networks interfaces, we need to define the following additional parameters for the firewall to recognize the other network interface and the private network behind it. # Interface 1 MYSQL outgoing client request INTERFACE1_MYSQL_CLIENT="yes" INTERFACE1_MYSQL_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR INTERFACE1_MYSQL_OUT_DST_IPADDR[0]=$NETWORK1 In the above example, we enable MySQL outgoing client request on the internal interface (INTERFACE1_MYSQL_CLIENT="yes"). We instruct the system that the parameters apply to internal interface 1 (INTERFACE1) for the MySQL service (MYSQL) to outgoing requests (OUT) with source IP address (SRC_IPADDR) coming from our internal interface IP address ($INTERFACE1_IPADDR). Therefore, any packets having our internal interface IP address, as source IP address will be able to go out and/or start a new connection. Next, we inform the system that the parameters also apply to internal interface 1 (INTERFACE1) for the MySQL service (MYSQL) for outgoing requests (OUT) with a destination IP address (DST_IPADDR) going from our internal subnet IP address range ($NETWORK1). Therefore, any packets from our internal subnet will be able to go out and/or start new connections. Using the connection tracking capability of IPTables, the related MySQL incoming packets are automatically allowed back in by the firewall. In this case, our machine can be a MySQL client that is allowed to access any MySQL server from our internal subnet. # Network 1 MYSQL forwarded outgoing client request NETWORK1_MYSQL_CLIENT="yes" NETWORK1_MYSQL_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_MYSQL_OUT_DST_IPADDR[0]=$ANY_IPADDR
- 280. GIPTables Firewall 1 CHAPTER0 280 Here, we enable MySQL outgoing client requests on our internal subnet (NETWORK1_MYSQL_CLIENT="yes"). We instruct the system that the parameters apply to our internal subnet (NETWORK1) for the MySQL service (MYSQL) for outgoing requests (OUT) with the source IP address (SRC_IPADDR) coming from our internal subnet IP address range ($NETWORK1). In the second line, we inform the system that the parameters also apply to our internal subnet (NETWORK1) for the MySQL service (MYSQL) to outgoing requests (OUT) with destination IP address (DST_IPADDR) going to anywhere ($ANY_IPADDR). Using the connection tracking capability of IPTables, the related MySQL incoming packets are automatically allowed back in by the firewall. In this case, our machines from our internal subnet are the MySQL clients and are allowed to access any MySQL server on the Internet. NOTE: The requests are automatically SNATed (MASQUERADEd) by the GIPTables Firewall, so that the MySQL server from the Internet thinks that talks with our Gateway server. In general, you should only replace MYSQL with the name of the service that you want to define for the parameters to work for other type of services. In our example, we use MYSQL; it is to you to change it for the service of your choice. Enabling incoming client requests As we can see, all of the above parameters apply only to outgoing client requests for the MySQL service on a Gateway Server. Now for incoming server requests, we should add the related lines to the configuration file to allow them in. In the example below, we define and enable MySQL incoming client requests for a Gateway Server. The difference with parameters for other types of servers is that here we need to define an additional network interface (INTERFACE1) and a network (NETWORK1) for a Gateway Server set-up. # ---------------------------------------------------------------------------- # MYSQL incoming client request # # Interface 0 MYSQL incoming client request INTERFACE0_MYSQL_SERVER="yes" INTERFACE0_MYSQL_IN_SRC_IPADDR[0]=$ANY_IPADDR INTERFACE0_MYSQL_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR In the above example, we first enable incoming client request for MySQL on the external interface (INTERFACE0_MYSQL_SERVER="yes"). Next, we instruct the system that the parameters apply to external interface 0 (INTERFACE0) for the MySQL service (MYSQL) for incoming requests (IN) with the source IP address (SRC_IPADDR) coming from anywhere ($ANY_IPADDR). This mean that we permit the firewall to receive packets coming from anywhere on our external interface to start a new connection.
- 281. GIPTables Firewall 1 CHAPTER0 281 Finally, we inform the system that parameters also apply to external interface 0 (INTERFACE0) for MySQL service (MYSQL) on incoming requests (IN) with destination IP address (DST_IPADDR) coming from our external IP address ($INTERFACE0_IPADDR). In other terms, incoming packets having our external interface, as destination IP address will be able to come in and/or start a new connection. Using the connection tracking capability of IPTables, the related MySQL outgoing packets are automatically allowed back out by the firewall. In this case, our machine is a MySQL server that is allowed to receive requests from any MySQL client from the Internet. If we want to allow access to only one external client machine on the MySQL server, the parameters should be configured like in the example below: # Interface 0 MYSQL incoming client request INTERFACE0_MYSQL_SERVER="yes" INTERFACE0_MYSQL_IN_SRC_IPADDR[0]="x.x.x.x" INTERFACE0_MYSQL_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR In this case, "x.x.x.x" is the IP address of the external client machine that is allowed to access our MySQL server. For a second external client machine allowed, another set of parameters should be added, like in the example below: # Interface 0 MYSQL incoming client request INTERFACE0_MYSQL_SERVER="yes" INTERFACE0_MYSQL_IN_SRC_IPADDR[0]="x.x.x.x" INTERFACE0_MYSQL_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_MYSQL_IN_SRC_IPADDR[1]="y.y.y.y" INTERFACE0_MYSQL_IN_DST_IPADDR[1]=$INTERFACE0_IPADDR "x.x.x.x" is the IP address of the first external client machine that is allowed to access our MySQL server and "y.y.y.y" is the IP address of the second external client machine that is allowed to access our MySQL server. Please note that the index of parameters has been increased, so that the first set of parameters have the index 0, and the second set of parameters have the index 1. NOTE: This rule is the same for all GIPTables Firewall parameters that have an index. If you would like to add a second set of parameters, just copy/paste them, make the required changes and do not forget to increase the index. Don’t forget that we need to add all of the lines below for a Gateway Server set-up for the firewall to recognize the second network interface and our internal subnet. The definitions and explanations are the same as for outgoing client requests explained earlier. # Interface 1 MYSQL incoming client request INTERFACE1_MYSQL_SERVER="yes" INTERFACE1_MYSQL_IN_SRC_IPADDR[0]=$NETWORK1 INTERFACE1_MYSQL_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR
- 282. GIPTables Firewall 1 CHAPTER0 282 In the above example, we enable MySQL incoming client request on the internal interface (INTERFACE1_MYSQL_SERVER="yes"). Next, we instruct the firewall that on the internal interface (INTERFACE1), all MySQL (MYSQL) incoming packets (IN) with source IP address (SRC_IPADDR) from our internal subnet IP address range ($NETWORK1) and with destination IP address (DST_IPADDR) coming from our internal interface IP address ($INTERFACE1_IPADDR) will be allowed to come in and/or start a new connection. In other terms, any incoming MySQL packets with source IP address from our internal subnet IP address range and with our internal interface IP address as destination IP address will be allowed to come in and/or start a new connection. Using the connection tracking capability of IPTables, the related MySQL outgoing packets are automatically allowed back out by the firewall. In this case, our machine is a MySQL server that is allowed to receive requests from any MySQL client from our internal subnet. There might be a situation when we would like to access the MySQL server from our internal subnet using the external interface IP address ($INTERFACE0_IPADDR) as destination IP address (DST_IPADDR). This is the case when we connect to the MySQL server using its host name instead of the IP address. Our DNS server might resolve the MySQL server's IP address as the external interface IP address. In this case, the parameters should be configured like in the example below: # Interface 1 MYSQL incoming client request INTERFACE1_MYSQL_SERVER="yes" INTERFACE1_MYSQL_IN_SRC_IPADDR[0]=$NETWORK1 INTERFACE1_MYSQL_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR INTERFACE1_MYSQL_IN_SRC_IPADDR[1]=$NETWORK1 INTERFACE1_MYSQL_IN_DST_IPADDR[1]=$INTERFACE0_IPADDR As you can see, we have copy/paste the first set of parameters, then changes the destination IP address (DST_IPADDR) to our external interface IP address ($INTERFACE0_IPADDR) and also increase the index number. # Network 1 MYSQL forwarded incoming server request NETWORK1_MYSQL_SERVER="yes" NETWORK1_MYSQL_IN_CLI_IPADDR[0]=$ANY_IPADDR NETWORK1_MYSQL_IN_SRV_IPADDR[0]="192.168.1.1" In the above example, we enable MySQL incoming client request on our internal subnet (NETWORK1_MYSQL_SERVER="yes"). Next, we instruct the firewall that in our internal subnet (NETWORK1), all MySQL (MYSQL) incoming packets (IN) with source IP address (SRC_IPADDR) of any IP address ($ANY_IPADDR) and with destination IP address (DST_IPADDR) "192.168.1.1" will be allowed to come in and/or start a new connection.
- 283. GIPTables Firewall 1 CHAPTER0 283 In other terms, any incoming MySQL packets with any IP address as source IP address and with 192.168.1.1 as destination IP address will be allowed to come in and/or start a new connection. Using the connection tracking capability of IPTables, the related MySQL outgoing packets are automatically allowed back out by the firewall. In this case, our machine from our internal subnet that has the IP address 192.168.1.1 is the MySQL server and it is allowed to receive requests from any MySQL client from the Internet. NOTE: The MySQL client from the Internet thinks that it talks to our Gateway server, so the actual destination IP address of the packet is our external interface IP address ($INTERFACE1_IPADDR), but the packet is automatically DNATed to 192.168.1.1. Pay special attention to the above parameters. We noted that IP address “192.168.1.1” is used as the value for the incoming client requests with the forwarding feature. This is important, if your internal workstation IP address is different, you will have to adjust the setting to fit your own IP address for each of the forwarding definitions. Step3 Now that our parameters for MySQL service have been correctly entered in the GIPTables configuration file, we need to restart our GIPTables firewall for the changes to take effect. • To restart GIPTables on your system, use the following command: [root@deep /]# /etc/init.d/giptables restart Well, now we have a better idea about what these cryptic definitions do and how to change them to fit our needs depending of the type of firewall that we need for our server. Human error is inevitable and if we entered all the additional parameters into GIPTables by hand, we could in inadvertently make some errors. To avoid this risk, GIPTables provides through it’s “giptables.conf.README” file all the possible definitions for available services that can be used with it. Therefore, if you need to add some additional services, which do not exist by default in the giptables.conf file, you can refer to this file to get the parameters to make your service run with GIPTables Firewall. All you’ll need to do is to cut and paste the required lines into your GIPTables configuration file and set up each parameter by answering “yes” or “no” to the questions. Running the type of GIPTables firewall that you need All servers should be configured to block the unused ports, even if they are not a firewall server. This is required for increased security. Imagine that someone gains access to your main firewall server: if your other servers are not configured to block unused ports, this can result a serious network security risk. The same is true for local connections; unauthorized employees can gain access to your other servers from inside the network. As you should know now, before running GIPTables in your system, you must create a symbolic link under the /etc directory that points to the GIPTables configuration file suitable for your system. Once this configuration file exists under your /etc directory, all you have to do is to edit it and set-up your networking configuration to make it work for you. This is true with all of server types except for a Gateway Server which differs as explained below.
- 284. GIPTables Firewall 1 CHAPTER0 284 1) You may need to forward external traffic to your internal network. 2) You may need some specific services not available by default. 3) You need to use the SNAT feature of Linux. 4) You need to use the DNAT feature of Linux. The GIPTables configuration file for a Gateway Server allows you to accomplish these special requirements but requires more work from your part. This is the reason why we will show you later both a complete example configuration file and the required steps for a Gateway/Proxy Server GIPTables configuration that should work for most users. It is important to note that the below example is only a base starting point since every ones needs are different, and the number of services running on specific servers may change from one person to another. All the following steps and explanations are valid for a Gateway/Proxy Server. For any other type of server, you only need to create the symbolic link under your /etc directory that points to your type of server configuration and then start your firewall after setting up your networking configuration in the giptables.conf file. Unlike other types of GIPTables firewall configuration file, e.g. a Web, Mail, DNS Servers, etc., configuring a Linux Server to masquerade and forward traffic from the inside private network that has unregistered IP addresses (i.e. 192.168.1.0/24) to the outside network (i.e. the Internet) requires a special setup of your kernel and your GIPTables firewall configuration file. This kind of configuration is also known as a Gateway Server or Proxy Server (a machine that serves as a gateway for internal traffic to external traffic). This configuration must be set only if you have the need for this kind of service. Some Points to Consider You can assume that you are at risk if you connect your system to the Internet. Your gateway to the Internet is your greatest exposure, so we recommend the following: The Gateway should not run more applications than are absolutely necessary. The Gateway should strictly limit the type and number of protocols allowed to flow through it (protocols potentially provide security holes, such as FTP and telnet). Any system containing confidential or sensitive information should not be directly accessible from the Internet. A Proxy program like Squid is highly recommended on the Gateway Server. The GIPTables configuration file for a Gateway/Proxy Server Masquerading means that if one of the computers on your local network for which your Linux machine (or Gateway/Proxy) acts as a firewall wants to send something to the outside, your machine can "masquerade" as that computer. In other words, it forwards the traffic to the intended outside destination, but makes it look like it came from the firewall machine itself. It works both ways: if the outside host replies, the Linux firewall will silently forward the traffic to the corresponding local computer. This way, the computers on your local network are completely invisible to the outside world, even though they can reach outside and can receive replies. This makes it possible to have the computers on the local network participate on the Internet even if they don’t have officially registered IP addresses.
- 285. GIPTables Firewall 1 CHAPTER0 285 Step1 The IP masquerading code will only work if IP forwarding is enabled on your system. This feature is by default disabled and you can enable it with the following command: • To enable IPv4 forwarding on your Linux system, use the following command: Edit the sysctl.conf file (vi /etc/sysctl.conf) and add the following lines: # Enable packet forwarding (required only for Gateway, VPN, Proxy, PPP) net.ipv4.ip_forward = 1 You must restart your network for the change to take effect. The command to restart the network is: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] Step 2 Create the symbolic link to the giptables.conf file that points to the right GIPTables configuration file suitable for our setup of a Gateway Server. • This can be accomplished with the following command: [root@deep /]# ln -s /lib/giptables/conf/giptables.conf.gateway /etc/giptables.conf Step3 Once the symbolic link is created, we will edit it to suit our requirements. The text in bold are the parts of the configuration that must be modified to satisfy your needs. This is the configuration script file for a Gateway/Proxy Server, it will: 1 Log and limit the amount of incoming dropped packets. 2 Implement the Syn-flood protection. 3 Make sure that all NEW incoming tcp connections are SYN packets. 4 Protect from Spoofing and bad addresses. 5 Allow DNS client requests on internal & external interfaces. 6 Allow FTP client requests on internal & external interfaces. 7 Allow SSH client requests on internal & external interfaces. 8 Allow SSH server requests on the external interface. 9 Allow SMTP client requests on internal & external interfaces. 10 Allow POP3 client requests on the internal interface. 11 Allow POP3S client requests on the internal interface. 12 Allow HTTP client requests on internal & external interfaces. 13 Allow HTTPS client requests on internal & external interfaces. 14 Allow SQUID client requests on internal & external interfaces. 15 Allow SQUID server requests on the external interface. 16 Allow NETBIOS client requests on the internal interface. 17 Allow NETBIOS server requests on the internal interface. 18 Allow TRACEROUTE client requests on internal & external interfaces. 19 Allow TRACEROUTE server requests on internal & external interfaces. 20 Allow ICMP client requests on internal & external interfaces. 21 Allow DHCP client requests on the internal interface.
- 286. GIPTables Firewall 1 CHAPTER0 286 If you don’t want some of the services listed in the firewall rules files for the Gateway/Proxy Server, disable them by saying “no” to the questions. If you want some other services that are not enabled, simply say, “yes” to the questions. If the service does not exist, add it to your configuration based on the available examples from the giptables.conf.README file. • To edit the giptales.conf file, use the following command: [root@deep /]# vi /etc/giptables.conf # ---------------------------------------------------------------------------- # GIPTables Firewall v1.1 http://www.giptables.org # Copyright (C) 2002 Adrian Pascalau <apascalau@openna.com> # GATEWAY main configuration file # # ---------------------------------------------------------------------------- # This file is part of GIPTables Firewall # # GIPTables Firewall is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or # (at your option) any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA # ---------------------------------------------------------------------------- # DEBUG # DEBUG="off" # ---------------------------------------------------------------------------- # Some definitions for easy maintenance # Edit these to suit your system # MONOLITIC_KERNEL="no" # Interface 0: This is our external network interface # It is directly connected to Internet INTERFACE0="eth0" INTERFACE0_IPADDR="x.x.x.x" ANY_IPADDR="0/0" # Interface 1: This is our internal network interface # It is directly connected to our internal Network 1 INTERFACE1="eth1" INTERFACE1_IPADDR="192.168.1.254" NETWORK1="192.168.1.0/24" # Do you need Network Address Translation for your internal network? NETWORK1_NAT="yes"
- 287. GIPTables Firewall 1 CHAPTER0 287 # Your name servers ip address ISP_PRIMARY_DNS_SERVER="y.y.y.y" ISP_SECONDARY_DNS_SERVER="z.z.z.z" # SYSLOG server ip address SYSLOG_SERVER="c.c.c.c" # Loopback interface LOOPBACK_INTERFACE="lo" # Loopback interface # Port declarations, do not change them PRIV_PORTS="0:1023" UNPRIV_PORTS="1024:65535" # ---------------------------------------------------------------------------- # Loading custom firewall rules from /etc/rc.d/rc.giptables.custom # LOAD_CUSTOM_RULES="yes" # ---------------------------------------------------------------------------- # Logging # Limit the amount of incoming dropped packets that gets sent to the logs # # We log & drop all the packets that are not expected. In order to avoid # our logs beeing flooded, we rate limit the logging # Interface 0 log dropped packets INTERFACE0_LOG_DROPPED_PACKETS="yes" INTERFACE0_LOG_LIMIT="5/m" INTERFACE0_LOG_LIMIT_BURST="7" # Interface 1 log dropped packets INTERFACE1_LOG_DROPPED_PACKETS="yes" INTERFACE1_LOG_LIMIT="7/m" INTERFACE1_LOG_LIMIT_BURST="9" # Network 1 log forwarded dropped packets NETWORK1_LOG_DROPPED_PACKETS="yes" NETWORK1_LOG_LIMIT="9/m" NETWORK1_LOG_LIMIT_BURST="11" # ---------------------------------------------------------------------------- # Network Ghouls # Refuse any connection from problem sites # # The /etc/rc.d/rc.giptables.blocked file contains a list of ip addresses that # will be blocked from having any kind of access to your server on all your # interfaces if the next option is "yes" NETWORK_GHOULS="yes"
- 288. GIPTables Firewall 1 CHAPTER0 288 # ---------------------------------------------------------------------------- # Syn-flood protection # Limit the number of incoming tcp connections # SYN_FLOOD_PROTECTION="yes" # Interface 0 incoming syn-flood protection INTERFACE0_IN_SYN_FLOOD_PROTECTION="yes" INTERFACE0_IN_TCP_CONN_LIMIT="1/s" INTERFACE0_IN_TCP_CONN_LIMIT_BURST="3" # Interface 1 incoming syn-flood protection INTERFACE1_IN_SYN_FLOOD_PROTECTION="yes" INTERFACE1_IN_TCP_CONN_LIMIT="3/s" INTERFACE1_IN_TCP_CONN_LIMIT_BURST="5" # Network 1 forwarded incoming syn-flood protection NETWORK1_IN_SYN_FLOOD_PROTECTION="yes" NETWORK1_IN_TCP_CONN_LIMIT="5/s" NETWORK1_IN_TCP_CONN_LIMIT_BURST="7" # ---------------------------------------------------------------------------- # Sanity check # SANITY_CHECK="yes" # Make sure NEW incoming tcp connections are SYN packets INTERFACE0_IN_DROP_NEW_WITHOUT_SYN="yes" INTERFACE1_IN_DROP_NEW_WITHOUT_SYN="yes" NETWORK1_IN_DROP_NEW_WITHOUT_SYN="yes" # Drop all incoming fragments INTERFACE0_IN_DROP_ALL_FRAGMENTS="yes" INTERFACE1_IN_DROP_ALL_FRAGMENTS="yes" NETWORK1_IN_DROP_ALL_FRAGMENTS="yes" # Drop all incoming malformed XMAS packets INTERFACE0_IN_DROP_XMAS_PACKETS="yes" INTERFACE1_IN_DROP_XMAS_PACKETS="yes" NETWORK1_IN_DROP_XMAS_PACKETS="yes" # Drop all incoming malformed NULL packets INTERFACE0_IN_DROP_NULL_PACKETS="yes" INTERFACE1_IN_DROP_NULL_PACKETS="yes" NETWORK1_IN_DROP_NULL_PACKETS="yes"
- 289. GIPTables Firewall 1 CHAPTER0 289 # ---------------------------------------------------------------------------- # Spoofing and bad addresses # REFUSE_SPOOFING="yes" # Refuse incoming packets claiming to be from the ip addresses of our interfaces REFUSE_SPOOFING_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_IN_REFUSE_SPOOFING[0]="yes" INTERFACE1_IN_REFUSE_SPOOFING[0]="no" NETWORK1_IN_REFUSE_SPOOFING[0]="yes" REFUSE_SPOOFING_IPADDR[1]=$INTERFACE1_IPADDR INTERFACE0_IN_REFUSE_SPOOFING[1]="no" INTERFACE1_IN_REFUSE_SPOOFING[1]="yes" NETWORK1_IN_REFUSE_SPOOFING[1]="no" # Refuse incoming packets claiming to be from broadcast-src address range REFUSE_SPOOFING_IPADDR[2]="0.0.0.0/8" # If you provide DHCP services on one of your interfaces, do not forget to # set the following option related to that interface to "no" INTERFACE0_IN_REFUSE_SPOOFING[2]="yes" INTERFACE1_IN_REFUSE_SPOOFING[2]="no" NETWORK1_IN_REFUSE_SPOOFING[2]="yes" # Refuse incoming packets claiming to be from reserved loopback address range REFUSE_SPOOFING_IPADDR[3]="127.0.0.0/8" INTERFACE0_IN_REFUSE_SPOOFING[3]="yes" INTERFACE1_IN_REFUSE_SPOOFING[3]="yes" NETWORK1_IN_REFUSE_SPOOFING[3]="yes" # Refuse incoming packets claiming to be from class A private network REFUSE_SPOOFING_IPADDR[4]="10.0.0.0/8" INTERFACE0_IN_REFUSE_SPOOFING[4]="yes" INTERFACE1_IN_REFUSE_SPOOFING[4]="yes" NETWORK1_IN_REFUSE_SPOOFING[4]="yes" # Refuse incoming packets claiming to be from class B private network REFUSE_SPOOFING_IPADDR[5]="172.16.0.0/12" INTERFACE0_IN_REFUSE_SPOOFING[5]="yes" INTERFACE1_IN_REFUSE_SPOOFING[5]="yes" NETWORK1_IN_REFUSE_SPOOFING[5]="yes" # Refuse incoming packets claiming to be from class C private network REFUSE_SPOOFING_IPADDR[6]="192.168.0.0/16" INTERFACE0_IN_REFUSE_SPOOFING[6]="yes" INTERFACE1_IN_REFUSE_SPOOFING[6]="no" NETWORK1_IN_REFUSE_SPOOFING[6]="yes"
- 290. GIPTables Firewall 1 CHAPTER0 290 # Refuse incoming packets claiming to be from class D, E, and unallocated REFUSE_SPOOFING_IPADDR[7]="224.0.0.0/3" INTERFACE0_IN_REFUSE_SPOOFING[7]="yes" INTERFACE1_IN_REFUSE_SPOOFING[7]="yes" NETWORK1_IN_REFUSE_SPOOFING[7]="yes" # **************************************************************************** # * # A N Y * # * # **************************************************************************** ACCEPT_ANY="no" # **************************************************************************** # * # D N S * # * # **************************************************************************** ACCEPT_DNS="yes" # ---------------------------------------------------------------------------- # DNS outgoing client request # # Interface 0 DNS outgoing client request INTERFACE0_DNS_CLIENT="yes" INTERFACE0_DNS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_DNS_OUT_DST_IPADDR[0]=$ISP_PRIMARY_DNS_SERVER INTERFACE0_DNS_OUT_UDP_REQUEST[0]="yes" INTERFACE0_DNS_OUT_TCP_REQUEST[0]="yes" INTERFACE0_DNS_OUT_SPORT53_REQUEST[0]="no" INTERFACE0_DNS_OUT_SRC_IPADDR[1]=$INTERFACE0_IPADDR INTERFACE0_DNS_OUT_DST_IPADDR[1]=$ISP_SECONDARY_DNS_SERVER INTERFACE0_DNS_OUT_UDP_REQUEST[1]="yes" INTERFACE0_DNS_OUT_TCP_REQUEST[1]="yes" INTERFACE0_DNS_OUT_SPORT53_REQUEST[1]="no" # Network 1 DNS forwarded outgoing client request NETWORK1_DNS_CLIENT="yes" NETWORK1_DNS_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_DNS_OUT_DST_IPADDR[0]=$ISP_PRIMARY_DNS_SERVER NETWORK1_DNS_OUT_UDP_REQUEST[0]="yes" NETWORK1_DNS_OUT_TCP_REQUEST[0]="yes" NETWORK1_DNS_OUT_SPORT53_REQUEST[0]="no" # **************************************************************************** # * # F T P * # * # **************************************************************************** ACCEPT_FTP="yes"
- 291. GIPTables Firewall 1 CHAPTER0 291 # ---------------------------------------------------------------------------- # FTP outgoing client request # # Interface 0 FTP outgoing client request INTERFACE0_FTP_CLIENT="yes" INTERFACE0_FTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_FTP_OUT_DST_IPADDR[0]=$ANY_IPADDR INTERFACE0_FTP_OUT_PASIVE[0]="yes" INTERFACE0_FTP_OUT_ACTIVE[0]="no" # Network 1 FTP forwarded outgoing client request NETWORK1_FTP_CLIENT="yes" NETWORK1_FTP_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_FTP_OUT_DST_IPADDR[0]=$ANY_IPADDR NETWORK1_FTP_OUT_PASIVE[0]="yes" NETWORK1_FTP_OUT_ACTIVE[0]="yes" # **************************************************************************** # * # S S H * # * # **************************************************************************** ACCEPT_SSH="yes" # ---------------------------------------------------------------------------- # SSH outgoing client request # # Interface 0 SSH outgoing client request INTERFACE0_SSH_CLIENT="yes" INTERFACE0_SSH_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_SSH_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 SSH forwarded outgoing client request NETWORK1_SSH_CLIENT="yes" NETWORK1_SSH_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_SSH_OUT_DST_IPADDR[0]=$ANY_IPADDR # ---------------------------------------------------------------------------- # SSH incoming client request # # Interface 0 SSH incoming client request INTERFACE0_SSH_SERVER="yes" INTERFACE0_SSH_IN_SRC_IPADDR[0]=$ANY_IPADDR INTERFACE0_SSH_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR
- 292. GIPTables Firewall 1 CHAPTER0 292 # Interface 1 SSH incoming client request INTERFACE1_SSH_SERVER="yes" INTERFACE1_SSH_IN_SRC_IPADDR[0]=$NETWORK1 INTERFACE1_SSH_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR # **************************************************************************** # * # T E L N E T * # * # **************************************************************************** ACCEPT_TELNET="no" # ---------------------------------------------------------------------------- # TELNET outgoing client request # # Interface 0 TELNET outgoing client request INTERFACE0_TELNET_CLIENT="yes" INTERFACE0_TELNET_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_TELNET_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 TELNET forwarded outgoing client request NETWORK1_TELNET_CLIENT="yes" NETWORK1_TELNET_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_TELNET_OUT_DST_IPADDR[0]=$ANY_IPADDR # ---------------------------------------------------------------------------- # TELNET incoming client request # # Interface 1 TELNET incoming client request INTERFACE1_TELNET_SERVER="no" INTERFACE1_TELNET_IN_SRC_IPADDR[0]=$NETWORK1 INTERFACE1_TELNET_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR # **************************************************************************** # * # T E L N E T S * # * # **************************************************************************** ACCEPT_TELNETS="no" # **************************************************************************** # * # S M T P * # * # **************************************************************************** ACCEPT_SMTP="yes"
- 293. GIPTables Firewall 1 CHAPTER0 293 # ---------------------------------------------------------------------------- # SMTP outgoing client request # # Interface 0 SMTP outgoing client request INTERFACE0_SMTP_CLIENT="yes" INTERFACE0_SMTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_SMTP_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 SMTP forwarded outgoing client request NETWORK1_SMTP_CLIENT="yes" NETWORK1_SMTP_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_SMTP_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # S M T P S * # * # **************************************************************************** ACCEPT_SMTPS="no" # ---------------------------------------------------------------------------- # SMTPS outgoing client request # # Interface 0 SMTPS outgoing client request INTERFACE0_SMTPS_CLIENT="yes" INTERFACE0_SMTPS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_SMTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 SMTPS forwarded outgoing client request NETWORK1_SMTPS_CLIENT="yes" NETWORK1_SMTPS_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_SMTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # P O P 3 * # * # **************************************************************************** ACCEPT_POP3="yes" # ---------------------------------------------------------------------------- # POP3 outgoing client request # # Network 1 POP3 forwarded outgoing client request NETWORK1_POP3_CLIENT="yes" NETWORK1_POP3_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_POP3_OUT_DST_IPADDR[0]=$ANY_IPADDR
- 294. GIPTables Firewall 1 CHAPTER0 294 # **************************************************************************** # * # P O P 3 S * # * # **************************************************************************** ACCEPT_POP3S="yes" # ---------------------------------------------------------------------------- # POP3S outging client request # # Network 1 POP3S forwarded outging client request NETWORK1_POP3S_CLIENT="yes" NETWORK1_POP3S_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_POP3S_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # I M A P * # * # **************************************************************************** ACCEPT_IMAP="no" # ---------------------------------------------------------------------------- # IMAP outgoing client request # # Network 1 IMAP forwarded outgoing client request NETWORK1_IMAP_CLIENT="yes" NETWORK1_IMAP_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_IMAP_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # I M A P S * # * # **************************************************************************** ACCEPT_IMAPS="no" # ---------------------------------------------------------------------------- # IMAPS outgoing client request # # Network 1 IMAPS forwarded outgoing client request NETWORK1_IMAPS_CLIENT="yes" NETWORK1_IMAPS_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_IMAPS_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # H T T P * # * # ****************************************************************************
- 295. GIPTables Firewall 1 CHAPTER0 295 ACCEPT_HTTP="yes" # ---------------------------------------------------------------------------- # HTTP outgoing client request # # Interface 0 HTTP outgoing client request INTERFACE0_HTTP_CLIENT="yes" INTERFACE0_HTTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_HTTP_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 HTTP forwarded outgoing client request NETWORK1_HTTP_CLIENT="yes" NETWORK1_HTTP_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_HTTP_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # H T T P S * # * # **************************************************************************** ACCEPT_HTTPS="yes" # ---------------------------------------------------------------------------- # HTTPS outgoing client request # # Interface 0 HTTPS outgoing client request INTERFACE0_HTTPS_CLIENT="yes" INTERFACE0_HTTPS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_HTTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 HTTPS forwarded outgoing client request NETWORK1_HTTPS_CLIENT="yes" NETWORK1_HTTPS_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_HTTPS_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # S Q U I D * # * # **************************************************************************** ACCEPT_SQUID="yes" # Squid in Proxy-Caching Mode # ---------------------------------------------------------------------------- # SQUID outgoing client request # # Interface 0 SQUID outgoing client request INTERFACE0_SQUID_CLIENT="yes"
- 296. GIPTables Firewall 1 CHAPTER0 296 INTERFACE0_SQUID_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_SQUID_OUT_DST_IPADDR[0]=$ANY_IPADDR # Interface 1 SQUID outgoing client request INTERFACE1_SQUID_CLIENT="yes" INTERFACE1_SQUID_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR INTERFACE1_SQUID_OUT_DST_IPADDR[0]=$NETWORK1 # Network 1 SQUID forwarded outgoing client request NETWORK1_SQUID_CLIENT="yes" NETWORK1_SQUID_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_SQUID_OUT_DST_IPADDR[0]=$ANY_IPADDR # ---------------------------------------------------------------------------- # SQUID incoming client request # # Interface 0 SQUID incoming client request INTERFACE0_SQUID_SERVER="yes" INTERFACE0_SQUID_IN_SRC_IPADDR[0]=$ANY_IPADDR INTERFACE0_SQUID_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR # Interface 1 SQUID incoming client request INTERFACE1_SQUID_SERVER="yes" INTERFACE1_SQUID_IN_SRC_IPADDR[0]=$NETWORK1 INTERFACE1_SQUID_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR # **************************************************************************** # * # W E B C A C H E * # * # **************************************************************************** ACCEPT_WEBCACHE="no" # Squid in HTTPD-Accelerator Mode # **************************************************************************** # * # N N T P * # * # **************************************************************************** ACCEPT_NNTP="no" # ---------------------------------------------------------------------------- # NNTP outgoing client request # # Network 1 NNTP forwarded outgoing client request NETWORK1_NNTP_CLIENT="yes" NETWORK1_NNTP_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_NNTP_OUT_DST_IPADDR[0]=$ANY_IPADDR
- 297. GIPTables Firewall 1 CHAPTER0 297 # **************************************************************************** # * # N N T P S * # * # **************************************************************************** ACCEPT_NNTPS="no" # **************************************************************************** # * # M Y S Q L * # * # **************************************************************************** ACCEPT_MYSQL="no" # **************************************************************************** # * # P O S T G R E S * # * # **************************************************************************** ACCEPT_POSTGRES="no" # **************************************************************************** # * # O R A C L E * # * # **************************************************************************** ACCEPT_ORACLE="no" # **************************************************************************** # * # L D A P * # * # **************************************************************************** ACCEPT_LDAP="no" # **************************************************************************** # * # L D A P S * # * # **************************************************************************** ACCEPT_LDAPS="no" # **************************************************************************** # * # A U T H * # * # **************************************************************************** ACCEPT_AUTH="no" # ---------------------------------------------------------------------------- # AUTH outgoing client request # # Interface 0 AUTH outgoing client request
- 298. GIPTables Firewall 1 CHAPTER0 298 INTERFACE0_AUTH_CLIENT="yes" INTERFACE0_AUTH_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_AUTH_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 AUTH forwarded outgoing client request NETWORK1_AUTH_CLIENT="yes" NETWORK1_AUTH_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_AUTH_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # W H O I S * # * # **************************************************************************** ACCEPT_WHOIS="no" # ---------------------------------------------------------------------------- # WHOIS outgoing client request # # Interface 0 WHOIS outgoing client request INTERFACE0_WHOIS_CLIENT="yes" INTERFACE0_WHOIS_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_WHOIS_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 WHOIS forwarded outgoing client request NETWORK1_WHOIS_CLIENT="yes" NETWORK1_WHOIS_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_WHOIS_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # F I N G E R * # * # **************************************************************************** ACCEPT_FINGER="no" # ---------------------------------------------------------------------------- # FINGER outgoing client request # # Interface 0 FINGER outgoing client request INTERFACE0_FINGER_CLIENT="yes" INTERFACE0_FINGER_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_FINGER_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 FINGER forwarded outgoing client request NETWORK1_FINGER_CLIENT="yes" NETWORK1_FINGER_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_FINGER_OUT_DST_IPADDR[0]=$ANY_IPADDR
- 299. GIPTables Firewall 1 CHAPTER0 299 # **************************************************************************** # * # N T P * # * # **************************************************************************** ACCEPT_NTP="no" # ---------------------------------------------------------------------------- # NTP outgoing client request # # Interface 0 NTP outgoing client request INTERFACE0_NTP_CLIENT="yes" INTERFACE0_NTP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_NTP_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 NTP forwarded outgoing client request NETWORK1_NTP_CLIENT="no" NETWORK1_NTP_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_NTP_OUT_DST_IPADDR[0]=$ANY_IPADDR # **************************************************************************** # * # N E T B I O S * # * # **************************************************************************** ACCEPT_NETBIOS="yes" # ---------------------------------------------------------------------------- # NETBIOS outgoing client request # # Interface 1 NETBIOS outgoing client request INTERFACE1_NETBIOS_CLIENT="yes" INTERFACE1_NETBIOS_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR INTERFACE1_NETBIOS_OUT_DST_IPADDR[0]=$NETWORK1 # ---------------------------------------------------------------------------- # NETBIOS incoming client request # # Interface 1 NETBIOS incoming client request INTERFACE1_NETBIOS_SERVER="yes" INTERFACE1_NETBIOS_IN_SRC_IPADDR[0]=$NETWORK1 INTERFACE1_NETBIOS_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR # **************************************************************************** # * # S Y S L O G * # * # ****************************************************************************
- 300. GIPTables Firewall 1 CHAPTER0 300 ACCEPT_SYSLOG="no" # ---------------------------------------------------------------------------- # SYSLOG outgoing client request # # Interface 1 SYSLOG outgoing client request INTERFACE1_SYSLOG_CLIENT="yes" INTERFACE1_SYSLOG_OUT_SRC_IPADDR[0]=$INTERFACE1_IPADDR INTERFACE1_SYSLOG_OUT_DST_IPADDR[0]=$SYSLOG_SERVER # **************************************************************************** # * # T R A C E R O U T E * # * # **************************************************************************** ACCEPT_TRACEROUTE="yes" # ---------------------------------------------------------------------------- # TRACEROUTE outgoing client request # # Interface 0 TRACEROUTE outgoing client request INTERFACE0_TRACEROUTE_CLIENT="yes" INTERFACE0_TRACEROUTE_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_TRACEROUTE_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 TRACEROUTE forwarded outgoing client request NETWORK1_TRACEROUTE_CLIENT="yes" NETWORK1_TRACEROUTE_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_TRACEROUTE_OUT_DST_IPADDR[0]=$ANY_IPADDR # ---------------------------------------------------------------------------- # TRACEROUTE incoming client request # # Interface 1 TRACEROUTE incoming client request INTERFACE1_TRACEROUTE_SERVER="no" INTERFACE1_TRACEROUTE_IN_SRC_IPADDR[0]=$NETWORK1 INTERFACE1_TRACEROUTE_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR # **************************************************************************** # * # I C M P * # * # **************************************************************************** ACCEPT_ICMP="yes" # ---------------------------------------------------------------------------- # ICMP outgoing client request # # Interface 0 ICMP outgoing client request
- 301. GIPTables Firewall 1 CHAPTER0 301 INTERFACE0_ICMP_CLIENT="yes" INTERFACE0_ICMP_OUT_SRC_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE0_ICMP_OUT_DST_IPADDR[0]=$ANY_IPADDR # Network 1 ICMP forwarded outgoing client request NETWORK1_ICMP_CLIENT="yes" NETWORK1_ICMP_OUT_SRC_IPADDR[0]=$NETWORK1 NETWORK1_ICMP_OUT_DST_IPADDR[0]=$ANY_IPADDR # ---------------------------------------------------------------------------- # ICMP incoming client request # # Interface 1 ICMP incoming client request INTERFACE1_ICMP_SERVER="no" INTERFACE1_ICMP_IN_SRC_IPADDR[0]=$NETWORK1 INTERFACE1_ICMP_IN_DST_IPADDR[0]=$INTERFACE0_IPADDR INTERFACE1_ICMP_IN_SRC_IPADDR[1]=$NETWORK1 INTERFACE1_ICMP_IN_DST_IPADDR[1]=$INTERFACE1_IPADDR # **************************************************************************** # * # D H C P * # * # **************************************************************************** ACCEPT_DHCP="yes" # ---------------------------------------------------------------------------- # DHCP incoming client request # # Interface 1 DHCP incoming client request INTERFACE1_DHCP_SERVER="yes" # If above option is "yes", do not forget to configure the following # lines in the "Spoofing and bad addresses" section # REFUSE_SPOOFING_IPADDR[2]="0.0.0.0/8" # INTERFACE1_IN_REFUSE_SPOOFING[2]="no" INTERFACE1_DHCP_IN_SRC_IPADDR[0]=$NETWORK1 INTERFACE1_DHCP_IN_DST_IPADDR[0]=$INTERFACE1_IPADDR # **************************************************************************** # * # E N D * # * # **************************************************************************** DROP_EVERYTHING_FROM_HERE="yes" # ---------------------------------------------------------------------------- # LOG & DROP everything from here... just in case. #
- 302. GIPTables Firewall 1 CHAPTER0 302 INTERFACE0_IN_DROP_EVERYTHING_FROM_HERE="yes" INTERFACE1_IN_DROP_EVERYTHING_FROM_HERE="yes" NETWORK1_IN_DROP_EVERYTHING_FROM_HERE="yes" # ---------------------------------------------------------------------------- # End of file Step 4 Once the configuration file has been configured, it is time to start the firewall on your system. • To start the firewall on your system, use the following command: [root@deep /]# /etc/init.d/giptables start Starting Firewalling Services: [OK] GIPTables-Firewall Administrative Tools The commands listed below are some that we use often, but many more exist. Check the manual page and documentation for more information. IPTables The iptables tool is used for the firewall packet filter administration of the system. We can use it to set up a firewall rules file, as we are doing in this book. Once firewall rules have been created we can play with its many commands to maintain, and inspect the rules in the kernel. • To list all rules in the selected chain, use the command: [root@deep /]# iptables –L Chain INPUT (policy DROP) target prot opt source destination ACCEPT all -- anywhere anywhere Chain FORWARD (policy DROP) target prot opt source destination Chain OUTPUT (policy DROP) target prot opt source destination ACCEPT all -- anywhere anywhere This command will list all rules in the selected chain. If no chain is selected, all chains are listed. • To list all INPUT rules in the selected chain, use the command: [root@deep /]# iptables -L INPUT Chain INPUT (policy DROP) target prot opt source destination ACCEPT all -- anywhere anywhere ACCEPT all -- 192.168.1.0/24 anywhere DROP all -- 204.254.45.9 anywhere DROP all -- 187.231.11.5 anywhere DROP all -- 207.35.78.5 anywhere
- 303. GIPTables Firewall 1 CHAPTER0 303 • To list all OUTPUT rules in the selected chain, use the command: [root@deep /]# iptables -L OUTPUT Chain OUTPUT (policy DROP) target prot opt source destination ACCEPT all -- anywhere anywhere ACCEPT all -- anywhere 192.168.1.0/24 ACCEPT udp -- 207.35.78.5 207.35.78.3 udp spt:domain dpt:domain ACCEPT tcp -- 207.35.78.5 207.35.78.3 tcp spts:1024:65535 dpt:domain • To list all FORWARD rules in the selected chain, use the command: [root@deep /]# iptables -L FORWARD Chain FORWARD (policy DROP) target prot opt source destination DROP tcp -- anywhere anywhere tcp DROP tcp -- anywhere anywhere tcp DROP all -- !192.168.0.0/24 anywhere ACCEPT all -- 192.168.0.0/24 anywhere state NEW ACCEPT all -- !192.168.0.0/24 anywhere state This of course works only if you have configured Masquerading on your server (for Gateway servers in general). • To list all rules in numeric OUTPUT in the selected chain, use the command: [root@deep /]# iptables –nL Chain INPUT (policy DROP) target prot opt source destination ACCEPT all -- 0.0.0.0/0 0.0.0.0/0 ACCEPT all -- 192.168.1.0/24 0.0.0.0/0 DROP all -- 204.254.45.9 0.0.0.0/0 Chain FORWARD (policy DROP) target prot opt source destination Chain OUTPUT (policy DROP) target prot opt source destination ACCEPT all -- 0.0.0.0/0 0.0.0.0/0 ACCEPT all -- 0.0.0.0/0 192.168.1.0/24 ACCEPT udp -- 207.35.78.5 207.35.78.5 All the IP addresses and port numbers will be printed in numeric format.
- 305. Squid Proxy Server INTHIS CHAPTER 1. Compiling - Optimizing & Installing Squid 2. Configuring Squid 3. Running Squid with Users Authentication Support 4. Securing Squid 5. Optimizing Squid 6. Squid Administrative Tools 7. The cachemgr.cgi program utility of Squid
- 307. Squid 1 CHAPTER 1 307 LinuxSquid Abstract Another important program to consider is Squid especially for those of you that want to configure and run a Gateway Server for computers on your internal network and I know that you are numerous. Therefore, there is no doubt that for a Gateway Server set-up, IPTables, our Packet Filter software, and Squid, which will become our Application Gateway software, is required. In general, IPTables will protect our Gateway Server and Squid our private internal hosts. Do not install Squid on a Gateway Server without IPTables; both are very important and must be installed together if you want to have a secure Gateway system. IPTables is necessary to manage the legitimate opened ports on our server that Squid users will use to access the Internet or the network. Proxy-servers like Squid, with their capability to save bandwidth, improve security, and increase web-surfing speeds are becoming more popular than ever. Currently only a few proxy-server programs are on the market. These proxy-servers have two main drawbacks: they are commercial, and they don’t support ICP (ICP is used to exchange hints about the existence of URLs in neighbor caches). Squid is the best choice for a proxy-cache server since it is robust, free, and can use ICP features. Derived from the “cached” software from the ARPA-funded Harvest research project, developed at the National Laboratory for Applied Network Research and funded by the National Science Foundation, Squid offers high-performance caching of web clients, and also supports FTP, Gopher, HTTP and HTTPS data objects. It stores hot objects in RAM, maintains a robust database of objects on disk, has a complex access control mechanism (ACL), and supports the SSL protocol for proxying secure connections. In addition, it can be hierarchically linked to other Squid-based proxy servers for streamlined caching of pages through its unique ICP feature. In our compilation and configuration we’ll show you how to configure Squid depending on your needs. Two different set-ups are available. The first will be to configure it to run as an httpd-accelerator to get more performance out of our Web Server. In accelerator mode, the Squid server acts as a reverse proxy cache: it accepts client requests, serves them out of cache, if possible, or requests them from the original Web Server for which it is the reverse proxy. However, this set-up is not what we need for a Gateway Server. It is only useful on a Web Server where you want better performance. The second, the one suitable for our Gateway Server set-up will be to configure Squid as a proxy-caching server to be able to let all users on your corporate network use Squid to access the Internet. This is a very interesting addition when you run a Gateway Server your corporate network. A Gateway Server with IPTables as described earlier in this book plus a Squid Server mounted on it will highly improve the security and performance speed of the system. This is also the solution to control and restrict what can be viewed on the Internet. With a Squid Server configured as a proxy-caching server on a Gateway Server, you will be able to block for example porno sites, underground sites, warez (if you want ☺), etc. many different possibilities exist, like authorizing access to the Internet based on specific hours or days.
- 308. Squid 1 CHAPTER 1 308 Theseinstallation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest Squid version number is 2.4.STABLE6 Packages The following are based on information listed by Squid as of 2002/03/20. Please regularly check at www.squid-cache.org for the latest status. We chose to install the required component from source file because it provides the facility to fine tune the installation. Source code is available from: Squid Homepage: http://www.squid-cache.org/ Squid FTP Site: 206.168.0.9 You must be sure to download: squid-2.4.STABLE7-src.tar.gz Though the procedures given in this chapter are likely to work on all Linux platforms, we have only tested it on OpenNA Linux and Red Hat Linux. Pristine source If you don’t use the RPM package to install this program, it will be difficult for you to locate all files installed into the system if the package is updated in the future. To solve this problem, it is a good idea to make a list of files on the system before you install Squid, and one afterwards, and then compares them using the diff utility of Linux to find out what files are placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > Squid1 • And the following one after you install the software: [root@deep root]# find /* > Squid2 • Then use the following command to get a list of what changed: [root@deep root]# diff Squid1 Squid2 > Squid-Installed By doing this, if in the future any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. We use the /root directory of the system to store all generated list files.
- 309. Squid 1 CHAPTER 1 309 Compiling- Optimizing & Installing Squid Below are the steps that you must make to configure, compile and optimize the Squid server software before installing it on your system. First off, we install the program as user 'root' so as to avoid any authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp squid-version-src.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf squid-version-src.tar.gz Step 2 To avoid security risks, we must create a new user account called “squid” to be the owner of the Squid database cache files and daemon. • To create this special Squid user on OpenNA Linux, use the following command: [root@deep tmp]# groupadd -g 23 squid > /dev/null 2>&1 || : [root@deep tmp]# useradd -c "Squid Server" -d /var/spool/squid -g 23 -s /bin/false -u 23 squid > /dev/null 2>&1 || : • To create this special Squid user on Red Hat Linux, use the following command: [root@deep tmp]# useradd -c "Squid Server" -u 23 -s /bin/false -r -d /var/spool/squid squid 2>/dev/null || : The above command will create a null account, with no password, no valid shell, no files owned- nothing but a UID and a GID for the program. Remember that Squid daemon does not need to have a shell account on the server. Step 3 After that, move into the newly created Squid source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created Squid source directory use the command: [root@deep tmp]# cd squid-2.4.STABLE7/ Step 4 There are some source files to modify before going into the configuration and compilation of the program; the changes allow us to fix some problems and to configure the program for our PATH environment variable under Linux. • Edit the acl.c file (vi +651 src/acl.c) and change the line: *Top = splay_insert(xstrdup(t), *Top, aclDomainCompare); To read: *Top = splay_insert(xstrdup(t), *Top, aclHostDomainCompare);
- 310. Squid 1 CHAPTER 1 310 Thisfixes a small bug for our version of Linux. • Edit the Makefile.in file (vi +18 icons/Makefile.in) and change the line: DEFAULT_ICON_DIR = $(sysconfdir)/icons To read: DEFAULT_ICON_DIR = $(libexecdir)/icons We change the variable (sysconfdir) to become (libexecdir). With this modification, the /icons directory of Squid will be located under the /usr/lib/squid directory. • Edit the Makefile.in file (vi +40 src/Makefile.in) and change the lines: DEFAULT_CACHE_LOG = $(localstatedir)/logs/cache.log To read: DEFAULT_CACHE_LOG = $(localstatedir)/log/squid/cache.log DEFAULT_ACCESS_LOG = $(localstatedir)/logs/access.log To read: DEFAULT_ACCESS_LOG = $(localstatedir)/log/squid/access.log DEFAULT_STORE_LOG = $(localstatedir)/logs/store.log To read: DEFAULT_STORE_LOG = $(localstatedir)/log/squid/store.log DEFAULT_PID_FILE = $(localstatedir)/logs/squid.pid To read: DEFAULT_PID_FILE = $(localstatedir)/run/squid.pid DEFAULT_SWAP_DIR = $(localstatedir)/cache To read: DEFAULT_SWAP_DIR = $(localstatedir)/spool/squid DEFAULT_ICON_DIR = $(sysconfdir)/icons To read: DEFAULT_ICON_DIR = $(libexecdir)/icons
- 311. Squid 1 CHAPTER 1 311 Wechange the default location of “cache.log”, “access.log”, and “store.log” files to be located under the /var/log/squid directory. Then, we put the pid file of Squid under the /var/run directory, and finally, locate the /icons directory of Squid under /usr/lib/squid/icons with the variable (libexecdir) as shown above. One important note here is the location of the Squid cache directory. As we can see, we relocate it under the /var/spool/squid directory since the file system (/var/spool) should be on its own partition. This allows us to isolate this file system from the rest of our operating system and to eliminate possible buffer overflow attacks. Also having the directory where the Squid cache will reside on its own partition will allow us to improve performance by tuning parameters of this separate partition with Linux commands like ulimit, etc. Step 5 Once the modifications have been made to the related Squid source files, it is time configure and optimize Squid for our system. • To configure and optimize Squid use the following compilation lines: CFLAGS="-O2 -march=i686 -funroll-loops" ./configure --exec_prefix=/usr --bindir=/usr/sbin --libexecdir=/usr/lib/squid --localstatedir=/var --sysconfdir=/etc/squid --enable-dlmalloc --enable-gnuregex --enable-xmalloc-statistics --with-pthreads --enable-removal-policies="heap" --enable-storeio=diskd,ufs --enable-delay-pools --enable-cache-digests --enable-err-language=English --enable-poll --enable-linux-netfilter --enable-truncate This tells Squid to set itself up for this particular configuration setup with: - Link Squid with an external malloc library to improve its cache performance. - Compile Squid with the GNUregex feature enable. - Show malloc statistics in status page (cachemgr.cgi). - Use POSIX Threads to improve Squid performance on Linux. - Use the heap-replacement feature of Squid to have the choice of various cache replacement algorithms, instead of the standard LRU algorithm for better performance. - Build support for ufs & diskd I/O modules for better performance. - Use the delay pools feature of Squid to limit and control bandwidth usage for users. - Use Squid Cache Digests feature to improve client response time and network utilization. - Select which default language will be used and installed by Squid for Error pages report. - Enable poll() instead of select() since it’s preferred over select. - Enable transparent proxy support for Linux kernel 2.4. - Enable truncate to clean some performance improvements when removing cached files.
- 312. Squid 1 CHAPTER 1 312 Step6 Now, we must make a list of all existing files on the system before installing the software, and one afterwards, then compare them using the diff utility tool of Linux to find out what files are placed where and finally install the Squid Proxy Server: [root@deep squid-2.4.STABLE7]# make all [root@deep squid-2.4.STABLE7]# cd auth_modules [root@deep auth_modules]# cd NCSA [root@deep NCSA]# make [root@deep NCSA]# cd ../PAM [root@deep PAM]# make [root@deep PAM]# cd ../SMB [root@deep SMB]# make SAMBAPREFIX=/usr [root@deep SMB]# cd ../getpwnam [root@deep getpwnam]# make [root@deep getpwnam]# cd [root@deep root]# find /* > Squid1 [root@deep root]# cd /var/tmp/squid-2.4.STABLE7/ [root@deep squid-2.4.STABLE7]# make install [root@deep squid-2.4.STABLE7]# cd auth_modules [root@deep auth_modules]# install -m 4511 PAM/pam_auth /usr/lib/squid/ [root@deep auth_modules]# install -m 0511 NCSA/ncsa_auth /usr/lib/squid/ [root@deep auth_modules]# install -m 0511 SMB/smb_auth /usr/lib/squid/ [root@deep auth_modules]# install –m 0511 getpwnam/getpwnam_auth /usr/lib/squid/ [root@deep auth_modules]# mkdir -p /var/spool/squid [root@deep auth_modules]# mkdir -p /var/log/squid [root@deep auth_modules]# chown -R squid.squid /var/spool/squid/ [root@deep auth_modules]# chown -R squid.squid /var/log/squid/ [root@deep auth_modules]# chmod 0750 /var/spool/squid/ [root@deep auth_modules]# chmod 0750 /var/log/squid/ [root@deep auth_modules]# rm -rf /var/logs/ [root@deep auth_modules]# rm -f /usr/sbin/RunCache [root@deep auth_modules]# rm -f /usr/sbin/RunAccel [root@deep auth_modules]# strip /usr/sbin/squid [root@deep auth_modules]# strip /usr/sbin/client [root@deep auth_modules]# strip /usr/lib/squid/* [root@deep auth_modules]# /sbin/ldconfig [root@deep auth_modules]# cd [root@deep root]# find /* > Squid2 [root@deep root]# diff Squid1 Squid2 > Squid-Installed The make all command will compile all source files into executable binaries that can be installed, and make install will install the binaries and any supporting files into the appropriate locations. Pay special attention to the authenticator module directory of Squid, we move into this directory (auth_modules) and compile all authenticator modules that may be needed with Squid. Squid authenticator module is required when you want to authorize and authenticate users before allowing them an access to the Internet or the network. Different authenticator modules using different techniques are available with Squid. In our compilation, we compile Squid authenticator modules for PAM, NCSA, SMB, and getpwnam. You don’t need to compile all of them but only the one that you want to use or nothing if you are not intending to provide user authentication for Proxy access. The mkdir command will create two new directories named “squid” under /var/spool and /var/log directory.
- 313. Squid 1 CHAPTER 1 313 Therm command will remove the /var/logs directory since it has been created to handle the log files for Squid that we have relocated during compile time into the /var/log/squid directory. The chown command will change the owner of the /var/spool/squid and /var/log/squid directories to be owned by the user squid, and the chmod command will make the mode of both squid directories (0750/drwxr-x---) for security reasons. This means that only squid owner and group will be able to access these directories and others will not. Note that we remove the small scripts named “RunCache” and “RunAccel” which take care of starting Squid in either caching mode or accelerator mode, since we use a better script named “squid” located under /etc/init.d directory that takes advantage of Linux system V. Finally, the strip command will reduce the size of the specified binaries for optimum performance. Step 7 Once we’ve configured, optimized, compiled, and installed the Squid Proxy Server software, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • This is done by using the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf squid-version/ [root@deep tmp]# rm -f squid-version-src.tar.gz The rm command as used above will remove all the source files we have used to compile and install Squid. It will also remove the Squid compressed archive from the /var/tmp directory. Configuring Squid After Squid has been built and installed successfully on your system, your next step is to configure and customize all the required parameters in the different Squid configuration files. Parameters entered into the Squid configuration file (squid.conf) will decide how the Squid software will run on the server and in which mode (either httpd-accelerator mode or in proxy- caching mode). This shows us that the installation of Squid under Linux does not care and that only the configuration of the squid.conf file will decide whether Squid will run in httpd- accelerator or proxy-caching mode. /etc/squid/squid.conf: (The Squid Configuration File) /etc/sysconfig/squid: (The Squid System Configuration File) /etc/logrotate.d/squid: (The Squid Log Rotation File) /etc/init.d/squid: (The Squid Initialization File)
- 314. Squid 1 CHAPTER 1 314 RunningSquid in a httpd-accelerator mode The squid.conf file is used to set all the different options for your Squid proxy server. In the Squid configuration file, we’ll configure the /etc/squid/squid.conf file to be in httpd- accelerator mode. In this mode, if the Web Server runs on the same server where Squid is installed, you must set its daemon to run on port 81. With the Apache Web Server, you can do it by changing the line (Port 80) to (Port 81) in its httpd.conf file. If the Web Server runs on other servers on your network, like we do, you can keep the same port number (80) for Apache, since Squid will bind on a different IP number where port (80) is not already in use.
- 315. Squid 1 CHAPTER 1 315 /etc/squid/squid.conf:The Squid Configuration File The /etc/squid/squid.conf file is the main configuration file for squid. Though there are hundred of option tags in this file, you should only need to change a few options to get Squid up and running. The other options give you amazing flexibility, but you can learn about them once you have Squid running. The text in bold are the parts of the configuration file that must be customized and adjusted to meet our needs. This configuration is suitable when you want to run Squid in httpd-accelerator mode only. Please see later in this chapter for the configuration of Squid in proxy caching mode. • Edit the squid.conf file (vi /etc/squid/squid.conf) and add/change the following options. Below is what we recommend you: http_port 80 icp_port 0 hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin ? no_cache deny QUERY cache_mem 128 MB redirect_rewrites_host_header off cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF cache_dir diskd /var/spool/squid 1000 16 256 cache_store_log none emulate_httpd_log on acl all src 0.0.0.0/0.0.0.0 http_access allow all cache_mgr sysadmin@openna.com cache_effective_user squid cache_effective_group squid httpd_accel_host 207.35.78.3 httpd_accel_port 80 logfile_rotate 0 log_icp_queries off cachemgr_passwd my-secret-pass all buffered_logs on This tells the squid.conf file to set itself up for this particular configuration with: http_port 80 The option “http_port” specifies the port number where Squid will listen for HTTP client requests. If you set this option to port 80, the client will have the illusion of being connected to the Apache Web Server. Since we are running Squid in accelerator mode and our Web Server on other hosts, we must listen on port 80. icp_port 0 The option “icp_port” specifies the port number where Squid will send and receive ICP requests from neighbouring caches. We must set the value of this option to “0” to disable it, since we are configuring Squid to be in accelerator mode for the Web Server. The ICP feature is needed only in a multi-level cache environment with multiple siblings and parent caches (a feature that only Squid supports compared to other proxy servers on the market). Using ICP in an accelerator mode configuration would add unwanted overhead to Squid. This is an optimization feature. hierarchy_stoplist cgi-bin ? The option “hierarchy_stoplist cgi-bin ?” is used to not query neighbor cache for certain objects. The above line is recommended.
- 316. Squid 1 CHAPTER 1 316 aclQUERY urlpath_regex cgi-bin ? no_cache deny QUERY The options “acl QUERY urlpath_regex cgi-bin ?” and “no_cache deny QUERY” are used to force certain objects to never be cached, like files under “cgi-bin” directory. This is a security feature. cache_mem 128 MB The option “cache_mem” specifies the amount of memory (RAM) to be used for caching the so called: In-Transit objects, Hot Objects, Negative-Cached objects. It’s important to note that Squid can use much more memory than the value you specify in this parameter. For example, if you have 384 MB free for Squid, you must put 384/3 = 128 MB here. This is an optimization feature. redirect_rewrites_host_header off The option “redirect_rewrites_host_header”, if set to “off”, tells Squid to not rewrites any Host: header in redirected requests. It’s recommended to set this option to “off” if you are running Squid in httpd-accelerator mode. cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF The options “cache_replacement_policy” and “memory_replacement_policy” specify the cache policy Squid will use to determine which objects in the cache must be replaced when the proxy needs to free disk space and which objects are purged from memory when memory space is needed. In our configuration, we choose the GDSF (Greedy-Dual Size Frequency) policy as our default policy. See http://www.hpl.hp.com/techreports/1999/HPL-1999-69.html and http://fog.hpl.external.hp.com/techreports/98/HPL-98-173.html for more information. cache_dir diskd /var/spool/squid 1000 16 256 The option “cache_dir” specifies in order: which kind of storage system to use and in our case we choose to use the new DISKD storage format of Squid, the name of the cache directory (/var/spool/squid), the disk space in megabytes to use under this directory (1000 MB), the number of first-level subdirectories to be created under the cache directory (16), and the number of second-level subdirectories to be created under each first-level cache directory (256). In accelerator mode, this option is directly related to the size and number of files that you want to serve with your Apache web server. In our example, we suppose that the total size of your web directory will be 1000 MB. Don’t forget to change this value to fit the size of your web directory. cache_store_log none The option “cache_store_log” logs the activity of the storage manager to the specified file. It shows which objects are ejected from the Squid cache, which objects are saved and for how long. We can safely set this option to “none” to disable the feature because there are not really any utilities to analyze this data. emulate_httpd_log on The option “emulate_httpd_log” if set to “on” specifies that Squid should emulate the log file format of the Apache Web Server. This is very useful if you want to use a third party program like Webalizer to analyze and produce static report of the Squid Server. acl all src 0.0.0.0/0.0.0.0 http_access allow all The options “acl” and “http_access” specify and define an access control list to be applied on the Squid Proxy Server in httpd-accelerator mode. Our “acl” and “http_access” option are not restricted, and allows everyone to connect to the proxy server since we use this proxy to accelerate the public Apache Web Server. See your Squid documentation for more information when using Squid in non-httpd-accelerator mode.
- 317. Squid 1 CHAPTER 1 317 cache_mgrsysadmin@openna.com The option “cache_mgr” specifies the email-address of the administrator responsible for the Squid Proxy Server. This person is the one who will receive mail if Squid encounter problems. You can specify the name or the complete email address in this option. In our example, we specify the complete email address to be more verbose when errors are encounter. cache_effective_user squid cache_effective_group squid The options “cache_effective_user” and “cache_effective_group” specify the UID/GID that the cache will run on. Don’t forget to never run Squid as “root”. In our configuration we use the UID “squid” and the GID “squid” that we have created previously in this chapter. This is a security feature. httpd_accel_host 207.35.78.3 httpd_accel_port 80 The options “httpd_accel_host” and “httpd_accel_port” specify to Squid the IP address and port number where the real HTTP Server (i.e. Apache) resides. These are some of the most important parameters when configuring Squid to run in httpd-accelerator mode. In our configuration, the real HTTP Web Server is on IP address 207.35.78.3 (www.openna.com) and on port (80). “www.openna.com” is another FDQN on our network, and since the Squid Proxy Server doesn’t reside on the same host where our Apache HTTP Web Server runs, we can use port (80) for our Squid Proxy Server, and port (80) for our Apache Web Server, and the illusion is perfect. logfile_rotate 0 The option “logfile_rotate” specifies the number of logfile rotations that we want the Squid program to make. Setting the value to 0 will disable the default rotation and will let us control this feature through our personal logrotate script file. This is what we need to do on Linux since we use our own log script file to make the appropriate rotation of Squid log files. log_icp_queries off The option “log_icp_queries” specifies if you want ICP queries (remember, ICP is used to exchange hints about the existence of URLs in neighbor caches) to be logged to the “access.log” file or not. Since we don’t use the ICP feature of Squid in httpd-accelerator mode configuration, we can safely set this option to “off”. cachemgr_passwd my-secret-pass all The option “cachemgr_passwd” specifies a password that will be required for accessing the operations of the “cachemgr.cgi” utility program that comes with Squid. This CGI program is designed to run through a web interface and outputs statistics about the Squid configuration and performance. The <my-secret-pass> is the password that you have chosen, and the keyword <all> specifies to set this password to be the same for all actions you can perform with this program. See “The cachemgr.cgi program utility of Squid”, below in this chapter for more information. buffered_logs on The option “buffered_logs”, if turned “on”, can speed up the writing of some log files slightly. This is an optimization feature.
- 318. Squid 1 CHAPTER 1 318 RunningSquid in proxy-caching mode With some minor modifications to the squid.conf file we defined earlier to run in httpd- accelerator mode, we can run Squid as a proxy-caching server. With a proxy-caching server, all users in your corporate network will use Squid to access the Internet. This is the configuration that you must use for a Gateway Server running Squid and it is the most commonly used configuration by Linux administrators who install Squid on their servers. With this configuration, you can have complete control, apply special policies on what can be viewed, accessed, and downloaded. You can also control bandwidth usage, connection time, and so on. A proxy caching server can be configured to run as stand-alone server for your corporation, or to use and share caches hierarchically with other proxy servers around the Internet.
- 319. Squid 1 CHAPTER 1 319 /etc/squid/squid.conf:The Squid Configuration File To set up Squid as a proxy-caching server, we use the same configuration file as before but with some additional modifications to the default in relation to Squid in httpd-accelerator mode. The text in bold are the parts of the configuration file that must be customized and adjusted to satisfy our needs. The rest of the parameters are the same as for Squid in httpd-accelerator mode and I recommend you to read the configuration section related to Squid in httpd-accelerator mode for more information on each option. This configuration is suitable when you want to run Squid in proxy-caching mode only. Please see the information earlier in this chapter for the configuration of Squid in httpd-accelerator mode. • Edit the squid.conf file (vi /etc/squid/squid.conf) and add/change the following options for Squid in proxy caching mode that run as a stand-alone server. Below is what we recommend: icp_port 0 hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin ? no_cache deny QUERY cache_mem 128 MB cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF cache_dir diskd /var/spool/squid 2000 16 256 cache_store_log none acl localnet src 192.168.1.0/255.255.255.0 acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 210 70 21 1025-65535 acl CONNECT method CONNECT acl all src 0.0.0.0/0.0.0.0 http_access allow localnet http_access allow localhost http_access deny !Safe_ports http_access deny CONNECT http_access deny all cache_mgr sysadmin@openna.com cache_effective_user squid cache_effective_group squid logfile_rotate 0 log_icp_queries off cachemgr_passwd my-secret-pass all buffered_logs on NOTE: In the above configuration example, the default Proxy port ‘3128’ will be used. If you prefer to use another port like ‘8080’, all you will have to do will be to add the parameter “http_port 8080” and configure your clients accordingly. One of the big differences with the Squid httpd-accelerator mode configuration file is the use of Access Control Lists (ACL). For Squid in Proxy-Caching mode, this feature allows you to restrict access based on source IP address (src), destination IP address (dst), source domain, destination domain, time, and so on. Many types exist with this feature, and you should consult the “squid.conf” file for a complete list.
- 320. Squid 1 CHAPTER 1 320 Thefour most commonly used types are as follows: acl name type data | | | | acl some-name src a.b.c.d/e.f.g.h # ACL restrict access based on source IP address acl some-name dst a.b.c.d/e.f.g.h # ACL restrict access based on destination IP address acl some-name srcdomain foo.com # ACL restrict access based on source domain acl some-name dstdomain foo.com # ACL restrict access based on destination domain For example, to restrict access to your Squid proxy server to only your internal clients, and to a specific range of designated ports, something like the following will do the job: # Our ACL Elements acl localnet src 192.168.1.0/255.255.255.0 acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 210 70 21 1025-65535 acl CONNECT method CONNECT acl all src 0.0.0.0/0.0.0.0 # Our Access Lists http_access allow localnet http_access allow localhost http_access deny !Safe_ports http_access deny CONNECT http_access deny all Let’s explain what’s going on. First we can see that there are two distinct groups acl and http_access; all the ‘acl’ parts with their different types are called “ACL elements” and all the ‘http_access’ parts with their different types are called “Access Lists”. We use “ACL elements” to define our names, source IP addresses, destination IP addresses, source domain, destination domain, port, etc and “Access Lists” to define the action that must be associated with the “ACL elements”. The action can be to deny or allow the “ACL elements” rules. In our example above, we define five “ACL elements”: acl localnet src 192.168.1.0/255.255.255.0 acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 210 70 21 1025-65535 acl CONNECT method CONNECT acl all src 0.0.0.0/0.0.0.0 and five “Access Lists” pertaining to the “ACL elements”: http_access allow localnet http_access allow localhost http_access deny !Safe_ports http_access deny CONNECT http_access deny all
- 321. Squid 1 CHAPTER 1 321 TheSquid program reads the access lines in the order that there are appearing. Pertaining to our example, Squid will interpret all the access lines as follow: 1) Read acl localnet src 192.168.1.0/255.255.255.0 2) Read acl localhost src 127.0.0.1/255.255.255.255 3) Read acl Safe_ports port 80 443 210 70 21 1025-65535 4) Read acl CONNECT method CONNECT 5) Read acl all src 0.0.0.0/0.0.0.0 6) Apply http_access allow localnet 7) Apply http_access allow localhost 8) Apply http_access deny !Safe_ports 9) Apply http_access deny CONNECT 10) Apply http_access deny all This ACL configuration will allow all internal clients from the private class C 192.168.1.0 to access the proxy server; it’s also recommended that you allow the localhost IP (a special IP address used by your own server) to access the proxy. After we choose a range of ports (80=http, 443=https, 210=wais, 70=gopher, and 21=ftp) which our internal clients can use to access the Internet, we deny the CONNECT method to prevent outside people from trying to connect to the proxy server, and finally, we deny all source IP address and ports on the proxy server. Multi-level Web Caching The second method of proxy caching is the so-called “Multi-level Web Caching” where you choose to share and cooperate with more proxy-cache servers on the Internet. With this method, your organization uses the cache of many others proxy cache servers, and to compensate, the other cache server can use yours. It’s important to note that in this situation, the proxy cache can play two different roles in the hierarchy. It can be configured as a sibling cache, and be able to only serve documents it already has, or it can be configured as a parent cache, and be able to get documents from another cache or from the source directly.
- 322. Squid 1 CHAPTER 1 322 NOTE:A good strategy to avoid generating more network traffic than without web caching is to choose to have several sibling caches and only a small number of parent caches. /etc/sysconfig/squid: The Squid System Configuration File The /etc/sysconfig/squid file is used to specify Squid system configuration information, such as if Squid should enable initial DNS checks at start-up, and the value of time to wait for Squid to shut down when asked. • Create the squid file (touch /etc/sysconfig/squid) and add the following lines: # If you most likely will not to have an Internet connection when you # start Squid, uncomment this. The -D option disables initial dns checks. #SQUID_OPTS="-D" # Time to wait for Squid to shut down when asked. Should not be necessary # most of the time. SQUID_SHUTDOWN_TIMEOUT=100
- 323. Squid 1 CHAPTER 1 323 /etc/logrotate.d/squid:The Squid Log Rotation Configuration File The /etc/logrotate.d/squid file is responsible for rotating log files related to Squid software automatically each week via syslog. If you are not familiar with syslog, look at the syslog.conf (5) manual page for a description of the syslog configuration file, or the syslogd (8) manual page for a description of the syslogd daemon. • Create the squid file (touch /etc/logrotate.d/squid) and add the following lines: /var/log/squid/access.log { weekly rotate 5 copytruncate compress notifempty missingok } /var/log/squid/cache.log { weekly rotate 5 copytruncate compress notifempty missingok } /var/log/squid/store.log { weekly rotate 5 copytruncate compress notifempty missingok # This script asks Squid to rotate its logs on its own. Restarting Squid # is a long process and it is not worth doing it just to rotate logs. postrotate /usr/sbin/squid -k rotate endscript } /etc/init.d/squid: The Squid Initialization File The /etc/init.d/squid script file is responsible for automatically stopping and starting the Squid Internet Object Cache on your server. Loading the squid daemon, as a standalone daemon will eliminate load time and will even reduce swapping, since non-library code will be shared. Please note that the following script is suitable for Linux operating systems that use SystemV. If your system uses some other method like BSD, you’ll have to adjust the script below to make it work for you. Step 1 Create the squid script file (touch /etc/init.d/squid) and add the following lines: #!/bin/bash # This shell script takes care of starting and stopping Squid (Proxy server). # # chkconfig: 345 90 25 # description: Squid - Internet Object Cache. Internet object caching is # a way to store requested Internet objects (i.e., data available
- 324. Squid 1 CHAPTER 1 324 #via the HTTP, FTP, and gopher protocols) on a system closer to the # requesting site than to the source. Web browsers can then use the # local Squid cache as a proxy HTTP server, reducing access time as # well as bandwidth consumption. # # processname: squid # pidfile: /var/run/squid.pid # config: /etc/squid/squid.conf PATH=/usr/bin:/sbin:/bin:/usr/sbin export PATH # Source function library. . /etc/rc.d/init.d/functions # Source networking configuration. . /etc/sysconfig/network # Check that networking is up. [ ${NETWORKING} = "no" ] && exit 0 # Check if the squid.conf file is present. [ -f /etc/squid/squid.conf ] || exit 0 # Source Squid configureation. if [ -f /etc/sysconfig/squid ]; then . /etc/sysconfig/squid else SQUID_OPTS="-D" SQUID_SHUTDOWN_TIMEOUT=100 fi # Determine the name of the squid binary. [ -f /usr/sbin/squid ] && SQUID=squid [ -z "$SQUID" ] && exit 0 prog="$SQUID" # Determine which one is the cache_swap directory CACHE_SWAP=`sed -e 's/#.*//g' /etc/squid/squid.conf | grep cache_dir | awk '{ print $3 }'` [ -z "$CACHE_SWAP" ] && CACHE_SWAP=/var/spool/squid RETVAL=0 start() { for adir in $CACHE_SWAP; do if [ ! -d $adir/00 ]; then echo -n "init_cache_dir $adir... " $SQUID -z -F 2>/dev/null fi done echo -n $"Starting $prog: " $SQUID $SQUID_OPTS 2> /dev/null & # Trap and prevent certain signals from being sent to the Squid process. trap '' 1 2 3 18 RETVAL=$? [ $RETVAL -eq 0 ] && touch /var/lock/subsys/$SQUID [ $RETVAL -eq 0 ] && echo_success [ $RETVAL -ne 0 ] && echo_failure echo return $RETVAL
- 325. Squid 1 CHAPTER 1 325 } stop(){ echo -n $"Stopping $prog: " $SQUID -k check >/dev/null 2>&1 RETVAL=$? if [ $RETVAL -eq 0 ] ; then $SQUID -k shutdown & rm -f /var/lock/subsys/$SQUID timeout=0 while : ; do [ -f /var/run/squid.pid ] || break if [ $timeout -ge $SQUID_SHUTDOWN_TIMEOUT ]; then echo return 1 fi sleep 2 && echo -n "." timeout=$((timeout+2)) done echo_success echo else echo_failure echo fi return $RETVAL } reload() { $SQUID $SQUID_OPTS -k reconfigure } restart() { stop start } condrestart() { [ -e /var/lock/subsys/squid ] && restart || : } # See how we were called. case "$1" in start) start ;; stop) stop ;; reload) reload ;; restart) restart ;; condrestart) condrestart ;; *) echo $"Usage: $0 {start|stop|reload|restart|condrestart}" exit 1 esac exit $?
- 326. Squid 1 CHAPTER 1 326 Step2 Once the /etc/init.d/squid script file has been created, it is important to make it executable, change its default permissions, create the necessary links and start it. Making this file executable will allow the system to run it, changing its default permission is to allow only the root user to change this file for security reason, and creation of the symbolic links will let the process control initialization of Linux which is in charge of starting all the normal and authorized processes that need to run at boot time on your system to start the program automatically for you at each system reboot. • To make this script executable and to change its default permissions, use the commands: [root@deep /]# chmod 700 /etc/init.d/squid [root@deep /]# chown 0.0 /etc/init.d/squid • To create the symbolic rc.d links for Squid, use the following commands: [root@deep /]# chkconfig --add squid [root@deep /]# chkconfig --level 345 squid on • To start Squid software manually, use the following command: [root@deep /]# /etc/init.d/squid start Starting squid: [OK] Running Squid with Users Authentication Support Squid is a complete Proxy software solution that provides many useful features for the administrator and it is up to us to use and configure them. In this section, we’ll discuss the Proxy Authentication mechanism that provides a way to authenticate internal users who can use it to access the Internet. This allows us to add an additional layer of security on which users need to have rights (the authorization) to access the Internet via the Gateway server in the enterprise. The Squid source code comes with a few authentication processes. These include: LDAP: Uses the Lightweight Directory Access Protocol. NCSA: Uses an NCSA-style username and password file. MSNT: Uses a Windows NT authentication domain. PAM: Uses the Linux Pluggable Authentication Modules scheme. SMB: Uses a SMB server like Windows NT or Samba. getpwam: Uses the old-fashioned Unix password file. In order to authenticate users, you need to compile and install one of the above supplied authentication modules. In our compilation of Squid, we have already included the most interesting authentication modules, which were NCSA, PAM, SMB, and getpwam. One problem with all of these authentication modules is the fact that the supplied username and password are essentially sent in clear text between the browser and the proxy. Therefore, administrators should not set-up the same username and password that users would use for account login on the server (if they are allowed) or for email accounts.
- 327. Squid 1 CHAPTER 1 327 Thismeans that we have to create a null account, with no valid shell, no files owned-nothing but a UID and a GID for every user that will use the Squid Proxy Server, with authentication, to access the Internet. The best authentication module to accomplish this will be the PAM authentication module because it will allow us to manage proxy users’ authentication access through the /etc/passwd file in the easiest and fastest manner available. It would also allow us to create the null account without problem. Below, we will show you, how to use and configure the PAM authentication module with Squid. Step 1 The first step in our procedure will be to create a PAM configured authentication service called "squid" under the /etc/pam.d directory to allow us to authenticate Squid users. • Create the squid file (touch /etc/pam.d/squid) and add the following lines: #%PAM-1.0 auth required /lib/security/pam_stack.so service=system-auth account required /lib/security/pam_stack.so service=system-auth Step 2 Now, it is time to let Squid know which authentication program to use in squid.conf. In our case, we have to tell it to use the PAM authentication module. • Edit the squid.conf file (vi /etc/squid/squid.conf) and add the following line. Text in bold is what we have added to our default Squid example configuration file. Below is what we recommend you enter: icp_port 0 hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin ? no_cache deny QUERY cache_mem 128 MB cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF cache_dir diskd /var/spool/squid 2000 16 256 cache_store_log none authenticate_program /usr/lib/squid/pam_auth /etc/passwd acl localnet src 192.168.1.0/255.255.255.0 acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 210 70 21 1025-65535 acl CONNECT method CONNECT acl all src 0.0.0.0/0.0.0.0 http_access allow localnet http_access allow localhost http_access deny !Safe_ports http_access deny CONNECT http_access deny all cache_mgr sysadmin@openna.com cache_effective_user squid cache_effective_group squid logfile_rotate 0 log_icp_queries off cachemgr_passwd my-secret-pass all buffered_logs on In the above line, we specify the name of the program (pam_auth) to use for user authentication, plus any command line options if necessary (/etc/passwd).
- 328. Squid 1 CHAPTER 1 328 Step3 Next, we have to add some proxy_auth ACL entries to our squid.conf configuration file to control and authorize the access. • Edit the squid.conf file (vi /etc/squid/squid.conf) and add the following options to the squid.conf file to be able to authenticate and control users access. Again the text in bold is what we have added to the previous Squid example configuration file in step 2. Below is what we recommend you enter: icp_port 0 hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin ? no_cache deny QUERY cache_mem 128 MB cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF cache_dir diskd /var/spool/squid 2000 16 256 cache_store_log none authenticate_program /usr/lib/squid/pam_auth /etc/passwd acl authenticated proxy_auth REQUIRED acl localnet src 192.168.1.0/255.255.255.0 acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 210 70 21 1025-65535 acl CONNECT method CONNECT acl all src 0.0.0.0/0.0.0.0 http_access allow authenticated http_access allow localnet http_access allow localhost http_access deny !Safe_ports http_access deny CONNECT http_access deny all cache_mgr sysadmin@openna.com cache_effective_user squid cache_effective_group squid logfile_rotate 0 log_icp_queries off cachemgr_passwd my-secret-pass all buffered_logs on The added lines mean that any authenticated user will match the ACL named "authenticated". The string REQUIRED is used to accept any valid username. NOTE: Don’t forget to restart your Squid Proxy Server for the changes to take effect. The order in which each line appears in the Squid configuration file is important and you have to respect them. You can’t just add ‘acl’ or ‘http_access’ parameters, wherever you want. Because the program reads and interprets each access line in the order that they appear. The above configurations CAN’T be used in conjunction with the ACL configuration for Banning all Destination addresses except one (see further down in this chapter for more information).
- 329. Squid 1 CHAPTER 1 329 Step4 One of the last steps is to create accounts for all users who will be allowed to access the Internet with Squid after proper authentication with a username and password. Remember, we have to create null account, with no valid shell for our users. • To create null user account, use the following command: [root@deep /]# useradd -s /bin/false gmourani The above command will create a null account, with no password, no valid shell, no files owned- nothing but a UID and a GID. • To set a password for this new user, use the following command: [root@deep /]# passwd gmourani Changing password for user gmourani New UNIX password: Retype new UNIX password: passwd: all authentication tokens updated successfully NOTE: It is NOT important to create a home directory for the users (i.e. /home/gmourani). Squid with Users Authentication Support can run even if home directories are not created for the users. All we need for authentication is a username and password. Therefore, a home directory is futile and since we do not give shell access, there is really no reason for users to have a home directory on the system. Step 5 Finally, open your favorite browser and enter the username and password to access the Internet with Squid as your Proxy Caching Gateway Server.
- 330. Squid 1 CHAPTER 1 330 SecuringSquid This section deals specifically with actions we can take to improve and tighten security under Squid. As with the other chapters, the interesting points here are that we refer to the features available within the base installed program and not any additional software. More control on mounting the cache directory of Squid If you have created the cache directory of Squid in a separate partition your system (i.e. /var/spool), like we have done during the initial set-up of Linux, then you can use the noexec, nodev, and nosuid features of Linux to improve and consolidate the cache security. These features can be set up in the /etc/fstab file to inform the system to not allow execution of any binaries (noexec), to not interpret character or block special devices (nodev), and to not allow set-user-identifier or set-group-identifier bits to take effect (nosuid) on the mounted file system (/var/spool in our example). Applying this procedure on the partition where the Squid Cache resides will help to eliminate the possibility of DEV, SUID/SGID, and execution of any binaries that may be in the Squid cache. Step 1 • Edit the fstab file (vi /etc/fstab) and add in the line that refer to /var/spool file system the following options after the defaults option as show below: LABEL=/var/spool /var/spool ext3 defaults,noexec,nodev,nosuid 1 2 Step 2 Once you have made the necessary adjustments to the /etc/fstab file, it is time to inform the system about the modification. • This can be accomplished with the following commands: [root@deep /]# mount /var/spool -oremount Each file system that has been modified must be remounted with the command as shown previously. In our example we have modified the /var/spool file system and it is for this reason that we remount this file system with the above command. NOTE: If /var/spool is not a file system but just a directory, then the above command obviously will not work. The ‘-oremount’ option of the Linux ‘mount’ command is used to remount a file system, which resides on its own partition on your computer.
- 331. Squid 1 CHAPTER 1 331 Step3 • You can verify if the modifications have been correctly applied to the Linux system with the following command: [root@deep /]# cat /proc/mounts /dev/root / ext2 rw 0 0 /proc /proc proc rw 0 0 /dev/sda1 /boot ext3 rw 0 0 /dev/sda9 /chroot ext3 rw 0 0 /dev/sda8 /home ext3 rw 0 0 /dev/sda13 /tmp ext3 rw 0 0 /dev/sda7 /usr ext3 rw 0 0 /dev/sda11 /var ext3 rw 0 0 /dev/sda12 /var/spool ext3 rw,noexec,nodev,nosuid 0 0 none /dev/pts devpts rw 0 0 This command will show you all file system in your Linux server with parameters applied to them. If you see something like the following, congratulations! /var/spool /var/spool ext3 rw,noexec,nodev,nosuid 0 0 Immunize the Squid configuration file As we already know, the immutable bit can be used to prevent deletion, overwriting, or creation of a symbolic link to a file. Once your squid.conf file has been configured, it’s a good idea to immunize it with the following command: [root@deep /]# chattr +i /etc/squid/squid.conf Banning all Destination addresses except one In the university libraries, we can often see computers available to students that want to, for their studies, search for a specific author. Administrators have the task of configuring the computer to allow only searches on one site where the database of all books and authors are stored. Therefore, they don’t want to give students access to other sites on the Internet but just to the database site. With Squid as the Proxy Server, this can be accomplished easily by adding the right ACL to its existing configuration file. In the next example, we introduce new ACL rules to our Squid example configuration file to do just this. • Edit the squid.conf file (vi /etc/squid/squid.conf) and add/change the following options. Text in bold are what we have added/changed to our default Squid example configuration file. Below is what we recommend you enter: icp_port 0 hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin ? no_cache deny QUERY cache_mem 128 MB cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF cache_dir diskd /var/spool/squid 2000 16 256 cache_store_log none acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 1025-65535 acl CONNECT method CONNECT acl DATABASE dst 207.78.0.1
- 332. Squid 1 CHAPTER 1 332 aclINTERNET dst 0.0.0.0/0.0.0.0 acl all src 0.0.0.0/0.0.0.0 http_access allow localhost http_access allow DATABASE http_access deny !Safe_ports http_access deny CONNECT http_access deny INTERNET http_access deny all cache_mgr sysadmin@openna.com cache_effective_user squid cache_effective_group squid logfile_rotate 0 log_icp_queries off cachemgr_passwd my-secret-pass all buffered_logs on This new ACL configuration allows the localhost and any internal clients to access the Proxy Server on the standard ports HTTP, HTTPS and all non-privileged ports, only when they want to connect to the destination IP address (207.78.0.1), which runs our database site. In this way, we limit web access to only one site and students cannot access the Internet. NOTE: Don’t forget to restart your Squid Proxy Server for the changes to take effect. The order in which each line appears in the Squid configuration file is important and you have to respect them. You can’t just add ‘acl’ or ‘http_access’ parameters, wherever you want. The program reads and interprets each access line in the order that they appear. The above configurations CAN’T be used in conjunction with the ACL configuration for Users Authentication Support (see further up in this chapter for more information). Allowing access to the Internet at specific times Let's say you want all of your internal hosts only be allowed access to the Internet during working hours (8:30 - 17:30). You can use something like this. • Edit the squid.conf file (vi /etc/squid/squid.conf) and add the following options. Text in bold is what we have added to our default Squid example configuration file: icp_port 0 hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin ? no_cache deny QUERY cache_mem 128 MB cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF cache_dir diskd /var/spool/squid 2000 16 256 cache_store_log none acl staff src 192.168.1.0/255.255.255.0 acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 210 70 21 1025-65535 acl CONNECT method CONNECT acl WORKING time MTWHF 08:30-17:30 acl all src 0.0.0.0/0.0.0.0 http_access allow staff WORKING http_access allow localhost http_access deny !Safe_ports http_access deny CONNECT http_access deny staff
- 333. Squid 1 CHAPTER 1 333 http_accessdeny all cache_mgr sysadmin@openna.com cache_effective_user squid cache_effective_group squid logfile_rotate 0 log_icp_queries off cachemgr_passwd my-secret-pass all buffered_logs on This new ACL configuration allows all internal clients from the private class C 192.168.1.0 to access the Internet between 08:30 and 17:30. In this way, we limit the time when our staff can connect to the Internet to the working hours of the company only. NOTE: Don’t forget to restart your Squid Proxy Server for the changes to take effect. The order in which each line appears in the Squid configuration file is important and you have to respect them. You can’t just add ‘acl’ or ‘http_access’ parameters, wherever you want. Because the program reads and interprets each access line in the order that they appear. Optimizing Squid This section deals specifically with the actions we can take to improve and tighten the performance of Squid. Note that we refer to the features available within the base installed program only. The atime and noatime attributes The atime and noatime attributes of Linux can be used to get a measurable performance gain in the Squid cache directory (/var/spool/squid). See the chapter related to the kernel in this book for more information on this issue. Physical memory The most important resource for Squid is physical memory. Your processor does not need to be ultra-fast. Your disk system will be the major bottleneck, so fast disks are also important for high- volume caches. Therefore, our recommendation is to use a SCSI disk with at least 512 MB of physical memory. Squid Administrative Tools Now you’ve configured Squid, tightened its security and optimized it for maximum performance, we can start to play with its utilities. Stopping Squid process immediately There are some interesting command line options, especially when we want to stop Squid on the server. Unlike other services that run as daemons on the system, Squid cannot be stopped directly and we have to wait for existing connections to terminate before Squid shutdowns. Sometimes this is not appropriate and we have to stop Squid immediately. This is possible with the following command: • To stop Squid process immediately, use the following command: [root@deep /]# /usr/sbin/squid -k kill This command sends a KILL signal, which causes the Squid process to exit immediately, without closing any connections or log files.
- 334. Squid 1 CHAPTER 1 334 Purgingobject from your cache Sometimes when information stored in the Squid cache become inaccurate for some reason or when there’s a problem relating to the cached files, it is nice to have an option which purges the cache. We cannot just delete the cache directories and then expect that everything will be fine. There is a command to do it and we must use it. Step 1 By default, Squid does not allow you to purge objects unless it is configured with access controls in squid.conf. Below, we’ll show you the procedure to accomplish this action. • Edit the squid.conf file (vi /etc/squid/squid.conf) and add the following options to the squid.conf file so we can purge objects. The text in bold are what we have added to our default Squid example configuration file. Below is what we recommend you put in your file: icp_port 0 hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin ? no_cache deny QUERY cache_mem 128 MB cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF cache_dir diskd /var/spool/squid 2000 16 256 cache_store_log none acl localnet src 192.168.1.0/255.255.255.0 acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 210 70 21 1025-65535 acl CONNECT method CONNECT acl PURGE method PURGE acl all src 0.0.0.0/0.0.0.0 http_access allow localnet http_access allow localhost http_access allow PURGE localhost http_access deny !Safe_ports http_access deny CONNECT http_access deny PURGE http_access deny all cache_mgr sysadmin@openna.com cache_effective_user squid cache_effective_group squid logfile_rotate 0 log_icp_queries off cachemgr_passwd my-secret-pass all buffered_logs on This new ACL configuration allows only purge requests of the cache if the request is made from the localhost (on the terminal of your Gateway Server), and denies all other purge requests. NOTE: Don’t forget to restart your Squid Proxy Server for the changes to take effect. The order in which each line appears in the Squid configuration file is important and you have to respect that. You can’t just add ‘acl’ or ‘http_access’ parameters, wherever you want. Because the program reads and interprets each access line in the order that they appears.
- 335. Squid 1 CHAPTER 1 335 Step2 Once the correct ACL have been added to your squid.conf file to allow purge requests on the Proxy Server, we have to use the client program that comes with Squid to purge an object. • This procedure can be accomplished with the following commands: [root@deep /]# client -m PURGE http://www.mydomain.com/ Where <www.mydomain.com> is the object that you want to purge. If the purge was successful, you will see a “200 OK” response. If the object was not found in the cache, you will see a "404 Not Found" response. NOTE: The PURGE feature of Squid works only when Users Authentication Support is disabled in the Squid configuration file. The client program of Squid is not capable of using User Authentication because it doesn’t have the option to specify a username or password through its command line. The cachemgr.cgi program utility of Squid The cachemgr.cgi utility program, which is available by default when you compile and install Squid on your system, is designed to run through a web interface, and outputs various statistics about Squid configuration and performance. Personally, I don’t recommend you use it. The cachemgr.cgi is a buggy utility, which provides incomprehensible and cryptic results. Connection to its web interface is not always guaranteed even if you have the proper configuration. I think that more development and a complete revision of its functionality is required. Especially when we want to make a remote connection to its web interface. If you really want to use it, then here are the correct steps you must follow. This program, by default, is located under the /usr/lib/squid directory, and you have to put it in your “cgi-bin” directory (eg, /home/httpd/cgi-bin) to be able to use it. Follow the simple steps below to use this program. Step 1 The first step will be to move the “cachemgr.cgi” CGI file from the /usr/lib/squid directory to your /home/httpd/cgi-bin directory. • This procedure can be accomplished with the following command: [root@deep /]# mv /usr/lib/squid/cachemgr.cgi /home/httpd/cgi-bin/ Step 2 Once you’ve put the “cachemgr.cgi” program into your /cgi-bin directory, it is time to change its default mode permission and owner. • These procedures can be accomplished with the following commands: [root@deep /]# cd /home/httpd/cgi-bin/ [root@deep cgi-bin]# chown 0.0 cachemgr.cgi [root@deep cgi-bin]# chmod 0511 cachemgr.cgi
- 336. Squid 1 CHAPTER 1 336 Step3 Finally, you can point your web browser to the following address (http://my-web-server/cgi- bin/cachemgr.cgi) to use the various features of this program. The <my-web-server> is the address where your Apache web server lives, and <cachemgr.cgi> is the Squid utility program we have just placed in our “cgi-bin” directory to display information and the configuration of our Squid Proxy Linux server. If you have configured the squid.conf file to use password authentication for cachemgr.cgi (as we do), you‘ll be asked to enter the “Cache Host”, “Cache Port”, “Manager Name”, and “Password information” before you are able to access the cachemgr.cgi program. See the configuration of the /etc/squid/squid.conf file, shown earlier, for more information. WARNING: Please note that only a browser running on the Squid machine (the Gateway Server) that doesn’t use the proxy will be able to connect to the cachemgr.cgi web interface. If you try to access the web interface remotely via another system, then the authentication will fail.
- 337. SquidGuard Filter IN THISCHAPTER 1. Compiling - Optimizing & Installing SquidGuard 2. Configuring SquidGuard 3. Testing SquidGuard 4. Optimizing SquidGuard
- 339. SquidGuard 1 CHAPTER 2 339 LinuxSquidGuard Abstract As we saw in the previous chapter, the Squid ACL (Access Control Lists) has some limitations in its functionality and it can become very hard to configure a complex ACL. We need to find another way to simplify the procedure of configuring our ACL and this is possible with plug-in software called SquidGuard. SquidGuard is a combined filter, redirector and access controller plug-in for Squid. It allows us to improve, in many ways, the default ACL of Squid. We can use it to limit web access for users to a list of accepted/well known web servers and/or URLs like Squid does already but in an easier manner. We can use it to block access to particular listed or blacklisted web servers and/or URLs, block access to URLs matching a list of regular expressions or words, redirect blocked URLs to an "intelligent" CGI based info page, have different access rules based on time of day, day of the week, date etc, and much more. In general it is a good addition to run with Squid Proxy Server on your Gateway Server for additional security and power. In this chapter, we will show you how to install and configure it to block undesirable websites like porn sites, warez, etc and how to configure it to allow Internet access on specific days and times from our corporate network. We will also merge it with the Squid default ACL to get maximum efficiency and security. Thousands, even millions, of IP addresses, and URL’s can be added to different filters files without sacrificing too much performance of the Squid Proxy Server. This is possible since SquidGuard uses good programming techniques to achieve this, and it is far ahead of its competitors for speed. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest SquidGuard version number is 1.2.0 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages The following is based on information listed by SquidGuard as of 2001/12/18. Please check http://www.squidguard.org/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: SquidGuard Homepage: http://www.squidguard.org/ SquidGuard FTP Site: 195.70.164.135 You must be sure to download: squidGuard-1.2.0.tar.gz
- 340. SquidGuard 1 CHAPTER 2 340 Pristinesource If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install SquidGuard, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > SquidGuard1 • And the following one after you install the software: [root@deep root]# find /* > SquidGuard2 • Then use this command to get a list of what changed: [root@deep root]# diff SquidGuard1 SquidGuard2 > SquidGuard-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In our example above, we use the /root directory of the system to store all the generated file lists. Compiling - Optimizing & Installing SquidGuard Below are the steps that you must take to configure, compile and optimize the SquidGuard software before installing it onto your system. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • These procedures can be accomplished with the following commands: [root@deep /]# cp squidGuard-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf squidGuard-version.tar.gz Step 2 After that, move into the newly created SquidGuard source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created SquidGuard source directory use the command: [root@deep tmp]# cd squidGuard-1.2.0/ • To configure and optimize SquidGuard use the following compilation lines: CFLAGS="-O2 -march=i686 -funroll-loops" ./configure --prefix=/usr --sysconfdir=/etc --localstatedir=/var --with-sg-config=/etc/squid/squidGuard.conf --with-sg-logdir=/var/log/squid/squidGuard --with-sg-dbhome=/var/spool/squid/squidGuard --with-db-inc=/usr/include --with-db-lib=/usr/lib
- 341. SquidGuard 1 CHAPTER 2 341 Thistells SquidGuard to set itself up for this particular configuration setup with: - The location of where the squidGuard configuration file must be installed. - The location of where the squidGuard log file must be installed. - The location of where the squidGuard database directory must be installed. - The location of the Berkley DB includes files that SquidGuard need. - The location of the Berkley DB library that SquidGuard need. Step 3 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the SquidGuard software: [root@deep squidGuard-1.2.0]# make [root@deep squidGuard-1.2.0]# cd [root@deep root]# find /* > SquidGuard1 [root@deep root]# cd /var/tmp/squidGuard-1.2.0/ [root@deep squidGuard-1.2.0]# make install [root@deep squidGuard-1.2.0]# cd samples/ [root@deep samples]# install –m 511 squidGuard.cgi /home/httpd/cgi-bin/ [root@deep samples]# cd dest/ [root@deep dest]# mkdir -p /var/spool/squid/squidGuard [root@deep dest]# chown -R squid.squid /var/spool/squid/squidGuard/ [root@deep dest]# chmod 0750 /var/spool/squid/squidGuard/ [root@deep dest]# chown -R squid.squid /var/log/squid/squidGuard/ [root@deep dest]# chmod 0750 /var/log/squid/squidGuard/ [root@deep dest]# cp blacklists.tar.gz /var/spool/squid/squidGuard/ [root@deep dest]# cd /var/spool/squid/squidGuard/ [root@deep squidGuard]# mkdir -p aggressive [root@deep squidGuard]# mkdir -p gambling [root@deep squidGuard]# mkdir -p hacking [root@deep squidGuard]# mkdir -p porn [root@deep squidGuard]# chown -R squid.squid aggressive/ [root@deep squidGuard]# chown -R squid.squid gambling/ [root@deep squidGuard]# chown -R squid.squid hacking/ [root@deep squidGuard]# chown -R squid.squid porn/ [root@deep squidGuard]# tar xzpf blacklists.tar.gz [root@deep squidGuard]# cd blacklists [root@deep blacklists]# install -m 644 aggressive/domains ../aggressive/ [root@deep blacklists]# install -m 644 aggressive/urls ../aggressive/ [root@deep blacklists]# install -m 644 gambling/domains ../gambling/ [root@deep blacklists]# install -m 644 gambling/urls ../gambling/ [root@deep blacklists]# install -m 644 hacking/domains ../hacking/ [root@deep blacklists]# install -m 644 hacking/urls ../hacking/ [root@deep blacklists]# install -m 644 porn/domains ../porn/ [root@deep blacklists]# install -m 644 porn/urls ../porn/ [root@deep blacklists]# install -m 644 porn/expressions ../porn/ [root@deep blacklists]# cd .. [root@deep squidGuard]# chown -R squid.squid * [root@deep squidGuard]# strip /usr/bin/squidGuard [root@deep squidGuard]# /sbin/ldconfig [root@deep squidGuard]# rm -rf blacklists blacklists.tar.gz [root@deep squidGuard]# cd [root@deep root]# find /* > SquidGuard2 [root@deep root]# diff SquidGuard1 SquidGuard2 > SquidGuard-Installed
- 342. SquidGuard 1 CHAPTER 2 342 Themake command will compile all source files into executable binaries that can be installed, and make install will install the binaries and any supporting files into the appropriate locations. The “install -m 0511” command will install the CGI program called squidGuard.cgi (squidGuard.cgi is a small script, which is used to explain to the user that the URL is blocked and by which rule set) into your cgi-bin directory. The “mkdir -p” command will create the SquidGuard directory and subdirectories to store database filter files to run with squidGuard, the “chown and chmod” commands will set the appropriate mode and ownership permissions to the squidGuard directory and it’s subdirectories. The “tar” command will untar the blacklists.tar.gz compressed archive containing all the filter files and the “install -m 644” commands will install the entire filter files to their appropriate directories. Finally, the strip command will reduce the size of the specified binaries for optimum performance and the “rm -rf” commands will remove the blacklists directory and archive file that we no longer need on our system. Step 4 Once the configuration, optimization, compilation, and installation of the SquidGuard software have been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete SquidGuard and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf squidGuard-version/ [root@deep tmp]# rm -f squidGuard-version.tar.gz The rm command as used above will remove all the source files we have used to compile and install SquidGuard. It will also remove the SquidGuard compressed archive from the /var/tmp directory. Configuring SquidGuard After SquidGuard has been built and installed successfully on your system, your next step is to configure and customize the squidGuard.conf file to suit your needs. The parameters entered into the squidGuard configuration file (squidGuard.conf) will decide how the ACL should be applied and on which users, hosts, IP addresses, times, dates, destination, source, etc. /etc/squid/squidGuard.conf (The SquidGuard Configuration File) /home/httpd/cgi-bin/squidGuard.cgi (The squidGuard.cgi File) /etc/squid/squidGuard.conf: The SquidGuard Configuration File The /etc/squid/squidGuard.conf file is the main and only configuration file for squidGuard. It is not difficult to configure or understand but we have to understand the rules order if we want to have a working SquidGuard configuration file. The SquidGuard configuration file (squidGuard.conf) has a structure that must be followed during its configuration. Next we show you the recommended structure for the configuration file and the order in which declarations are supposed to appear. There are six different possible declarations where five are optional and one required.
- 343. SquidGuard 1 CHAPTER 2 343 1.Path declarations (i.e. logdir and dbhome) (optional) 2. Time space declarations (i.e. time zones) (optional) 3. Source group declarations (i.e. clients) (optional) 4. Destination group declarations (i.e. URLs) (optional) 5. Rewrite rule group declarations (optional) 6. Access control rule declaration (required) The “Path declarations (1)” is used to define the location of the SquidGuard logfiles directory and to define the base for relative list filenames, also known as filter files. This declaration is optional but recommended for clarity. dbhome /var/spool/squid/squidGuard logdir /var/log/squid/squidGuard In the above declaration we specify the database directory from where all list filenames/filter files and their possible subdirectories, which are used to handle source and/or destination group information, should be located (dbhome /var/spool/squid/squidGuard). In the second option, “logdir /var/log/squid/squidGuard”, we specify the directory from where the SquidGuard log files are stored. With the “Path declaration” lines, we can ensure that SquidGuard will find the both directories when it runs. The “Time space declarations (2)” is used to define time or date rules that can be used in our ACL to limit Internet access times based on this declaration. The “Time space declarations” is optional and should be defined only if you think that you’ll need to restrict Internet access based on time. In most enterprises and universities, this feature is useful to control Internet access to working hours. time workhours { weekly mtwhf 08:30 - 16:30 } In the above declaration we define a range of days and hours that we will later use in our configuration to limit employee access to the Internet. This is based on the days and hours defined in the time space declaration above. Many different specifications and combinations exist. In our example, we limit connection to days of the week (weekly mtwhf) between 08:30 AM to 16:30 PM (08:30 - 16:30) for everyone who is a member of the “workhours” name. Individual IP address, or an IP addresses range can also be put into the “workhours” name. We begin our definition with a reserved word called "time" that the software recognizes as the time declaration, we give this declaration a name of our choice, “workhours”, we then add another reserved word called “weekly”, which allows us to enter day parameters (mtwhf) for m=mon, t =tue, w=wed, h=thu, f=fri, and finally include the time constraint (08:30 - 16:30) for each day. NOTE: The numeric time formats are important. For example, if you want to define 8:30, you must use 08:30 and not 8:30 for HH:MM.
- 344. SquidGuard 1 CHAPTER 2 344 The“Source group declarations (3)” is used to define the source on which our rules and ACL will be applied. This declaration is again optional but used when we want to define a different source for our network. src internal { ip 192.168.1.0/24 } In the above declaration we define with an IP address and net prefix (192.168.1.0/24) what our source network is and where it comes from (here, they come from our internal network). We start our definition with a reserved word called "src" that the software recognizes as a source declaration, again we give this declaration a name of our choice “internal”, and we add another reserved word called “ip”, which allows us to specify the origin as an IP address. In our case the IP address is defined as an IP/net prefix. NOTE: Source group declarations are matched in the order they are defined. If you have defined only one source group (as we do in our example), then there is no problem, but if you have more than one source group declaration, you must consider the order they are defined”. The “Destination group declarations (4)” is used to define the destination on which the rules and ACL will be applied. This declaration is another optional declaration and is used to control what can be viewed on the Internet. It is in this declaration that we can associate with our ACL the filters file containing the IP addresses and/or domain names that must be blocked depending on their contents. dest aggressive { domainlist aggressive/domains urllist aggressive/urls } dest gambling { domainlist gambling/domains urllist gambling/urls } dest hacking { domainlist hacking/domains urllist hacking/urls } dest porn { domainlist porn/domains urllist porn/urls expressionlist porn/expressions } dest warez { domainlist warez/domains urllist warez/urls } The above declarations are not difficult to understand. We can observe that we have five different destination groups defined. The specifications are the same only paths and filter names change.
- 345. SquidGuard 1 CHAPTER 2 345 Let’slook at these in more detail. The reserved word called "dest" starts each of our groups and the software recognizes it as a destination declaration, we give this declaration a name of our choice, in this example it’s called “agressive”, and two other reserved words “domainlist” and “urllist”. The program interprets the “domainlist” specification as a path pointing to a file called “domains” located under the “/var/spool/squid/squidGuard/aggressive” directory, which contains all the domain names that must be blocked to users. The program also interprets the “urllist” specification as the path pointing to a file called “urls” which is located under the “/var/spool/squid/squidGuard/aggressive” directory, which contains all the URL’s that must be blocked and not accessible to the users. In the above example, another specification exists, which is “expressionlist” that lets us specify via the “/var/spool/squid/squidGuard/porn/expressions” file, a list of regular expressions to use in the scan for blocked sites. WARNING: As with the previous groups, declarations are matched in the order they are listed in the “pass” declaration. If you have defined only one destination group, then there is no problem, but if you have more than one destination group declaration, you must consider the order in which they will be listed during the configuration of your “Access control rule declaration”. Regular expressions can produce bogus result in a search; it is up to you to decide if you really want to use regular expressions via the “expressionlist” file to block sites. The “Rewrite rule group declarations (5)” is a special declaration option of SquidGuard that can be used to defined, for example via regular expressions, redirection to local copies within peek business hours of the most popular programs on the Internet. This declaration is optional and should be used with care since it can quite easily slow down SquidGuard on busy systems or produce bogus information. In our configuration, we don’t use it. The “Access control rule declaration (6)” is used to combine all of the previous declarations into distinct rulesets for each clientgroup. This is the place in our SquidGuard configuration, where our policies and ACL will take effect once properly defined. acl { internal within workhours { pass !aggressive !gambling !hacking !porn !warez all } default { pass none redirect http://gtw.openna.com/cgi- bin/squidGuard.cgi?clientaddr=%a&clientname=%n&clientuser=%i&clientgroup= %s&targetgroup=%t&url=%u } } In the above declaration, we inform the system what we want it to do when users try to connect to the Internet through the proxy server. This is our Access Control Lists declaration. As with any of the previous declarations, we can see that the definition begins with a reserved word.
- 346. SquidGuard 1 CHAPTER 2 346 Therefore,we begin our definition with the reserved word called "acl" that the software recognizes as the beginning of our ACL definition. Next, we inform the program that this ACL applies to the source group called “internal”, that we defined previously. We also inform it that this ACL applies within the company working hours we defined in the time space declaration section of the configuration. We use the reserved word called “pass” to instruct it to allow users to view all Internet sites except ones in the blacklisted files “domains, urls, or expressions”. In other words, the “pass” rules declares destination groups that should pass for the actual client group. The "!" sign is the NOT operator and indicates a destination group that should not pass. It is good practice to always end the “pass” rule(s) with either "all" or "none" to make it/them clear. We have another important section into our declaration called “default”. The “default” rule set is used for all clients that match no clientgroup and for clientgroups with no acls declared. This section must always end our “acl” declaration for security reasons, since it will deny by default anything not previously declared and allowed. The “redirect” rule is used to redirect blocked destination groups, sites, users, etc to an alternative URL, where they will get more information about the reason why they cannot access the requested Internet site. WARNING: You cannot define or use more than one acl block in the squidGuard.conf file. Everything must be defined in the same acl block. Step1 Now that we have a better idea about how the SquidGuard configuration file works, it’s time to think about what we need to define inside it. Let’s create the SquidGuard configuration file. Our example assumes that you want to permit Internet access during working hours for all internal client workstations coming from the IP address range 192.168.1.0/24, and that you want to deny access to aggressive, gambling, hacking, and porn sites and redirect any refused connections to an alternative URL. This configuration is suitable for most needs. If you have a specific requirement, then you have to adjust the configuration and read the SquidGuard documentation for more information. For optimum security, we will merge the SquidGuard ACL with the Squid ACL to force clients to enter a username and password before accessing the Internet. • Create the squidGuard.conf file (touch /etc/squid/squidGuard.conf) and add the following ACL lines: dbhome /var/spool/squid/squidGuard logdir /var/log/squid/squidGuard # TIME SPACE DECLARATIONS # The following declaration define a time rule from where clients are # allowed and can access the Internet. Outside this time, connections # will be denied. # time workhours { weekly mtwhf 08:30 - 17:30 }
- 347. SquidGuard 1 CHAPTER 2 347 #SOURCE GROUP DECLARATIONS # The following declaration define a source group, or client groups IP # addresses range from where connection to the Internet through the proxy # are allowed. # src internal { ip 192.168.1.0/24 } # DESTINATION GROUP DECLARATIONS # The following declaration define destination group, or target groups # websites where connection are forbiden. # dest aggressive { domainlist aggressive/domains urllist aggressive/urls } dest gambling { domainlist gambling/domains urllist gambling/urls } dest hacking { domainlist hacking/domains urllist hacking/urls } dest porn { domainlist porn/domains urllist porn/urls expressionlist porn/expressions } # REWRITE RULES GROUP DECLARATIONS # # ACCESS CONTROL LISTS # The Access Control List, ACL, combines the previous definitions into # distinct rulesets for each clientgroup. # acl { internal within workhours { pass !aggressive !gambling !hacking !porn all } default { pass none redirect http://my.proxy.com/cgi- bin/squidGuard.cgi?clientaddr=%a&clientname=%n&clientuser=%i&clientgroup= %s&targetgroup=%t&url=%u } }
- 348. SquidGuard 1 CHAPTER 2 348 Step2 Oncethe SquidGuard has been configured, we have to include in our default Squid configuration file some additional lines that will make Squid Proxy Server run with SquidGuard. In the configuration below, we use the default squid.conf file as described in the Squid chapter. The text in bold are the parts of the configuration file that we have added to the default Squid configuration file as used in the Squid chapter. • Edit the squid.conf file (vi /etc/squid/squid.conf) and add the following options to make Squid runs with SquidGuard: icp_port 0 hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin ? no_cache deny QUERY cache_mem 128 MB cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF cache_dir diskd /var/spool/squid 2000 16 256 cache_store_log none log_fqdn on redirect_program /usr/bin/squidGuard acl localnet src 192.168.1.0/255.255.255.0 acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 210 70 21 1025-65535 acl CONNECT method CONNECT acl all src 0.0.0.0/0.0.0.0 http_access allow localnet http_access allow localhost http_access deny !Safe_ports http_access deny CONNECT http_access deny all cache_mgr sysadmin@openna.com cache_effective_user squid cache_effective_group squid logfile_rotate 0 log_icp_queries off cachemgr_passwd my-secret-pass all buffered_logs on The option “redirect_program”, specifies the location of the URL redirector executable. The executable in our case is the squidguard binary program located under the “/usr/bin” directory. Once the “redirect_program” line is added into the squid.conf file, Squid will know that it must run and work with a new program called squidguard. In this way, Squid will continue its proxy job and SuidGuard will be in charge filtering, checking, authorizing and redirecting, if necessary, all Internet destinations. The option “log_fqdn”, enables reverse lookups with Squid. This is important with SquidGuard, since the use of domain matches for clientsgroups requires that Squid is set up to do reverse lookups on clients. Without this option, any domain specification parameters in the SquidGuard configuration file that point to a filter file will simply not work. Therefore, when SquidGuard is used with Squid, we have to check and enable this option in the squid.conf file.
- 349. SquidGuard 1 CHAPTER 2 349 Step3 Foradditional security or for those who want to authenticate users with a username and password before allowing Internet access, there are some previously shown options that we can add into our squid.conf file. Below, we use the squid.conf file used in step 2 and add the user authentication feature. The text in bold are the parts of the configuration file that we have added to the above Squid configuration file. • Edit the squid.conf file (vi /etc/squid/squid.conf) and add the following options to make Squid use the users authentication feature: icp_port 0 hierarchy_stoplist cgi-bin ? acl QUERY urlpath_regex cgi-bin ? no_cache deny QUERY cache_mem 128 MB cache_replacement_policy heap GDSF memory_replacement_policy heap GDSF cache_dir diskd /var/spool/squid 2000 16 256 cache_store_log none log_fqdn on redirect_program /usr/bin/squidGuard authenticate_program /usr/lib/squid/pam_auth /etc/passwd acl authenticated proxy_auth REQUIRED acl localnet src 192.168.1.0/255.255.255.0 acl localhost src 127.0.0.1/255.255.255.255 acl Safe_ports port 80 443 210 70 21 1025-65535 acl CONNECT method CONNECT acl all src 0.0.0.0/0.0.0.0 http_access allow authenticated http_access allow localnet http_access allow localhost http_access deny !Safe_ports http_access deny CONNECT http_access deny all cache_mgr sysadmin@openna.com cache_effective_user squid cache_effective_group squid logfile_rotate 0 log_icp_queries off cachemgr_passwd my-secret-pass all buffered_logs on NOTE: If you need more information about Users Authentication support with Squid, please see the previous Squid chapter.
- 350. SquidGuard 1 CHAPTER 2 350 /home/httpd/cgi-bin/squidGuard.cgi:The SquidGuard.cgi File The squidGuard.cgi program is what users will see if the sites they try to access are blocked by company policy. It is used to explain to the user that the URL is blocked and by which rule set. Step1 There are two important options to configure in this small cgi program to make it work for your site. Below we show you how to do it. • Edit the squidGuard.cgi program (vi /home/httpd/cgi-bin/squidGuard.cgi) and change the following options to make SquidGuard runs for your site: $proxy = "my.proxydomain.com"; $proxymaster = "sysadmin@proxydomain.com"; Where “my.proxydomain.com” is the FQDN of the Gateway Server where SquidGuard is running, and “sysadmin@proxydomain.com” is the email address of the administrator. NOTE: You can use any personal html page of your choice to replace the squidGuard.cgi script, if it does not fit with your requirements. There is no problem as long as your squidGuard configuration file is properly updated to point to the right file. Testing SquidGuard Now it is time to restart our Squid server for all the changes to take effect and connect to the Internet with our preferred browser to see if everything is working as expected. 1. First, we try to connect to a legitimate site. We should receive a new window asking us to enter username and password. 2. Now we will try to access a blocked warez site just to see if SquidGuard filtering works as expected. For example, try www.warez.com. We must be redirected to a new URL, which will give us the reason why we cannot access this site.
- 351. SquidGuard 1 CHAPTER 2 351 NOTE:If you receive an error message here, it is likely because you have forgot to configure the squidguard.cgi program to fit your domain name information. Please edit the “/home/httpd/cgi-bin/suidguard.cgi” program and make the appropriate changes. Optimizing SquidGuard This section deals specifically with the actions we can take to improve and tighten the performance of SquidGuard. Note that we refer to the features available within the base installed program only. Creating a prebuilt database Ok, your SquidGuard program is running and you’re happy to see that it works, but wait a minute; it is possible to make SquidGuard runs faster simply by converting its filter files (where blacklisted domains and urls reside) into a db filter file. The default filter files used with SquidGuard are in plain text format, and SquidGuard needs to parse all the lines inside the filter file to decide if domains/url’s can be allowed or not. There is a better method that gives the same result and also runs faster by converting all of its filter files into a db file. Step1 The first step to accomplish this conversion will be to use the “-C” command of SquidGuard. This command simply converts the text file into a db file. • To convert your filter text files into a db file,use the following commands: [root@deep /]# cd /var/spool/squid/squidGuard/ [root@deep /]# squidGuard -C aggressive/domains [root@deep /]# squidGuard -C aggressive/urls [root@deep /]# squidGuard -C gambling/domains [root@deep /]# squidGuard -C gambling/urls [root@deep /]# squidGuard -C hacking/domains [root@deep /]# squidGuard -C hacking/urls [root@deep /]# squidGuard -C porn/domains [root@deep /]# squidGuard -C porn/urls The above commands, will convert a domainlist or urllist from plain text file to a prebuilt database.
- 352. SquidGuard 1 CHAPTER 2 352 NOTE:There is one filter file that cannot and should not be converted into a db file. This filter file is the “expressions” file located under the “porn” directory. Step2 Once all of our filter files have been converted, now we have to edit our squidGuard.conf file to change the default extension for our filter files to reflect the change. The text in bold are the parts of the configuration file that we have changed in the default SquidGuard configuration file. • Edit the squidGuard.conf file (vi /etc/squid/squidGuard.conf) and change the following lines to make SquidGuard load and interpret the entire db file now. dbhome /var/spool/squid/squidGuard logdir /var/log/squid/squidGuard # TIME SPACE DECLARATIONS # The following declaration define a time rule from where clients are # allowed and can access the Internet. Outside this time, connections # will be denied. # time workhours { weekly mtwhf 08:30 - 17:30 } # SOURCE GROUP DECLARATIONS # The following declaration define a source group, or client groups IP # addresses range from where connection to the Internet through the proxy # are allowed. # src internal { ip 192.168.1.0/24 } # DESTINATION GROUP DECLARATIONS # The following declaration define destination group, or target groups # websites where connection are forbiden. # dest aggressive { domainlist aggressive/domains.db urllist aggressive/urls.db } dest gambling { domainlist gambling/domains.db urllist gambling/urls.db } dest hacking { domainlist hacking/domains.db urllist hacking/urls.db } dest porn { domainlist porn/domains.db urllist porn/urls.db expressionlist porn/expressions }
- 353. SquidGuard 1 CHAPTER 2 353 #REWRITE RULES GROUP DECLARATIONS # # ACCESS CONTROL LISTS # The Access Control List, ACL, combines the previous definitions into # distinct rulesets for each clientgroup. # acl { internal within workhours { pass !aggressive !gambling !hacking !porn all } default { pass none redirect http://my.proxy.com/cgi- bin/squidGuard.cgi?clientaddr=%a&clientname=%n&clientuser=%i&clientgroup= %s&targetgroup=%t&url=%u } } Step3 Finally, we have to restart our Squid Proxy server for the changes to take effect.
- 355. FreeS/WAN VPN IN THISCHAPTER 1. Compiling - Optimizing & Installing FreeS/WAN 2. Configuring FreeS/WAN 3. Configuring RSA private keys secrets 4. Requiring network setup for IPSec 5. Testing the FreeS/WAN installation
- 357. FreeS/WAN VPN 1 CHAPTER3 357 Linux FreeS/WAN Abstract First of, I would like to mention that this chapter about FreeSWAN is an unsupported chapter now. This because FreeSWAN is a very special piece of software that often required specific kernel versions to work on the system. Since kernel versions are updated frequently and more often than FreeSWAN versions, there is no guarantee that the kernel version you use when reading this chapter will be compatible with FreeSWAN. Also, FreeSWAN is not software that everyone uses daily on the Internet for proper operation of their servers. Usually, only experts and companies, which have specific needs for their network, will need to install and use it. For this reason, I’ve decided to not provide advanced information about FreeSWAN in this book but since some of you will certainly ask for it, I’ll provide some information about how to compile, configure and run it for Linux. Unlike other chapters in this book, there is no guarantee that the information provided here will work for your system. If you have problem getting FreeSWAN to work for you, then ask the FreeSWAN group for some help. Here is just some basic startup information about FreeSWAN now. Protection of client-to-server and vice versa with PGP for mail, SSH for remote login, and SSL solutions are an excellent choice but sometimes for enterprise environments establishing secure communication channels, assuring full privacy, authenticity and data integrity between two gateway machines, routers, or firewalls system over the Internet are vital. For this, IPSEC has been created. IPSEC is Internet Protocol SECurity. It uses strong cryptography to provide both authentication and encryption services. Authentication ensures that packets are from the right sender and have not been altered in transit. Encryption prevents unauthorized reading of packet contents. IPSEC can protect any protocol running above IP and any medium used below IP. IPSEC can also provide some security services "in the background", with no visible impact on users. More to the point, it can protect a mixture of protocols running over a complex combination of media (i.e. IMAP/POP etc.) without having to change them in any way, since the encryption occurs at the IP level. Three protocols are used with FreeS/WAN to archive this result: 1. AH (Authentication Header) provides a packet-level authentication service. 2. ESP (Encapsulating Security Payload) provides encryption plus authentication. 3. IKE (Internet Key Exchange) negotiates connection parameters. The FreeSWAN implementation has three main parts: 1. KLIPS (kernel IPsec) implements AH, ESP, and packet handling within the kernel. 2. Pluto (an IKE daemon) implements IKE, negotiating connections with other systems. 3. Various scripts provide an adminstrator's interface to the machinery. IPSEC services allow you to build secure tunnels through untrusted networks like the Internet. Everything passing through the untrusted net is encrypted by the IPSEC gateway machine and decrypted by the gateway server at the other end. The result is Virtual Private Network or VPN. This is a network, which is effectively private even though it includes machines at several different sites connected by the insecure Internet.
- 358.
- 359. FreeS/WAN VPN 1 CHAPTER3 359 These installation instructions assume Commands are Unix-compatible. The source path is /usr/src (note that other paths are possible, at personal discretion). Installations were tested on OpenNA Linux and Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: Yes Latest FreeS/WAN version number is 1.95 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages The following is based on information listed by FreeS/WAN as of 2001/02/04. Please check http://www.freeswan.org/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: FreeS/WAN Homepage Site: http://www.freeswan.org/ FreeS/WAN FTP Site: 194.109.6.26 You must be sure to download: freeswan-1.95.tar.gz Prerequisites Linux FreeS/WAN requires that the software below is already installed on your system to be able to run and work successfully. If this is not the case, you must install it from your Linux CD-ROM or source archive file. Please make sure you have this program installed on your machine before you proceed with this chapter. gmp is required to compile and make FreeS/WAN works in your system. NOTE: Not installing the GMP library will make pluto fail to compile on your server. Pristine source If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install FreeS/WAN, and then one afterwards, and then compares them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > Freeswan1 • And the following one after you install the software: [root@deep root]# find /* > Freeswan2 • Then use this command to get a list of what changed: [root@deep root]# diff Freeswan1 Freeswan2 > Freeswan-Installed
- 360. FreeS/WAN VPN 1 CHAPTER3 360 Compiling - Optimizing & Installing FreeS/WAN Below are the required steps that you must make to compile and optimize the FreeS/WAN software before installing it into your Linux system. Step 1 The installation of IPSEC FreeS/WAN software requires some modification of your original kernel since some parts of FreeS/WAN must be included and incorporated in your kernel before you can use it. For this reason the first step in installing FreeS/WAN is to go to the Linux Kernel section in this book and follow the instructions on how to install the Linux Kernel on your system (even if you have already done this before) and come back to this section after you have executed the “make dep; make clean” commands, but before the “make bzImage” command in the Linux Kernel chapter. Step 2 Once your kernel is configured and you download the FreeS/WAN program from the main software site you must copy it to the /usr/src directory and change to this location before expanding the archive. Putting FreeS/WAN under /usr/src/linux will confuse the links, therefore, expand the software under /usr/src and never under /usr/src/linux directory. • These procedures can be accomplished with the following commands: [root@deep /]# cp freeswan-version.tar.gz /usr/src/ [root@deep /]# cd /usr/src/ [root@deep src]# tar xzpf freeswan-version.tar.gz Step 3 After that, move into the newly created FreeS/WAN directory then configure, compile and optimize it. • To move into the top-level directory of FreeS/WAN distribution use the command: [root@deep src]# cd freeswan-1.95/ Step 4 You must modify the Makefile.inc under the FreeS/WAN source directory to specify installation paths and optimization parameters. We must modify this file to be compliant with Linux file system structure, add optimization parameters for our specific processor architecture and install FreeS/WAN files under our PATH environment variable. • Edit the Makefile.inc file (vi Makefile.inc) and change all of the targeted lines in the order shown below: PUBDIR=$(DESTDIR)/usr/local/sbin To read: INC_USRLOCAL=/usr REALPRIVDIR=/usr/local/lib/ipsec To read: INC_MANDIR=/share/man
- 361. FreeS/WAN VPN 1 CHAPTER3 361 MANTREE=$(DESTDIR)/usr/local/man To read: USERCOMPILE=-O2 CONFDIR=$(DESTDIR)/etc To read: KLIPSCOMPILE=-O2 All of the above changes, will relocate all files related to the FreeS/WAN software to the destination target directories we have chosen. We also add optimization parameters related to the type of processor that we use in our system for better performance. Step 5 Once the modifications have been made to the source file of FreeS/WAN as described in step 4, we need to patch the pre-configured Linux Kernel to include FreeS/WAN support. • This procedure can be accomplished with the following command: [root@deep freeswan-1.95]# make ogo echo "===============" >>out.kpatch echo "`date` `cd /usr/src/linux ; pwd`" >>out.kpatch make _patches2.3 >>out.kpatch ………… The make ogo command is what we use to patch the kernel. It will automatically start the kernel configuration part for the second time and will let you answer all kernel configuration questions before compilation and integration of its component into the kernel. During the second kernel configuration, be sure that your kernel has been built with FreeS/WAN support enabled. A new section related to Frees/WAN support named “IPSec options (FreeS/WAN)” should appear in your kernel configuration after you have patched the kernel with the FreeS/WAN program as described above. You need ensure that you have answered Y to the following questions under the new kernel section: IPSec options (FreeS/WAN). IP Security Protocol (FreeS/WAN IPSEC) (CONFIG_IPSEC) [Y/n/?] * * IPSec options (FreeS/WAN) * IPSEC: IP-in-IP encapsulation (tunnel mode) (CONFIG_IPSEC_IPIP) [Y/n/?] IPSEC: Authentication Header (CONFIG_IPSEC_AH) [Y/n/?] HMAC-MD5 authentication algorithm (CONFIG_IPSEC_AUTH_HMAC_MD5) [Y/n/?] HMAC-SHA1 authentication algorithm (CONFIG_IPSEC_AUTH_HMAC_SHA1) [Y/n/?] IPSEC: Encapsulating Security Payload (CONFIG_IPSEC_ESP) [Y/n/?] 3DES encryption algorithm (CONFIG_IPSEC_ENC_3DES) [Y/n/?] IPSEC: IP Compression (CONFIG_IPSEC_IPCOMP) [Y/n/?] IPSEC Debugging Option (CONFIG_IPSEC_DEBUG) [Y/n/?]
- 362. FreeS/WAN VPN 1 CHAPTER3 362 NOTE: All the customization you made to your kernel the first time you ran the make config, make dep, and make clean commands will be preserved, so you don’t need to reconfigure every part of your kernel; Just the new section added by FreeS/WAN named “IPSec options (FreeS/WAN)” is required, as shown above. Some networking options will get turned on automatically, even if you previously turned them off; This is because IPSEC needs them. Whichever configuration program you are using, you should pay careful attention to a few issues: in particular, do NOT disable any of the following under the “Networking Options” of your kernel configuration: Kernel/User netlink socket (CONFIG_NETLINK) [Y/n/?] Netlink device emulation (CONFIG_NETLINK_DEV) [Y/n/?] Step 6 Once the make ogo command is completed, your FreeS/WAN software and Linux kernel with FreeS/WAN support is ready to be installed on your server. We must make a list of files on the system before you install the software and the kernel, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally install FreeS/WAN and the new kernel with FreeS/WAN support in your server: [root@deep freeswan-1.95]# cd [root@deep root]# find /* > Freeswan1 [root@deep root]# cd /usr/src/freeswan-1.95/ [root@deep freeswan-1.95]# make install [root@deep freeswan-1.95]# cd [root@deep root]# find /* > Freeswan2 [root@deep root]# diff Freeswan1 Freeswan2 > Freeswan-Installed The make install command will install all FreeS/WAN and kernel components together to the appropriated location on your server. Step 7 At this stage of your installation of FreeS/WAN, you must follow the rest of the instructions in the Linux Kernel chapter of this book as normal to install the kernel. At this point, after you have copied and installed your new kernel image, system.map, or modules (if necessary), and set the lilo.conf file to load the new kernel, you must edit and customize the configuration files related to FreeS/WAN “ipsec.conf” and “ipsec.secrets” before rebooting your system. Step 8 Once the compilation, optimization and installation of the software have been finished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete FreeS/WAN and its related source directory, use the following commands: [root@deep /]# cd /usr/src/ [root@deep src]# rm -rf freeswan-version/ [root@deep src]# rm -f freeswan-version.tar.gz
- 363. FreeS/WAN VPN 1 CHAPTER3 363 Configuring FreeS/WAN After building FreeS/WAN, your next step is to verify or change, if necessary, the options in your FreeS/WAN configuration files. Those files are: /etc/ipsec.conf (The FreeS/WAN Configuration File) /etc/ipsec.secrets (The FreeS/WAN Configuration File to store secret keys) /etc/ipsec.conf: The FreeS/WAN Configuration File The configuration file for FreeS/WAN (/etc/ipsec.conf) allows you to configure your IPSEC configurations, control information and connections types. IPSEC currently supports two types of connections: Manually keyed and Automatically keyed. The difference is strictly in how they are keyed. Manually keyed connections use keys stored in the /etc/ipsec.conf file. This type of connection is less secure than automatically keyed. Automatically keyed connections use keys automatically generated by the Pluto key negotiation daemon. The key negotiation protocol, used by default and named IKE, authenticates the other system using shared secrets stored in /etc/ipsec.secrets file. For these reasons, we will use the automatically keyed connection that is more secure than the manually keyed connection (it is highly recommended that you use the automatically keyed connection). In our example configuration below, we configure a sample tunnel with a firewall-penetrating tunnel, and we assume that firewalling is being done on the left and right side. We choose to show you this configuration since we assume it is what most users and companies will use. Also, it allows us to play with more options in the configuration file ipsec.conf for automatically keyed connections. Different configurations exist and you may consult the “doc/examples” file under the subdirectory “doc” of the Frees/WAN source directory for more information and other possible configurations. We must edit the ipsec.conf file (vi /etc/ipsec.conf) and change the default values to fit our specifications for IPSEC configuration and communication. Currently there are two types of section in this file (/etc/ipsec.conf): a “config” section, which specifies general configuration information for IPSEC, and a “conn” section which specifies an IPSEC connection. Its contents are not security-sensitive unless manual keying is being done (recall, manual keying is not recommended for security reasons). The first section type, named config setup, is the only config section known to the IPSEC software containing overall setup parameters for IPSEC that applies to all connections, and information used when the software is being started. The second type, named conn, contains a connection specification defining a network connection to be made using IPSEC. The name it is given is arbitrary, and is simply used to identify the connection to ipsec_auto(8) and ipsec_manual(8).
- 364. FreeS/WAN VPN 1 CHAPTER3 364 # /etc/ipsec.conf - FreeS/WAN IPSEC configuration file # More elaborate and more varied sample configurations can be found # in doc/examples. # basic configuration config setup interfaces="ipsec0=eth0" klipsdebug=none plutodebug=none plutoload=%search plutostart=%search # sample connection conn deep-mail left=208.164.186.1 leftsubnet=192.168.1.0/24 leftnexthop=205.151.222.250 right=208.164.186.2 rightsubnet=192.168.1.0/24 rightnexthop=205.151.222.251 keyingtries=0 auth=ah auto=start This tells the ipsec.conf file to set itself up for this particular configuration setup with: interfaces="ipsec0=eth0" This option specifies which appropriate virtual and physical interfaces for IPSEC to use. The default setting, “interfaces=%defaultroute”, will look for your default connection to the Internet, or your corporate network. Also, you can name one or more specific interfaces to be used by FreeS/WAN. For example: interfaces="ipsec0=eth0" interfaces="ipsec0=eth0 ipsec1=ppp0" Both set the eth0 interface as ipsec0. The second one, however, also supports IPSEC over a PPP interface. If the default setting “interfaces=%defaultroute” is not used, then the specified interfaces will be the only ones this gateway machine can use to communicate with other IPSEC gateways. klipsdebug=none This option specifies the debugging output for KLIPS (the kernel IPSEC code). The default value none, means no debugging output and the value all means full output. plutodebug=none This option specifies the debugging output for the Pluto key. The default value, none, means no debugging output, and the value all means full output. plutoload=%search This option specifies which connections (by name) to load automatically into memory when Pluto starts. The default is none and the value %search loads all connections with auto=add or auto=start. plutostart=%search This option specifies which connections (by name) to automatically negotiate when Pluto starts. The default is none and the value %search starts all connections with auto=start.
- 365. FreeS/WAN VPN 1 CHAPTER3 365 conn deep-mail This option specifies the name given to identify the connection specification to be made using IPSEC. It’s a good convention to name connections by their ends to avoid mistakes. For example, the link between deep.openna.com and mail.openna.com gateways server can be named "deep-mail", or the link between your Montreal and Paris offices, "montreal-paris". Note that the name “deep-mail” or whatever you have chosen should be the same in the ipsec.conf file on both gateways. In other words, the only change you should make in the /etc/ipsec.conf file on the second gateway is changing the “interfaces=” line to match the interface the second gateway uses for IPSEC connection, if, of course, it’s different from the first gateway. For example, if the interface eth0 is used on the both gateways for IPSEC communication, you don’t need to change the line “interfaces=” on the second gateway. On the other hand, if the first gateway uses eth0 and the second uses eth1, you must change the line “interfaces=” on the second gateway to match the interface eth1. left=208.164.186.1 This option specifies the IP address of the gateway's external interface used to talk to the other gateway. leftsubnet=192.168.1.0/24 This option specifies the IP network or address of the private subnet behind the gateway. leftnexthop=205.151.222.250 This option specifies the IP address of the first router in the appropriate direction or ISP router. right=208.164.186.2 This is the same explanation as “left=” but for the right destination. rightsubnet=192.168.1.0/24 This is the same explanation as “leftsubnet=” but for the right destination. rightnexthop=205.151.222.251 This is the same explanation as “leftnexthop=” but for the right destination. keyingtries=0 This option specifies how many attempts (an integer) should be made in (re)keying negotiations. The default value 0 (retry forever) is recommended. auth=ah This option specifies whether authentication should be done separately using AH (Authentication Header), or be included as part of the ESP (Encapsulated Security Payload) service. This is preferable when the IP headers are exposed to prevent man-in-the-middle attacks. auto=start This option specifies whether automatic startup operations should be done at IPSEC startup. NOTE: A data mismatch anywhere in this configuration “ipsec.conf” will cause FreeS/WAN to fail and to log various error messages.
- 366. FreeS/WAN VPN 1 CHAPTER3 366 /etc/ipsec.secrets: The FreeS/WAN File to store Secret Keys The file ipsec.secrets stores the secrets used by the pluto daemon to authenticate communication between both gateways. Two different kinds of secrets can be configured in this file, preshared secrets and RSA private keys. You must check the permissions of this file to be sure that the super-user “root” owns the file, and its permissions are set to block all access by others. Step 1 An example secret is supplied in the ipsec.secrets file by default. You should change it by creating your own. With automatic keying you may have a shared secret up to 256 bits, which is then used during the key exchanges to make sure a man in the middle attack does not occur. • To create a new shared secret, use the following commands: [root@deep /]# ipsec ranbits 256 > temp New, random keys are created with the ranbits(8) utility in the file named “temp”. The ranbits utility may pause for a few seconds if not enough entropy is available immediately. Don’t forget to delete the temporary file as soon as you are done with it. Step 2 Now that our new shared secret key has been created in the “temp” file, we must put it in the /etc/ipsec.secrets file. When editing the ipsec.secrets file, you should see something like the following appearing in your text editor. Each line has the IP addresses of the two gateways plus the secret. It should look something like this: # This file holds shared secrets which are currently the only inter-Pluto # authentication mechanism. See ipsec_pluto(8) manpage. Each secret is # (oversimplifying slightly) for one pair of negotiating hosts. # The shared secrets are arbitrary character strings and should be both # long and hard to guess. # Note that all secrets must now be enclosed in quotes, even if they have # no white space inside them. 10.0.0.1 11.0.0.1 "jxVS1kVUTTulkVRRTnTujSm444jRuU1mlkklku2nkW3nnVu V2WjjRRnulmlkmU1Run5VSnnRT" • Edit the ipsec.secrets file (vi /etc/ipsec.secrets) and change the default secrets keys: 10.0.0.1 11.0.0.1 " jxVS1kVUTTulkVRRTnTujSm444jRuU1mlkklku2nkW3nnVu V2WjjRRnulmlkmU1Run5VSnnRT " To read: 208.164.186.1 208.164.186.2 "0x9748cc31_2e99194f_d230589b_cd846b57_dc070b01_74b66f34_19c40a1a_804906ed" Where “208.164.186.1" and “208.164.186.2" are the IP addresses of the two gateways and "0x9748cc31_2e99194f_d230589b_cd846b57_dc070b01_74b66f34_19c40a1a_804906 ed" (note that the quotes are required) is the shared secret we have generated above with the command “ipsec ranbits 256 > temp” in the “temp” file.
- 367. FreeS/WAN VPN 1 CHAPTER3 367 Step 3 The files ipsec.conf, and ipsec.secrets must be copied to the second gateway machine so as to be identical on both ends. The only exception to this is the ipsec.conf file, which must have in it a section labeled by the line config setup with the correct interface settings for the second gateway, if they differ from the first. The ipsec.secrets file, contrary to the RSA private key, should have the same-shared secrets on the two gateways. WARNING: The file /etc/ipsec.secrets should have permissions rw------- (600) and be owned by the super-user “root”. The file /etc/ipsec.conf is installed with permissions rw-r- -r— (644) and must be owned also by “root”. Configuring RSA private keys secrets Recall that currently with FreeSWAN software there are two kinds of secrets: preshared secrets and RSA private keys. The preshared secrets are what we have configured in our ipsec.conf and ipsec.secrets example, above. Some people may prefer to use RSA private keys for authentication by the Pluto daemon of the other hosts. If you are in this situation, you will have to make some minor modifications to your ipsec.conf and ipsec.secrets files as described in the following steps: You need to create a separate RSA key for *each* gateway. Each one gets its private key in its own ipsec.secrets file, and the public keys go in leftrsasigkey and rightrsasigkey parameters in the conn description of ipsec.conf file, which goes to both. Step 1 Create a separate RSA key for *each* gateway: • On the first gateway (e.i. deep), use the following commands: [root@deep /]# cd / [root@deep /]# ipsec rsasigkey --verbose 1024 > deep-keys computing primes and modulus... getting 64 random bytes from /dev/random looking for a prime starting there found it after 30 tries getting 64 random bytes from /dev/random looking for a prime starting there found it after 230 tries swapping primes so p is the larger computing (p-1)*(q-1)... computing d... computing exp1, exp1, coeff... output... • On the second gateway (e.i. mail), use the following commands: [root@mail /]# cd / [root@mail /]# ipsec rsasigkey --verbose 1024 > mail-keys computing primes and modulus... getting 64 random bytes from /dev/random looking for a prime starting there found it after 30 tries getting 64 random bytes from /dev/random looking for a prime starting there found it after 230 tries swapping primes so p is the larger computing (p-1)*(q-1)...
- 368. FreeS/WAN VPN 1 CHAPTER3 368 computing d... computing exp1, exp1, coeff... output... The rsasigkey utility generates an RSA public and private key pair of a 1024-bit signature, and puts it in the file deep-keys (mail-keys for the second command on the second gateway). The private key can be inserted verbatim into the ipsec.secrets file, and the public key into the ipsec.conf file. WARNING: The rsasigkey utility may pause for a few seconds if not enough entropy is available immediately. You may want to give it some bogus activity such as random mouse movements. The temporary RSA “deep-keys” and “mail-keys” files should be deleted as soon as you are done with it. Don’t forget to delete the deep-keys and mail-keys RSA files. Step 2 Modify your /etc/ipsec.conf files to use RSA public keys in *each* gateway: Edit you original ipsec.conf file (vi /etc/ipsec.conf) and add the following parameters related to RSA in the conn desciption of your ipsec.conf file on both gateway: # sample connection conn deep-mail left=208.164.186.1 leftsubnet=192.168.1.0/24 leftnexthop=205.151.222.250 right=208.164.186.2 rightsubnet=192.168.1.0/24 rightnexthop=205.151.222.251 keyingtries=0 auth=ah authby=rsasig leftrsasigkey=<Public key of deep> rightrsasigkey=<Public key of mail> auto=start authby=rsasig This parameter specifies how the two security gateways should authenticate each other. The default value is secret for shared secrets. We must specify rsasig for RSA since we have decided to use RSA digital signatures. leftrsasigkey=<Public key of deep> This parameter specifies the left participant's public key for RSA signature authentication. In our example, left is 208.164.186.1, and represents deep.openna.com, so we must put the RSA public key for deep on this line. rightrsasigkey=<Public key of mail> This parameter specifies the right participant's public key for RSA signature authentication. In our example, right is 208.164.186.2, and represents mail.openna.com, so we must put the RSA public key of mail on this line. You can retrieve the public key of deep in the RSA key file named “deep-keys”, and the public key of mail in the RSA key file named “mail-keys”, that we have created in step 1 above. These files will look like this:
- 369. FreeS/WAN VPN 1 CHAPTER3 369 RSA keys for gateway deep (deep-keys): [root@deep /]# cd / [root@deep /]# vi deep-keys # 1024 bits, Fri Feb 4 05:05:19 2000 # for signatures only, UNSAFE FOR ENCRYPTION #pubkey=0x010395daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74 bcfb51a6ecc08890d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb 4d7689f2da7a69199e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16 a76bab772888f1fd71aa08f08502a141b611f Modulus: 0x95daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74bcfb51a6ecc08890 d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb4d7689f2da7a6919 9e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16a76bab772888f1fd 71aa08f08502a141b611f PublicExponent: 0x03 # everything after this point is secret PrivateExponent: 0x63e74967eaea2025c98c69f6ef0753a6a3ff6764157dbdf1f50013471324dd352366f48805b0b 37f232384b2b52ce2ee85d173468b62eaa052381a9588a317b3a1324d01a531a41fa7add6c5efbd d88f4718feed2bc0246be924e81bb90f03e49ceedf7af0dd48f06f265b519600bd082c6e6bd27ea a71cc0288df1ecc3b062b Prime1: 0xc5b471a88b025dd09d4bd7b61840f20d182d9b75bb7c11eb4bd78312209e3aee7ebfe632304db 6df5e211d21af7fee79c5d45546bea3ccc7b744254f6f0b847f Prime2: 0xc20a99feeafe79767122409b693be75f15e1aef76d098ab12579624aec708e85e2c5dd62080c3 a64363f2f45b0e96cb4aef8918ca333a326d3f6dc2c72b75361 Exponent1: 0x83cda11b0756e935be328fcebad5f6b36573bcf927a80bf2328facb6c0697c9eff2a9976cade7 9ea3ec0be1674fff4512e8d8e2f29c2888524d818df9f5d02ff Exponent2: 0x815c66a9f1fefba44b6c2b124627ef94b9411f4f9e065c7618fb96dc9da05f03ec83e8ec055d7 c42ced4ca2e75f0f3231f5061086ccd176f37f9e81da1cf8ceb Coefficient: 0x10d954c9e2b8d11f4db1b233ef37ff0a3cecfffad89ba5d515449b007803f577e3bd7f0183ced dfd805466d62f767f3f5a5731a73875d30186520f1753a7e325 RSA keys for gateway mail (mail-keys): [root@mail /]# cd / [root@mail /]# vi mail-keys # 1024 bits, Fri Feb 4 04:46:59 2000 # for signatures only, UNSAFE FOR ENCRYPTION #pubkey=0x01037631b81f00d5e6f888c542d44dbb784cd3646f084ed96f942d341c7c4686c bd405b805dc728f8697475f11e8b1dd797550153a3f0d4ff0f2b274b70a2ebc88f073748d1c1c88 21dc6be6a2f0064f3be7f8e4549f8ab9af64944f829b014788dd202cf7d2e320cab666f5e7a197e 64efe0bfee94e92ce4dad82d5230c57b89edf Modulus: 0x7631b81f00d5e6f888c542d44dbb784cd3646f084ed96f942d341c7c4686cbd405b805dc728f8 697475f11e8b1dd797550153a3f0d4ff0f2b274b70a2ebc88f073748d1c1c8821dc6be6a2f0064f 3be7f8e4549f8ab9af64944f829b014788dd202cf7d2e320cab666f5e7a197e64efe0bfee94e92c e4dad82d5230c57b89edf PublicExponent: 0x03 # everything after this point is secret PrivateExponent: 0x4ecbd014ab3944a5b08381e2de7cfadde242f4b03490f50d737812fd8459dd3803d003e84c5fa f0f84ea0bf07693a64e35637c2a08dff5f721a324b1747db09f62c871d5e11711251b845ae76753 d4ef967c494b0def4f5d0762f65da603bc04c41b4c6cab4c413a72c633b608267ae2889c162a3d5 bc07ee083b1c6e038400b
- 370. FreeS/WAN VPN 1 CHAPTER3 370 Prime1: 0xc7f7cc8feaaac65039c39333b878bffd8f95b0dc22995c553402a5b287f341012253e9f25b839 83c936f6ca512926bebee3d5403bf9f4557206c6bbfd9aac899 Prime2: 0x975015cb603ac1d488dc876132d8bc83079435d2d3395c03d5386b5c004eadd4d7b01b3d86aad 0a2275d2d6b791a2abe50d7740b7725679811a32ca22db97637 Exponent1: 0x854fddb5471c84357bd7b777d0507ffe5fb92092c1bb92e37801c3cc5aa22b5616e29bf6e7ad1 028624a486e0c619d47f428e2ad2a6a2e3a159d9d2a911c85bb Exponent2: 0x64e00e87957c81385b3daf9621e5d302050d7937377b92ad38d04792aadf1e8de52012290471e 06c1a3e1e47a61171d435e4f807a4c39a6561177316c9264ecf Coefficient: 0x6f087591becddc210c2ee0480e30beeb25615a3615203cd3cef65e5a1d476fd9602ca0ef10d9b 858edb22db42c975fb71883a470b43433a7be57df7ace4a0a3f Extract and copy the public RSA key files of deep and mail to your ipsec.conf files as shown below. You can locate the line related to the public key by a sentence beginning with the commented-out: “#pubkey=” line. # sample connection conn deep-mail left=208.164.186.1 leftsubnet=192.168.1.0/24 leftnexthop=205.151.222.250 right=208.164.186.2 rightsubnet=192.168.1.0/24 rightnexthop=205.151.222.251 keyingtries=0 auth=ah authby=rsasig leftrsasigkey=0x010395daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801ce a9cb74bcfb51a6ecc08890d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef9 8a7f29edcb4d7689f2da7a69199e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5d eac3b19d561c16a76bab772888f1fd71aa08f08502a141b611f rightrsasigkey=0x01037631b81f00d5e6f888c542d44dbb784cd3646f084ed96f942d341c 7c4686cbd405b805dc728f8697475f11e8b1dd797550153a3f0d4ff0f2b274b70a2ebc88f07 3748d1c1c8821dc6be6a2f0064f3be7f8e4549f8ab9af64944f829b014788dd202cf7d2e320 cab666f5e7a197e64efe0bfee94e92ce4dad82d5230c57b89edf auto=start NOTE: Don’t forget that, in this example, the “leftrsasigkey=” parameter contains the public key of deep and the “rightrsasigkey=” parameter contains the public key of mail. Step 3 Modify your /etc/ipsec.secrets files to use RSA private keys in *each* gateway: Edit your original ipsec.secrets file (vi /etc/ipsec.secrets) and add the RSA private key for authentication on both gateways: The ipsec.secrets file for gateway deep: [root@deep /]# vi /etc/ipsec.secrets 208.164.186.1 208.164.186.2 "0x9748cc31_2e99194f_d230589b_cd846b57_dc070b01_74b66f34_19c40a1a_804906ed"
- 371. FreeS/WAN VPN 1 CHAPTER3 371 You must change your original ipsec.secrets file as shown above to look like the following on both gateways. It is important to note that the private keys are not the same on both gateways, deep and mail. The private key for deep comes from the RSA key file “deep-keys”, while the private key for mail comes from the RSA key file “mail-keys”: 208.164.186.1 208.164.186.2: RSA { Modulus: 0x95daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74bcfb51a6ecc08890 d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb4d7689f2da7a6919 9e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16a76bab772888f1fd 71aa08f08502a141b611f PublicExponent: 0x03 # everything after this point is secret PrivateExponent: 0x63e74967eaea2025c98c69f6ef0753a6a3ff6764157dbdf1f50013471324dd352366f48805b0b 37f232384b2b52ce2ee85d173468b62eaa052381a9588a317b3a1324d01a531a41fa7add6c5efbd d88f4718feed2bc0246be924e81bb90f03e49ceedf7af0dd48f06f265b519600bd082c6e6bd27ea a71cc0288df1ecc3b062b Prime1: 0xc5b471a88b025dd09d4bd7b61840f20d182d9b75bb7c11eb4bd78312209e3aee7ebfe632304db 6df5e211d21af7fee79c5d45546bea3ccc7b744254f6f0b847f Prime2: 0xc20a99feeafe79767122409b693be75f15e1aef76d098ab12579624aec708e85e2c5dd62080c3 a64363f2f45b0e96cb4aef8918ca333a326d3f6dc2c72b75361 Exponent1: 0x83cda11b0756e935be328fcebad5f6b36573bcf927a80bf2328facb6c0697c9eff2a9976cade7 9ea3ec0be1674fff4512e8d8e2f29c2888524d818df9f5d02ff Exponent2: 0x815c66a9f1fefba44b6c2b124627ef94b9411f4f9e065c7618fb96dc9da05f03ec83e8ec055d7 c42ced4ca2e75f0f3231f5061086ccd176f37f9e81da1cf8ceb Coefficient: 0x10d954c9e2b8d11f4db1b233ef37ff0a3cecfffad89ba5d515449b007803f577e3bd7f0183ced dfd805466d62f767f3f5a5731a73875d30186520f1753a7e325 } The ipsec.secrets file for gateway mail: [root@mail /]# vi /etc/ipsec.secrets 208.164.186.1 208.164.186.2: RSA { Modulus: 0x95daee1be05f3038ae529ef2668afd79f5ff1b16203c9ceaef801cea9cb74bcfb51a6ecc08890 d3eb4b5470c0fc35465c8ba2ce9d1145ff07b5427e04cf4a38ef98a7f29edcb4d7689f2da7a6919 9e4318b4c8d0ea25d33e4f084186a2a54f4b4cec12cca1a5deac3b19d561c16a76bab772888f1fd 71aa08f08502a141b611f PublicExponent: 0x03 # everything after this point is secret PrivateExponent: 0x63e74967eaea2025c98c69f6ef0753a6a3ff6764157dbdf1f50013471324dd352366f48805b0b 37f232384b2b52ce2ee85d173468b62eaa052381a9588a317b3a1324d01a531a41fa7add6c5efbd d88f4718feed2bc0246be924e81bb90f03e49ceedf7af0dd48f06f265b519600bd082c6e6bd27ea a71cc0288df1ecc3b062b Prime1: 0xc5b471a88b025dd09d4bd7b61840f20d182d9b75bb7c11eb4bd78312209e3aee7ebfe632304db 6df5e211d21af7fee79c5d45546bea3ccc7b744254f6f0b847f Prime2: 0xc20a99feeafe79767122409b693be75f15e1aef76d098ab12579624aec708e85e2c5dd62080c3 a64363f2f45b0e96cb4aef8918ca333a326d3f6dc2c72b75361 Exponent1: 0x83cda11b0756e935be328fcebad5f6b36573bcf927a80bf2328facb6c0697c9eff2a9976cade7 9ea3ec0be1674fff4512e8d8e2f29c2888524d818df9f5d02ff
- 372. FreeS/WAN VPN 1 CHAPTER3 372 Exponent2: 0x815c66a9f1fefba44b6c2b124627ef94b9411f4f9e065c7618fb96dc9da05f03ec83e8ec055d7 c42ced4ca2e75f0f3231f5061086ccd176f37f9e81da1cf8ceb Coefficient: 0x10d954c9e2b8d11f4db1b233ef37ff0a3cecfffad89ba5d515449b007803f577e3bd7f0183ced dfd805466d62f767f3f5a5731a73875d30186520f1753a7e325 } Authentication by RSA Signatures requires that each host have its own private key. The key part of an entry may start with a token indicating the kind of key. “RSA” signifies RSA private key and “PSK” (which is the default) signifies PreShared Key. Since “PSK” is the default, we must specify “RSA”, so that we’ll be able to use RSA private keys in this file (ipsec.secrets). The super-user “root” should own the file ipsec.secrets, and its permissions should be set to block all access by others. Requiring network setup for IPSec There are some considerations you must ensure are correct before running FreeS/WAN software. These considerations are important if you don’t want to receive error messages during start up of your VPN. The following are the steps to follow: Step1 You will need to enable TCP/IP forwarding on the both gateway servers. In Linux, this is accomplished by adding the following line: • To enable IPv4 forwarding on your Linux system, edit the /etc/sysctl.conf file (vi /etc/sysctl.conf) and add the following line: # Enable/Disable packet forwarding net.ipv4.ip_forward = 1 You must restart your network for the change to take effect. The command to restart the network is the following: • To restart all network devices manually on your system, use the following command: [root@deep /]# /etc/init.d/network restart Setting network parameters [OK] Bringing up interface lo [OK] Bringing up interface eth0 [OK] Bringing up interface eth1 [OK] Step 2 Recall that automatically keyed connections use keys automatically generated by the Pluto key negotiation daemon. The pluto daemon will start up, try to connect to the Pluto daemon at the other end of the tunnel, and establish a connection. For this reason, an IPSEC gateway should have packet filters rules (in the firewall script file) permitting the following protocols to traverse the gateway when talking to other IPSEC gateway: UDP port 500 for IKE implemented by the Pluto daemon Protocol 50 for ESP encryption and/or authentication Protocol 51 for AH packet-level authentication See the GIPTables chapter in this book and the GIPTables manual for the correct rules to add to your firewall on both gateway machines to allow IPSEC packets to traverse the remote network gateway to your network gateway and vice versa.
- 373. FreeS/WAN VPN 1 CHAPTER3 373 Step 3 The rp_filter subsystem (related to IP spoofing protection) must be turned off on both gateways for IPSEC to work properly. This is accomplished by checking if the value 0 (off) is set in the /proc/sys/net/ipv4/conf/ipsec0/rp_filter and /proc/sys/net/ipv4/conf/eth0/rp_filter files respectively: • To check if the value 0 (off) is set in the rp_filter files, use the commands: [root@deep /]# cat /proc/sys/net/ipv4/conf/ipsec0/rp_filter 0 [root@deep /]# cat /proc/sys/net/ipv4/conf/eth0/rp_filter 0 NOTE: The subdirectory “ipsec0” in our example will be created only after the reboot of your system. So you may check the value of the “rp_filter” file in the “ipsec0” directory after your system has been restarted. • To set the value 0 (off) in the both rp_filter files manually, use the commands: [root@deep /]# echo 0 > /proc/sys/net/ipv4/conf/ipsec0/rp_filter [root@deep /]# echo 0 > /proc/sys/net/ipv4/conf/eth0/rp_filter Also you can put lines like the following in your firewall script files /etc/rc.d/init.d/iptables on both gateways to automatically set these values to 0 (off) and avoid making them manually: # Disable IP spoofing protection to allow IPSEC to work properly echo 0 > /proc/sys/net/ipv4/conf/ipsec0/rp_filter echo 0 > /proc/sys/net/ipv4/conf/eth0/rp_filter NOTE: In the example of the firewall script file above, we assume that eth0 is the interface you use for your connection. Of course if you use eth1 you must change eth0 to eth1, and so on. If you forget this step you will receive error messages on your terminal such as the following during the start up of FreeSWAN IPSEC: ipsec_setup: WARNING: ipsec0 has route filtering turned on, KLIPS may not work ipsec_setup: (/proc/sys/net/ipv4/conf/ipsec0/rp_filter = `1', should be 0) ipsec_setup: WARNING: eth0 has route filtering turned on, KLIPS may not work ipsec_setup: (/proc/sys/net/ipv4/conf/eth0/rp_filter = `1', should be 0)
- 374. FreeS/WAN VPN 1 CHAPTER3 374 Testing the FreeS/WAN installation • Reboot the both gateways to get FreeS/WAN started. • Examine the /var/log/messages file for any signs of trouble. If all goes well you should see something like this in the /var/log/messages file: Feb 2 05:22:35 deep ipsec_setup: Starting FreeS/WAN IPSEC snap2000jan31b... Feb 2 05:22:35 deep ipsec_setup: KLIPS debug `none' Feb 2 05:22:35 deep ipsec_setup: KLIPS ipsec0 on eth0 192.168.1.1/255.255.255.0 broadcast 192.168.1.255 Feb 2 05:22:36 deep ipsec_setup: Disabling core dumps: Feb 2 05:22:36 deep ipsec_setup: Starting Pluto (debug `none'): Feb 2 05:22:37 deep ipsec_setup: Loading Pluto database `deep-mail': Feb 2 05:22:37 deep ipsec_setup: Enabling Pluto negotiation: Feb 2 05:22:37 deep ipsec_setup: Routing for Pluto conns `deep-mail': Feb 2 05:22:37 deep ipsec_setup: Initiating Pluto tunnel `deep-mail': Feb 2 05:22:39 deep ipsec_setup: 102 "deep-mail" #1: STATE_MAIN_I1: initiate Feb 2 05:22:39 deep ipsec_setup: 104 "deep-mail" #1: STATE_MAIN_I2: from STATE_MAIN_I1; sent MI2, expecting MR2 Feb 2 05:22:39 deep ipsec_setup: 106 "deep-mail" #1: STATE_MAIN_I3: from STATE_MAIN_I2; sent MI3, expecting MR3 Feb 2 05:22:39 deep ipsec_setup: 004 "deep-mail" #1: STATE_MAIN_I4: SA established Feb 2 05:22:39 deep ipsec_setup: 110 "deep-mail" #2: STATE_QUICK_I1: initiate Feb 2 05:22:39 deep ipsec_setup: 004 "deep-mail" #2: STATE_QUICK_I2: SA established Feb 2 05:22:39 deep ipsec_setup: ...FreeS/WAN IPSEC started • Examine the /var/log/secure file for any signs of trouble. If all goes well you should see something like the following: Feb 21 14:45:42 deep Pluto[432]: Starting Pluto (FreeS/WAN Version 1.3) Feb 21 14:45:43 deep Pluto[432]: added connection description "deep-mail" Feb 21 14:45:43 deep Pluto[432]: listening for IKE messages Feb 21 14:45:43 deep Pluto[432]: adding interface ipsec0/eth0 192.168.1.1 Feb 21 14:45:43 deep Pluto[432]: loading secrets from "/etc/ipsec.secrets" Feb 21 14:45:43 deep Pluto[432]: "deep-mail" #1: initiating Main Mode Feb 21 14:45:44 deep Pluto[432]: "deep-mail" #1: ISAKMP SA established Feb 21 14:45:44 deep Pluto[432]: "deep-mail" #2: initiating Quick Mode POLICY_RSASIG+POLICY_ENCRYPT+POLICY_AUTHENTICATE+POLICY_TUNNEL+POLICY_PFS Feb 21 14:45:46 deep Pluto[432]: "deep-mail" #2: sent QI2, IPsec SA established Feb 21 14:45:47 deep Pluto[432]: "deep-mail" #3: responding to Main Mode Feb 21 14:45:49 deep Pluto[432]: "deep-mail" #3: sent MR3, ISAKMP SA established Feb 21 14:45:49 deep Pluto[432]: "deep-mail" #4: responding to Quick Mode Feb 21 14:45:50 deep Pluto[432]: "deep-mail" #4: IPsec SA established
- 375. FreeS/WAN VPN 1 CHAPTER3 375 • On both gateways, the following entries should now exist in the /proc/net/ directory: [root@deep /]# ls -l /proc/net/ipsec_* -r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_eroute -r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_klipsdebug -r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_spi -r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_spigrp -r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_spinew -r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_tncfg -r--r--r-- 1 root root 0 Feb 2 05:30 /proc/net/ipsec_version • The IPSEC interfaces should be attached on top of the specified physical interfaces. Confirm that with: [root@deep /]# cat /proc/net/ipsec_tncfg ipsec0 -> eth0 mtu=16260 -> 1500 ipsec1 -> NULL mtu=0 -> 0 ipsec2 -> NULL mtu=0 -> 0 ipsec3 -> NULL mtu=0 -> 0 • Now execute the following command to show minimal debugging information and see if the output looks something like this: [root@deep /]# ipsec look deep.openna.com Fri Feb 4 17:25:17 EST 2000 ============-============ 192.168.1.1/32 -> 192.168.1.2/32 => tun0x106@192.168.1.2 esp0x4450894d@192.168.1.2 ah0x4450894c@192.168.1.2 ------------=------------ ah0x3350f551@192.168.1.1 AH_HMAC_MD5: dir=in ooowin=32 seq=115 bit=0xffffffff alen=128 aklen=16 life(c,s,h)=bytes(16140,0,0)add(51656,0,0)use(54068,0,0)packets(115,0,0) idle=499 ah0x4450894c@192.168.1.2 AH_HMAC_MD5: dir=out ooowin=32 seq=2828 alen=128 aklen=16 life(c,s,h)=bytes(449488,0,0)add(51656,0,0)use(51656,0,0)packets(2828,0,0 ) idle=6 esp0x3350f552@192.168.1.1 ESP_3DES: dir=in ooowin=32 seq=115 bit=0xffffffff eklen=24 life(c,s,h)=bytes(13380,0,0)add(51656,0,0)use(54068,0,0)packets(115,0,0) idle=499 esp0x4450894d@192.168.1.2 ESP_3DES: dir=out ooowin=32 seq=2828 eklen=24 life(c,s,h)=bytes(381616,0,0)add(51656,0,0)use(51656,0,0)packets(2828,0,0 ) idle=6 tun0x105@192.168.1.1 IPIP: dir=in 192.168.1.2 -> 192.168.1.1 life(c,s,h)=add(51656,0,0) tun0x106@192.168.1.2 IPIP: dir=out 192.168.1.1 -> 192.168.1.2 life(c,s,h)=bytes(327581,0,0)add(51656,0,0)use(51656,0,0)packets(2828,0,0 ) idle=6 192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 ipsec0 192.168.1.1 0.0.0.0 255.255.255.255 UH 0 0 0 eth0 192.168.1.2 192.168.1.2 255.255.255.255 UGH 0 0 0 ipsec0 Destination Gateway Genmask Flags MSS Window irtt Iface
- 376. FreeS/WAN VPN 1 CHAPTER3 376 • Try pinging 192.168.1.2 from the 192.168.1.1 client. If this works then you have set it up correctly. If it does not work check your network to make sure 208.164.186.1 can reach 208.164.186.2, and that TCP-IP forwarding is enabled, and make sure that no firewall rules are blocking the packets, or trying to masquerade them before the rules allowing IPSec related traffic. For this test to work, it is important to use pings that go from one subnet to the other. 208.164.186.1 ---- 205.151.222.250 ---- 205.151.222.251 ---- 208.164.186.2 | | 192.168.1.0/24 192.168.1.0/24 | | 192.168.1.1 192.168.1.2 A last note about testing the installation of FreeSWAN IPSEC, if you encounter a problem that you are unable to resolve, you can use the following command to view a collection of debugging information (contents of files, selections from logs, etc.) related to the IPSEC encryption/authentication system that you should send to the Linux-IPSEC Mailing List (linux- ipsec@clinet.fi) to help you. • Use the following command to make an output of a collection of debugging information: [root@deep /]# ipsec barf > result This command is primarily provided as a convenience for remote debugging; A single command which packages up (and labels) all information that might be relevant to diagnosing a problem in IPSEC. Further documentation For more details, there are several manual pages about FreeS/WAN that you could read: $ man ipsec (8) - invoke IPSEC utilities. $ man ipsec atoaddr, addrtoa (3) - convert Internet addresses to and from ASCII. $ man ipsec atoasr (3) - convert ASCII to Internet address, subnet, or range. $ man ipsec atobytes, bytestoa (3) - convert binary data bytes from and to ASCII formats. $ man ipsec atodata, datatoa (3) - convert binary data from and to ASCII formats. $ man ipsec atosa, satoa (3) - convert IPSEC Security Association IDs to and from ASCII. $ man ipsec atosubnet, subnettoa (3) - convert subnet/mask ASCII form to and from addresses. $ man ipsec atoul, ultoa (3) - convert unsigned-long numbers to and from ASCII. $ man ipsec auto (8) - control automatically-keyed IPSEC connections. $ man ipsec barf (8) - spew out collected IPSEC debugging information. $ man ipsec bitstomask (3) - convert bit count to Internet subnet mask. $ man ipsec eroute (8) - manipulate IPSEC extended routing tables. $ man ipsec goodmask (3) - is this Internet subnet mask a valid one? $ man ipsec hostof (3) - given Internet address and subnet mask, return host part. $ man ipsec klipsdebug (8) - set Klips (kernel IPSEC support) debug features and level. $ man ipsec look (8) - show minimal debugging information. $ man ipsec manual (8) - take manually-keyed IPSEC connections up and down. $ man ipsec masktobits (3) - convert Internet subnet mask to bit count. $ man ipsec optionsfrom (3) - read additional ``command-line'' options from file. $ man ipsec pluto (8) - IPsec IKE keying daemon. $ man ipsec ranbits (8) - generate random bits in ASCII form. $ man ipsec rangetoa (3) - convert Internet address range to ASCII. $ man ipsec rsasigkey (8) - generate RSA signature key. $ man ipsec setup (8) - control IPSEC subsystem. $ man ipsec spi (8) - manage IPSEC Security Associations. $ man ipsec spigrp (8) - group/ungroup IPSEC Security Associations. $ man ipsec subnetof (3) - given Internet address and subnet mask, return subnet number. $ man ipsec tncfg (8) - associate IPSEC virtual interface with real interface. $ man ipsec whack (8) - control interface for IPSEC keying daemon. $ man ipsec.conf (5) - IPSEC configuration and connections. $ man ipsec.secrets (5) - secrets for IKE/IPsec authentication.
- 379. GnuPG IN THIS CHAPTER 1.Compiling - Optimizing & Installing GnuPG 2. Using GnuPG under Linux terminal
- 381. GnuPG 1 CHAPTER 4 381 LinuxGnuPG Abstract At this point we are ready to compile, configure, optimize and install software on our Linux server. Yes it is time, and we will begin our adventure with the powerful and easy to install GnuPG tool. Why do we choose to begin with GnuPG? The answer is simple, we are playing with a highly secured server and the first action to take each time we want to install some new software on this secured machine is to be absolutely sure that the software in question comes from a trusted source and is unmodified. With the GnuPG tool we can verify the supplied signature and be sure that the software is original. So it is recommended that this program is installed before any others. Encryption of data sources is an invaluable feature that gives us a high degree of confidentiality for our work. A tool like GnuPG does much more than just encryption of mail messages. It can be used for all kinds of data encryption, and its utilization is only limited by the imagination. GnuPG is GNU's tool for secure data communication and storage. It can be used to encrypt data and to create digital signatures. It includes an advanced key management facility and is compliant with the proposed OpenPGP Internet standard as described in RFC2440. Because GnuPG does not use any patented algorithm it is not compatible with PGP2 versions. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest GnuPG version number is 1.0.7 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages Please check http://www.gnupg.org/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: GnuPG Homepage: http://www.gnupg.org/ GnuPG FTP Site: 217.69.76.44 You must be sure to download: gnupg-1.0.7.tar.gz
- 382. GnuPG 1 CHAPTER 4 382 Prerequisites GnuPGrequires that the listed software below be already installed on your system to be able to compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source archive files. Please make sure you have this program installed on your machine before you proceed with this chapter. gettext is required to run GnuPG on your system. python-1.5 is required to run GnuPG on your system. expat is required to run GnuPG on your system. gmp is required to run GnuPG on your system. Pristine source If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed on the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install GnuPG, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > GnuPG1 • And the following one after you install the software: [root@deep root]# find /* > GnuPG2 • Then use the following command to get a list of what changed: [root@deep root]# diff GnuPG1 GnuPG2 > GnuPG-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In our example above, we use the /root directory of the system to store all the generated file lists. Compiling - Optimizing & Installing GnuPG Below are the required steps that you must make to configure, compile and optimize the GnuPG software before installing it into your Linux system. First off, we install the program as user 'root' so as to avoid authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp gnupg-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf gnupg-version.tar.gz
- 383. GnuPG 1 CHAPTER 4 383 Step2 In order to check that the version of GnuPG, which you are going to install, is an original and unmodified one, use the commands described below and check the supplied signature. Since we don’t have GnuPG already installed in the system, we have to verify the MD5 checksum of the program. • To verify the MD5 checksum of GnuPG, use the following command: [root@deep /]# md5sum gnupg-version.tar.gz This should yield an output similar to this: d8b36d4dfd213a1a1027b1877acbc897 gnupg-1.0.7.tar.gz Now check that this checksum is exactly the same as the one published on the GnuPG website at the following URL: http://www.gnupg.org/download.html Step 3 Next, move into the newly created GnuPG source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created GnuPG directory use the following command: [root@deep tmp]# cd gnupg-1.0.7/ • To configure and optimize GnuPG use the following compilation lines: CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS ./configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --disable-nls WARNING: Pay special attention to the compile CFLAGS line above. We optimize GnuPG for an i686 CPU architecture with the parameter “-march=i686”. Please don’t forget to adjust the CFLAGS line to reflect your own system. Step 4 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the GnuPG software: [root@deep gnupg-1.0.7]# make [root@deep gnupg-1.0.7]# make check [root@deep gnupg-1.0.7]# cd [root@deep root]# find /* > GnuPG1 [root@deep root]# cd /var/tmp/gnupg-1.0.7/ [root@deep gnupg-1.0.7]# make install [root@deep gnupg-1.0.7]# strip /usr/bin/gpg [root@deep gnupg-1.0.7]# strip /usr/bin/gpgv [root@deep gnupg-1.0.7]# cd [root@deep root]# find /* > GnuPG2 [root@deep root]# diff GnuPG1 GnuPG2 > GnuPG-Installed
- 384. GnuPG 1 CHAPTER 4 384 Theabove commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then install the binaries and any supporting files into the appropriate locations. Step 5 Once the configuration, optimization, compilation, and installation of the GnuPG software have been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete GnuPG and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf gnupg-version/ [root@deep tmp]# rm -f gnupg-version.tar.gz The rm command as used above will remove all the source files we have used to compile and install GnuPG. It will also remove the GnuPG compressed archive from the /var/tmp/ directory. Using GnuPG under Linux terminal Here we show you how to use GnuPG using a terminal to manage GPG keys. The commands listed below are ones that we use often, but many more exist. Check the manual page gpg (1) for more information. Creating a key-pair: First of all, we must create a new key-pair (public and private) if this is a first use of the GnuPG software to be able to use its encryption features. Step 1 The “--gen-key” option of GnuPG is used to generate a new (public and private) key, we have to use it every time we need to create a new GnuPG key on the system. When we issue this command for the first time, GnuPG will create the required directory and options file for us. • To create a new key-pair, use the following command: [root@deep /]# gpg --gen-key gpg (GnuPG) 1.0.7; Copyright (C) 2000 Free Software Foundation, Inc. This program comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to redistribute it under certain conditions. See the file COPYING for details. gpg: /root/.gnupg: directory created gpg: /root/.gnupg/options: new options file created gpg: you have to start GnuPG again, so it can read the new options file
- 385. GnuPG 1 CHAPTER 4 385 Step2 Once the command has been executed, we have to run it again for a second time to create our public and private keys, because on first utilization, it just creates the required directory and options file for us. Therefore, it will now create the keys. • We start GnuPG again with the same command: [root@deep /]# gpg --gen-key gpg (GnuPG) 1.0.7; Copyright (C) 2002 Free Software Foundation, Inc. This program comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to redistribute it under certain conditions. See the file COPYING for details. gpg: keyring `/root/.gnupg/secring.gpg' created gpg: keyring `/root/.gnupg/pubring.gpg' created Please select what kind of key you want: (1) DSA and ElGamal (default) (2) DSA (sign only) (4) ElGamal (sign and encrypt) (5) RSA (sign only) Your selection? 1 DSA keypair will have 1024 bits. About to generate a new ELG-E keypair. minimum keysize is 768 bits default keysize is 1024 bits highest suggested keysize is 2048 bits What keysize do you want? (1024) Press Enter Requested keysize is 1024 bits Please specify how long the key should be valid. 0 = key does not expire <n> = key expires in n days <n>w = key expires in n weeks <n>m = key expires in n months <n>y = key expires in n years Key is valid for? (0) Press Enter Key does not expire at all Is this correct (y/n)? y You need a User-ID to identify your key; the software constructs the user id from Real Name, Comment and Email Address in this form: "Heinrich Heine (Der Dichter) <heinrichh@duesseldorf.de>" Real name: Gerhard Mourani Email address: gmourani@openna.com Comment: Press Enter You selected this USER-ID: "Gerhard Mourani <gmourani@openna.com>" Change (N)ame, (C)omment, (E)mail or (O)kay/(Q)uit? o You need a Passphrase to protect your secret key. Enter passphrase: mypassphrase Repeat passphrase: mypassphrase We need to generate a lot of random bytes. It is a good idea to perform some other action (type on the keyboard, move the mouse, utilize the disks) during the prime generation; this gives the random number generator a better chance to gain enough entropy. ++++++++++++++++++++.+++++.+++++.+++++++++++++++.+++++++++++++++..+++++++ ++++++++...+++++++++++++++.+++++.++++++++++++++++++++++++++++++++++++++++ ..+++++
- 386. GnuPG 1 CHAPTER 4 386 <+++++..>+++++.................<..+++++.................................. ..........+++++^^^^ publicand secret key created and signed. NOTE: A new key-pair is created (secret and public key) in the “root” home directory ~/root under the .gnupg subdirectory because we issued this GnuPG command as user “root”. If you run the above command under other user into the system, then the generated keys will be located under its home directory on the server. Exporting GPG key/s for a user: Once your own key-pair is created, you can expand your horizons by exporting and distributing your public key over the world (NEVER export you private key). This can be done by publishing it on your homepage, through an available key server on the Internet, or any other available method. GnuPG has some useful options to help you publish your public key. Step 1 First off, we have to extract our public key in ASCII text to be able to distribute it. ASCII text is a good format to use because it allows people to get it easily. In this way, anyone can just cut and past your public key and use it when they want to securely communicate with you. • To extract your public key in ASCII armored output, use the following command: [root@deep /]# gpg --export –ao UID As an example: [root@deep /]# gpg --export –ao Gerhard Mourani Where “--export” is for extracting Public-key from your pubring encrypted file, “a” is to create ASCII armored output that you can mail, publish or put it on a web page, “o” to put the result in a file and UID represents the user key you want to export, which is in our example the user “Gerhard Mourani” key that we have create previously. Step 2 Once your public key has been extracted, the resulting output will be a file called “Gerhard” under the directory where you are issuing the above command, representing the First name of the user key to extract. In our example, the file is called “Gerhard” because it is the name of the key we want to export in ASCII text format. Note that the file name will be different for your public ASCII text format key. • Edit your public key in ASCII armored output format, and distribute it: [root@deep /]# vi Gerhard -----BEGIN PGP PUBLIC KEY BLOCK----- Version: GnuPG v1.0.7 (GNU/Linux) Comment: For info see http://www.gnupg.org mQGiBDzGNQcRBAC+1NrjFMCEtyjcv5lhtFNMLHEQ0VdHObv0CMUdCkiDslJ9QT9v MtVG1d4r3+0RJan23Z+8fc11E7Q0wRjRO13efRGEbxaIushhRc/p11LsEubWMWC7 E1UCsMmniScEdoZLSq9/myjj7IJqAavgL0a7/VkVHjrX1j/pTTK1wUUsRwCgy0jp 0JzY1+dIK4ElfGxAQ7oHop8D/03MkyVhUZh9asLW4tyGlmMN8exqfRoMdeSv0jnz ftAAZ71sn8jDdviccvaJvj2eTdZ7J43BIhxALJZ8KMQdEDWQnW62FfV9uGWcB5HL c869XOD0so9LOJGsgF1XpnMKQhTRXXEIuN0THpGDSLdQtXelBzIusQuSmNBrx7A0 6/5xA/0W3H2NYzvMWnTuENpHUQR8KtIARcmis4bGIH/fEiPQyR7YWIAs9sPOE5Yr 3cQuUpZ3nwGcZ5CGOKm0qRBkhMI49SO25gsoaRVVatNZ1v1o07AaNDimmvE0hhO3 +/LTv9cJYMdm4ijp+XOhssO4zctgdg0bHISsTWqB1AJcSsdAirQpR2VyaGFyZCBN
- 387. GnuPG 1 CHAPTER 4 387 b3VyYW5pIDxzeXNhZG1pbkBkZXYub3Blbm5hLmNvbT6IVwQTEQIAFwUCPMY1BwUL BwoDBAMVAwIDFgIBAheAAAoJEOTyFOEuU3j3OB8AoJcMlZkGYlHBt013kjg6U7Xt e7muAJ9LBfIlSHtmR3aZAn/4yekA8jwkrbkBDQQ8xjULEAQAvA7lwVx/AUga4j3d yo4upmHClk4+rYW9bQQXdMGj9EO2gdrxXzbQ2AlQj0UXgDN8HzXHdcZ4TyGghNVm zq9k2+Ud4Gx0+q34tJI+ljDM7eGhBZbSMGs7kB75/DKIvqONV2JCYJMutrRQPBF1 8ZRf/FgJEtOcjOHu5UfpMresWXsAAwYEAKj2b7LmSfPpm9X/eTEoHAFbR5WPXkRP eNUEgN2nk2rzyA+7IL4Sg9OPz31qhKOCh/NhFHKcg5VCS4bG35p78eb9KHr8CO01 +h1lUmqCf+s9UvHLUGJahnfp3lnFul9qBqK9MXvWd2bXfovHzAObC1kWAXuYmfnw 8RxdVSgFD4VyiEYEGBECAAYFAjzGNQsACgkQ5PIU4S5TePeMrwCgslkWPnwc3aTY xQnMq9ml/PdIhS0An1P917iFxhfP2mneemt4N6ELcF4E =7bvq -----ENDPGP PUBLIC KEY BLOCK----- WARNING: Never export or distribute your private key to the world. I know, this seem to be a stupid warning, but I’ve been informed that some people do it. Importing GPG key/s from a user: When you receive someone's public key (or some trusted third partly keys) you have to add them to your key database in order to be able to use his/her keys for future encryption, verification and authentication. This is often the case, when we install software that has a GPG key available for verification. Therefore, here is what you should do before installing software that has a GPG key to your disposal for authenticity. Step 1 First off, we have to retrieve the GPG public key of the company, organization, etc that we want to import into our keyring database. In our example, we will retrieve the GPG key that OpenNA uses to sign RPM packages and other software. This GPG public key is available from: http://www.openna.com/about/openna.asc. Cut and past it into a file called “openna.asc” on your server machine where GnuPG is installed. Step 2 Now, we have to import the OpenNA GPG public key into our database. This procedure should be done for any GPG public keys that you want to use to verify authenticity of software you want to install on your server. Most organizations have GPG public keys for you to download. • To import Public Keys to your keyring database, use the following command: [root@deep /]# gpg --import filename As an example: [root@deep /]# gpg --import openna.asc gpg: key 3487965A: public key imported gpg: /root/.gnupg/trustdb.gpg: trustdb created gpg: Total number processed: 1 gpg: imported: 1 The above command will append the new key “filename” into the keyring database and will update all already existing keys. It is important to note that GnuPG does not import keys that are not self-signed (asc).
- 388. GnuPG 1 CHAPTER 4 388 SigningGPG key/s from a user: When you import keys into your public keyring database and are sure that the trusted third party is really the person they claim, you can start signing his/her keys. Signing a key certifies that you know the owner of the keys and this leads to the situation where the signature acknowledges that the user ID mentioned in the key is actually the owner of that key. • To sign the key for company OpenNA that we have added into our keyring database above, use the following command: [root@deep /]# gpg --sign-key UID As an example: [root@deep /]# gpg --sign-key OpenNA pub 1024D/3487965A created: 2001-07-02 expires: never trust: -/q sub 1024g/0146F594 created: 2001-07-02 expires: never (1). OpenNA Inc. <noc@openna.com> pub 1024D/3487965A created: 2001-07-02 expires: never trust: -/q Fingerprint: 7A3D 6871 2DF1 9210 8ABE AF36 D460 86D5 3487 965A OpenNA Inc. <noc@openna.com> Are you really sure that you want to sign this key with your key: "Gerhard Mourani <gmourani@openna.com>" Really sign? y You need a passphrase to unlock the secret key for user: "Gerhard Mourani <gmourani@openna.com>" 1024-bit DSA key, ID 2E5378F7, created 2002-04-24 Enter passphrase: WARNING: You should only sign a key as being authentic when you are ABSOLUTELY SURE that the key is really authentic! You should never sign a key based on any kind of assumption. Checking GPG signature: We have shown above how to sign a key, now we will explain how people can verify if the signature is really the good one. Once you have extracted your public key and exported it, everyone who knows or gets your public key should be able to check whether encrypted data from you is also really signed by you. • To check the signature of encrypted data, use the following command: [root@deep /]# gpg --verify Data The “--verify” option will check the signature where Data is the encrypted data/file you want to verify.
- 389. GnuPG 1 CHAPTER 4 389 Encryptingand decrypting GPG files: After installing, importing, signing and configuring everything in the way that we want, we can start encrypting and decrypting our files, software, etc. • To encrypt and sign data for the user OpenNA that we have added on our keyring database above, use the following command: [root@deep /]# gpg -sear OpenNA file As an example: [root@deep /]# gpg -sear OpenNA Message-to-OpenNA.txt You need a passphrase to unlock the secret key for user: "Gerhard Mourani <gmourani@openna.com>" 1024-bit DSA key, ID 2E5378F7, created 2002-04-24 Enter passphrase: Of the arguments passed, the “s” is for signing (To avoid the risk that somebody else claims to be you, it is very useful to sign everything you encrypt), “e” for encrypting, “a” to create ASCII armored output (“.asc” ready for sending by mail), “r” to encrypt the UID name and “file” is the message you want to encrypt. • To decrypt data, use the following command: [root@deep /]# gpg -d file For example: [root@deep /]# gpg -d Message-from-GerhardMourani.asc You need a passphrase to unlock the secret key for user: "Gerhard Mourani (Open Network Architecture) <gmourani@openna.com>" 1024-bit DSA key, ID 2E5378F7, created 2002-04-24 Enter passphrase: Where “d” is for decrypting and “file” is the message you want to decrypt. It is important that the public key of the sender of the message we want to decrypt be in our public keyring database or of course nothing will work. Further documentation For more details, there are some manual pages about GnuPG that you could read: $ man gpg (1) - GPG encryption and signing tool. $ man gpgv (1) - GPGV signature verification tool.
- 391. OpenSSL IN THIS CHAPTER 1.Compiling - Optimizing & Installing OpenSSL 2. Configuring OpenSSL 3. OpenSSL Administrative Tools 4. Securing OpenSSL
- 393. OpenSSL 1 CHAPTER 5 393 LinuxOpenSSL Abstract The majority of Internet protocols like IMAP, POP, SQL, SMTP, SMB, HTTP, FTP, and LDAP, provide now support for SSL encryption. The big problem in the past was that they asked users to authenticate themselves before allowing access to services, and then they would transmit the users’ login ID’s and passwords in plain text format over the network, allowing external crackers, using sniffer tools, to get the information and log in into the server themselves. Encryption mechanisms like SSL ensure safe and secure transactions to eliminate this problem. With this technology, data going over the network is point-to-point encrypted. OpenSSL is a free implementation of this SSL support for all Internet protocols that could run with it (most now do). Once OpenSSL has been installed on your Linux server you can use it as a third party tool to enable SSL functionality with other applications. The OpenSSL Project is a collaborative effort to develop a robust, commercial-grade, fully featured, and Open Source toolkit implementing the Secure Sockets Layer (SSL v2/v3) and Transport Layer Security (TLS v1) protocols with full-strength cryptography. In this chapter, we’ll show you how to install OpenSSL for your servers, and how to use it to create certificate keys used by third party software to provide SSL support for, and encryption of, usernames and passwords. Most of the software described in this book needs the presence of OpenSSL on the system to be able to be compiled with SSL support. Therefore, I strongly recommend that you install this encryption software on your Linux system. Summary of the Cryptographic Thechnology.
- 394. OpenSSL 1 CHAPTER 5 394 CryptographyAdvantages The main advantages gained by using encryption technology are: Data Confidentiality When a message is encrypted, an algorithm converts it into enciphered text that hides the meaning of the message, which can then be sent via any public mechanism, and transforms the input plain text. This process involves a secret key that is used to encrypt and later decrypt the data. Without the secret key, the encrypted data is meaningless. Data Integrity A cryptographic checksum, called a Message Authentication Code (MAC), can be calculated on an arbitrarily user-supplied text to protect the integrity of the data. The results (text and MAC) are then sent to the receiver who can verify the trial MAC appended to a message by recalculating the MAC for the message, using the appropriate secret key and verifying that it matches exactly the trial MAC. Authentication Personal identification is another use of cryptography, where the user/sender knows a secret, which can serve to authenticate his or her identity. Electronic Signature A digital signature assures the sender and receiver that the message is authentic and that only the owner of the key could have generated the digital signature. Disclaimer This software package uses strong cryptography, so even if it is created, maintained and distributed from liberal countries in Europe (where it is legal to do this), it falls under certain export/import and/or use restrictions in some other parts of the world. PLEASE REMEMBER THAT EXPORT/IMPORT AND/OR USE OF STRONG CRYPTOGRAPHY SOFTWARE, PROVIDING CRYPTOGRAPHY HOOKS OR EVEN JUST COMMUNICATING TECHNICAL DETAILS ABOUT CRYPTOGRAPHY SOFTWARE IS ILLEGAL IN SOME PARTS OF THE WORLD. SO, WHEN YOU IMPORT THIS PACKAGE TO YOUR COUNTRY, RE- DISTRIBUTE IT FROM THERE OR EVEN JUST EMAIL TECHNICAL SUGGESTIONS OR EVEN SOURCE PATCHES TO THE AUTHOR OR OTHER PEOPLE YOU ARE STRONGLY ADVISED TO PAY CLOSE ATTENTION TO ANY EXPORT/IMPORT AND/OR USE LAWS WHICH APPLY TO YOU. THE AUTHORS OF OPENSSL ARE NOT LIABLE FOR ANY VIOLATIONS YOU MAKE HERE. SO BE CAREFUL, RESPONSIBILITY IS YOURS.
- 395. OpenSSL 1 CHAPTER 5 395 Theseinstallation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, at personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest OpenSSL version number is 0.9.6d The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages The following is based on information listed by OpenSSL as of 2002/05/09. Please check http://www.openssl.org/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: OpenSSL Homepage: http://www.openssl.org/ OpenSSL FTP Site: 129.132.7.170 You must be sure to download: openssl-0.9.6d.tar.gz Pristine source If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install OpenSSL, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > OpenSSL1 • And the following one after you install the software: [root@deep root]# find /* > OpenSSL2 • Then use the following command to get a list of what changed: [root@deep root]# diff OpenSSL1 OpenSSL2 > OpenSSL-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In our example above, we use the /root directory of the system to store all the generated file lists.
- 396. OpenSSL 1 CHAPTER 5 396 Compiling- Optimizing & Installing OpenSSL Below are the steps that you must make to configure, compile and optimize the OpenSSL software before installing it into your Linux system. First off, we install the program as user 'root' so as to avoid authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp openssl-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf openssl-version.tar.gz Step 2 Next, move into the newly created OpenSSL source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created OpenSSL directory use the following command: [root@deep tmp]# cd openssl-0.9.6d/ Step 3 With OpenSSL, the optimization FLAGS should be changed in the “Configure” file of the program. It is in this file that we define the GCC optimizations we want to use related to the type of processor running in our system. OpenSSL is cryptography software and there are some optimization hacks that we can make that can significantly increase the performance of the program, therefore take the time to modify the “Configure” file of the software. This will be a benefit for you. a) Edit the Configure file (vi +337 Configure) and change the following lines: "linux-elf", "gcc:-DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -m486 -Wall::- D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:- fPIC:.so.$(SHLIB_MAJOR).$(SHLIB_MINOR)", To read: "linux-elf", "gcc:-DL_ENDIAN -DTERMIO -O3 -march=i686 -funroll-loops -fomit- frame-pointer -Wall::-D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn:linux-shared:- fPIC:.so.$(SHLIB_MAJOR).$(SHLIB_MINOR)", b) Edit the Configure file (vi +338 Configure) and change the following lines: "debug-linux-elf","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG - DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -m486 -Wall::-D_REENTRANT:-lefence - ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn", To read: "debug-linux-elf","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG - DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -O3 -march=i686 -funroll-loops -fomit-frame- pointer -Wall::-D_REENTRANT:-lefence -ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn",
- 397. OpenSSL 1 CHAPTER 5 397 c)Edit the Configure file (vi +339 Configure) and change the following lines: "debug-linux-elf-noefence","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -g -m486 -Wall::-D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn", To read: "debug-linux-elf-noefence","gcc:-DBN_DEBUG -DREF_CHECK -DCONF_DEBUG -DBN_CTX_DEBUG -DCRYPTO_MDEBUG -DL_ENDIAN -DTERMIO -O3 -march=i686 -funroll-loops -fomit-frame- pointer -Wall::-D_REENTRANT:-ldl:BN_LLONG ${x86_gcc_des} ${x86_gcc_opts}:${x86_elf_asm}:dlfcn", Step 4 By default, OpenSSL source files assume that our “perl” binary program is located under /usr/local/bin/perl. We must change this to reflect our environment variable. • To point all OpenSSL script files to our “perl” binary, use the following command: [root@deep openssl-0.9.6d]# perl util/perlpath.pl /usr/bin/perl Step 5 At this stage, it is time to configure and compile OpenSSL for our system. • To configure and optimize OpenSSL use the following compilation lines: ./Configure linux-elf no-asm shared --prefix=/usr --openssldir=/usr/share/ssl Step 6 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the OpenSSL software: [root@deep openssl-0.9.6d]# LD_LIBRARY_PATH=`pwd` make all build-shared [root@deep openssl-0.9.6d]# LD_LIBRARY_PATH=`pwd` make test apps tests [root@deep openssl-0.9.6d]# cd [root@deep root]# find /* > OpenSSL1 [root@deep root]# cd /var/tmp/openssl-0.9.6d/ [root@deep openssl-0.9.6d]# make install build-shared [root@deep openssl-0.9.6d]# cd /usr/lib/ [root@deep lib]# mv libcrypto.so.0.9.6 ../../lib/ [root@deep lib]# mv libssl.so.0.9.6 ../../lib/ [root@deep lib]# ln -sf ../../lib/libcrypto.so.0.9.6 libcrypto.so [root@deep lib]# ln -sf ../../lib/libcrypto.so.0.9.6 libcrypto.so.0 [root@deep lib]# ln -sf ../../lib/libssl.so.0.9.6 libssl.so [root@deep lib]# ln -sf ../../lib/libssl.so.0.9.6 libssl.so.0 [root@deep lib]# mv /usr/share/ssl/man/man1/* /usr/share/man/man1/ [root@deep lib]# mv /usr/share/ssl/man/man3/* /usr/share/man/man3/ [root@deep lib]# mv /usr/share/ssl/man/man5/* /usr/share/man/man5/ [root@deep lib]# mv /usr/share/ssl/man/man7/* /usr/share/man/man7/ [root@deep lib]# rm -rf /usr/share/ssl/man/ [root@deep lib]# rm -rf /usr/share/ssl/lib/ [root@deep lib]# strip /usr/bin/openssl [root@deep lib]# mkdir -p /usr/share/ssl/crl [root@deep lib]# cd [root@deep root]# find /* > OpenSSL2 [root@deep root]# diff OpenSSL1 OpenSSL2 > OpenSSL-Installed
- 398. OpenSSL 1 CHAPTER 5 398 Theabove commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then test the OpenSSL libraries to finally install the binaries and any supporting files into the appropriate locations. Step 7 Once the configuration, optimization, compilation, and installation of the OpenSSL software has completed, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete OpenSSL and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf openssl-version/ [root@deep tmp]# rm -f openssl-version.tar.gz Configuring OpenSSL After OpenSSL has been built and installed successfully on your system, your next step is to configure and customize the openssl.cnf and sign files to suit your needs. /usr/shared/ssl/openssl.cnf (The OpenSSL Configuration File) /usr/shared/ssl/misc/sign (A CA script file to sign certificates) /usr/shared/ssl/openssl.cnf: The OpenSSL Configuration File This is the general configuration file for OpenSSL, where you can configure the expiration date of your keys, the name of your organization, address and so on. The most important parameters you may need to change will be in the [ CA_default ] and especially the [ req_distinguished_name ] sections of the file. We must change the default one to fit our requirements and operating system. The text in bold is the parts of the configuration file that must be customized and adjusted to satisfy our needs. • Edit the openssl.cnf file (vi /usr/share/ssl/openssl.cnf) and set your needs. # # OpenSSL example configuration file. # This is mostly being used for generation of certificate requests. # # This definition stops the following lines choking if HOME isn't # defined. HOME = . RANDFILE = $ENV::HOME/.rnd # Extra OBJECT IDENTIFIER info: #oid_file = $ENV::HOME/.oid oid_section = new_oids # To use this configuration file with the "-extfile" option of the # "openssl x509" utility, name here the section containing the # X.509v3 extensions to use: # extensions = # (Alternatively, use a configuration file that has only # X.509v3 extensions in its main [= default] section.)
- 399. OpenSSL 1 CHAPTER 5 399 [new_oids ] # We can add new OIDs in here for use by 'ca' and 'req'. # Add a simple OID like this: # testoid1=1.2.3.4 # Or use config file substitution like this: # testoid2=${testoid1}.5.6 #################################################################### [ ca ] default_ca = CA_default # The default ca section #################################################################### [ CA_default ] dir = /usr/share/ssl # Where everything is kept certs = $dir/certs # Where the issued certs are kept crl_dir = $dir/crl # Where the issued crl are kept database = $dir/ca.db.index # database index file. new_certs_dir = $dir/ca.db.certs # default place for new certs. certificate = $dir/certs/ca.crt # The CA certificate serial = $dir/ca.db.serial # The current serial number crl = $dir/crl.pem # The current CRL private_key = $dir/private/ca.key # The private key RANDFILE = $dir/ca.db.rand # private random number file x509_extensions = usr_cert # The extentions to add to the cert # Extensions to add to a CRL. Note: Netscape communicator chokes on V2 CRLs # so this is commented out by default to leave a V1 CRL. # crl_extensions = crl_ext default_days = 365 # how long to certify for default_crl_days= 30 # how long before next CRL default_md = md5 # which md to use. preserve = no # keep passed DN ordering # A few difference way of specifying how similar the request should look # For type CA, the listed attributes must be the same, and the optional # and supplied fields are just that :-) policy = policy_match # For the CA policy [ policy_match ] countryName = match stateOrProvinceName = match organizationName = match organizationalUnitName = optional commonName = supplied emailAddress = optional # For the 'anything' policy # At this point in time, you must list all acceptable 'object' # types. [ policy_anything ] countryName = optional stateOrProvinceName = optional localityName = optional organizationName = optional organizationalUnitName = optional commonName = supplied emailAddress = optional
- 400. OpenSSL 1 CHAPTER 5 400 #################################################################### [req ] default_bits = 1024 default_keyfile = privkey.pem distinguished_name = req_distinguished_name attributes = req_attributes x509_extensions = v3_ca # The extentions to add to the self signed cert # Passwords for private keys if not present they will be prompted for # input_password = secret # output_password = secret # This sets a mask for permitted string types. There are several options. # default: PrintableString, T61String, BMPString. # pkix : PrintableString, BMPString. # utf8only: only UTF8Strings. # nombstr : PrintableString, T61String (no BMPStrings or UTF8Strings). # MASK:XXXX a literal mask value. # WARNING: current versions of Netscape crash on BMPStrings or UTF8Strings # so use this option with caution! string_mask = nombstr # req_extensions = v3_req # The extensions to add to a certificate request [ req_distinguished_name ] countryName = Country Name (2 letter code) countryName_default = CA countryName_min = 2 countryName_max = 2 stateOrProvinceName = State or Province Name (full name) stateOrProvinceName_default = Quebec localityName = Locality Name (eg, city) localityName_default = Montreal 0.organizationName = Organization Name (eg, company) 0.organizationName_default = OpenNA, Inc. # we can do this but it is not needed normally :-) #1.organizationName = Second Organization Name (eg, company) #1.organizationName_default = World Wide Web Pty Ltd organizationalUnitName = Organizational Unit Name (eg, section) organizationalUnitName_default = Open Network Architecture commonName = Common Name (eg, YOUR name) commonName_default = www.openna.com commonName_max = 64 emailAddress = Email Address emailAddress_default = noc@openna.com emailAddress_max = 40 # SET-ex3 = SET extension number 3 [ req_attributes ] challengePassword = A challenge password challengePassword_min = 8 challengePassword_max = 20 unstructuredName = An optional company name
- 401. OpenSSL 1 CHAPTER 5 401 [usr_cert ] # These extensions are added when 'ca' signs a request. # This goes against PKIX guidelines but some CAs do it and some software # requires this to avoid interpreting an end user certificate as a CA. basicConstraints=CA:FALSE # Here are some examples of the usage of nsCertType. If it is omitted # the certificate can be used for anything *except* object signing. # This is OK for an SSL server. # nsCertType = server # For an object signing certificate this would be used. # nsCertType = objsign # For normal client use this is typical # nsCertType = client, email # and for everything including object signing: # nsCertType = client, email, objsign # This is typical in keyUsage for a client certificate. # keyUsage = nonRepudiation, digitalSignature, keyEncipherment # This will be displayed in Netscape's comment listbox. nsComment = "OpenSSL Generated Certificate" # PKIX recommendations harmless if included in all certificates. subjectKeyIdentifier=hash authorityKeyIdentifier=keyid,issuer:always # This stuff is for subjectAltName and issuerAltname. # Import the email address. # subjectAltName=email:copy # Copy subject details # issuerAltName=issuer:copy #nsCaRevocationUrl = http://www.domain.dom/ca-crl.pem #nsBaseUrl #nsRevocationUrl #nsRenewalUrl #nsCaPolicyUrl #nsSslServerName [ v3_req ] # Extensions to add to a certificate request basicConstraints = CA:FALSE keyUsage = nonRepudiation, digitalSignature, keyEncipherment [ v3_ca ] # Extensions for a typical CA # PKIX recommendation.
- 402. OpenSSL 1 CHAPTER 5 402 subjectKeyIdentifier=hash authorityKeyIdentifier=keyid:always,issuer:always #This is what PKIX recommends but some broken software chokes on critical # extensions. #basicConstraints = critical,CA:true # So we do this instead. basicConstraints = CA:true # Key usage: this is typical for a CA certificate. However since it will # prevent it being used as an test self-signed certificate it is best # left out by default. # keyUsage = cRLSign, keyCertSign # Some might want this also # nsCertType = sslCA, emailCA # Include email address in subject alt name: another PKIX recommendation # subjectAltName=email:copy # Copy issuer details # issuerAltName=issuer:copy # DER hex encoding of an extension: beware experts only! # obj=DER:02:03 # Where 'obj' is a standard or added object # You can even override a supported extension: # basicConstraints= critical, DER:30:03:01:01:FF [ crl_ext ] # CRL extensions. # Only issuerAltName and authorityKeyIdentifier make any sense in a CRL. # issuerAltName=issuer:copy authorityKeyIdentifier=keyid:always,issuer:always WARNING: You don’t need to change all of the default options set in the file openssl.cnf; The configurations you usually change will be into the [ CA_default ] and [ req_distinguished_name ] sections of the file. /usr/share/ssl/misc/sign: The CA Script File to Sign Certificates OpenSSL CA command has some strange requirements and the default OpenSSL config doesn't allow one to easily use OpenSSL CA directly. It is for this reason that we don’t use the files CA.pl or CA.sh to sign certificates. Step 1 To solve the problem, we’ll create and customize the sign script file below to replace them. Text in bold are the parts of the script that must be customized and adjusted to satisfy our needs. • Create the sign script file (touch /usr/share/ssl/misc/sign) and add the following lines: #!/bin/sh ##
- 403. OpenSSL 1 CHAPTER 5 403 ##sign.sh -- Sign a SSL Certificate Request (CSR) ## Copyright (c) 1998-1999 Ralf S. Engelschall, All Rights Reserved. ## # argument line handling CSR=$1 if [ $# -ne 1 ]; then echo "Usage: sign.sign <whatever>.csr"; exit 1 fi if [ ! -f $CSR ]; then echo "CSR not found: $CSR"; exit 1 fi case $CSR in *.csr ) CERT="`echo $CSR | sed -e 's/.csr/.crt/'`" ;; * ) CERT="$CSR.crt" ;; esac # make sure environment exists if [ ! -d ca.db.certs ]; then mkdir ca.db.certs fi if [ ! -f ca.db.serial ]; then echo '01' >ca.db.serial fi if [ ! -f ca.db.index ]; then cp /dev/null ca.db.index fi # create an own SSLeay config cat >ca.config <<EOT [ ca ] default_ca = CA_own [ CA_own ] dir = /usr/share/ssl certs = /usr/share/ssl/certs new_certs_dir = /usr/share/ssl/ca.db.certs database = /usr/share/ssl/ca.db.index serial = /usr/share/ssl/ca.db.serial RANDFILE = /usr/share/ssl/ca.db.rand certificate = /usr/share/ssl/certs/ca.crt private_key = /usr/share/ssl/private/ca.key default_days = 365 default_crl_days = 30 default_md = md5 preserve = no policy = policy_anything [ policy_anything ] countryName = optional stateOrProvinceName = optional localityName = optional organizationName = optional organizationalUnitName = optional commonName = supplied emailAddress = optional EOT # sign the certificate echo "CA signing: $CSR -> $CERT:" openssl ca -config ca.config -out $CERT -infiles $CSR echo "CA verifying: $CERT <-> CA cert" openssl verify -CAfile /usr/share/ssl/certs/ca.crt $CERT # cleanup after SSLeay
- 404. OpenSSL 1 CHAPTER 5 404 rm-f ca.config rm -f ca.db.serial.old rm -f ca.db.index.old # die gracefully exit 0 Step 2 Once the script file has been created, it is important to make it executable and change its default permissions. Making this file executable will allow the system to run it, changing its default permission is to allow only the root user to change this file for security reason. • To make this script executable and to change its default permissions, use the commands: [root@deep /]# chmod 700 /usr/share/ssl/misc/sign [root@deep /]# chown 0.0 /usr/share/ssl/misc/sign OpenSSL Administrative Tools Once your configuration options have been set in the openssl.cnf file, we can play with the OpenSSL utility. As an example, we’ll show you how to create certificates for the Apache web server and your own CA (Certifying Authority) to sign your “Certificate Signing Request” yourself. All commands listed below are to be made in the /usr/share/ssl directory. The Apache Key & CSR Generation: The utility “openssl” that you use to generate RSA Private Keys (Key) and Certificate Signing Requests (CSR) comes with OpenSSL and is usually installed under the directory /usr/bin on our Linux distribution. Below are the steps to create certificates for Apache. Step 1 First you have to know the Fully Qualified Domain Name (FQDN) of the website/server for which you want to request a certificate. When you want to access your website/server through https://www.mydomain.com/ then the FQDN of your website is www.mydomain.com. Step 2 Second, select five large and relatively random files from your hard drive (compressed log files are a good start) and put them under your /usr/share/ssl directory. These will act as your random seed enhancers. We refer to them as random1: random2:...: random5 below. • To select five random files and put them under /usr/share/ssl, use the commands: [root@deep /]# cp /var/log/boot.log /usr/share/ssl/random1 [root@deep /]# cp /var/log/cron /usr/share/ssl/random2 [root@deep /]# cp /var/log/dmesg /usr/share/ssl/random3 [root@deep /]# cp /var/log/messages /usr/share/ssl/random4 [root@deep /]# cp /var/log/secure /usr/share/ssl/random5
- 405. OpenSSL 1 CHAPTER 5 405 Step3 Third, create the RSA private key protected with a pass-phrase for your web server. The command below will generate 1024 bit RSA Private Key and store it in www.mydomain.com.key. It will ask you for a pass-phrase: use something secure and remember it. Your certificate will be useless without the key. If you don't want to protect your key with a pass-phrase (only if you absolutely trust that server machine, and you make sure the permissions are carefully set so only you can read that key) you can leave out the -des3 option below. • To generate the Key, use the following command: [root@deep /]# cd /usr/share/ssl/ [root@deep ssl]# openssl genrsa -des3 –rand random1:random2:random3:random4:random5 -out www.mydomain.com.key 1024 123600 semi-random bytes loaded Generating RSA private key, 1024 bit long modulus ......................+++++ .....+++++ e is 65537 (0x10001) Enter PEM pass phrase: Verifying password - Enter PEM pass phrase: WARNING: Please backup your www.mydomain.com.key file and remember the pass-phrase you had to enter at a secure location. A good choice is to backup this information onto a diskette or other removable media. Step 4 Finally, generate a Certificate Signing Request (CSR) with the server RSA private key. The command below will prompt you for the X.509 attributes of your certificate. Remember to give the name “www.mydomain.com” when prompted for ‘Common Name'. Do not enter your personal name here. We are requesting a certificate for a web server, so the ‘Common Name’ has to match the FQDN of your website (a requirement of the browsers). • To generate the CSR, use the following command: [root@deep ssl]# openssl req -new -key www.mydomain.com.key -out www.mydomain.com.csr Using configuration from /usr/share/ssl/openssl.cnf Enter PEM pass phrase: You are about to be asked to enter information that will be incorporated into your certificate request. What you are about to enter is what is called a Distinguished Name or a DN. There are quite a few fields but you can leave some blank For some fields there will be a default value, If you enter '.', the field will be left blank. ----- Country Name (2 letter code) [CA]: State or Province Name (full name) [Quebec]: Locality Name (eg, city) [Montreal]: Organization Name (eg, company) [OpenNA, Inc.]: Organizational Unit Name (eg, section) [Open Network Architecture]: Common Name (eg, YOUR name) [www.openna.com]: Email Address [noc@openna.com]: Please enter the following 'extra' attributes
- 406. OpenSSL 1 CHAPTER 5 406 tobe sent with your certificate request A challenge password []:. An optional company name []:. WARNING: Make sure you enter the FQDN (Fully Qualified Domain Name) of the server when OpenSSL prompts you for the “Common Name” (i.e. when you generate a CSR for a website which will be later accessed via https://www.mydomain.com/, enter “www.mydomain.com” here). After the generation of your Certificate Signing Request (CSR), you must send this certificate to a commercial Certifying Authority (CA) like Thawte or Verisign for signing. You usually have to post the CSR into a web form, pay for the signing, await the signed certificate and store it into a “www.mydomain.com.crt” file. The result is then a real certificate, which can be used with Apache. The CA Key & CRT Generation: If you don’t want to pay a commercial Certifying Authority (CA) to sign you certificates, you can use your own CA and now have to sign the CSR yourself by this CA. This solution is economical, and allows an organization to host their own CA server and generate as many certificates as they need for internal use without paying a cent to a commercial CA. Unfortunately using your own CA to generate certificates causes problems in electronic commerce, because customers need to have some trust in your organization by the use of a recognized commercial CA. See below on how to sign a CSR with your CA yourself. Step 1 As for the Apache web server above, the first step is to create the RSA private key protected with a pass-phrase for your CA. The command below will generate 1024 bit RSA Private Key and stores it in the file “ca.key”. It will ask you for a pass-phrase: use something secure and remember it. Your certificate will be useless without the key. • To create the RSA private key for your (CA), use the following command: [root@deep /]# cd /usr/share/ssl/ [root@deep ssl]# openssl genrsa -des3 -out ca.key 1024 Generating RSA private key, 1024 bit long modulus ...........................+++++ ............................................+++++ e is 65537 (0x10001) Enter PEM pass phrase: Verifying password - Enter PEM pass phrase: WARNING: Please backup your “ca.key” file and remember the pass-phrase you had to enter at a secure location. A good choice is to backup this information onto a diskette or other removable media.
- 407. OpenSSL 1 CHAPTER 5 407 Step2 Now, we must create a self-signed (CA) certificate (x509 structure) with the RSA key of the CA. The req command creates a self-signed certificate when the -x509 switch is used. • To create a self-signed (CA) certificate, use the following command: [root@deep ssl]# openssl req -new -x509 -days 365 -key ca.key -out ca.crt Using configuration from /usr/share/ssl/openssl.cnf Enter PEM pass phrase: You are about to be asked to enter information that will be incorporated into your certificate request. What you are about to enter is what is called a Distinguished Name or a DN. There are quite a few fields but you can leave some blank For some fields there will be a default value, If you enter '.', the field will be left blank. ----- Country Name (2 letter code) [CA]: State or Province Name (full name) [Quebec]: Locality Name (eg, city) [Montreal]: Organization Name (eg, company) [OpenNA, Inc.]: Organizational Unit Name (eg, section) [Open Network]:Sales Dept Common Name (eg, YOUR name) [www.openna.com]: Email Address [noc@openna.com]:sales@openna.com Step 3 Once the self-signed (CA) certificate has been created, we must place all certificates and CA files into their appropriate directories. • To place the files into their appropriate directories, use the following commands: [root@deep ssl]# mv www.mydomain.com.key private/ [root@deep ssl]# mv ca.key private/ [root@deep ssl]# mv ca.crt certs/ Step 4 Finally, you can use this CA to sign all the servers CSR's in order to create real SSL Certificates for use inside the web server (assuming you already have a www.mydomain.com.csr at hand). We must also prepare the script “sign” for signing. • To sign server CSR's in order to create real SSL Certificates, use the following command: [root@deep ssl]# /usr/share/ssl/misc/sign www.mydomain.com.csr CA signing: www.mydomain.com.csr -> www.mydomain.com.crt: Using configuration from ca.config Enter PEM pass phrase: Check that the request matches the signature Signature ok The Subjects Distinguished Name is as follows countryName :PRINTABLE:'CA' stateOrProvinceName :PRINTABLE:'Quebec' localityName :PRINTABLE:'Montreal' organizationName :PRINTABLE:'OpenNA, Inc.' organizationalUnitName :PRINTABLE:'Open Network Architecture' commonName :PRINTABLE:'www.openna.com' emailAddress :IA5STRING:'noc@openna.com' Certificate is to be certified until Oct 18 14:59:29 2001 GMT (365 days) Sign the certificate? [y/n]:y
- 408. OpenSSL 1 CHAPTER 5 408 1out of 1 certificate requests certified, commit? [y/n]y Write out database with 1 new entries Data Base Updated CA verifying: www.mydomain.com.crt <-> CA cert www.mydomain.com.crt: OK This signs the CSR and results in a “www.mydomain.com.crt” file. Move this file to its appropriate directory as follows. • To move the CRT file to its appropriate directory, use the following command: [root@deep ssl]# mv www.mydomain.com.crt certs/ Now you have two files: “www.mydomain.com.key” and “www.mydomain.com.crt”. These can now, for example, be used as follows, inside the virtual host section of your Apache server's httpd.conf file: SSLCertificateFile /usr/share/ssl/certs/www.mydomain.com.crt SSLCertificateKeyFile /usr/share/ssl/private/www.mydomain.com.key In this example, www.mydomain.com.crt is our web server Certificate Signing Request Public Key, and www.mydomain.com.key is our web server RSA Private Key. The www.mydomain.com.csr file is no longer needed; we can remove it from the system. • To remove this file from the system, use the following command: [root@deep ssl]# rm -f www.mydomain.com.csr WARNING: If you receive an error message during the signing of the certificate, it’s probably because you’ve entered the wrong FQDN (Fully Qualified Domain Name) for the server when OpenSSL prompted you for the “Common Name”; the “Common Name” must be something like “www.mydomain.com” and not “mydomain.com”. Also, since you generate both the certificate and the CA certificate, it’s important that at least ONE piece of information differs between both files, or you may encounter problems during the signature of the certificate request.
- 409. OpenSSL 1 CHAPTER 5 409 SecuringOpenSSL This small section deals specifically with actions we can take to improve and tighten security under OpenSSL. It’s important to note that we refer to the features available within the base installed program and not to any additional software. Changing the default mode of OpenSSL keys: Make your keys “Read and Write” only by the super-user “root”. This is important because no one needs to touch these files. • To make your keys “read and Write” only by “root”, use the following commands: [root@deep /]# chmod 750 /usr/share/ssl/private/ [root@deep /]# chmod 400 /usr/share/ssl/certs/ca.crt [root@deep /]# chmod 400 /usr/share/ssl/certs/www.mydomain.com.crt [root@deep /]# chmod 400 /usr/share/ssl/private/ca.key [root@deep /]# chmod 400 /usr/share/ssl/private/www.mydomain.com.key Some possible uses of OpenSSL software OpenSSL can be used to: 1. Creation of your own Certifying Authority Server. 2. Creation of RSA, DH and DSA key parameters. 3. Creation of X.509 certificates, CSRs and CRLs. 4. Calculation of Message Digest. 5. Encryption and Descryptiion with Ciphers. 6. SSL/TLS Client and Server Tests. 7. Handling of S/MIME signed or encrypted mail. 8. Provide data confidentiality, integrity, authentication, and electronic signature in transmission for the users. 9. Secure electronic commerce transactions.
- 411. OpenSSH IN THIS CHAPTER 1.Compiling - Optimizing & Installing OpenSSH 2. Configuring OpenSSH 3. Running OpenSSH in a chroot jail 4. Creating OpenSSH private & public keys 5. OpenSSH Users Tools
- 413. OpenSSH 1 CHAPTER 6 413 LinuxOpenSSH Abstract As illustrated in the chapter related to the Linux installation, many network services including, but not limited to, telnet, rsh, rlogin, or rexec are vulnerable to electronic eavesdropping. As a consequence, anyone who has access to any machine connected to the network can listen in on its network communications and get your password, as well as any other private information that is sent over the network in plain text. Currently the Telnet program is indispensable for daily administration tasks, but it is insecure since it transmits your password in plain text over the network and allows any listener to capture your password and then use your account to do anything he likes. To solve this problem we must find either another way, or another program, to replace it. Fortunately OpenSSH is a truly seamless and secure replacement of old, insecure and obsoletes remote login programs such as telnet, rlogin, rsh, rdist, or rcp. SSH (Secure Shell) is a program to log into another computer over a network, to execute commands on a remote machine, and to move files from one machine to another. It provides strong authentication and secure communications over insecure channels. It is intended as a replacement for rlogin, rsh, rcp, and rdist. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, at personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest OpenSSH version number is 3.4p1 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages The following is based on information listed by OpenSSH as of 2002/06/26. Please check http://www.openssh.com/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: OpenSSH Homepage: http://www.openssh.com/ OpenSSH FTP Site: 129.128.5.191 You must be sure to download: openssh-3.4p1.tar.gz NOTE: Don't forget to download the portable version (the p suffix) of OpenSSH tarball for Linux. There is strictly OpenBSD-based development of this software and another one known as portable version, which runs on many operating systems (these are known as the p releases, and named like "OpenSSH 3.4p1").
- 414. OpenSSH 1 CHAPTER 6 414 Prerequisites OpenSSHrequires that the listed software below be already installed on your system to be able to compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source archive files. Please make sure you have this program installed on your machine before you proceed with this chapter. OpenSSL is required to run OpenSSH on your system. NOTE: For more information on OpenSSL software, see its related chapter in this book. Even if you don’t need to use OpenSSL software to create or hold encrypted key files, it’s important to note that OpenSSH requires its libraries files to be able to work. Pristine source As we don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install OpenSSH, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > OpenSSH1 • And the following one after you install the software: [root@deep root]# find /* > OpenSSH2 • Then use the following command to get a list of what changed: [root@deep root]# diff OpenSSH1 OpenSSH2 > OpenSSH-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In our example above, we use the /root directory of the system to store all the generated file lists. Compiling - Optimizing & Installing OpenSSH Below are the steps that you must make to configure, compile and optimize the OpenSSH server software before installing it on your system. First off, we install the program as user 'root' so as to avoid any authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp openssh-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf openssh-version.tar.gz
- 415. OpenSSH 1 CHAPTER 6 415 Step2 In order to check that the version of OpenSSH, which you are, going to install, is an original and unmodified one, please check the supplied signature with the GPG key of OpenSSH available on the OpenSSH website. To get a GPG key copy of OpenSSH, please point your browser to the following URL: ftp://ftp.openbsd.org/pub/OpenBSD/OpenSSH/portable/openssh-3.4p1.tar.gz.sig. For more information about how to use this key for verification, see the GnuPG chapter in this book. Step 3 OpenSSH needs a UID and GID to properly run on the system but this UID/GID cannot run as super-user root; for this reason we must create a special user with no shell privileges on the system for running sshd daemon. This is required by the privilege separation feature of OpenSSH by which operations that require root privilege are performed by a separate privileged monitor process. • To create this special OpenSSH user on OpenNA Linux, use the following command: [root@deep tmp]# groupadd -g 39 sshd > /dev/null 2>&1 || : [root@deep tmp]# useradd -c "SSH Server" -d /var/empty -g 39 -s /bin/false -u 39 sshd > /dev/null 2>&1 || : • To create this special OpenSSH user on Red Hat Linux, use the following command: [root@deep tmp]# groupadd -g 39 sshd > /dev/null 2>&1 || : [root@deep tmp]# useradd -u 39 -g 39 -s /bin/false -M -r -d /var/empty sshd > /dev/null 2>&1 || : The above command will create a null account, with no password, no valid shell, no files owned- nothing but a UID and a GID for the program. Remember that OpenSSH daemon does not need to have a shell account on the server. Step 4 Now, edit the shells file (vi /etc/shells) and add a non-existent shell name “/bin/false”, which is the one we used in the useradd command above. [root@deep tmp]# vi /etc/shells /bin/bash2 /bin/bash /bin/sh /bin/false This is our added no-existent shell Patching OpenSSH to run in chroot jail mode for some users: There is an external patch available for OpenSSH that allow us to compile OpenSSH with chroot jail support. If you are interested in compiling OpenSSH to run in chroot jail environment for some of your users, then I recommend you to follow these steps. If you don’t want to compile OpenSSH with chroot jail support, you can simply skip these steps and go directly to next section where we will compile the software for our system. For OpenSSH to run and work in chroot jail mode, you have to be sure that you have recompiled your Linux kernel without the Grsecurity option that allows us to enable chroot jail restrictions protection on the system. You should be sure that “Chroot jail restrictions (CONFIG_GRKERNSEC_CHROOT) [N/y/?]” is NOT enable or nothing will work.
- 416. OpenSSH 1 CHAPTER 6 416 Step1 First of, we have to retrieve the OpenSSH chroot patch available on the Internet. This patch can be downloaded from the following location: http://chrootssh.sourceforge.net/ Step 2 Once you have a copy of this patch, you should move it under the /var/tmp directory and patch your OpenSSH source files. • This can be done with the following commands: [root@deep /]# mv osshChroot-3.4.diff /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# patch -p0 < osshChroot-3.4.diff NOTE: It’s important to note that the version number of the OpenSSH chroot patch that you have to download from the Internet must match the version number of the OpenSSH software you intended to install. For example, if the version number of OpenSSH is 3.4p1, you should download the newer OpenSSH chroot patch that matches this number. Step 3 After that, move into the newly created OpenSSH source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created OpenSSH directory use the following command: [root@deep tmp]# cd openssh-3.4p1/ • To configure and optimize OpenSSH use the following compilation lines: CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS ./configure --prefix=/usr --sysconfdir=/etc/ssh --libexecdir=/usr/libexec/openssh --mandir=/usr/share/man --with-pam --with-ipaddr-display --with-ipv4-default --with-md5-passwords --with-zlib This tells OpenSSH to set itself up for this particular configuration setup with: - Enable PAM support. - Use the IP address instead of hostname. - Use IPv4 by connections. - Enable use of MD5 passwords. - Use zlib for transport compression. NOTE: Pay special attention to the compile CFLAGS line above. We optimize OpenSSH for an i686 CPU architecture with the parameter “-march=i686”. Please don’t forget to adjust this CFLAGS line to reflect your own system and architecture.
- 417. OpenSSH 1 CHAPTER 6 417 Step4 Now, we must make a list of all existing files on the system before installing the software and one afterwards then compare them using the diff utility to find out what files are placed where and finally install the OpenSSH Server: [root@deep openssh-3.4p1]# make [root@deep openssh-3.4p1]# cd [root@deep root]# find /* > OpenSSH1 [root@deep root]# cd /var/tmp/openssh-3.4p1/ [root@deep openssh-3.4p1]# make install [root@deep openssh-3.4p1]# mkdir /var/empty [root@deep openssh-3.4p1]# chown root.sys /var/empty [root@deep openssh-3.4p1]# cd [root@deep root]# find /* > OpenSSH2 [root@deep root]# diff OpenSSH1 OpenSSH2 > OpenSSH-Installed The above commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then install the binaries and any supporting files into the appropriate locations. Step 5 Once the configuration, optimization, compilation, and installation of the OpenSSH software has been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete OpenSSH and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf openssh-version/ [root@deep tmp]# rm -f openssh-version.tar.gz The rm command as used above will remove all the source files we have used to compile and install OpenSSH. It will also remove the OpenSSH compressed archive from the /var/tmp directory. Configuring OpenSSH After OpenSSH has been built and installed successfully in your system, your next step is to configure and customize its configuration files to fit your needs. /etc/ssh/sshd_config: (The OpenSSH Server Configuration File) /etc/ssh/ssh_config: (The OpenSSH Client Configuration File) /etc/pam.d/sshd: (The OpenSSH PAM Support Configuration File) /etc/init.d/sshd: (The OpenSSH Initialization File) /etc/ssh/sshd_config: The OpenSSH Server Configuration File The sshd_config file is the system-wide server configuration file for OpenSSH which allows you to set options that modify the operation of the sshd daemon. This file contains keyword-value pairs, one per line, with keywords being case insensitive. Here are the most important keywords to configure your sshd server for maximum security; a complete listing and/or special requirements are available in the manual page for sshd (8). We must change the default one to fit our requirements and operating system. The text in bold are the parts of the configuration file that must be customized and adjusted to satisfy our needs.
- 418. OpenSSH 1 CHAPTER 6 418 •Edit the sshd_config file (vi /etc/ssh/sshd_config). Below is what we recommend you enter: # This is ssh server systemwide configuration file. Port 22 Protocol 2,1 ListenAddress 207.35.78.3 HostKey /etc/ssh/ssh_host_key HostKey /etc/ssh/ssh_host_rsa_key HostKey /etc/ssh/ssh_host_dsa_key ServerKeyBits 768 LoginGraceTime 60 KeyRegenerationInterval 3600 PermitRootLogin no IgnoreRhosts yes IgnoreUserKnownHosts yes StrictModes yes X11Forwarding no X11DisplayOffset 10 PrintMotd yes KeepAlive yes SyslogFacility AUTHPRIV LogLevel INFO RhostsAuthentication no RhostsRSAAuthentication no RSAAuthentication yes PasswordAuthentication no PermitEmptyPasswords no AllowUsers sysadmin UsePrivilegeSeparation yes Subsystem sftp /usr/libexec/openssh/sftp-server This tells the sshd_config file to set itself up for this particular configuration with: Port 22 The option “Port” specifies on which port number the sshd daemon listens for incoming connections. The default port is 22. Protocol 2,1 This option “Protocol” specifies the protocol versions sshd should support in order of preference. In our configuration the default is “2,1”. This means that sshd tries version 2 and falls back to version 1 if version 2 is not available. Depending of the ssh client version you use to connect, you may need to invert this order but you can connect with ssh client version 1 even if the order is “2,1”. ListenAddress 207.35.78.3 The option “ListenAddress” specifies the IP address of the interface network on which the sshd daemon server socket is bound. The default is “0.0.0.0”; to improve security you may specify only the required ones to limit possible IP addresses. This is a security feature. HostKey /etc/ssh/ssh_host_key HostKey /etc/ssh/ssh_host_dsa_key HostKey /etc/ssh/ssh_host_rsa_key These options specify the location containing the different private host keys. If you have compiled OpenSSH as described in this book, then the default ones are correct.
- 419. OpenSSH 1 CHAPTER 6 419 ServerKeyBits768 The option “ServerKeyBits” specifies how many bits to use in the server key. These bits are used when the daemon starts to generate its RSA key. LoginGraceTime 60 The option “LoginGraceTime” specifies how long in seconds after a connection request the server will wait before disconnecting, if the user has not successfully logged in. A low value is recommended for this setting. Imagine what 1024 simulated connections at the same time can do to the other processes on your server. KeyRegenerationInterval 3600 The option “KeyRegenerationInterval” specifies how long in seconds the server should wait before automatically regenerated its key. This is a security feature to prevent decrypting captured sessions. PermitRootLogin no The option “PermitRootLogin” specifies whether super-user “root” can log in using ssh. Never say “yes” to this option. It is safer to log in with a regular UID and then su or sudo to super-user “root”. This is a security feature. IgnoreRhosts yes The option “IgnoreRhosts” specifies whether the rhosts or shosts files should not be used in authentication. For security reasons it is recommended to NOT use rhosts or shosts files for authentication. This is a security feature. IgnoreUserKnownHosts yes The option “IgnoreUserKnownHosts” specifies whether the sshd daemon should ignore the user's $HOME/.ssh/known_hosts file during RhostsRSAAuthentication. Since we don’t allow .rhosts files on our server, it is safe to say “yes” here. This is a security feature. StrictModes yes The option “StrictModes” specifies whether sshd should check user's permissions in their home directory and rhosts files before accepting login. This option must always be set to “yes” because sometimes users may accidentally leave their directory or files world-writable. This is a security feature. X11Forwarding no The option “X11Forwarding” specifies whether X11 forwarding should be enabled or not on this server. Since we setup a server without a GUI installed on it, we can safely turn this option off. PrintMotd yes The option “PrintMotd” specifies whether the sshd daemon should print the contents of the /etc/motd file when a user logs in interactively. The /etc/motd file is also known as “the message of the day”. SyslogFacility AUTHPRIV The option “SyslogFacility” specifies the facility code used when logging messages from sshd. The facility specifies the subsystem that produced the message--in our case, AUTHPRIV. LogLevel INFO The option “LogLevel” specifies the level that is used when logging messages from sshd. INFO is a good choice. See the manual page for sshd for more information on other possibilities.
- 420. OpenSSH 1 CHAPTER 6 420 RhostsAuthenticationno The option “RhostsAuthentication” specifies whether sshd can try to use rhosts based authentication. Because rhosts authentication is insecure you shouldn’t use this option. This is a security feature. RhostsRSAAuthentication no The option “RhostsRSAAuthentication” specifies whether to try rhosts authentication in concert with RSA host authentication. This is a security feature. RSAAuthentication yes The option “RSAAuthentication” specifies whether to try RSA authentication. It is important to note that it is reserved for the SSH1 protocol only. This option must be set to “yes” for enhanced security in your sessions if you use SSH1 and only SSH1, since it doesn’t apply for the SSH2 protocol (SSH2 use DSA instead of RSA). RSA uses public and private key pairs created with the ssh-keygen utility for authentication purposes. PasswordAuthentication no The option “PasswordAuthentication” specifies whether we should use password-based authentication. For strong security, this option must always be set to “no”. You should put ‘PasswordAuthentication no’ in the sshd_config file, otherwise people might try to guess the password for the user. With ‘PasswordAuthentication no’, your public key must be on the computer or no login is allowed: that's what we want. This is a security feature. PermitEmptyPasswords no This option “PermitEmptyPasswords” is closely related with the above option “PasswordAuthentication” and specifies whether, if password authentication is allowed, the server should allow logging in to accounts with a null password. Since we do not allow password authentication in the server, we can safety turn off this option. This is a security feature. AllowUsers sysadmin This option “AllowUsers” specifies and controls which users can access ssh services. Multiple users can be specified, separated by spaces. This is a security feature. UsePrivilegeSeparation yes This option “UsePrivilegeSeparation” is used to contain and restrict the effects of programming errors. A bug in the unprivileged child process does not result in a system compromise. Previously any corruption in the sshd daemon could lead to an immediate remote root compromise if it happened before authentication and to local root compromise if it happened after authentication. The “Privilege Separation” feature of OpenSSH will make such compromise very difficult if not impossible. This is a security feature.
- 421. OpenSSH 1 CHAPTER 6 421 /etc/ssh/ssh_config:The OpenSSH Client Configuration File The ssh_config file is the system-wide client configuration file for OpenSSH which allows you to set options that modify the operation of the SSH client programs. The file contains keyword-value pairs, one per line, with keywords being case insensitive. Here are the most important keywords to configure your ssh client for maximum security; a complete listing and/or special requirements is available in the manual page for ssh (1). We must change the default ones to fit our requirements and operating system. The text in bold is the parts of the configuration file that must be customized and adjusted to satisfy your needs. • Edit the ssh_config file (vi /etc/ssh/ssh_config) and set your needs. Below is what we recommend you enter: # Site-wide defaults for various options Host * ForwardAgent no ForwardX11 no RhostsAuthentication no RhostsRSAAuthentication no RSAAuthentication yes PasswordAuthentication no FallBackToRsh no UseRsh no BatchMode no CheckHostIP yes StrictHostKeyChecking yes IdentityFile ~/.ssh/identity IdentityFile ~/.ssh/id_dsa IdentityFile ~/.ssh/id_rsa Port 22 Protocol 2,1 Cipher blowfish EscapeChar ~ This tells the ssh_config file to set itself up for this particular configuration with: Host * This option “Host” restricts all forwarded declarations and options in the configuration file to be only for those hosts that match one of the patterns given after the keyword. The pattern “*” means for all hosts up to the next “Host” keyword. With this option you can set different declarations for different hosts in the same ssh_config file. In particular, I find it useful when you want to automate backups over the network with SSH and don’t want to supply the users password. In this way we can build a new section reserved for this and disable functions that ask for passwords for the specified host in question. ForwardAgent no This option “ForwardAgent” specifies which connection authentication agent (if any) should be forwarded to the remote machine. ForwardX11 no This option “ForwardX11” is for people that use the Xwindow GUI and want to automatically redirect X11 sessions to the remote machine. Since we have a server and it doesn’t have GUI installed on it, we can safely turn this option off.
- 422. OpenSSH 1 CHAPTER 6 422 RhostsAuthenticationno This option “RhostsAuthentication” specifies whether we can try to use rhosts based authentication. Because rhosts authentication is insecure you shouldn’t use this option. This is a security feature. RhostsRSAAuthentication no This option “RhostsRSAAuthentication” specifies whether or not to try rhosts authentication in concert with RSA host authentication. Evidently our answer is “no”. This is a security feature. RSAAuthentication yes This option “RSAAuthentication” specifies whether to try RSA authentication. It is important to note that it is reserved for the SSH1 protocol only. This option must be set to yes for better security in your sessions if you use SSH1 and only SSH1 since it doesn’t applies for SSH2 protocol (SSH2 use DSA instead of RSA). RSA use public and private key pairs created with the ssh- keygen utility for authentication purposes. Enable only if you connect to OpenSSH with client software that use SSH1 protocol. PasswordAuthentication no This option “PasswordAuthentication” specifies whether we should use password-based authentication. For strong security, this option must always be set to no. You should put ‘PasswordAuthentication no’ in the sshd_config file, otherwise people might try to guess the password for the user. With ‘PasswordAuthentication no’, your public key must be on the computer or no login is allowed: that's what we want. This is a security feature. FallBackToRsh no This option “FallBackToRsh” specifies that if a connection with ssh daemon fails rsh should automatically be used instead. Recalling that rsh service is insecure, this option must always be set to “no”. This is a security feature. UseRsh no This option “UseRsh” specifies that rlogin/rsh services should be used on this host. As with the FallBackToRsh option, it must be set to “no” for obvious reasons. This is a security feature. BatchMode no This option “BatchMode” specifies whether a username and password querying on connect will be disabled. This option is useful when you create scripts and don’t want to supply the password. (e.g. Scripts that use the scp command to make backups over the network). CheckHostIP yes This option “CheckHostIP” specifies whether or not ssh will additionally check the host IP address that connect to the server to detect DNS spoofing. It’s recommended that you set this option to “yes” but on the other hand you can lose some performance doing this. StrictHostKeyChecking yes This option “StrictHostKeyChecking” specifies whether or not ssh will automatically add new host keys to the $HOME/.ssh/known_hosts file. This option, when set to “yes”, provides the maximum protection against Trojan horse attacks. One interesting procedure with this option is to set it to “no” at the beginning, allow ssh to add automatically all common hosts to the host file as they are connected to, and then return to set it to “yes” to take advantage of its feature. This is a security feature.
- 423. OpenSSH 1 CHAPTER 6 423 IdentityFile~/.ssh/identity IdentityFile ~/.ssh/id_dsa IdentityFile ~/.ssh/id_rsa These options specify alternate multiple authentication identity files to read. Port 22 This option “Port” specifies on which port number ssh connects to on the remote host. The default port is 22. Protocol 2,1 This option “Protocol” specifies the protocol versions ssh should support in order of preference. In our configuration the default is “2,1”. This means that ssh tries version 2 and falls back to version 1 if version 2 is not available. Depending of the ssh client version you use to connect, you may need to invert this order but you can connect with ssh client version 1 even if the order is “2,1”. Cipher blowfish This option “Cipher” specifies what cipher should be used for encrypting sessions. The blowfish use 64-bit blocks and keys of up to 448 bits. EscapeChar ~ This option “EscapeChar” specifies the session escape character for suspension. /etc/pam.d/sshd: The OpenSSH PAM Support Configuration File For increased security of OpenSSH, we have compiled it to use the PAM mechanism for password authentication. Step 1 To be able to use this feature, we must create the /etc/pam.d/sshd file and add the following parameters inside it. • Create the sshd file (touch /etc/pam.d/sshd) and add the following lines: #%PAM-1.0 auth required /lib/security/pam_stack.so service=system-auth auth required /lib/security/pam_nologin.so account required /lib/security/pam_stack.so service=system-auth account required /lib/security/pam_access.so account required /lib/security/pam_time.so password required /lib/security/pam_stack.so service=system-auth session required /lib/security/pam_stack.so service=system-auth session required /lib/security/pam_limits.so Step2 Now, set the permissions of the sshd file to be (0640/-rw-r-----) and owned by the super- user ‘root’ for security reasons. • To change the permissions and ownership of the sshd file, use the commands: [root@deep /]# chmod 640 /etc/pam.d/sshd [root@deep /]# chown 0.0 /etc/pam.d/sshd
- 424. OpenSSH 1 CHAPTER 6 424 /etc/init.d/sshd:The OpenSSH Initialization File The /etc/init.d/sshd script file is responsible to automatically starting and stopping the OpenSSH server on your Linux system. Loading the sshd daemon as a standalone daemon will eliminate load time and will even reduce swapping since non-library code will be shared. Please note that the following script is suitable for Linux operating systems that use SystemV. If you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to make it work for you. Step 1 Create the sshd script file (touch /etc/init.d/sshd) and add the following lines: #!/bin/bash # This shell script takes care of starting and stopping OpenSSH. # # chkconfig: 2345 55 25 # description: OpenSSH is a program thas allows to establish a secure remote # connection to a server. # # processname: sshd # config: /etc/ssh/ssh_host_key # config: /etc/ssh/ssh_host_key.pub # config: /etc/ssh/ssh_random_seed # config: /etc/ssh/sshd_config # pidfile: /var/run/sshd.pid # Source function library. . /etc/init.d/functions # Source networking configuration. . /etc/sysconfig/network # Source OpenSSH configureation. if [ -f /etc/sysconfig/sshd ] ; then . /etc/sysconfig/sshd fi RETVAL=0 # Some functions to make the below more readable. KEYGEN=/usr/bin/ssh-keygen RSA1_KEY=/etc/ssh/ssh_host_key RSA_KEY=/etc/ssh/ssh_host_rsa_key DSA_KEY=/etc/ssh/ssh_host_dsa_key PID_FILE=/var/run/sshd.pid my_success() { local msg if [ $# -gt 1 ]; then msg="$2" else msg="done" fi case "`type -type success`" in function) success "$1" ;; *) echo -n "${msg}" ;; esac
- 425. OpenSSH 1 CHAPTER 6 425 } my_failure(){ local msg if [ $# -gt 1 ]; then msg="$2" else msg="FAILED" fi case "`type -type failure`" in function) failure "$1" ;; *) echo -n "${msg}" ;; esac } do_rsa1_keygen() { if ! test -f $RSA1_KEY ; then echo -n "Generating SSH1 RSA host key: " if $KEYGEN -q -t rsa1 -f $RSA1_KEY -C '' -N '' >&/dev/null; then my_success "RSA1 key generation" echo else my_failure "RSA1 key generation" echo exit 1 fi fi } do_rsa_keygen() { if ! test -f $RSA_KEY ; then echo -n "Generating SSH2 RSA host key: " if $KEYGEN -q -t rsa -f $RSA_KEY -C '' -N '' >&/dev/null; then my_success "RSA key generation" echo else my_failure "RSA key generation" echo exit 1 fi fi } do_dsa_keygen() { if ! test -f $DSA_KEY ; then echo -n "Generating SSH2 DSA host key: " if $KEYGEN -q -t dsa -f $DSA_KEY -C '' -N '' >&/dev/null; then my_success "DSA key generation" echo else my_failure "DSA key generation" echo exit 1 fi fi } do_restart_sanity_check() { sshd -t RETVAL=$? if [ ! "$RETVAL" = 0 ]; then my_failure "Configuration file or keys" echo
- 426. OpenSSH 1 CHAPTER 6 426 exit$RETVAL fi } case "$1" in start) # Create keys if necessary do_rsa1_keygen; do_rsa_keygen; do_dsa_keygen; echo -n "Starting SSHD: " if [ ! -f $PID_FILE ] ; then sshd $OPTIONS RETVAL=$? if [ "$RETVAL" = "0" ] ; then my_success "sshd startup" "sshd" touch /var/lock/subsys/sshd else my_failure "sshd startup" "" fi fi echo ;; stop) echo -n "Shutting down SSHD: " if [ -f $PID_FILE ] ; then killproc sshd RETVAL=$? [ $RETVAL -eq 0 ] && rm -f /var/lock/subsys/sshd fi echo ;; restart) do_restart_sanity_check $0 stop $0 start RETVAL=$? ;; condrestart) if [ -f /var/lock/subsys/sshd ] ; then do_restart_sanity_check $0 stop $0 start RETVAL=$? fi ;; status) status sshd RETVAL=$? ;; *) echo "Usage: sshd {start|stop|restart|status|condrestart}" exit 1 ;; esac exit $RETVAL
- 427. OpenSSH 1 CHAPTER 6 427 Step2 Once the /etc/init.d/sshd script file has been created, it is important to make it executable, change its default permissions, create the necessary links and start it. Making this file executable will allow the system to run it, changing its default permission is to allow only the root user to change this file for security reason, and creation of the symbolic links will let the process control initialization of Linux to start the program automatically for you at each system boot. • To make this script executable and to change its default permissions, use the commands: [root@deep /]# chmod 700 /etc/init.d/sshd [root@deep /]# chown 0.0 /etc/init.d/sshd • To create the symbolic rc.d links for OpenSSH, use the following commands: [root@deep /]# chkconfig --add sshd [root@deep /]# chkconfig --level 2345 sshd on • To start OpenSSH software manually, use the following command: [root@deep /]# /etc/init.d/sshd start Starting SSHD: [OK] Running OpenSSH in a chroot jail This section applies only if you want to run OpenSSH in chroot jail environment for some of your users. This kind of setup is useful for web hosting companies that want to provide shell access via remote secure connection with OpenSSH but don’t want to allow full access to the server and just limit users to their own web directory. By default, OpenSSH does not support the chroot jail mode and we have to compile it with an external patch to enable the chroot mode extensions. The patch is available from the following site: http://chrootssh.sourceforge.net/ Remember that you have to download the version number equal to the OpenSSH version number you use in order for chroot jail support to work. At the beginning of this chapter, we have already patched the software with the chroot jail mode extensions patch, therefore, we only need to create the required skeleton environment and copy the necessary tools into this chroot jail directory to enable chroot jail support. Below are the steps to follow if you want to run OpenSSH with chroot jail support for the specified users. The main benefit of a chroot jail is that the jail will limit the portion of the file system the daemon can see to the root directory of the jail. Additionally, since the jail only needs to support OpenSSH, the programs available in the jail can be extremely limited. More importantly, there is no need for setuid-root programs, which can be used to gain root access and break out of the jail.
- 428.
- 429. OpenSSH 1 CHAPTER 6 429 Necessarysteps to run OpenSSH in a chroot jail: What you're essentially doing is creating a skeleton root file system with enough components (binaries, libraries, etc.) to allow Unix to do a chroot when the user logs in. Step 1 With OpenSSH, it’s important to give to your strictly SSH users a real shell account on the Linux system because we want to allow remote shell access, even if limited to running just a few commands on the server. First, create the new users for this purpose; these users will be the users allowed to connect to your OpenSSH server running in chroot jail mode. These have to be separate from regular user accounts with unlimited access because of how the "chroot" environment works. Chroot makes it appear from the user's perspective as if the level of the file system you've placed them in is the top level of the file system. Here we create a new SSH user called “gmourani” as an example and set it’s the home directory under the /home/httpd/gmourani directory since it is the place where it’s the users web directory and web pages will be located. • Use the following command to create a new SSH user. This step must be done for each additional new user you allow to access your OpenSSH server on OpenNA Linux. [root@deep /]# useradd -m -d /home/httpd/gmourani gmourani [root@deep /]# passwd gmourani Changing password for user gmourani New UNIX password: Retype new UNIX password: passwd: all authentication tokens updated successfully • Use the following command to create a new SSH user. This step must be done for each additional new user you allow to access your OpenSSH server on Red Hat Linux. [root@deep /]# useradd -g users -d /home/httpd/gmourani gmourani [root@deep /]# passwd gmourani Changing password for user gmourani New UNIX password: Retype new UNIX password: passwd: all authentication tokens updated successfully The useradd command will add the new SSH user called “gmourani” to our Linux server and will set it’s the home directory under the /home/httpd/gmourani directory on the system since it is a useful location for remote clients to maintain their web accounts. The passwd command will set the password for this user “gmourani”.
- 430. OpenSSH 1 CHAPTER 6 430 Step2 Once the new SSH users have been created, we must edit the /etc/passwd file and make the appropriated changes to the accounts to allow OpenSSH to chroot when the users login on the system. In general, the sshd daemon will chroot when it encounters the magic token '/./' in a users home directory. Therefore this is what we will add to the passwd file for the SSH user in question. • Edit the passwd file (vi /etc/passwd) and change the following line: gmourani:x:501:100::/home/httpd/gmourani:/bin/bash To read: gmourani:x:501:100::/home/httpd/gmourani/./:/bin/bash NOTE: Don’t forget to make the same modification for each additional SSH user for whom you want to chroot. Step 3 Now, we have to create all the necessary chrooted environment subdirectories where we will copy tools we want to allow this SSH user to use on the system. • Use the following command to create all the necessary chroot subdirectories. [root@deep /]# mkdir /home/httpd/gmourani/bin [root@deep /]# mkdir /home/httpd/gmourani/dev [root@deep /]# mkdir /home/httpd/gmourani/etc [root@deep /]# mkdir /home/httpd/gmourani/lib [root@deep /]# mkdir /home/httpd/gmourani/usr [root@deep /]# mkdir /home/httpd/gmourani/usr/bin [root@deep /]# mkdir /home/httpd/gmourani/usr/lib • For Red Hat Linux 7.3 users, you should create the following additional directory: [root@deep /]# mkdir /home/httpd/gmourani/lib/i686 Step 4 Next, we must change the permissions on all the chroot glue subdirectories to mode (0111/d-- x--x--x) for security reasons. • Use the following command to change the permissions of all the subdirectories. [root@deep /]# chmod -R 0111 /home/httpd/gmourani/*
- 431. OpenSSH 1 CHAPTER 6 431 Step5 Once all permissions of the supporting glues have been changed, it is time to copy the required binary programs to the related subdirectories in the chroot area for OpenSSH to work. These programs are necessary to allow the SSH users to list, create directory, copy, remove, and edit files on the SSH chroot jail directory. If there are features you don’t want the user to be able to use, then don’t copy them to the chroot area. • Use the following commands to copy the require binaries programs into the chroot area. [root@deep /]# cp /bin/bash /home/httpd/gmourani/bin/ [root@deep /]# cp /bin/cp /home/httpd/gmourani/bin/ [root@deep /]# cp /bin/ls /home/httpd/gmourani/bin/ [root@deep /]# cp /bin/mkdir /home/httpd/gmourani/bin/ [root@deep /]# cp /bin/grep /home/httpd/gmourani/bin/ [root@deep /]# cp /bin/rm /home/httpd/gmourani/bin/ [root@deep /]# cp /bin/vi /home/httpd/gmourani/bin/ [root@deep /]# cp /usr/bin/dircolors /home/httpd/gmourani/usr/bin/ [root@deep /]# chmod 0111 /home/httpd/gmourani/bin/* [root@deep /]# chmod 0111 /home/httpd/gmourani/usr/bin/* NOTE: The above chmod commands will change default permissions of those programs under the /bin directories of the chroot jail area to be (0111 ---x--x—x) because we don’t want users to be able to modify or read binaries in the chroot area but just to execute them if necessary. Step 6 The binaries we have copied into the chroot area have been compiled with shared libraries by default and for this reason it is important to find the shared libraries dependencies associated with them and copy them into the /lib subdirectory in the chroot jail area that we created earlier. To find the shared library dependencies of binaries, you have to use the ldd command of Linux. You must copy all the libraries below to the /home/httpd/gmourani/lib directory of the chroot area. These libraries are part of libc, and needed by various programs. • Use the following commands to copy the require libraries into the chroot area. [root@deep /]# cp /lib/libtermcap.so.2 /home/httpd/gmourani/lib/ [root@deep /]# cp /lib/libdl.so.2 /home/httpd/gmourani/lib/ [root@deep /]# cp /lib/libc.so.6 /home/httpd/gmourani/lib/ [root@deep /]# cp /lib/libgcc_s.so.1 /home/httpd/gmourani/lib/ [root@deep /]# cp /lib/ld-linux.so.2 /home/httpd/gmourani/lib/ [root@deep /]# cp /usr/lib/libncurses.so.5 /home/httpd/gmourani/usr/lib/ [root@deep /]# strip -R .comment /home/httpd/gmourani/lib/* • For Red Hat Linux 7.3 users, you should copy the following additional library: [root@deep /]# cp /lib/i686/libc.so.6 /home/httpd/gmourani/lib/i686/
- 432. OpenSSH 1 CHAPTER 6 432 WARNING:Depending on what has been compiled, the required shared libraries may be different than the ones illustrated above. Please use the ldd command on each binary under /bin subdirectories of the chroot jail to find out the ones you need and copy them to the /lib subdirectory of the chroot area. The “strip -R .comment” command will remove all the named section “.comment” from the libraries files under the /lib subdirectory and will make them smaller in size and can help in the performance of them. Step 7 One of the last step to do, is to make a copy of the “DIR_COLORS” and “passwd” files located under the /etc directory to our chroot jail for SSH to be able to find it. • Use the following commands to copy the file into the chroot area. [root@deep /]# cp /etc/DIR_COLORS /home/httpd/gmourani/etc/ [root@deep /]# cp /etc/passwd /home/httpd/gmourani/etc/ Step 8 Finally, we have to create the /home/httpd/gmourani/dev/null device file and set its mode appropriately. • Use the following commands to create the null device into the chroot area. [root@deep /]# mknod /home/httpd/gmourani/dev/null c 1 3 [root@deep /]# chmod 666 /home/httpd/gmourani/dev/null Creating OpenSSH private & public keys This section deals with actions that need to be performed to create new private/public keys for our users to establish a secure connection on the server. There are cryptosystems where encryption and decryption are done using separate keys, and it is not possible to derive the decryption key from the encryption key. The idea is that each user creates a public/private key pair for authentication purposes. The server knows the public key, and only the user knows the private key. The file $HOME/.ssh/authorized_keys lists the public keys that are permitted for logging in. When the user logs in, the ssh program tells the server which key pair it would like to use for authentication. The server checks if this key is permitted, and if so, sends the user (actually the ssh program running on behalf of the user) a challenge, a random number, encrypted by the user's public key. The challenge can only be decrypted using the proper private key. The user's client then decrypts the challenge using the private key, proving that he/she knows the private key but without disclosing it to the server. Step 1 Below, are the steps to follow to create a new SSH private & public key for one user. This example assumes that secure encrypted connections will be made between Linux servers. • To create your (RSA) private & public keys for SSH2 of LOCAL, use the commands: [root@deep /]# su gmourani [gmourani@deep /]$ ssh-keygen -t rsa Generating public/private rsa key pair.
- 433. OpenSSH 1 CHAPTER 6 433 Enterfile in which to save the key (/home/gmourani/.ssh/id_rsa): Created directory '/home/gmourani/.ssh'. Enter passphrase (empty for no passphrase): Enter same passphrase again: Your identification has been saved in /home/gmourani/.ssh/id_rsa. Your public key has been saved in /home/gmourani/.ssh/id_rsa.pub. The key fingerprint is: ba:0c:08:6d:9d:51:4f:b3:32:68:9b:0d:83:ce:be:bd gmourani@deep WARNING: The above example assumes that you want to generate (RSA) private & public keys for SSH protocol 2 (highly recommended). If you want to generate (RSA) private & public keys for SSH protocol 1, then you must use the ‘-t rsa1’ option to the key generation command as follows: [root@deep /]# su gmourani [gmourani@deep /]$ ssh-keygen -t rsa1 Using the ‘-t rsa1’ option will generate SSH1 instead of SSH2 private & public keys. The SSH1 private key will be named ”identity” and the public key will be “identity.pub”. The '-t' option is used to specify the type of the key to create. The possible values are "rsa1" for protocol version 1 and "rsa" or "dsa" for protocol version 2. If you have multiple accounts you might want to create a separate key on each of them. You may want to have separate keys for: • Your server (1) • Your server (2) • Your server (3) This allows you to limit access between these servers, e.g. not allowing the first server (1) account to access your second server (2) account or the third server (3). This enhances the overall security in the case any of your authentication keys are compromised for any reason. Step 2 Copy your local public key id_rsa.pub for SSH2 or identity.pub for SSH1 from the /home/gmourani/.ssh directory remotely under the name, say, “authorized_keys”. One way to copy the file is to use the ftp command or you might need to send your public key in electronic mail to the administrator of the other system. Just include the contents of the ~/.ssh/id_rsa.pub or ~/.ssh/identity.pub file in the message. To resume the required steps: 1) The user creates his/her RSA keys pair by running ssh-keygen. This stores the private key in id_rsa (SSH2) or in identity (SSH1) and the public key in id_rsa.pub (SSH2) or in identity.pub (SSH1) into the user's home directory on the LOCAL machine. 2) The user should then copy the id_rsa.pub key (SSH2) or identity.pub key (SSH1) to $HOME/.ssh/authorized_keys into his/her home directory on the REMOTE machine.
- 434. OpenSSH 1 CHAPTER 6 434 -------------------------------------- | | | | | | | | | Server 1 | | Server 2 | | | | | | | | | ------------------- ------------------- User: gmourani User: gmourani Pass-phrase: qwerty1 Pass-phrase: qwerty2 Private key: id_rsa Private key: id_rsa Public key: id_rsa.pub --------------- authorized_keys authorized_keys ----------------------- Public key: id_rsa.pub Public key of user gmourani on the first server (1) is sending to the second server (2) under the $HOME directory of user gmourani and become ‘authorized_keys’; the same action is made on the second server (2). The public key of user gmourani on server (2) is sending to server (1) under the $HOME directory of user gmourani and become ‘authorized_keys’. NOTE: OpenSSH's public key is a one-line string. Adding public keys from commercial SSH tools which stretch the public key over several lines, will not be recognized by OpenSSH. OpenSSH Users Tools The commands listed below are some that we use regularly, but many more exist, and you should check the manual pages and documentation of OpenSSH for more details. ssh The ssh (Secure Shell) command provides secure encrypted communications between two untrusted hosts over an insecure network. It is a program for securely logging into a remote machine and executing commands from there. It is a suitable replacement for insecure programs like telnet, rlogin, rcp, rdist, and rsh. • To login to a remote machine, use the following command: [root@deep /]# ssh -l <login_name> <hostname> For example: [root@deep /]# ssh -l gmourani deep.openna.com gmourani@deep.openna.com’s password: Last login: Tue Oct 19 1999 18:13:00 -0400 from deep.openna.com No mail. [gmourani@deep gmourani]$ Where <login_name> is the name you use to connect to the ssh server and <hostname> is the remote address (you can use IP address here) of your ssh server.
- 435. OpenSSH 1 CHAPTER 6 435 scp Thescp (Secure Copy) utility copies files from the local system to a remote system or vice versa, or even between two remote systems using the scp command. • To copy files from remote to local system, use the following commands: [root@deep /]# su gmourani [gmourani@deep /]$ scp -p <login_name@hostname>:/dir/for/file localdir/to/filelocation For example: [gmourani@deep /]$ scp -p gmourani@mail:/etc/test1 /tmp Enter passphrase for RSA key 'gmourani@mail.openna.com': test1 | 2 KB | 2.0 kB/s | ETA: 00:00:00 | 100% • To copy files from local to remote system, use the following commands: [root@deep /]# su gmourani [gmourani@deep /]$ scp -p localdir/to/filelocation <username@hostname>:/dir/for/file For example: [gmourani@deep /]$ scp -p /usr/bin/test2 gmourani@mail:/var/tmp gmourani@mail's password: test2 | 7 KB | 7.9 kB/s | ETA: 00:00:00 | 100% WARNING: The “-p” option indicates that the modification and access times, as well as modes of the source file, should be preserved on the copy. Usually this is desirable. Please check the chapter related to backups in this book for more information about other possible uses of SSH technology with Linux. Changing your pass-phrase You can change the pass-phrase at any time by using the -p option of ssh-keygen. • To change the pass-phrase, use the following commands: [root@deep /]# su gmourani [gmourani@deep /]$ ssh-keygen -p Enter file in which the key is (/home/gmourani/.ssh/id_rsa): Enter old passphrase: Key has comment '/home/gmourani/.ssh/id_rsa' Enter new passphrase (empty for no passphrase): Enter same passphrase again: Your identification has been saved with the new passphrase. Further documentation For more details, there are several manual pages about OpenSSH that you can read: $ man ssh (1) - OpenSSH secure shell client (remote login program). $ man ssh [slogin] (1) - OpenSSH secure shell client (remote login program). $ man ssh-add (1) - Adds identities for the authentication agent. $ man ssh-agent (1) - Authentication agent. $ man ssh-keygen (1) - Authentication key generation. $ man sshd (8) - Secure shell daemon. $ sftp-server (8) - SFTP server subsystem.
- 437. Sudo IN THIS CHAPTER 1.Compiling - Optimizing & Installing Sudo 2. Configuring Sudo 3. A more complex sudoers configuration file 4. Securing Sudo 5. Sudo Users Tools
- 439. Sudo 1 CHAPTER 7 439 LinuxSudo Abstract Sudo (superuser do) is a security program designed to allow a system administrator to give limited root privileges to users and log root activity. The basic philosophy is to give as few privileges as possible, but still allow people to get their work done. It operates on a per-command basis and it is not a replacement for the shell. This means that you have to use it every time you need to execute some commands with “root” privilege on the server. In general, it does the same function as the command 'su' does on the Linux but with a big difference that we have full control about what should be done by which users, what commands a user may run, etc. Here is some of its feature: It provides ability to restrict what commands a user may run on a per-host basis. It does copious logging of each command, providing a clear audit trail. It can log all commands to a central host (as well as on the local host). It uses timestamp files to implement a "ticketing" system “root” time access. Its configuration file is setup in such a way that you could use it on many machines. Imagine that your boss asks you to create a new account for the new webmaster of your company. This webmaster will be responsible of the web server, but you don’t know if this person will stay with the company for a long time or not. You don’t want to give him full “root” privileges via the ‘su’ command of Linux because you don’t trust him or he doesn’t need to have full “root” access to manage a web server. This is where a program like sudo will help you to same time and protect your server. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest Sudo version number is 1.6.6 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages Please check http://www.sudo.ws/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: Sudo Homepage: http://www.sudo.ws/ Sudo FTP Site: 128.138.243.20 You must be sure to download: sudo-1.6.6.tar.gz
- 440. Sudo 1 CHAPTER 7 440 Pristinesource If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install Sudo, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > Sudo1 • And the following one after you install the software: [root@deep root]# find /* > Sudo2 • Then use this command to get a list of what changed: [root@deep root]# diff Sudo1 Sudo2 > Sudo-Installed Compiling - Optimizing & Installing Sudo Below are the steps that you must make to configure, compile and optimize the Sudo software before installing it on your system. First off, we install the program as user 'root' so as to avoid any authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp sudo-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf sudo-version.tar.gz Step 2 Next, move into the newly created Sudo source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created Sudo source directory use the command: [root@deep tmp]# cd sudo-1.6.6/ • To configure and optimize Sudo use the following compilation lines: CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS ./configure --prefix=/usr --sbindir=/usr/sbin --with-logging=syslog --with-logfac=authpriv --with-pam --with-env-editor --with-ignore-dot --with-tty-tickets --disable-root-mailer --disable-root-sudo --disable-path-info --with-mail-if-noperms --with-mailsubject="*** Sudo SECURITY information for %h ***"
- 441. Sudo 1 CHAPTER 7 441 Thistells Sudo to set itself up for this particular configuration setup with: - Log sudo messages via syslog. - The syslog facility to log with is authpriv. - Enable PAM support. - Use the environment variable EDITOR for visudo. - Ignore '.' in the PATH. - Use a different ticket file for each user/tty combo. - Don't run the mailer as root, run as the user since it’s safer. - Don't allow root to run sudo command. - Print 'command not allowed' instead of 'command not found'. - Send mail to sysadmin if user not allowed to runs command. - Change subject of sudo mail result. Step 3 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the Sudo software: [root@deep sudo-1.6.6]# make [root@deep sudo-1.6.6]# cd [root@deep root]# find /* > Sudo1 [root@deep root]# cd /var/tmp/sudo-1.6.6/ [root@deep sudo-1.6.6]# make install [root@deep sudo-1.6.6]# strip /usr/bin/sudo [root@deep sudo-1.6.6]# strip /usr/sbin/visudo [root@deep sudo-1.6.6]# mkdir –p –m0700 /var/run/sudo [root@deep sudo-1.6.6]# cd [root@deep root]# find /* > Sudo2 [root@deep root]# diff Sudo1 Sudo2 > Sudo-Installed The above commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then install the binaries and any supporting files into the appropriate locations. Step 4 Once the configuration, optimization, compilation, and installation of the Sudo software have been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete Sudo and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf sudo-version/ [root@deep tmp]# rm -f sudo-version.tar.gz
- 442. Sudo 1 CHAPTER 7 442 ConfiguringSudo After Sudo has been built and installed successfully in your system, your next step is to configure and customize its configuration files. /etc/sudoers: (The Sudo Configuration File) /etc/pam.d/sudo: (The Sudo PAM Support Configuration File) /etc/sudoers: The Sudo Configuration File The /etc/sudoers file is the main configuration file for Sudo. It is in this configuration file that Sudo gets all of its information on the way it should run on your system. The parameters entered in the sudoers configuration file will decide how regular users should use sudo to get “root” privileges to accomplish their works. On production servers where shell access and “root” privilege are limited to just some trusted regular users, the sudoers configuration file should be very simple to configure. All we need is to define a group under which trusted people are allowed to run all commands as “root” and put the related people into this group name. This kind of configuration file works a little bit like PAM to control who can have “root” access to the system via the ‘su’ command but in a more secure and natural way. In the sudoers configuration file below, we will show you the correct setup to limit some users of a specific group name (wheel) to sudo and get “root” access on the system. This is the most used configuration file for the majority of users. We only want to allow some regular users with shell access on the server to sudo to “root” account. Later, we will explain how to configure the sudoers file on server where many shell accesses are available for users. These kinds of servers are generally development servers where developers work and need special “root” privileges depending on their work and tasks to accomplish. Finally, I will inform you that with sudo, we must edit the sudoers configuration file with the “visudo” program which comes installed on your system for this purpose. Never edit the sudoers file with other editor like “vi”, always use the “visudo” command when you want to change information inside the sudoers configuration file. Step1 Ok, it’s time to configure sudoers to allow users who are members of the group “wheel” to get “root” access on the server. First, we have to edit sudoers and make the changes. • Edit the sudoers file (visudo) and set your needs. Below is what we recommend you use for production servers: # This file MUST be edited with the 'visudo' command as root. # Defaults specification Defaults rootpw # User privilege specification # Super-user root can run anything as any user. root ALL=(ALL) ALL # Comment if you don't want to allow people in group wheel to # run all commands as super-user root. %wheel ALL=(ALL) ALL
- 443. Sudo 1 CHAPTER 7 443 Thistells the sudoers file to set itself up for this particular configuration with: Defaults rootpw With sudo, certain configuration options may be changed from their default values at runtime via one or more “Default_Entry” options defined in the sudoers file. This is what we do here. In our case, we inform sudo with the “Defaults rootpw” option that we want it to prompt any allowed user who wants to sudo to super-user “root” to enter the “root” password instead of the password of the invoking user. By default sudo asks for the users password instead of super-user password when someone wants to sudo to “root” user. Because in this configuration file we want to allow full “root” access for users that are members of the “wheel” group and because we trust these users, we decide to change the default sudo setting and ask for “root” password before having access to “root” privilege on the server. This setting is useful when you make secure remote connections on the server with SSH software and want to sudo to “root” user. root ALL=(ALL) ALL Here we inform sudo that the super-user “root” can run anything as any user on the server. This option is required for the software to work or there is no need to use sudo. The “ALL=(ALL) ALL” parameter means that everything is allowed for the super-user “root”. %wheel ALL=(ALL) ALL Here we allow every user who is a member of the group “wheel” to run all commands as super- user “root” on the system. This is the equivalence of what we achieve with the PAM security feature, but in a most efficient and secure way. Step 2 Once the sudoers file has been configured, it is time to add some users who will be allowed to sudo to “root” account. • If you want to make, for example, the user “sysadmin” a member of the “wheel” group, and thus be able to sudo to “root”, use the following command: [root@deep /]# usermod -G10 sysadmin This means “G” is a list of supplementary groups that the user is also a member of. “10” is the numeric value of the user’s ID “wheel”, and “sysadmin” is the user we want to add to the “wheel” group. Use the same command above for all users on your system you want to be able to sudo to “root” account.
- 444. Sudo 1 CHAPTER 7 444 /etc/pam.d/sudo:The Sudo PAM Support Configuration File For better security of Sudo, we have compiled it to use the PAM mechanism for password authentication. Step 1 To be able to use this feature, we must create the /etc/pam.d/sudo file and add the following parameters inside it. • Create the sudo file (touch /etc/pam.d/sudo) and add the following lines: #%PAM-1.0 auth required /lib/security/pam_stack.so service=system-auth account required /lib/security/pam_stack.so service=system-auth password required /lib/security/pam_stack.so service=system-auth session required /lib/security/pam_stack.so service=system-auth Step2 Now, set the permissions of the sudo file to be (0640/-rw-r-----) and owned by the super- user ‘root’ for security reasons. • To change the permissions and ownership of the sudo file, use the commands: [root@deep /]# chmod 640 /etc/pam.d/sudo [root@deep /]# chown 0.0 /etc/pam.d/sudo A more complex sudoers configuration file For those who want to have complete control on who can use the “root” account on the server when there are many users with shell access doing different tasks, here is the sudoers configuration file to go with. As you’ll see it is more complex and covers some basic definitions. It is important to understand how the sudo policy works. To be as clear as possible, I will simply say that when you allow full access to user with the “ALL” definition, you cannot deny other access or privileges. Therefore, the best way is to allow only what you want user to be able to run with “root” privilege through the “Cmnd alias specification” part of the configuration file and use the defined aliases rules under the “User privilege specification” part of the configuration file. Here is a working sudoers configuration file to better understand what I mean. • Edit the sudoers file (visudo) and set your needs. Below is what we recommend you use for servers that have many users with shell access: # /etc/sudoers: OpenNA, Inc. (last updated 2002 Apr 19) # This file MUST be edited with the 'visudo' command as root. # User alias specification User_Alias FULLTIME_USERS = sysadmin, gmourani User_Alias PARTTIME_USERS = zeljko, mary # Cmnd alias specification Cmnd_Alias HTTP = /etc/init.d/httpd, /bin/vi /etc/httpd/* Cmnd_Alias FTP = /etc/init.d/proftpd, /bin/vi /etc/proftpd.conf Cmnd_Alias SMTP = /etc/init.d/exim, /bin/vi /etc/mail/* Cmnd_Alias SQL = /etc/init.d/mysqld, /usr/bin/mysql, /usr/bin/mysqladmin Cmnd_Alias BIND = /etc/init.d/named, /bin/vi /chroot/named/*
- 445. Sudo 1 CHAPTER 7 445 #Defaults specification Defaults:FULLTIME_USERS rootpw Defaults:FULLTIME_USERS !lecture # User privilege specification # Super-user root can run anything as any user. root ALL=(ALL) ALL # Every users member of the group wheel will be allowed # to run all commands as super-user root. %wheel ALL=(ALL) ALL # Full time users may run anything but need a password. FULLTIME_USERS ALL = ALL # Part time users may administrate httpd, ftpd, smtp, sql, and bind servers. PARTTIME_USERS ALL = HTTP, FTP, SMTP, SQL, BIND This tells the sudoers file to set itself up for this particular configuration with: User_Alias FULLTIME_USERS = sysadmin, gmourani User_Alias PARTTIME_USERS = zeljko, mary The “User_Alias” definition lines are used to define local users who may sudo to the “root” account on the server. The definition works on the following way: Alias_Type NAME = item1, item2, ... Where “Alias_Type” is the type of alias to use, in our case, we use “User_Alias” to define local users aliases on the system. A NAME is a string of uppercase letters, numbers, or underscores characters ('_'). A NAME must always start with an uppercase letter. You can use as any name as you like to define the NAME. In our example, we use “FULLTIME_USERS” to define local users on the system who have full time access to the server and “PARTTIME_USERS” to define local users on the server who have part time access to the server for different reason. Item represents usernames to add in each category separated by [,]. NOTE: It is important to note that users added to the “User_Alias” definition will be able to sudo to super-user “root” account even if their names do not appear under the group “wheel”. Cmnd_Alias HTTP = /etc/init.d/httpd, /bin/vi /etc/httpd/* Cmnd_Alias FTP = /etc/init.d/proftpd, /bin/vi /etc/proftpd.conf Cmnd_Alias SMTP = /etc/init.d/exim, /bin/vi /etc/mail/* Cmnd_Alias SQL = /etc/init.d/mysqld, /usr/bin/mysql, /usr/bin/mysqladmin Cmnd_Alias BIND = /etc/init.d/named, /bin/vi /chroot/named/* Here we use another “Alias_Type” definition. This type of alias is used to define commands, programs, files, and directories that we want to allow certain users to be able to use when issuing the sudo command on the system. The definition works on the following way: Alias_Type NAME = item1, item2, ...
- 446. Sudo 1 CHAPTER 7 446 Where“Alias_Type” is the type of alias to use, in our case, we use “Cmnd_Alias” to define command aliases on the system. A NAME is a string of uppercase letters, numbers, or underscores characters ('_'). A NAME must always start with an uppercase letter. You can use any name you like to define the NAME. In our example, we use “HTTP, FTP, SMTP, SQL, and BIND” to define command names aliases associated with the commands we will allow local users to run when issuing the sudo command on the system. Item represent the commands, files, programs, or directories to add in each category separated by [,]. Defaults:FULLTIME_USERS rootpw Defaults:FULLTIME_USERS !lecture With sudo, certain configuration options may be changed from their default values at runtime via one or more “Default_Entry” options. Again, this is what we do here. In our case, we inform sudo with the “Defaults:FULLTIME_USERS rootpw” option that we want it to prompt any users with full time access on the server “FULLTIME_USERS” to enter the “root” password instead of their own password. Remember that by default sudo asks for the users password instead of super-user password when someone wants to sudo to “root”. Because we allow full “root” access for users under the “FULLTIME_USERS” alias, we decide to change the default sudo setting and ask for the “root” password before giving access to “root” privileges on the server. This also means that users under the “PARTTIME_USERS” alias will have to enter their own password and not the “root” password. This is a security feature to separate trusted users with full time access on the server from semi- trusted users with part time access on the system. Users having part time access could be students, limited administrators, or anything else that you think about. In this way, users with part time access do not know the “root” password since they enter their own passwords to get some “root” privileges and we don’t need to change “root” password every time these users leave the company. The second “Default_Entry” option “Defaults:FULLTIME_USERS !lecture”, simply informs sudo to not send a short lecture message about the policy use of the program to users with full time access on the server. FULLTIME_USERS ALL = ALL Here we allow every user who is listed in the “FULLTIME_USERS” alias of the configuration file to run all commands as super-user “root” on the system. This is the equivalence of what we achieve with the PAM security feature. PARTTIME_USERS ALL = HTTP, FTP, SMTP, SQL, BIND Here we allow every user who is listed in the “PARTTIME_USERS” alias of the configuration file to only run the specified commands as super-user “root”. These users will be able to edit, run and restart the specified services and nothing else. Therefore, they only have limited “root” access on the server to accomplish their tasks.
- 447. Sudo 1 CHAPTER 7 447 SecuringSudo This section deals specifically with actions we can take to improve and tighten security under Sudo. Sudo is very good and well-written software with high security in mind. Once properly compiled, installed, and configured, there are only some little things that we can do to better secure it. Most important of all the security measures are already made within the software. Disabling su access on the server Once sudo is configured and running on your server, you really don’t need to continue to use ‘su’ to get “root” access on the system. Here, I show you how to disable ‘su’ command to provide “root” access. The simpler approach will be to remove the SUID bit on the ‘su’ program. In this way, no one will be able to use it to gain “root” anymore. • To remove the SUID bit on the ‘su’ command, use the following command: [root@deep /]# chmod 0511 /bin/su Sudo Users Tools The commands and options listed bellows are some that we use regularly, but many more exist, and you should check the manual pages and documentation of Sudo for more details. Running sudo for user with full “root” access Basically, sudo is used by prepending "sudo" (followed by a space) to your command. It then prompts you for your personal password or root password depending of the configuration and then checks the /etc/sudoers configuration file to make sure you have "sudo" permission to run that command on the server. Sudo runs the command as root (or another user) and logs the details to syslog. It also logs problems and other invalid uses. When you have full “root” privileges on the system because you are listed in the sudoers file as one user with all “root” access right, you can run the sudo command with the -s (shell) option to become the super-user “root” on the server. • To sudo as super-user “root” with shell access, use the following command: [sysadmin@deep /]# sudo –s Password: To be able to use the above command, you should have all “root” access rights in the sudoers configuration file. Please note that, in this example, you have to enter the super-user “root” password and not the password of the user “sysadmin”.
- 448. Sudo 1 CHAPTER 7 448 Runningsudo for user with limited “root” access If you are a user with limited “root” access rights on the server, you cannot use the previous sudo command to sudo as super-user “root” on the server. Instead, you should prepend the sudo command with the command you want to run as “root” user. This supposes that commands you expect to run are listed as allowed commands to use inside the sudoers configuration file. • To sudo as super-user “root” with limited access, use the following command: [mary@deep /]# sudo /etc/init.d/httpd restart Password: The above command will restart the httpd web server daemon on the system even if you are the user called “mary” because the sudoers file allows you to do it as super-user “root”. Please note that in this example you have to enter the password of user “mary” and not the password of the super-user “root”. Further documentation For more details, there are some manual pages you can read: $ man sudoers (5) - List of which users may execute what. $ man sudo (8) - Execute a command as another user. $ man visudo (8) - Used to edit the sudoers file.
- 451. sXid IN THIS CHAPTER 1.Compiling - Optimizing & Installing sXid 2. Configuring sXid 3. sXid Administrative Tools
- 453. sXid 1 CHAPTER 8 453 LinuxsXid Abstract SUID/SGID files can be a security hazard. To reduce the risks, we have previously removed the 's' bits from root-owned programs that won't require such privileges (See chapter related to General System Security), but future and existing files may be set with these ‘s’ bits enabled without you being notified. sXid is an all in one suid/sgid monitoring program designed to be run by “cron” on a regular basis. Basically, it tracks any changes in your s[ug]id files and folders. If there are any new ones, ones that aren't set any more, or they have changed bits or other modes then it reports the changes in an easy to read format via email or on the command line. sXid will automate the task to find all SUID/SGID on your server and report them to you. Once installed you can forget it and it will do the job for you. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest sXid version number is 4.0.2 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages The following is based on information listed by sXid as of 2002/06/24. Please check ftp://marcus.seva.net/pub/sxid/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: sXid Homepage: ftp://marcus.seva.net/pub/sxid/ sXid FTP Site: 137.155.111.51 You must be sure to download: sxid_4.0.2.tar.gz
- 454. sXid 1 CHAPTER 8 454 Pristinesource If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install sXid, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > sXid1 • And the following one after you install the software: [root@deep root]# find /* > sXid2 • Then use the following command to get a list of what changed: [root@deep root]# diff sXid1 sXid2 > sXid-Installed Compiling - Optimizing & Installing sXid Below are the steps that you must make to configure, compile and optimize the sXid software before installing it on your system. First off, we install the program as user 'root' so as to avoid any authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp sxid_version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf sxid_version.tar.gz Step 2 Now move into the newly created sXid source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created sXid directory use the following command: [root@deep tmp]# cd sxid-4.0.2/ • To configure and optimize sXid use the following compilation lines: CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS ./configure --prefix=/usr --sysconfdir=/etc --mandir=/usr/share/man WARNING: Pay special attention to the compile CFLAGS line above. We optimize sXid for an i686 CPU architecture with the parameter “-march=i686”. Please don’t forget to adjust this CFLAGS line to reflect your own system.
- 455. sXid 1 CHAPTER 8 455 Step3 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the sXid software: [root@deep sXid-4.0.2]# cd [root@deep root]# find /* > sXid1 [root@deep root]# cd /var/tmp/sxid-4.0.2/ [root@deep sxid-4.0.2]# make install [root@deep sxid-4.0.2]# cd [root@deep root]# find /* > sXid2 [root@deep root]# diff sXid1 sXid2 > sXid-Installed The above commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then install the binaries and any supporting files into the appropriate locations. Step 4 Once the configuration, optimization, compilation, and installation of the sXid software have been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete sXid and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf sxid-version/ [root@deep tmp]# rm -f sxid_version_tar.gz The rm command as used above will remove all the source files we have used to compile and install sXid. It will also remove the sXid compressed archive from the /var/tmp directory. Configuring sXid After sXid has been built and installed successfully in your system, your next step is to configure and customize its configuration files to fit your needs. /etc/sxid.conf: (The sXid Configuration File) /etc/cron.daily/sxid: (The sXid Cron File) /etc/sxid.conf: The sXid Configuration File The configuration file for sXid allows you to set options that modify the operation of the program. It is well commented and very basic. Step 1 We must change the default one to fit our requirements and operating system. The text in bold are the parts of the configuration file that must be customized and adjusted to satisfy our needs. • Edit the sxid.conf file (vi /etc/sxid.conf) and set your needs. Below is what we recommend you. # Configuration file for sXid # Note that all directories must be absolute with no trailing /'s # Where to begin our file search SEARCH = "/"
- 456. sXid 1 CHAPTER 8 456 #Which subdirectories to exclude from searching EXCLUDE = "/proc /mnt /cdrom /floppy" # Who to send reports to EMAIL = "root" # Always send reports, even when there are no changes? ALWAYS_NOTIFY = "no" # Where to keep interim logs. This will rotate 'x' number of # times based on KEEP_LOGS below LOG_FILE = "/var/log/sxid.log" # How many logs to keep KEEP_LOGS = "5" # Rotate the logs even when there are no changes? ALWAYS_ROTATE = "no" # Directories where +s is forbidden (these are searched # even if not explicitly in SEARCH), EXCLUDE rules apply FORBIDDEN = "/home /tmp" # Remove (-s) files found in forbidden directories? ENFORCE = "yes" # This implies ALWAYS_NOTIFY. It will send a full list of # entries along with the changes LISTALL = "no" # Ignore entries for directories in these paths # (this means that only files will be recorded, you # can effectively ignore all directory entries by # setting this to "/"). The default is /home since # some systems have /home g+s. IGNORE_DIRS = "/home" # File that contains a list of (each on it's own line) # other files that sxid should monitor. This is useful # for files that aren't +s, but relate to system # integrity (tcpd, inetd, apache...). # EXTRA_LIST = "/etc/sxid.list" # Mail program. This changes the default compiled in # mailer for reports. You only need this if you have changed # it's location and don't want to recompile sxid. MAIL_PROG = "/bin/mail" Step 2 Now, for security reasons, change the mode of this file to be 0400. • This can be done with the following command: [root@deep /]# chmod 400 /etc/sxid.conf
- 457. sXid 1 CHAPTER 8 457 /etc/cron.daily/sxid:The sXid Cron File The sxid file is a small script executed automatically by the “crond” program of your server each day to tracks any changes in your s[ug]id files and folders. Step 1 If there are any new ones, ones that aren't set any more, or they have changed bits or other modes then it reports the changes. If you intend to automate this task, follow the simple steps below. • Create the sxid script file (touch /etc/cron.daily/sxid) and add the following lines: #!/bin/sh SXID_OPTS= if [ -x /usr/bin/sxid ]; then /usr/bin/sxid ${SXID_OPTS} fi Step2 Now, make this script executable and change its permissions to be 0510. • This can be done with the following command: [root@deep /]# chmod 510 /etc/cron.daily/sxid sXid Administrative Tools After your desired configuration options have been set and the program is running, we can play with its utility. The sXid software is meant to run as a cronjob. It must run once a day, but busy shell boxes may want to run it twice a day. You can also run this manually for spot-checking. • To run sXid manually, use the command: [root@deep /]# sxid -k sXid Vers : 4.0.2 Check run : Thu Apr 25 19:35:36 2002 This host : deep.openna.com Spotcheck : /root Excluding : /proc /mnt /cdrom /floppy Ignore Dirs: /home Forbidden : /home /tmp (enforcing removal of s[ug]id bits in forbidden paths) No changes found This checks for changes by recursing the current working directory. Log files will not be rotated and no email sent. All output will go to stdout. Further documentation For more details, there are some manual pages you can read: $ man sxid.conf (5) - Configuration settings for sxid. $ man sxid (1) - Check for changes in s[ug]id files and directories.
- 459. LogSentry IN THIS CHAPTER 1.Compiling - Optimizing & Installing LogSentry 2. Configuring LogSentry
- 461. LogSentry 1 CHAPTER 9 461 LinuxLogSentry Abstract One of the most important tasks in the security world is to regularly check the log files. Often the daily activities of an administrator doesn’t allow them the time to do this task and this can bring about problems. Don’t let the media image fool you, most hackers you’ll run across are not very crafty and make a lot of noise ratting your system’s door knob…then again they can be as noisy as they want really because there is a 99.99% chance the system administrator won’t know anyway <Craig>. Auditing and logging system events is important! What’s more important is that system administrators be aware of these events, so they can prevent problems that will inevitably occur if you have a system connected to the Internet. Unfortunately for most UNIX administrators, it doesn't matter how much you log activity if nobody ever checks the logs, which is often the case. This is where LogSentry also knows in the past as Logcheck will help. LogSentry automates the auditing process and weeds out "normal" log information to give you a condensed look at problems and potential troublemakers and then mailed to wherever you please. It is a software package that is designed to automatically run and check system log files for security violations and unusual activity by utilizing a program called “logtail” that remembers the last position it read from in a log file and uses this position on subsequent runs to process new information. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest LogSentry version number is 1.1.1 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages Please check http://www.psionic.com/products/logsentry.html regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: LogSentry Homepage: http://www.psionic.com/ You must be sure to download: logsentry-1.1.1.tar.gz
- 462. LogSentry 1 CHAPTER 9 462 Pristinesource If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install LogSentry, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > LogSentry1 • And the following one after you install the software: [root@deep root]# find /* > LogSentry2 • Then use the following command to get a list of what changed: [root@deep root]# diff LogSentry1 LogSentry2 > LogSentry-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In our example above, we use the /root directory of the system to store all the generated file lists. Compiling - Optimizing & Installing LogSentry Below are the steps that you must make to configure, compile and optimize the LogSentry software before installing it on your system. First off, we install the program as user 'root' so as to avoid any authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp logsentry-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf logsentry-version.tar.gz Step 2 In order to check that the version of LogSentry, which you are going to install, is an original and unmodified one, use the command described below to check its MD5 hashes checksum. • To verify the MD5 checksum of LogSentry, use the following command: [root@deep tmp]# md5sum logsentry-1.1.1.tar.gz This should yield an output similar to this: e97c2f096e219e20310c1b80e9e1bc29 logsentry-1.1.1.tar.gz Now check that this checksum is exactly the same as the one published on the LogSentry website at the following URL: http://www.psionic.com/downloads/checksums.md5
- 463. LogSentry 1 CHAPTER 9 463 Step3 There are some source files to modify before going into the configuration and compilation of the program; the changes allow us to configure the program for our PATH environment variable under Linux. Therefore, move into the newly created LogSentry source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created LogSentry directory use the following command: [root@deep tmp]# cd logcheck-1.1.1/ Step 4 Here, we have to change default locations of different LogSentry configuration files on the system. To archive these modifications, we must edit the logcheck.sh script file as follow. • Edit the logcheck.sh file and change all of the targeted lines in the order shown below: a) vi +47 systems/linux/logcheck.sh and change the line: LOGTAIL=/usr/local/bin/logtail To read: LOGTAIL=/usr/bin/logtail b) vi +55 systems/linux/logcheck.sh and change the line: TMPDIR=/usr/local/etc/tmp To read: TMPDIR=/var/logsentry c) vi +92 systems/linux/logcheck.sh and change the line: HACKING_FILE=/usr/local/etc/logcheck.hacking To read: HACKING_FILE=/etc/logsentry/hacking d) vi +101 systems/linux/logcheck.sh and change the line: VIOLATIONS_FILE=/usr/local/etc/logcheck.violations To read: VIOLATIONS_FILE=/etc/logsentry/violations
- 464. LogSentry 1 CHAPTER 9 464 e)vi +118 systems/linux/logcheck.sh and change the line: VIOLATIONS_IGNORE_FILE=/usr/local/etc/logcheck.violations.ignore To read: VIOLATIONS_IGNORE_FILE=/etc/logsentry/violations.ignore f) vi +125 systems/linux/logcheck.sh and change the line: IGNORE_FILE=/usr/local/etc/logcheck.ignore To read: IGNORE_FILE=/etc/logsentry/ignore Step 5 The Makefile file of LogSentry needs some modifications too. As for the previous step, we will change default locations of some LogSentry files, binary and will add the required optimization FLAGS for our CPU architecture. • Edit the Makefile file and change all of the targeted lines in the order shown below: a) vi +14 Makefile and change the line: CFLAGS = -O To read: CFLAGS = -O2 -march=i686 -funroll-loops b) vi +22 Makefile and change the line: INSTALLDIR = /usr/local/etc To read: INSTALLDIR = /etc/logsentry c) vi +25 Makefile and change the line: INSTALLDIR_BIN = /usr/local/bin To read: INSTALLDIR_BIN = /usr/bin
- 465. LogSentry 1 CHAPTER 9 465 d)vi +30 Makefile and change the line: INSTALLDIR_SH = /usr/local/etc To read: INSTALLDIR_SH = /usr/sbin e) vi +30 Makefile and change the line: TMPDIR = /usr/local/etc/tmp To read: TMPDIR = /var/logsentry Step 6 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the LogSentry software: [root@deep logcheck-1.1.1]# cd [root@deep root]# find /* > LogSentry1 [root@deep root]# cd /var/tmp/logcheck-1.1.1/ [root@deep logcheck-1.1.1]# mkdir –m0700 /etc/logsentry [root@deep logcheck-1.1.1]# make linux [root@deep logcheck-1.1.1]# strip /usr/bin/logtail [root@deep logcheck-1.1.1]# cd /etc/logsentry/ [root@deep logsentry]# mv logcheck.hacking hacking [root@deep logsentry]# mv logcheck.violations violations [root@deep logsentry]# mv logcheck.violations.ignore violations.ignore [root@deep logsentry]# mv logcheck.ignore ignore [root@deep logsentry]# cd [root@deep root]# find /* > LogSentry2 [root@deep root]# diff LogSentry1 LogSentry2 > LogSentry-Installed The above commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then install the binaries and any supporting files into the appropriate locations. Step 7 Once the configuration, optimization, compilation, and installation of the LogSentry software have been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete LogSentry and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf logcheck-version/ [root@deep tmp]# rm -f logsentry-version.tar.gz The rm command as used above will remove all the source files we have used to compile and install LogSentry. It will also remove the LogSentry compressed archive from the /var/tmp directory.
- 466. LogSentry 1 CHAPTER 9 466 ConfiguringLogSentry After LogSentry has been built and installed successfully in your system, your next step is to check its configuration files to see if they fit your needs. /etc/logsentry/hacking: (The LogSentry Hacking File) /etc/logsentry/ignore: (The LogSentry Ignore File) /etc/logsentry/violations: (The LogSentry Violation File) /etc/logsentry/violations.ignore: (The LogSentry Violation Ignore File) From the default install, there are no LogSentry configuration files to modify, the default entries look fine and if you want to make some personal adjustment, all you have to do is to edit the related LogSentry configuration files located under /etc/logsentry directory. More information about the operation of each one is contained in the INSTALL file of LogSentry under its uncompressed source directory. Although the fact that there is no LogSentry configuration files to change, the last action to make before using the program is to automate it. Step 1 Create a file called logsentry under the /etc/cron.daily directory and add the following lines to set LogSentry to run once per day. • To create the logsentry file under /etc/cron.daily directory, type the following lines in your terminal (as root): cat <<EOF > /etc/cron.daily/logsentry #!/bin/sh # Hourly check Log files for security violations and unusual activity. /usr/bin/logcheck.sh EOF Step 2 Now, make this script executable and change its permissions to be 0510. • This can be done with the following commands: [root@deep /]# chmod 510 /etc/cron.daily/logsentry WARNING: Remember, in our configuration and installation, LogSentry does not report anything via email if it has nothing useful to say.
- 467. HostSentry IN THIS CHAPTER 1.Compiling - Optimizing & Installing HostSentry 2. Configuring HostSentry
- 469. HostSentry 2 CHAPTER 0 469 LinuxHostSentry Abstract On Linux servers to accomplish various administrative tasks it is important to have shell access. This shell access can be made from a remote connection or from a local connection but it doesn’t matter, we always need to have some shell access on the system and it’s rare, if not impossible, to never have the requirement to login in to the server. At least, the super-user “root” will be allowed to get access to the system and for this reason it becomes clear that a tool which can help us to monitor who’s connected on the Linux server is important. Fortunately, a tool exists and it’s called “HostSentry” from Psionic Technologies again. HostSentry is a host based intrusion detection tool that performs Login Anomaly Detection (LAD). This tool allows administrators to spot strange login behavior and quickly respond to compromised accounts and unusual behavior. We can use it on all servers where shell access is allowed on the system, for known and trusted, users to spot a login problem before it becomes an embarrassing incident. When HostSentry is installed on your server, a large number of useful possibilities begin to emerge from a single login record and we can track and avoid an anomalous event that seems just a little out of place for a known user. For example imagine the following: - Carol the web master, suddenly logs into her shell account at 5:00 AM from China, Sweden, Malaysia and South Korea. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest HostSentry version number is 0.02 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages Please check http://www.psionic.com/products/hostsentry.html regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: HostSentry Homepage: http://www.psionic.com/ You must be sure to download: hostsentry-0.02.tar.gz
- 470. HostSentry 2 CHAPTER 0 470 Prerequisites HostSentryrequires that the listed software below be already installed on your system to be able to compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source archive files. Please make sure you have this program installed on your machine before you proceed with this chapter. Python, which allows HostSentry to run, must already be installed on your system to be able to compile and use the HostSentry software. Pristine source If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install HostSentry, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > HostSentry1 • And the following one after you install the software: [root@deep root]# find /* > HostSentry2 • Then use the following command to get a list of what changed: [root@deep root]# diff HostSentry1 HostSentry2 > HostSentry-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In our example above, we use the /root directory of the system to store all the generated file lists. Compiling - Optimizing & Installing HostSentry Below are the steps that you must make to configure, compile and optimize the HostSentry software before installing it on your system. First off, we install the program as user 'root' so as to avoid any authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp hostsentry-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf hostsentry-version.tar.gz
- 471. HostSentry 2 CHAPTER 0 471 Step2 In order to check that the version of HostSentry, which you are going to install, is an original and unmodified one, use the command described below to check its MD5 hashes checksum. • To verify the MD5 checksum of HostSentry, use the following command: [root@deep tmp]# md5sum hostsentry-0.02.tar.gz This should yield an output similar to this: 3de0bbb7d456bb53683de56dfdf98362 hostsentry-0.02.tar.gz Now check that this checksum is exactly the same as the one published on the HostSentry website at the following URL: http://www.psionic.com/downloads/checksums.md5 Step 3 There are some source files to modify before going into the configuration and compilation of the program; the changes allow us to configure the program for our PATH environment variable under Linux. Therefore, move into the newly created HostSentry source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created HostSentry directory use the following command: [root@deep tmp]# cd hostsentry-0.02/ Step 4 First, we have to define directories where we want HostSentry to be installed on our system. Editing the Makefile script file as follows does this: • Edit the Makefile file and change all of the targeted lines in the order shown below: a) vi +7 Makefile and change the line: INSTALLDIR = /usr/local/abacus/hostsentry To read: INSTALLDIR = /etc/hostsentry LIBDIR= /usr/lib/hostsentry b) vi +21 Makefile and change the lines: @echo "Installing HostSentry in: $(INSTALLDIR)" install -d -g 0 -o root -m 0700 $(INSTALLDIR) install -d -g 0 -o root -m 0700 $(INSTALLDIR)/modules install -g 0 -o root -m 0700 host* $(INSTALLDIR) install -g 0 -o root -m 0700 module* $(INSTALLDIR)/modules To read: install -d -m 0700 $(INSTALLDIR) install -d -m 0700 $(LIBDIR)/modules install -m 0400 host* $(LIBDIR) install -m 0400 module* $(LIBDIR)/modules
- 472. HostSentry 2 CHAPTER 0 472 Step5 Once we have defined directories where we want to install the program, we have to change the default locations of some HostSentry files, and modules. • Edit the hostSentryConfig.py file (vi +38 hostSentryConfig.py) and change the line: CONFIG='/usr/local/abacus/hostsentry/hostsentry.conf' To read: CONFIG='/etc/hostsentry/hostsentry.conf' • Edit the hostSentryStat.py file (vi +141 hostSentryStat.py) and change the line: db = '/usr/local/abacus/hostsentry/hostsentry.db' To read: db = '/var/hostsentry/hostsentry.db' • Edit the moduleForeignDomain.py file (vi +45 moduleForeignDomain.py) and change the line: ALLOW_FILE = '/moduleForeignDomain.allow' To read: ALLOW_FILE = 'moduleForeignDomain.allow' • Edit the moduleForeignDomain.py file (vi +63 moduleForeignDomain.py) and change the line: allowPath = config.parseToken('MODULE_PATH') To read: allowPath = '/etc/hostsentry/' • Edit the moduleMultipleLogins.py file (vi +49 moduleMultipleLogins.py) and change the line: ALLOW_FILE = '/moduleMultipleLogins.allow' To read: ALLOW_FILE = 'moduleMultipleLogins.allow'
- 473. HostSentry 2 CHAPTER 0 473 •Edit the moduleMultipleLogins.py file (vi +78 moduleMultipleLogins.py) and change the line: allowPath = config.parseToken('MODULE_PATH') To read: allowPath = '/etc/hostsentry/' Step 6 Finally, we have to edit the hostsentry.py file and add a new line at the BEGINNING of the file to set the environment variable of the python binary for the program to find and use it when it runs. • Edit the hostsentry.py file (vi hostsentry.py) and add the line: #!/usr/bin/env python Step 7 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the HostSentry software: [root@deep hostsentry-0.02]# cd [root@deep root]# find /* > HostSentry1 [root@deep root]# cd /var/tmp/hostsentry-0.02/ [root@deep hostsentry-0.02]# make install [root@deep hostsentry-0.02]# mkdir -m0700 /var/hostsentry [root@deep logsentry]# cd [root@deep root]# find /* > HostSentry2 [root@deep root]# diff HostSentry1 HostSentry2 > HostSentry-Installed The above commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then install the binaries and any supporting files into the appropriate locations. Step 8 Once the configuration, optimization, compilation, and installation of the HostSentry software have been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete HostSentry and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf hostsentry-version/ [root@deep tmp]# rm -f hostsentry-version.tar.gz
- 474. HostSentry 2 CHAPTER 0 474 ConfiguringHostSentry After HostSentry has been built and installed successfully in your system, your next step is to configure and customize its configuration files. /etc/hostsentry/hostsentry.conf: (The HostSentry Configuration File) /etc/hostentry/hostsentry.ignore: (The HostSentry Ignore File) /etc/hostentry/hostsentry.action: (The HostSentry Action File) /etc/hostentry/hostsentry.modules: (The HostSentry Modules File) /etc/hostentry/moduleForeignDomain.allow /etc/hostentry/moduleMultipleLogins.allow /etc/init.d/hostsentry: (The HostSentry Initialization File) /etc/hostsentry/hostsentry.conf: The HostSentry Config File The hostsentry.conf file is the main configuration file for HostSentry. It is in this file that HostSentry gets all of its information about the way it should run on your system. Step 1 By default, the hostsentry.conf file does not exist after installation and we have to create it. • Create the hostsentry.conf file (touch /etc/hostsentry/hostsentry.conf) and set your needs. Below is what we recommend you have in your file. IGNORE_FILE = "/etc/hostsentry/hostsentry.ignore" ACTION_FILE = "/etc/hostsentry/hostsentry.action" MODULE_FILE = "/etc/hostsentry/hostsentry.modules" MODULE_PATH = "/usr/lib/hostsentry/modules" WTMP_FILE = "/var/log/wtmp" DB_FILE = "/var/hostsentry/hostsentry.db" DB_TTY_FILE = "/var/hostsentry/hostsentry.tty.db" WTMP_FORMAT = "384/8:32/44:32/76:256" Step2 Now, set the permissions of the hostsentry.conf file to be (0600/-rw-------) and owned by the super-user ‘root’ for security reasons. • To change the permissions and ownership of the hostsentry.conf file, use: [root@deep /]# chmod 600 /etc/hostsentry/hostsentry.conf [root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.conf
- 475. HostSentry 2 CHAPTER 0 475 /etc/hostsentry/hostsentry.ignore:The HostSentry Ignore File The hostsentry.ignore file contains a list of users you want HostSentry to never process or never take action against with the modules. This is useful for users such as "ftp" who show up in wtmp but would cause a large number of false alarms because of the anonymous access. It is important to note that each user must be placed one per line. Step 1 By default, the hostsentry.ignore file doesn’t exist after installation and we have to create it. • Create the hostsentry.ignore file (touch /etc/hostsentry/hostsentry.ignore) and put in any user account you want to be ignored if it logs in to the system. Below is what we recommend you put in the file. By default, we have no users defined. # Place usernames in this file that you want to ignore (ftp, etc.) Step2 Now, set the permissions of the hostsentry.ignore file to be (0600/-rw-------) and owned by the super-user ‘root’ for security reasons. • To change the permissions and ownership of the hostsentry.ignore file, use: [root@deep /]# chmod 600 /etc/hostsentry/hostsentry.ignore [root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.ignore /etc/hostsentry/hostsentry.action: The HostSentry Action File The hostsentry.action file is used to define actions we want HostSentry to take on the specified module. In our example we inform it to logs, blocks the route connection, blocks the TCP connection and disables the user access. It seems that this feature is not implemented but we define and configure it for future version. Step 1 By default, the hostsentry.action file doesn’t exist after installation, so we have to create it manually. • Create the hostsentry.action file (touch /etc/hostsentry/hostsentry.action) and add in any action you want to be taken. Below is what we recommend you enter. moduleFirstLogin=log,blockRoute,blockTCP,disable Step2 Now, set the permissions of the hostsentry.action file to be (0600/-rw-------) and owned by the super-user ‘root’ for security reasons. • To change the permission mode and ownership of the hostsentry.action file, use: [root@deep /]# chmod 600 /etc/hostsentry/hostsentry.action [root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.action
- 476. HostSentry 2 CHAPTER 0 476 /etc/hostsentry/hostsentry.modules:The HostSentry Modules File The hostsentry.modules file tells HostSentry what modules to execute on login/logout and in what order. If you don't want a particular module to run for whatever reason (false alarms, not interested, etc.) then delete it here. Step 1 By default, the hostsentry.modules file doesn’t exist after installation, so we have to create it. • Create hostsentry.modules file (touch /etc/hostsentry/hostsentry.modules) and add the following module lines to the file. Below is what we recommend. moduleLoginLogout moduleFirstLogin moduleForeignDomain moduleMultipleLogins moduleRhostsCheck moduleHistoryTruncated moduleOddDirnames Step2 Now, set the permissions of the hostsentry.modules file to be (0600/-rw-------) and owned by the super-user ‘root’ for security reasons. • To change the permission and ownership of the hostsentry.modules file, use: [root@deep /]# chmod 600 /etc/hostsentry/hostsentry.modules [root@deep /]# chown 0.0 /etc/hostsentry/hostsentry.modules /etc/hostsentry/moduleForeignDomain.allow The moduleForeignDomain.allow file is used to list all domains from which we don’t want an alert to be sent to the administrator when they log in to the system. Every domain listed in this file will be processed as allowed domains. I recommend you only add your localhost to this file. Step 1 By default, the moduleForeignDomain.allow file doesn’t exist after installation and we have to create it. • Create the moduleForeignDomain.allow file (touch /etc/hostsentry/moduleForeignDomain.allow) and add the following line. :0.0 Step2 Now, set the permissions of the moduleForeignDomain.allow file to be (0600/-rw-------) and owned by the super-user ‘root’ for security reasons. • To change the permission mode and ownership of the moduleForeignDomain.allow file, use the following commands: [root@deep /]# chmod 600 /etc/hostsentry/moduleForeignDomain.allow [root@deep /]# chown 0.0 /etc/hostsentry/moduleForeignDomain.allow
- 477. HostSentry 2 CHAPTER 0 477 /etc/hostsentry/moduleMultipleLogins.allow ThemoduleMultipleLogins.allow file is used to list all hosts from which multiple loggings are allowed. This mean that all hosts listed in this file will be allowed to make multiple connections from different place without an alert to be send to the administrator. Again, I recommend you to only add your localhost to this file as we do below. Step 1 By default, the moduleMultipleLogins.allow file does not exist after installation; therefore we have to create it. • Create the moduleMultipleLogins.allow file (touch /etc/hostsentry/moduleMultipleLogins.allow) and add the following line. Below is what we recommend. # Place hosts in here you want this module to disregard logins from. localhost Step2 Now, set the permissions of the moduleMultipleLogins.allow file to be (0600/-rw-------) and owned by the super-user ‘root’ for security reasons. • To change the permission mode and ownership of the moduleMultipleLogins.allow file, use the following commands: [root@deep /]# chmod 600 /etc/hostsentry/moduleMultipleLogins.allow [root@deep /]# chown 0.0 /etc/hostsentry/moduleMultipleLogins.allow /etc/init.d/hostsentry: The HostSentry Initialization File The /etc/init.d/hostsentry script file is responsible to automatically starting and stopping the HostSentry server on your Linux system. Please note that the following script is suitable for Linux operating systems that use SystemV. If you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to make it work for you. Step 1 Create the hostsentry script file (touch /etc/init.d/hostsentry) and add the following lines inside it: #!/bin/bash # This shell script takes care of starting and stopping HostSentry. # # chkconfig: 345 98 85 # description: HostSentry is a host based intrusion detection tool that # performs Login Anomaly Detection (LAD). This tool allows # administrators to spot strange login behavior and quickly # respond to compromised accounts and unusual behavior. # # processname: hostsentry # config: /etc/hostsentry/hostsentry.conf # pidfile: /var/run/hostsentry.pid # Source function library. . /etc/init.d/functions
- 478. HostSentry 2 CHAPTER 0 478 #Source networking configuration. . /etc/sysconfig/network # Check that networking is up. [ ${NETWORKING} = "no" ] && exit 0 RETVAL=0 prog="HostSentry" start() { if [ -f /var/run/hostsentry.pid ] ; then pid=`cat /var/run/hostsentry.pid` if [ "$pid" != "" ] ; then echo $"HostSentry is already running" exit 0 fi fi echo -n $"Starting $prog: " cd /usr/lib/hostsentry daemon python hostsentry.py RETVAL=$? echo echo `ps aux | grep "python hostsentry.py" | cut --delimiter=" " -f 7` > /var/run/hostsentry.pid [ $RETVAL -eq 0 ] && touch /var/lock/subsys/hostsentry return $RETVAL } stop() { echo -n $"Shutting down $prog: " cd /usr/lib/hostsentry killproc python hostsentry.py RETVAL=$? echo [ $RETVAL -eq 0 ] && rm -f /var/lock/subsys/hostsentry && rm -f /var/run /hostsentry.pid return $RETVAL } restart() { stop start } condrestart() { if [ -f /var/lock/subsys/hostsentry ]; then restart fi } # See how we were called. case "$1" in start) start ;; stop) stop ;; restart) restart
- 479. HostSentry 2 CHAPTER 0 479 ;; condrestart) condrestart ;; *) echo$"Usage: $0 {start|stop|restart|condrestart}" exit 1 ;; esac Step 2 Once the /etc/init.d/hostsentry script file has been created, it is important to make it executable, change its default permissions, create the necessary links and start it. Making this file executable will allow the system to run it, changing its default permission is to allow only the root user to change this file for security reason, and creation of the symbolic links will let your system start the program automatically for you at each system boot. • To make this script executable and to change its default permissions, use the commands: [root@deep /]# chmod 700 /etc/init.d/hostsentry [root@deep /]# chown 0.0 /etc/init.d/hostsentry • To create the symbolic rc.d links for HostSentry, use the following commands: [root@deep /]# chkconfig --add hostsentry [root@deep /]# chkconfig --level 345 hostsentry on • To start HostSentry software manually, use the following command: [root@deep /]# /etc/init.d/hostsentry start Starting HostSentry: [OK]
- 481. PortSentry IN THIS CHAPTER 1.Compiling - Optimizing & Installing PortSentry 2. Configuring Portsentry 3. Removing hosts that have been blocked by PortSentry
- 483. PortSentry 2 CHAPTER 1 483 LinuxPortSentry Abstract Firewalls help us to protect our network from intruders. With them, we can choose which ports we want to open and which ones we don’t. This information is kept private by your organization. Nobody on the outside knows this information, but attackers, as well as spammers, know that for some kinds of attacks you can use a special program to scan all the ports on a server to gleam this valuable information (what is open and what is not). A port scan is a symptom of a larger problem coming your way. It is often the pre-cursor for an attack and is a critical piece of information for properly defending your information resources. PortSentry is a program designed to detect and respond to port scans against a target host in real-time and has a number of options to detect port scans. When it finds one it can react in the following ways: A log indicating the incident is made via syslog(). The target host is automatically dropped. The local host is automatically re-configured to route all traffic to the target to a dead host to make the target system disappear. The local host is automatically re-configured to drop all packets from the target via a local packet filter. The purpose of this is to give to a system administrator a heads up that its host is being probed. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest PortSentry version number is 1.1 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages Please check http://www.psionic.com/products/portsentry.html regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: PortSentry Homepage: http://www.psionic.com/ You must be sure to download: portsentry-1.1.tar.gz
- 484. PortSentry 2 CHAPTER 1 484 Pristinesource If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install PortSentry, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > PortSentry1 • And the following one after you install the software: [root@deep root]# find /* > PortSentry2 • Then use the following command to get a list of what changed: [root@deep root]# diff PortSentry1 PortSentry2 > PortSentry-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In our example above, we use the /root directory of the system to store all the generated file lists. Compiling - Optimizing & Installing PortSentry Below are the steps that you must make to configure, compile and optimize the PortSentry software before installing it on your system. First off, we install the program as user 'root' so as to avoid any authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp portsentry-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf portsentry-version.tar.gz Step 2 In order to check that the version of PortSentry, which you are going to install, is an original and unmodified one, use the command described below to check its MD5 hashes checksum. • To verify the MD5 checksum of PortSentry, use the following command: [root@deep tmp]# md5sum portsentry-1.1.tar.gz This should yield an output similar to this: 782839446b7eca554bb1880ef0882670 portsentry-1.1.tar.gz Now check that this checksum is exactly the same as the one published on the PortSentry website at the following URL: http://www.psionic.com/downloads/checksums.md5
- 485. PortSentry 2 CHAPTER 1 485 Step3 There are some source files to modify before going into the configuration and compilation of the program; the changes allow us to configure the program for our PATH environment variable under Linux. Therefore, move into the newly created PortSentry source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created PortSentry directory use the following command: [root@deep tmp]# cd portsentry-1.1/ Step 4 Here, we have to change default locations of different PortSentry configuration files on the system and add the required optimization FLAGS for our CPU architecture. To make these modifications, we must edit the Makefile script file as follows. • Edit the Makefile file and change all of the targeted lines in the order shown below: a) vi +29 Makefile and change the line: CFLAGS = -O -Wall To read: CFLAGS = -O2 -march=i686 -funroll-loops -Wall b) vi +40 Makefile and change the lines: INSTALLDIR = /usr/local/psionic CHILDDIR=/portsentry To read: INSTALLDIR = /usr/sbin CONFIGDIR = /etc/portsentry c) vi +71 Makefile and change the line: @if [ ! -d $(INSTALLDIR) ]; then /bin/mkdir $(INSTALLDIR); fi To read: @if [ ! -d $(INSTALLDIR) ]; then /bin/mkdir -p $(INSTALLDIR); fi d) vi +73 Makefile and change the lines: @if [ "$(INSTALLDIR)" = "/usr/local/psionic" ]; then /bin/chmod 700 $(INSTALLDIR) ; fi @echo "Creating portsentry directory $(INSTALLDIR)$(CHILDDIR)" @if [ ! -d $(INSTALLDIR)$(CHILDDIR) ]; then /bin/mkdir $(INSTALLDIR)$(CHILDDIR); fi To read: @if [ "$(INSTALLDIR)" = "/usr/sbin" ]; then /bin/chmod 700 $(INSTALLDIR) ; fi @echo "Creating portsentry directory $(CONFIGDIR)" @if [ ! -d $(CONFIGDIR) ]; then /bin/mkdir -p $(CONFIGDIR); fi
- 486. PortSentry 2 CHAPTER 1 486 e)vi +77 Makefile and change the line: chmod 700 $(INSTALLDIR)$(CHILDDIR) To read: chmod 700 $(CONFIGDIR) f) vi +79 Makefile and change the lines: cp ./portsentry.conf $(INSTALLDIR)$(CHILDDIR) cp ./portsentry.ignore $(INSTALLDIR)$(CHILDDIR) cp ./portsentry $(INSTALLDIR)$(CHILDDIR) To read: cp ./portsentry.conf $(CONFIGDIR) cp ./portsentry.ignore $(CONFIGDIR) cp ./portsentry $(INSTALLDIR) g) vi +83 Makefile and change the lines: chmod 600 $(INSTALLDIR)$(CHILDDIR)/portsentry.ignore chmod 600 $(INSTALLDIR)$(CHILDDIR)/portsentry.conf chmod 700 $(INSTALLDIR)$(CHILDDIR)/portsentry To read: chmod 600 $(CONFIGDIR)/portsentry.ignore chmod 600 $(CONFIGDIR)/portsentry.conf chmod 500 $(INSTALLDIR)/portsentry h) vi +88 Makefile and change the lines: @echo "Edit $(INSTALLDIR)$(CHILDDIR)/portsentry.conf and change" To read: @echo "Edit $(CONFIGDIR)/portsentry.conf and change" i) vi +94 Makefile and change the lines: @echo "and config files ($(INSTALLDIR)$(CHILDDIR))." To read: @echo "and config files $(CONFIGDIR)."
- 487. PortSentry 2 CHAPTER 1 487 Step5 The second file that we will modify is the portsentry_config.h header file. In this file, we will change the default install location of the configuration file for PortSentry. • Edit the portsentry_config.h file (vi +32 portsentry_config.h) and change the following line: #define CONFIG_FILE "/usr/local/psionic/portsentry/portsentry.conf" To read: #define CONFIG_FILE "/etc/portsentry/portsentry.conf" Step 6 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the PortSentry software: [root@deep portsentry-1.1]# cd [root@deep root]# find /* > PortSentry1 [root@deep root]# cd /var/tmp/portsentry-1.1/ [root@deep portsentry-1.1]# make linux [root@deep portsentry-1.1]# make install [root@deep portsentry-1.1]# strip /usr/sbin/portsentry [root@deep portsentry-1.1]# mkdir -m0700 /var/portsentry [root@deep portsentry-1.1]# cd [root@deep root]# find /* > PortSentry2 [root@deep root]# diff PortSentry1 PortSentry2 > PortSentry-Installed The above commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then install the binaries and any supporting files into the appropriate locations. Step 7 Once the configuration, optimization, compilation, and installation of the PortSentry software have been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete PortSentry and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf portsentry-version/ [root@deep tmp]# rm -f portsentry-version.tar.gz Configuring PortSentry After PortSentry has been built and installed successfully on your system, your next step is to configure and customize its configuration files to fit your needs. /etc/portsentry/portsentry.conf: (The PortSentry Configuration File) /etc/portsentry/portsentry.ignore: (The PortSentry Ignore File) /etc/portsentry/portsentry.modes: (The PortSentry Modes File) /etc/init.d/portsentry: (The PortSentry Initialization File)
- 488. PortSentry 2 CHAPTER 1 488 /etc/portsentry/portsentry.conf:The PortSentry Config File The portsentry.conf file is the main configuration file for PortSentry. It is in this file that PortSentry gets all of its information about the way it should runs your system. You can specify which ports you want PortSentry to listen to, which IP addresses are to be denied, monitored, ignored, or have their automatic responses disabled, and so on. • Edit the portsentry.conf file (vi /etc/portsentry/portsentry.conf) and set your needs. Below is what we recommend. TCP_PORTS="1,11,81,82,83,1080,1720,1863,5190,8080" UDP_PORTS="1,7,9,81,82,83,1080,1720,1863,5190,8080" ADVANCED_PORTS_TCP="1024" ADVANCED_PORTS_UDP="1024" ADVANCED_EXCLUDE_TCP="113,139" ADVANCED_EXCLUDE_UDP="520,138,137,67" IGNORE_FILE="/etc/portsentry/portsentry.ignore" HISTORY_FILE="/var/portsentry/portsentry.history" BLOCKED_FILE="/var/portsentry/portsentry.blocked" RESOLVE_HOST="0" BLOCK_UDP="0" BLOCK_TCP="1" KILL_ROUTE="/sbin/route add -host $TARGET$ reject" SCAN_TRIGGER="0" PORT_BANNER="** UNAUTHORIZED ACCESS PROHIBITED **" This tells the posrtsentry.conf file to set itself up for this particular configuration with: TCP_PORTS="1,11,81,82,83,1080,1720,1863,5190,8080" The option “TCP_PORTS” specifies which TCP ports we want PortSentry to listen to for scan attacks. It is important to note that this option is used by all PortSentry modes except the Advanced Stealth Scan Detection mode that completely ignores this option because it uses a more advanced and a more secure method to monitor ports. Remember that the Advanced Stealth Scan Detection is what we use in this configuration; therefore we don’t really need to define this option. With the other scan detection modes; you have to define all the TCP ports from which you want PortSentry to monitor here. UDP_PORTS="1,7,9,81,82,83,1080,1720,1863,5190,8080" This option “UDP_PORTS” specifies which UTP ports we want PortSentry to listen for scan attacks on. As with the above option, it is important to note that this option is used by all PortSentry modes except the Advanced Stealth Scan Detection mode which completely ignores this option because it uses a more advanced and a more secure method to monitor ports. Again, Advanced Stealth Scan Detection is what we use in this configuration; therefore we don’t really need to define this option. On other scan detection modes, you have to define here all the UDP ports from which you want PortSentry to monitor. ADVANCED_PORTS_TCP="1024" The option “ADVANCED_PORTS_TCP” specifies the highest TCP port number to monitor down from. Any port *below* this number is then monitored by PortSentry in all detection modes. The default is 1024 (reserved port range), and the one we use here for TCP. ADVANCED_PORTS_UDP="1024" The option “ADVANCED_PORTS_UDP” specifies the highest UDP port number to monitor down from. Any port *below* this number is then monitored by PortSentry in all detection modes. The default is 1024 (reserved port range), and the one we use here for UDP.
- 489. PortSentry 2 CHAPTER 1 489 ADVANCED_EXCLUDE_TCP="113,139" Theoption “ADVANCED_EXCLUDE_TCP” specifies the TCP ports that should be manually excluded from monitoring in advanced mode. These are normally ports that may get hit by mistake by remote clients and shouldn't cause alarms. The above TCP ports should be ok for most of us. ADVANCED_EXCLUDE_UDP="520,138,137,67" The option “ADVANCED_EXCLUDE_UDP” specifies the UDP ports that should be manually excluded from monitoring in advanced mode. Again, these are normally ports that may get hit by mistake by remote clients and shouldn't cause alarms. The above UDP ports should be ok for most of us. IGNORE_FILE="/etc/portsentry/portsentry.ignore" The option “IGNORE_FILE” specifies the path to the file that contains IP addresses of hosts you want to always be ignored by PortSentry. See later in this chapter for more information about his file. HISTORY_FILE="/var/portsentry/portsentry.history" The option “HISTORY_FILE” specifies the path to the file that contains hosts that have been denied by PortSentry. BLOCKED_FILE="/var/portsentry/portsentry.blocked" The option “BLOCKED_FILE” specifies the path to the file that contains the IP addresses of blocked hosts by PortSentry. It is important to note that all IP addresses listed in this file are blocked by PortSentry until the program restarts. RESOLVE_HOST="0" The option “RESOLVE_HOST” specifies if we want PortSentry to make DNS resolution or not. In our configuration, we turn off DNS resolution for better performance. The number “1” enable the option and number “0” disable it. This is a performance feature. BLOCK_UDP="0" The option “BLOCK_UDP” is used to disable all automatic responses to UDP probes. Because UDP can be easily forged, it may allow an attacker to start a denial of service attack against the protected host, causing it to block all manner of hosts that should normally be left alone. Setting this option to "0" will disable all responses, although the connections are still logged. BLOCK_TCP="1" The option “BLOCK_TCP” is the same as above, but for TCP. Packet forgery is not as big a problem though, because PortSentry waits for a full connect to occur and this is much harder to forge in the basic modes. Leave this enabled, even for Internet connected hosts. KILL_ROUTE="/sbin/route add -host $TARGET$ reject" The option “KILL_ROUTE” specifies the command to run to drop the offending route if an attack is detected. We can use IPTables here as the command to block attack but it is better to go with the “route” command as we do because IPTables will block the attacker only when connection is closed by the remote host where the “route” command will directly block the attacker. Therefore the above option is more effective and secure than using IPTables command. SCAN_TRIGGER="0" PortSentry has a state engine that will remember hosts that connected to it. Setting this value will tell PortSentry to allow X number of grace port hits before it reacts. This will detect both sequential and random port sweeps. The default is 0, which will react immediately.
- 490. PortSentry 2 CHAPTER 1 490 PORT_BANNER="**UNAUTHORIZED ACCESS PROHIBITED **" The option “PORT_BANNER” specifies a text banner you want displayed to the connecting host if the PortSentry is activated. /etc/portsentry/portsentry.ignore: The PortSentry Ignore File The portsentry.ignore file is where you add any host you wanted to be ignored if it connects to a tripwired port. This should always contain at least the localhost (127.0.0.1) and the IP's of the local interfaces (lo). It is not recommend that you put in every IP on your network. It is well commented and very simple to understand. • Edit the portsentry.ignore file (vi /etc/portsentry/portsentry.ignore) and add in any host you want to be ignored if it connects to a tripwired port. Below is what we recommend. # Put hosts in here you never want blocked. This includes the IP # addresses of all local interfaces on the protected host # (i.e virtual host, mult-home). Keep 127.0.0.1 and 0.0.0.0 to keep # people from playing games. # # PortSentry can support full netmasks for networks as well. Format is: # # <IP Address>/<Netmask> # # Example: # # 192.168.2.0/24 # 192.168.0.0/16 # 192.168.2.1/32 # Etc. # # If you don't supply a netmask it is assumed to be 32 bits. # # 127.0.0.1/32 0.0.0.0 NOTE: Don’t forget to add the IP address of your server to the above list. For example, if I’ve installed PortSentry on one of my server, which has IP address of 207.35.78.3, then I’ll add this IP to the above list.
- 491. PortSentry 2 CHAPTER 1 491 /etc/portsentry/portsentry.modes:The PortSentry Modes File The PortSentry program can be configured in six different modes of operation but be aware that only one protocol mode type can be started at a time. To be more accurate, you can start one TCP mode and one UDP mode, so two TCP modes and one UDP mode, for example, won’t work. • The available PortSentry modes are: portsentry –tcp (Basic port-bound TCP mode) portsentry –udp (Basic port-bound UDP mode) portsentry –stcp (Stealth TCP scan detection mode) portsentry –sudp ("Stealth" UDP scan detection mode) portsentry –atcp (Advanced "Stealth" TCP scan detection mode) portsentry –audp (Advanced "Stealth" UDP scan detection mode) For the best use of this software it is preferable to start PortSentry in Advanced TCP stealth scan detection mode and Advanced UDP stealth scan detection mode. For information about the other modes available, please refer to the README.install and README.stealth file under the PortSentry source directory. With the Advanced TCP stealth scan detection mode “-atcp”, PortSentry will first check to see what ports you have running on your server, then remove these ports from monitoring and will begin watching the remaining ports. This is very powerful and reacts exceedingly quickly for port scanners. It also uses very little CPU time. This mode is the most sensitive and the most effective of all the protection options. The six different modes of operation under which PortSentry can operate must be specified in the configuration file named portsentry.modes located in the /etc/portsentry/ directory. We can add inside this file all the six possible modes of PortSentry, and then uncomment the two we want to use for our server. Step 1 By default, the portsentry.modes file does not exist after installation, and we have to create it. • Create the portsentry.modes file (touch /etc/portsentry/portsentry.modes) and add the following lines inside the file. Below is what we recommend. # These are the available startup modes for PortSentry. Uncomment the # modes you want PortSentry to run in. For information about each # available mode, please see the PortSentry documentation. # # Normal TCP/UDP scanning: #tcp #udp # # Steal TCP/UDP scanning: #stcp #sudp # # Advanced Stealth TCP/UDP scanning: atcp audp
- 492. PortSentry 2 CHAPTER 1 492 Step2 Now,set the permissions of the portsentry.modes file to be (0600/-rw-------) and owned by the super-user ‘root’ for security reasons. • To change the permission mode and ownership of the portsentry.modes file, use: [root@deep /]# chmod 600 /etc/portsentry/portsentry.modes [root@deep /]# chown 0.0 /etc/portsentry/portsentry.modes /etc/init.d/portsentry: The PortSentry Initialization File The /etc/init.d/portsentry script file is responsible to automatically starting and stopping the PortSentry server on your Linux system. Please note that the following script is suitable for Linux operating systems that use SystemV. If you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to make it work for you. Step 1 Create the portsentry script file (touch /etc/init.d/portsentry) and add the following lines inside it: #!/bin/bash # This shell script takes care of starting and stopping the Port Scan Detector. # # chkconfig: 345 98 05 # description: PortSentry Port Scan Detector is part of the Abacus Project # suite of tools. The Abacus Project is an initiative to release # low-maintenance, generic, and reliable host based intrusion # detection software to the Internet community. # # processname: portsentry # config: /etc/portsentry/portsentry.conf # pidfile: /var/run/portsentry.pid # Source function library. . /etc/rc.d/init.d/functions # Source networking configuration. . /etc/sysconfig/network # Check that networking is up. [ ${NETWORKING} = "no" ] && exit 0 RETVAL=0 prog="PortSentry" start() { SENTRYDIR=/etc/portsentry if [ -s $SENTRYDIR/portsentry.modes ] ; then modes=`cut -d "#" -f 1 $SENTRYDIR/portsentry.modes` else modes="tcp udp" fi for i in $modes ; do action "Starting $prog -$i: " /usr/sbin/portsentry -$i RETVAL=$?
- 493. PortSentry 2 CHAPTER 1 493 done echo [$RETVAL = 0 ] && touch /var/lock/subsys/portsentry return $RETVAL } stop() { echo -n "Shutting down $prog: " killproc portsentry RETVAL=$? echo [ $RETVAL = 0 ] && rm -f /var/lock/subsys/portsentry return $RETVAL } # See how we were called. case "$1" in start) start ;; stop) stop ;; restart|reload) stop start RETVAL=$? ;; condrestart) [ -f /var/lock/subsys/portsentry ] && restart || : ;; status) status portsentry ;; *) echo "Usage: portsentry {start|stop|restart|reload|condrestart|status}" exit 1 esac Step 2 Once the /etc/init.d/portsentry script file has been created, it is important to make it executable, change its default permissions, create the necessary links and start it. Making this file executable will allow the system to run it, changing its default permission is to allow only the root user to change this file for security reasons, and creation of the symbolic links to start the program automatically for you at each system boot. • To make this script executable and to change its default permissions, use the command: [root@deep /]# chmod 700 /etc/init.d/portsentry [root@deep /]# chown 0.0 /etc/init.d/portsentry • To create the symbolic rc.d links for PortSentry, use the following command: [root@deep /]# chkconfig --add portsentry [root@deep /]# chkconfig --level 345 portsentry on • To start PortSentry software manually, use the following command: [root@deep /]# /etc/init.d/portsentry start Starting PortSentry: [OK]
- 494. PortSentry 2 CHAPTER 1 494 Removinghosts that have been blocked by PortSentry When PortSentry want to blocks attackers it uses the “route” command of Linux to do it. Every host that has been blocked by PortSentry will appear under the ‘route” command. Below, I show you how to remove hosts that have been blocked by PortSentry from your system. Step 1 We have to use the “route” command to list which hosts are presently blocked by the program. The “route” command also lists other important information about your network routing but we use it in this example to get the list of blocked hosts and to unlock them from the system. • To list which hosts are presently blocked by PortSentry, use the command: [root@deep /]# route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface www.hack.com – 255.255.255.255 !H 0 – 0 - 207.35.78.0 * 255.255.255.224 U 0 0 0 eth0 127.0.0.0 * 255.0.0.0 U 0 0 0 lo default rt.openna.c 0.0.0.0 UG 0 0 0 eth0 In the above example, we can see that “www.hack.com” is listed into our routing table as a domain that has been blocked by PortSentry because it tried to scan our system. The “-” string inform us about the fact that this host is locked. Every host in the routing table with this string “-” is marked as blocked by the system. Step2 Now that we know about the host that has been blocked by PortSentry, we can decide to remove it from the list of blocked hosts on our system. • To remove the blocked host in question, use the following command: [root@deep /]# route del –host www.hack.com reject The above command will remove www.hack.com from the list of blocked hosts in the routing table of our system. The option “del” in the “route” command is what makes it possible to remove the host from the list. Your have to use the above command for any additional hosts that you want to remove from the routing table. Step 3 Finally, we have to edit the portsentry.history file and remove the line corresponding to www.hack.com from the file. This is important for PortSentry to be able to add the site into the list of blocked host in the event that the corresponding host tries to scan your system again. • Edit the portsentry.history file and remove the line corresponding to the host: [root@deep /]# vi /var/portsentry/portsentry.history 1020371099 - 05/02/2002 16:24:59 Host: 1.2.3.4/1.2.3.4 Port: 80 TCP Blocked
- 495. Snort IN THIS CHAPTER 1.Compiling - Optimizing & Installing Snort 2. Configuring Snort 3. Running Snort in a chroot jail
- 497. Snort 2 CHAPTER 2 497 LinuxSnort Abstract From the point of view of security, information is vital and we have to get as much information as we can to quickly discover problem and possible attack on our network. In previous chapters, we have already installed many useful security programs to help us gather information and stop attacks but this is not enough and we have to add to our arsenal another security tool which can scan our network and report possible problems and attacks. This is where Snort will help us. Snort is a flexible libpcap-based packet sniffer/logger tool, which can be used in the most classic sense as a lightweight network intrusion detection system (NIDS) but it is also useful for a wide variety of other uses. It features rules based logging and can perform protocol analysis, content searching/matching and can be used to detect a variety of attacks and probes, such as buffer overflows, stealth port scans, CGI attacks, SMB probes, OS fingerprinting attempts, and much more. Snort has a real-time alerting capability, with alerts being sent to syslog, a separate "alert" file, or as a WinPopup message via Samba's smbclient. Network intrusion detection systems (NIDS) are an important part of any network security architecture. They provide a layer of defense, which monitors network traffic for predefined suspicious activity or patterns, and alert system administrators when potential hostile traffic is detected. This is exactly what we are looking for here, a lightweight network intrusion detection tool that can be deployed to monitor TCP/IP networks and detect a wide variety of suspicious network traffic as well as outright attacks and can provide administrators with enough data to make informed decisions on the proper course of action in the face of suspicious activity. Some could say that PortSentry, which we have installed previously, does the same thing. This is NOT true; PortSentry can be used to block unauthorized ports that have been scanned by attackers and nothing else. Snort goes more deeply with the TCP/IP protocol and provides myriad of security information related to many services running on your server. In general, it is a very good security tool to use with all other security tools as discussed on this book. I highly recommend you to install it on your system if you want to be informed about hostile activities and also methods used by spammers, crackers, etc to probe your network. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest Snort version number is 1.8.7 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux.
- 498. Snort 2 CHAPTER 2 498 Packages Thefollowing is based on information listed by Snort as of 2002/07/08. Please check http://www.snort.org/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: Snort Homepage: http://www.snort.org/ You must be sure to download: snort-1.8.7.tar.gz Prerequisites Snort requires that the listed software below be already installed on your system to be able to compile successfully. If this is not the case, you must install it from your Linux CD-ROM or source archive files. Please make sure you have this program installed on your machine before you proceed with this chapter. Libpcap, which is used extensively by Snort, must already be installed on your system. Tcpdump, which allows some additional features with Snort. Pristine source If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install Snort, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > Snort1 • And the following one after you install the software: [root@deep root]# find /* > Snort2 • Then use the following command to get a list of what changed: [root@deep root]# diff Snort1 Snort2 > Snort-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In our example above, we use the /root directory of the system to store all the generated file lists.
- 499. Snort 2 CHAPTER 2 499 Compiling- Optimizing & Installing Snort Below are the steps that you must make to configure, compile and optimize the Snort software before installing it on your system. First off, we install the program as user 'root' so as to avoid any authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp snort-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf snort-version.tar.gz Step 2 In order to check that the version of Snort, which you are going to install, is an original and unmodified one, use the command described below to check its MD5 hashes checksum. • To verify the MD5 checksum of Snort, use the following command: [root@deep tmp]# md5sum snort-1.8.7.tar.gz This should yield an output similar to this: 29c81d0bc243edb21ba4ab33ee80457e snort-1.8.7.tar.gz Now check that this checksum is exactly the same as the one published on the Snort website at the following URL: http://www.snort.org/dl/snort-1.8.7.tar.gz.md5 Step 3 Snort needs a UID and GID to properly run on the system but this UID/GID cannot run as super-user root; therefore we must create a special user with no shell privileges on the system for running Snort daemon. • To create this special Snort user on OpenNA Linux, use the following command: [root@deep tmp]# groupadd -g 69 snort > /dev/null 2>&1 || : [root@deep tmp]# useradd -c "Snort NIDS" -d /var/log/snort -g 69 -s /bin/false -u 69 snort > /dev/null 2>&1 || : • To create this special Snort user on Red Hat Linux, use the following command: [root@deep tmp]# groupadd -g 69 snort > /dev/null 2>&1 || : [root@deep tmp]# useradd -u 69 -g 69 -s /bin/false -M -r -d /var/log/snort snort > /dev/null 2>&1 || : The above command will create a null account, with no password, no valid shell, no files owned- nothing but a UID and a GID for the program. Remember that Snort daemon does not need to have a shell account on the server.
- 500. Snort 2 CHAPTER 2 500 Step4 Now, edit the shells file (vi /etc/shells) and add a non-existent shell name “/bin/false”, which is the one we used in the passwd command above. [root@deep tmp]# vi /etc/shells /bin/bash2 /bin/bash /bin/sh /bin/false This is our added no-existent shell Step 5 Next, move into the newly created Snort source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created Snort directory use the following command: [root@deep tmp]# cd snort-1.8.7/ • To configure and optimize Snort use the following compilation lines: CFLAGS="-O2 -march=i686 -funroll-loops"; export CFLAGS ./configure --prefix=/usr --sysconfdir=/etc --localstatedir=/var --mandir=/usr/share/man --with-openssl Step 6 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the Snort software: [root@deep snort-1.8.7]# make [root@deep snort-1.8.7]# cd [root@deep root]# find /* > Snort1 [root@deep root]# cd /var/tmp/snort-1.8.7/ [root@deep snort-1.8.7]# make install [root@deep snort-1.8.7]# mkdir -p /var/log/snort [root@deep snort-1.8.7]# mkdir -p /etc/snort [root@deep snort-1.8.7]# chown -R snort.snort /var/log/snort/ [root@deep snort-1.8.7]# install classification.config /etc/snort/ [root@deep snort-1.8.7]# install snort.conf *.rules /etc/snort/ [root@deep snort-1.8.7]# chmod 0644 /etc/snort/* [root@deep snort-1.8.7]# strip /usr/bin/snort [root@deep snort-1.8.7]# cd [root@deep root]# find /* > Snort2 [root@deep root]# diff Snort1 Snort2 > Snort-Installed The above commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then install the binaries and any supporting files into the appropriate locations.
- 501. Snort 2 CHAPTER 2 501 Step7 Once the configuration, optimization, compilation, and installation of the Snort software have been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete Snort and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf snort-version/ [root@deep tmp]# rm -f snort-version.tar.gz Configuring Snort After Snort has been built and installed successfully in your system, your next step is to configure and customize its configuration files to fit your needs. /etc/snort/snort.conf: (The Snort Configuration File) /etc/init.d/snort: (The Snort Initialization File) /etc/snort/snort.conf: The Snort Config File The snort.conf file is the main configuration file for Snort. It is in this file that Snort gets all of its startup information and the way it should run on your system. You can edit it to specify your network variables, preprocessors parameters, output plugging to use and Snort rules files. The Snort configuration file is divided into four different sections. The first section is used to define network variables, the second section is used to configure the preprocessor parameters that Snort should use, the third section is used to configure output plugging to activate, and the last section is used to enable specific Snort rule set. Below, we will explain each section and how you should use them to configure Snort for your server. • Edit the snort.conf file (vi /etc/snort/snort.conf) and set your needs. Below is what we recommend you. var HOME_NET $eth0_ADDRESS var EXTERNAL_NET any var SMTP $HOME_NET var HTTP_SERVERS $HOME_NET var SQL_SERVERS $HOME_NET var DNS_SERVERS $HOME_NET var RULE_PATH ./ preprocessor frag2 preprocessor stream4: detect_scans,detect_state_problems preprocessor stream4_reassemble: both,ports all preprocessor http_decode: 80 preprocessor rpc_decode: 111 32771 preprocessor bo preprocessor telnet_decode preprocessor portscan: $HOME_NET 4 3 portscan.log preprocessor portscan-ignorehosts: 207.35.78.40 207.35.78.41 output alert_syslog: LOG_AUTH LOG_ALERT include classification.config This tells the snort.conf file to set itself up for this particular configuration with:
- 502. Snort 2 CHAPTER 2 502 Thenetwork variables section The first section of the Snort configuration file refers to all network parameters specific to your networking architecture and IP information. This section is used by Snort to get information about the way it should monitor your network. Other options are available as shown below. var HOME_NET $eth0_ADDRESS The option “HOME_NET” is used to specify the network interface on which you want Snort to run and listen. The above parameter allows us to initialize Snort for the IP address and netmask of the network interface which we run Snorton. This is very useful for dialup users and even for static IP addresses. var EXTERNAL_NET any The option “EXTERNAL_NET” is used to specify the external network addresses that Snort should monitor. Here we keep the default setting to monitor any external network addresses meaning that if any IP addresses/hosts try to do something with our server, we will know it because Snort will monitor any external network addresses trying to connect, scan, attack, etc our server. var SMTP $HOME_NET The option “SMTP” is used to specify all SMTP servers that Snort should monitor. The default setting is correct for most of us. Here, we simply inform Snort to monitor SMTP servers running on our network. The variable “$HOME_NET” redirects Snort to the network interface and IP address of the server where it runs meaning that if SMTP services are running on our network, Snort will monitor any connection to them. var HTTP_SERVERS $HOME_NET The option “HTTP_SERVERS” is used to specify all HTTP servers that Snort should monitor. Here again, the default setting is good enough for most of us. We simply inform Snort to monitor HTTP servers running on our network. The variable “$HOME_NET” redirects Snort to the network interface and IP address of the server where it runs, meaning that if HTTP services are running on our network, Snort will monitor any connection to them. var SQL_SERVERS $HOME_NET The option “SQL_SERVERS” is used to specify all SQL servers that Snort should monitor. Again, the default setting is the right one for most of us. We inform Snort to monitor SQL servers running on our network. The variable “$HOME_NET” redirects Snort to the network interface and IP address of the server where it runs meaning that if SQL services are running on our network, Snort will monitor any connection to them. var DNS_SERVERS $HOME_NET The option “DNS_SERVERS” is used to specify all DNS servers that Snort should monitor. Again, the default setting is good for most of us. We simply inform Snort to monitor DNS servers running on our network. The variable “$HOME_NET” redirects Snort to the network interface and IP address of the server where it runs meaning that if DNS services are running on our network, Snort will monitor any connection to them. var RULE_PATH ./ The option “RULE_PATH” simply specifies the path where all Snort rules files are located on the system. You don’t need to change the default setting. Snort use many rules files to get information about what actions to take when an attack is detected. Rule files handle signatures, etc about the specified service. More information about Snort rules can be found later in this chapter.
- 503. Snort 2 CHAPTER 2 503 Thepreprocessors section This section of the Snort configuration file is used to define general configuration for preprocessors. Preprocessors are used to define parameters and options that Snort should use when running. preprocessor frag2 The preprocessor “frag2” enables support for IP defragmentation and fragmentation attacks with Snort. This plug-in will allow Snort to perform IP defragmentation and detect people launching fragmentation attacks (usually DoS) against hosts. The preprocessor has two options associated with it. The options are “timeout” and “memcap”. The ”timeout” option could be used to change the default number of seconds an unfinished fragment will be kept around waiting for completion. The second option “memcap” could be used to limit memory usage of IP defragmentation. The default value for both options are correct and we don’t need to change them. This is a security feature. preprocessor stream4: detect_scans,detect_state_problems The preprocessor “stream4” enables support for full TCP stream reassembly, stateful inspection of TCP streams, etc with Snort. This plug-in will allow Snort to statefully detect various types of portscan, fingerprinting, ECN, etc and will help to defeat stick/snot against TCP rules. This preprocessor has seven options associated with it. The options are “detect_scans”, “detect_state_problems”, “keepstats”, “noinspect”, “timeout”, “memcap”, and “log_flushed_streams”. detect_scans: Used to detect stealth portscans and generate alerts. detect_state_problems: Used to detect TCP state problems. keepstats: Used to keep session statistics. noinspect: Used to turn off stateful inspection only. timeout: Used to set or change the default session timeout counter. memcap: Used to limit memory usage by changing default setting. log_flushed_streams: Used to cause all packets that are stored in the packet buffers to be flushed to disk. In our configuration of this preprocessor, we use “detect_scans” to detect stealth portscans and generate alerts and “detect_state_problems” to detect possible TCP state problems. We don’t need the other options. This is a security feature. preprocessor stream4_reassemble: both,ports all The preprocessor “stream4_reassemble” is a continuation of the above preprocessor parameter and specifies the tcp stream reassembly directive to use with Snort. This preprocessor has five options associated with it. The options are “clientonly”, “serveronly”, “both”, “noalerts”, and “ports”. clientonly: Used to reassemble traffic for the client side of a connection only. serveronly: Used to reassemble traffic for the server side of a connection only. both: Used to reassemble both sides of a session. noalerts: Used to turn off alerts from the stream reassembly stage. ports: Used to specify the ports number to use for reassembly. In our configuration of this preprocessor, we use “both” to reassemble both sides of a session, and “ports all” to turn on reassembly for all ports. We don’t need the other options. This is a security feature.
- 504. Snort 2 CHAPTER 2 504 preprocessorhttp_decode: 80 The preprocessor “http_decode” enables support for normalized HTTP requests with Snort. This preprocessor allow us to defeat hostile attackers trying to stealth themselves from IDSs by mixing these substitutions in with the HTTP request. It has three arguments that you can associate with it. The first is the port number you want it to analyze; this argument should always be present with this preprocessor. The second argument is “-unicode” and you can use it to turn off detection of UNICODE directory traversal attacks. By default, this argument (-unicode) is set with Snort and we remove it in our configuration to make the preprocessor use it. The last argument “-cginull” related to detection of CGI NULL code attacks with the HTTP protocol. If you add “-cginull” to this preprocessor parameter, you will turn off detection of CGI NULL code attacks. In our configuration we don’t specify this argument (-cginull) because we want to use this feature and let Snort detect all possible CGI NULL code attacks on the server. 80: Used to specify the port number you want the preprocessor to analyze. -unicode: Used to turn off detection of UNICODE directory traversal attacks. -cginull: Used to turn off detection of CGI NULL code attacks. In our configuration with this preprocessor, we only specify the port numbers (80) we want the preprocessor to analyze for HTTP services. We don’t need the other arguments. This is a security feature. preprocessor rpc_decode: 111 32771 The preprocessor “rpc_decode” enables support for normalized RPC traffic in much the same way as the http_decode preprocessor. This plug-in takes the port numbers that RPC services are running on as arguments. In our configuration of this preprocessor, we define ports 111 and 32771. You don’t need to change the default setting. preprocessor bo The preprocessor “bo” is used to detect Back Orifice (bo) traffic on the network. It uses the Back Orifice "encryption" algorithm to search for traffic conforming to the Back Orifice protocol. It provides two arguments that you can associate with it. The first is "-nobrute" which turns off the plugin's brute forcing routine and the second argument is a number to use as the default key when trying to decrypt the traffic. -nobrute: Used to turns off the plugin's brute forcing routine. 31337: Used as the default key when trying to decrypt the traffic. In our configuration of this preprocessor, we use the default setting and don’t specify any additional arguments. This is a performance feature. preprocessor telnet_decode The preprocessor “telnet_decode” enables support to normalize telnet negotiation strings from telnet and ftp traffic with Snort. It works in much the same way as the http_decode preprocessor, searching for traffic that breaks up the normal data stream of a protocol and replacing it with a normalized representation of that traffic. This preprocessor requires no arguments.
- 505. Snort 2 CHAPTER 2 505 preprocessorportscan: $HOME_NET 4 3 portscan.log The preprocessor “portscan” enables support to detect UDP packets or TCP SYN packets going to four different ports in less than three seconds. In our configuration, we log all reports to the “portscan.log” file locate under the /var/log/snort directory. preprocessor portscan-ignorehosts: 207.35.78.40 207.35.78.41 The preprocessor “portscan-ignorehosts” is used to list particular host(s) from which we want Snort to ignore traffic. This mean that any host(s) IP address(es) listed in this line will be ignored by Snort. Values are defined in a white space delimited list. The output plug-in section This section of the Snort configuration file is used to configure the output plug-in you decide to use with Snort. Snort can be configured to log all reports to an SQL database of your choice or to run with an external security program or even to log all reports to a local file on the system and many other features. In general, we only need to specify some options as shown below to make Snort work. output alert_syslog: LOG_AUTH LOG_ALERT The option “alert_syslog” is used to log all Snort alerts to syslog on your server. This is what we have to use to get readable Snort reports. include classification.config The option “include classification.config” is used to inform Snort to include the “classification.config” file to its configuration. The Snort “classification.config” file is used to classify and prioritize alerts. We use it to specify what priority each classification should have. The default setting is suitable for all of us. The rule set section This section of the Snort configuration file is used to enable rules files that we hope to use with Snort. Most of the predefined rules are already enabled in the configuration file. Rules are command lines or signatures used to generate alerts based on suspicious activity. You can keep the default setting and Snort will work on your server. You can also get latest rules files from the Snort web site and update your rules if necessary. In general, you only need to comment or uncomment rules that you expect to use with Snort in this section. As we said earlier, Snort uses rule sets to generate and get information about the way it should detect and interpret attacks on your network. Each common service has its own rule set available to use with Snort. In the configuration file, we use and enable all default Snort rules files except some that may provide false alarms. It is up to you to decide which additional rules you want to include with Snort. You can also write your own rule files to use since the software allows us to do it, but this is another story. Please see the Snort website for more information about how to create and use your own rules with Snort.
- 506. Snort 2 CHAPTER 2 506 /etc/init.d/snort:The Snort Initialization File The /etc/init.d/snort script file is responsible to automatically starting and stopping the Snort server on your Linux system. Please note that the following script is suitable for Linux operating systems that use SystemV. If you Linux system use some other methods like BSD, you’ll have to adjust the script bellow to make it work for you. Step 1 Create the snort script file (touch /etc/init.d/snort) and add the following lines inside it: #!/bin/bash # This shell script takes care of starting and stopping the snort IDS daemon. # # chkconfig: 2345 40 60 # description: Snort is a lightweight network intrusion detection tool that # currently detects more than 1100 host and network # vulnerabilities, portscans, backdoors, and more. # Source function library. . /etc/init.d/functions # Source networking configuration. . /etc/sysconfig/network # Check that networking is up. [ ${NETWORKING} = "no" ] && exit 0 # Specify your network interface here INTERFACE=eth0 RETVAL=0 prog="Snort" start() { echo -n $"Starting $prog: " daemon /usr/bin/snort -A fast -u snort -g snort -b -s -z -d -D -i $INTERFACE -c /etc/snort/snort.conf RETVAL=$? echo [ $RETVAL = 0 ] && touch /var/lock/subsys/snort return $RETVAL } stop() { echo -n $"Shutting down $prog: " killproc snort RETVAL=$? echo [ $RETVAL = 0 ] && rm -f /var/lock/subsys/snort return $RETVAL } # See how we were called. case "$1" in start) start ;; stop) stop ;;
- 507. Snort 2 CHAPTER 2 507 status) statussnort ;; restart) stop start ;; condrestart) [ -f /var/lock/subsys/snort ] && restart ;; *) echo $"Usage: $prog {start|stop|status|restart|condrestart}" exit 1 esac exit $RETVAL Step 2 Once the /etc/init.d/snort script file has been created, it is important to make it executable, change its default permissions, create the necessary links and start it. Making this file executable will allow the system to run it, changing its default permission to allow only the root user to change this file for security reasons, and creation of the symbolic links will let the process control initialization of Linux start the program automatically for you at each system boot. • To make this script executable and to change its default permissions, use the command: [root@deep /]# chmod 700 /etc/init.d/snort [root@deep /]# chown 0.0 /etc/init.d/snort • To create the symbolic rc.d links for Snort, use the following command: [root@deep /]# chkconfig --add snort [root@deep /]# chkconfig --level 2345 snort on • To start Snort software manually, use the following command: [root@deep /]# /etc/init.d/snort start Starting Snort: [OK] Running Snort in a chroot jail This section applies only if you want to run Snort in chroot jail environment. To do it, we need to create the required skeleton environment and copy necessary files into this chroot jail. Below are the steps to follow if you want to run Snort with chroot jail support. The main benefit of a chroot jail is that the jail will limit the portion of the file system the daemon can see to the root directory of the jail. Additionally, since the jail only needs to support Snort, the programs available in the jail can be extremely limited. Most importantly, there is no need for setuid-root programs, which can be used to gain root access and break out of the jail.
- 508. Snort 2 CHAPTER 2 508 Necessarysteps to run Snort in a chroot jail: What you're essentially doing is creating a skeleton root file system with just enough components necessary (files, directories) to allow UNIX to do a chroot when Snort starts. The procedures to run Snort in chroot jail are really easy to accomplish as follows. Step 1 First, we have to create all the necessary chrooted environment subdirectories where we will move Snort files and directories. • Use the following command to create all the necessary chroot subdirectories. [root@deep /]# mkdir -p /chroot/snort/etc/snort [root@deep /]# mkdir -p /chroot/snort/var/log/snort [root@deep /]# chown -R snort.snort /chroot/snort/var/log/snort Step 2 Now, it is time to move the required Snort files to the related subdirectories in the chroot area for Snort to work. We can copy these files to the chroot jail but it’s better to move them to avoid unnecessary duplication of Snort files on the server. • Use the following commands to move the require files into the chroot area. [root@deep /]# mv /etc/snort/* /chroot/snort/etc/snort/ [root@deep /]# chmod 0644 /chroot/snort/etc/snort/* Step 3 Once the Snort files have been moved to the chroot location, we can remove the old Snort directories from the system since they are no longer required. • This can be done with the following commands. [root@deep /]# rm -rf /etc/snort/ [root@deep /]# rm -rf /var/log/snort/ Step 4 Next, we have to recreate a new snort initialization script file which starts Snort in the chroot environment. Please note that the following script is suitable for Linux operating systems that use SystemV. If you Linux system use some other method like BSD, you’ll have to adjust the script below to make it work for you. The only difference with the previous Snort initialization script file is that we use the “-t” option of Snort to specify the chroot location. Edit the snort script file (vi /etc/init.d/snort) and add the following lines inside it: #!/bin/bash # This shell script takes care of starting and stopping the snort IDS daemon. # # chkconfig: 2345 40 60 # description: Snort is a lightweight network intrusion detection tool that # currently detects more than 1100 host and network # vulnerabilities, portscans, backdoors, and more. # Source function library. . /etc/init.d/functions # Source networking configuration.
- 509. Snort 2 CHAPTER 2 509 ./etc/sysconfig/network # Check that networking is up. [ ${NETWORKING} = "no" ] && exit 0 # Specify your network interface here INTERFACE=eth0 RETVAL=0 prog="Snort" start() { echo -n $"Starting $prog: " daemon /usr/bin/snort -A fast -u snort -g snort -b -s -z -d -D -i $INTERFACE -c /etc/snort/snort.conf -t /chroot/snort/ RETVAL=$? echo [ $RETVAL = 0 ] && touch /var/lock/subsys/snort return $RETVAL } stop() { echo -n $"Shutting down $prog: " killproc snort RETVAL=$? echo [ $RETVAL = 0 ] && rm -f /var/lock/subsys/snort return $RETVAL } # See how we were called. case "$1" in start) start ;; stop) stop ;; status) status snort ;; restart) stop start ;; condrestart) [ -f /var/lock/subsys/snort ] && restart ;; *) echo $"Usage: $prog {start|stop|status|restart|condrestart}" exit 1 esac exit $RETVAL
- 510. Snort 2 CHAPTER 2 510 Step5 Finally, we must test the new chrooted jail configuration of our Snort program. • Start the new chrooted jail Snort with the following command: [root@deep /]# /etc/init.d/snort start Starting Snort: [OK] • If you don't get any errors, do a ps ax | grep snort and see if we're running: [root@deep /]# ps ax | grep snort 16295 ? R 0:38 /usr/bin/snort -A fast -u snort -g snort -b -s -z -d If so, lets check to make sure it's chrooted by picking its process number and doing ls -la /proc/that_process_number/root/. [root@deep /]# ls -la /proc/16295/root/ If you see something like: total 4 drwxr-xr-x 4 root root 4096 May 7 19:10 ./ drwxr-xr-x 5 root root 4096 May 7 19:12 ../ drwxr-xr-x 3 root root 4096 May 7 19:12 etc/ drwxr-xr-x 3 root root 4096 May 7 19:12 var/ Congratulations! Your Snort in a chroot jail is working. Further documentation For more details, there is one manual page about Snort that you should read: $ man snort (8) - Open source network intrusion detection system.
- 511. Tripwire IN THIS CHAPTER 1.Compiling - Optimizing & Installing Tripwire 2. Configuring Tripwire 3. Running Tripwire for the first time 4. Securing Tripwire 5. Tripwire Administrative Tools
- 513. Tripwire 2 CHAPTER 3 513 LinuxTripwire Abstract With the advent of increasingly sophisticated and subtle account break-ins on Unix systems, the need for tools to aid in the detection of unauthorized modification of files becomes clear. Tripwire is a tool that aids system administrators and users in monitoring a designated set of files for any changes. Used with system files on a regular (e.g., daily) basis, Tripwire can notify system administrators of corrupted or tampered files, so damage control measures can be taken in a timely manner. Tripwire data and network integrity software was originally developed in 1992 at Purdue University by world-renowned computer security expert, Dr. Eugene Spafford, and by master's degree student, Gene Kim. It was quickly embraced by computer security experts and actively used by thousands of corporate, government, and educational organizations worldwide. Tripwire is a file and directory integrity checker, a utility that compares a designated set of files and directories against information stored in a previously generated database. Any differences are flagged and logged, including added or deleted entries. When run against system files on a regular basis, any changes in critical system files will be spotted -- and appropriate damage control measures can be taken immediately. With Tripwire, system administrators can conclude with a high degree of certainty that a given set of files remain free of unauthorized modifications if Tripwire reports no changes. Tripwire is a very valuable security tool for Linux systems, if it is installed to a clean system. Tripwire should be installed right after the OS installation, and before you have connected your system to a network (i.e., before any possibility exists that someone could alter files on your system). These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest Tripwire version number is 2.3.1-2 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages The following is based on information listed by Tripwire as of 2001/03/03. Please check http://sourceforge.net/projects/tripwire/ regularly for the latest status. We chose to install from source because it provides the facility to fine tune the installation. Source code is available from: Tripwire Homepage: http://sourceforge.net/projects/tripwire/ You must be sure to download: tripwire-2.3.1-2.tar.gz
- 514. Tripwire 2 CHAPTER 3 514 Pristinesource If you don’t use the RPM package to install this program, it will be difficult for you to locate all the files installed onto the system if you want to update the package in the future. To solve this problem, it’s a good idea to make a list of files on the system before you install Tripwire, and then one afterwards, and then compare them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > Tripwire1 • And the following one after you install the software: [root@deep root]# find /* > Tripwire2 • Then use the following command to get a list of what changed: [root@deep root]# diff Tripwire1 Tripwire2 > Tripwire-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were added or changed by the program and remove them manually from your system before installing the new software. In our example above, we use the /root directory of the system to store all the generated file lists. Compiling - Optimizing & Installing Tripwire Below are the steps that you must make to configure, compile and optimize the Tripwire software before installing it on your system. First off, we install the program as user 'root' so as to avoid any authorization problems. Step 1 Once you get the program from the main software site you must copy it to the /var/tmp directory and change to this location before expanding the archive. • This can be done with the following commands: [root@deep /]# cp tripwire-version.tar.gz /var/tmp/ [root@deep /]# cd /var/tmp/ [root@deep tmp]# tar xzpf tripwire-version.tar.gz Step 2 There are some source files to modify before going into the configuration and compilation of the program; the changes allow us to fix many bugs with Tripwire. Therefore, move into the newly created Tripwire source directory and perform the following steps to configure and optimize the software for your system. • To move into the newly created Tripwire directory use the following command: [root@deep tmp]# cd tripwire-2.3.1-2/
- 515. Tripwire 2 CHAPTER 3 515 Step3 The first source file to modify is called “mailmessage.cpp”. • Edit the mailmessage.cpp file (vi +244 src/tripwire/mailmessage.cpp) and change: const TCHAR* szFormat = _T("%a %d %b %Y %H:%M:%S %z"); To read: const TCHAR* szFormat = _T("%a, %d %b %Y %H:%M:%S %z"); Step 4 The second file is called “platform.h” and we have to edit it and add a new line as follows. • Edit the platform.h file (vi +294 src/core/platform.h) and change the line: #define USES_FHS IS_LINUX To read: #define USE_FHS IS_LINUX Step 5 The last file to modify is very important for Linux systems with GCC version 3; which should be the default compiler for most Linux system now. The modifications are important to allow Tripwire to compile with GCC v3. There is one problem, the modifications are too big to be listed in a book and we have to retrieve it from the Internet as a patch file and patch our sources code. The patch is available from the OpenNA website at the following URL: http://www.openna.com/products/books/securing-optimizing-linux/3rdedition/index.htm Please, download the patch and patch your Tripwire source codes as follow: • To patch your Tripwire source codes, use the command: [root@deep /]# cp tripwire-gcc3.patch /var/tmp/ [root@deep /]# cd /var/tmp/tripwire-2.3.1-2/ [root@deep tripwire-2.3.1-2]# patch -p1 < ../tripwire-gcc3.patch
- 516. Tripwire 2 CHAPTER 3 516 Step6 Now, we must make a list of all files on the system before installing the software, and one afterwards, then compare them using the diff utility to find out what files are placed where and finally we install the Tripwire software: [root@deep tripwire-2.3.1-2]# cd src/ [root@deep src]# rm -rf STLport* [root@deep src]# touch STLport_r STLport_d [root@deep src]# export CXXFLAGS="-O2 -march=i686 -funroll-loops" [root@deep src]# make release [root@deep src]# cd [root@deep root]# find /* > Tripwire1 [root@deep root]# cd /var/tmp/tripwire-2.3.1-2/bin/i686-pc-linux_r/ [root@deep i686-pc-linux_r]# install -m0500 siggen /usr/sbin/ [root@deep i686-pc-linux_r]# install -m0500 tripwire /usr/sbin/ [root@deep i686-pc-linux_r]# install -m0500 twadmin /usr/sbin/ [root@deep i686-pc-linux_r]# install -m0500 twprint /usr/sbin/ [root@deep i686-pc-linux_r]# cd ../../man/ [root@deep man]# install -m0440 man4/*.4 /usr/share/man/man4/ [root@deep man]# install -m0440 man5/*.5 /usr/share/man/man5/ [root@deep man]# install -m0440 man8/*.8 /usr/share/man/man8/ [root@deep man]# mkdir -m0700 /etc/tripwire [root@deep man]# mkdir -p /var/lib/tripwire/report [root@deep man]# chmod -R 0700 /var/lib/tripwire/ [root@deep man]# cd [root@deep root]# find /* > Tripwire2 [root@deep root]# diff Tripwire1 Tripwire2 > Tripwire-Installed The above commands will configure the software to ensure your system has the necessary libraries to successfully compile the package, compile all source files into executable binaries, and then install the binaries and any supporting files into the appropriate locations. Step 7 Once the configuration, optimization, compilation, and installation of the Tripwire software have been accomplished, we can free up some disk space by deleting the program tar archive and the related source directory since they are no longer needed. • To delete Tripwire and its related source directory, use the following commands: [root@deep /]# cd /var/tmp/ [root@deep tmp]# rm -rf tripwire-version/ [root@deep tmp]# rm -f tripwire-version.tar.gz The rm command as used above will remove all the source files we have used to compile and install Tripwire. It will also remove the Tripwire compressed archive from the /var/tmp directory.
- 517. Tripwire 2 CHAPTER 3 517 ConfiguringTripwire After Tripwire has been built and installed successfully in your system, your next step is to configure and customize its configuration files and policy file to fit your needs. /etc/tripwire/twcfg.txt: (The Tripwire Configuration File) /etc/tripwire/twpol.txt: (The Tripwire Policy File) /etc/tripwire/twinstall.sh: (The Tripwire Cryptographic File) /etc/cron.weekly/tripwire.cron: (The Tripwire Cron File) /etc/tripwire/twcfg.txt: The Tripwire Configuration File The twcfg.txt file is a small Tripwire configuration file used by the program the first time you run it to get information about locations of different files and the way Tripwire runs and reports on the integrity of the system. It stores system-specific information, such as the location of Tripwire data files. In general, we use this file ONE time and REMOVE it from the server once Tripwire is configured. Step 1 By default, the twcfg.txt file do not exist after installation, we have to create it as follow. • Create the twcfg.txt file (touch /etc/tripwire/twcfg.txt) and add the following lines inside the file. Below is what we recommend you. ROOT =/usr/sbin POLFILE =/etc/tripwire/tw.pol DBFILE =/var/lib/tripwire/$(HOSTNAME).twd REPORTFILE =/var/lib/tripwire/report/$(HOSTNAME)-$(DATE).twr SITEKEYFILE =/etc/tripwire/site.key LOCALKEYFILE =/etc/tripwire/$(HOSTNAME)-local.key EDITOR =/bin/vi LATEPROMPTING =true LOOSEDIRECTORYCHECKING =true MAILNOVIOLATIONS =false EMAILREPORTLEVEL =3 REPORTLEVEL =3 MAILMETHOD =SENDMAIL SYSLOGREPORTING =true MAILPROGRAM =/usr/sbin/sendmail -oi -t Step2 Now, set the permissions of the twcfg.txt file to be (0640/-rw-r-----) and owned by the super-user ‘root’ for security reasons. • To change the permission mode and ownership of the twcfg.txt file, use: [root@deep /]# chmod 640 /etc/tripwire/twcfg.txt [root@deep /]# chown 0.0 /etc/tripwire/twcfg.txt
- 518. Tripwire 2 CHAPTER 3 518 /etc/tripwire/twpol.txt:The Tripwire Policy File The twpol.txt file is the Tripwire policy file where you decide and set which system files and directories that you want monitored. It consists of a series of rules specifying the system objects the Tripwire should monitor, and the data for each object that should be collected and stored in the database files. Note that extensive testing and experience are necessary when editing this file before you get working file reports. The following is a working example from where you can start you own customization. We must create, edit or change it to fit our requirements and operating system. Step 1 By default, the twpol.txt file does not exist after installation; we have to create it as follow. The text in bold are the parts of the configuration file that must be customized and adjusted to fit your own system. • Create the twpol.txt file (touch /etc/tripwire/twpol.txt) and add in this file all the files and directories that you want monitored. The format of the configuration file is described in its header and in the manual page twpolicy (4). Below is what we recommend you enter: # This is the example Tripwire Policy file. It is intended as a place to # start creating your own custom Tripwire Policy file. Referring to it as # well as the Tripwire Policy Guide should give you enough information to # make a good custom Tripwire Policy file that better covers your # configuration and security needs. # # Because it is impossible to be one policy file for all machines, your # Linux configuration will most likey differ from the one our policy file # was tuned to, and will therefore require some editing of the default # Tripwire Policy file. # Global Variable Definitions # These are defined at install time by the installation script. You may # Manually edit these if you are using this file directly and not from the # installation script itself. @@section GLOBAL TWROOT=/usr/sbin; TWBIN=/usr/sbin; TWPOL="/etc/tripwire"; TWDB="/var/lib/tripwire"; TWSKEY="/etc/tripwire"; TWLKEY="/etc/tripwire"; TWREPORT="/var/lib/tripwire/report"; # NOTE: Change the following parameter to reflect your own host name. # For example, if your host is called 'www', then change 'localhost' to 'www'. HOSTNAME=localhost; @@section FS SEC_CRIT = $(IgnoreNone)-SHa ; SEC_SUID = $(IgnoreNone)-SHa ; SEC_BIN = $(ReadOnly) ; SEC_CONFIG = $(Dynamic) ; SEC_LOG = $(Growing) ; SEC_INVARIANT = +tpug ; SIG_LOW = 33 ; SIG_MED = 66 ; SIG_HI = 100 ;
- 519. Tripwire 2 CHAPTER 3 519 #SEC_CRIT are critical files that cannot change. # SEC_SUID are binaries with the SUID or SGID flags set. # SEC_BIN are binaries that should not change. # SEC_CONFIG are config files that are changed infrequently but accessed often. # SEC_LOG are files that grow, but that should never change ownership. # SEC_INVARIANT are directories that should never change permission or owners. # SIG_LOW are non-critical files that are of minimal security impact. # SIG_MED are non-critical files that are of significant security impact. # SIG_HI are critical files that are significant points of vulnerability. ( rulename = "Tripwire binaries", severity = $(SIG_HI) ) { $(TWBIN)/siggen -> $(SEC_BIN) ; $(TWBIN)/tripwire -> $(SEC_BIN) ; $(TWBIN)/twadmin -> $(SEC_BIN) ; $(TWBIN)/twprint -> $(SEC_BIN) ; } ( rulename = "Tripwire data files", severity = $(SIG_HI) ) { $(TWDB) -> $(SEC_CONFIG) -i ; $(TWPOL)/tw.pol -> $(SEC_BIN) -i ; $(TWPOL)/tw.cfg -> $(SEC_BIN) -i ; $(TWLKEY)/$(HOSTNAME)-local.key -> $(SEC_BIN) ; $(TWSKEY)/site.key -> $(SEC_BIN) ; $(TWREPORT) -> $(SEC_CONFIG) (recurse=0) ; } ( rulename = "Invariant directories", severity = $(SIG_MED) ) { / -> $(SEC_INVARIANT) (recurse = 0) ; /home -> $(SEC_INVARIANT) (recurse = 0) ; } ( rulename = "/root directory", severity = $(SIG_HI) ) { /root -> $(SEC_CRIT) (recurse = -1) ; /root/.bashrc -> $(SEC_CONFIG) (recurse = 0) ; /root/.bash_profile -> $(SEC_CONFIG) (recurse = 0) ; /root/.bash_logout -> $(SEC_CONFIG) (recurse = 0) ; /root/.bash_history -> $(SEC_CONFIG) (recurse = 0) ; } ( rulename = "/boot directory", severity = $(SIG_HI)
- 520. Tripwire 2 CHAPTER 3 520 ) { /boot-> $(SEC_CRIT) (recurse = -1) ; !/boot/System.map ; } ( rulename = "/etc directory", severity = $(SIG_HI) ) { /etc -> $(SEC_CRIT) (recurse = -1) ; } ( rulename = "/dev & /proc directories", severity = $(SIG_HI), ) { /dev -> $(Device) (recurse = -1) ; /proc/bus -> $(Device) (recurse = 0) ; /proc/cmdline -> $(Device) (recurse = 0) ; /proc/cpuinfo -> $(Device) (recurse = 0) ; /proc/devices -> $(Device) (recurse = 0) ; /proc/dma -> $(Device) (recurse = 0) ; /proc/driver -> $(Device) (recurse = 0) ; /proc/execdomains -> $(Device) (recurse = 0) ; /proc/filesystems -> $(Device) (recurse = 0) ; /proc/fs -> $(Device) (recurse = 0) ; /proc/ide -> $(Device) (recurse = 0) ; /proc/interrupts -> $(Device) (recurse = 0) ; /proc/iomem -> $(Device) (recurse = 0) ; /proc/ioports -> $(Device) (recurse = 0) ; /proc/irq -> $(Device) (recurse = 0) ; /proc/kcore -> $(Device) (recurse = 0) ; /proc/kmsg -> $(Device) (recurse = 0) ; /proc/ksyms -> $(Device) (recurse = 0) ; /proc/loadavg -> $(Device) (recurse = 0) ; /proc/locks -> $(Device) (recurse = 0) ; /proc/meminfo -> $(Device) (recurse = 0) ; /proc/misc -> $(Device) (recurse = 0) ; /proc/mounts -> $(Device) (recurse = 0) ; /proc/partitions -> $(Device) (recurse = 0) ; /proc/pci -> $(Device) (recurse = 0) ; /proc/self -> $(Device) (recurse = 0) ; /proc/slabinfo -> $(Device) (recurse = 0) ; /proc/stat -> $(Device) (recurse = 0) ; /proc/sys -> $(Device) (recurse = 0) ; /proc/sysvipc -> $(Device) (recurse = 0) ; /proc/tty -> $(Device) (recurse = 0) ; /proc/uptime -> $(Device) (recurse = 0) ; /proc/version -> $(Device) (recurse = 0) ; !/dev/pts ; !/dev/shm ; } ( rulename = "/bin & /sbin directories", severity = $(SIG_HI) )
- 521. Tripwire 2 CHAPTER 3 521 { /bin-> $(SEC_CRIT) (recurse = -1) ; /sbin -> $(SEC_CRIT) (recurse = -1) ; } ( rulename = "/lib directory", severity = $(SIG_HI) ) { /lib -> $(SEC_CRIT) (recurse = -1) ; } ( rulename = "/tmp directories", severity = $(SIG_LOW) ) { /usr/tmp -> $(SEC_INVARIANT) (recurse = 0) ; /var/tmp -> $(SEC_INVARIANT) (recurse = 0) ; /tmp -> $(SEC_INVARIANT) (recurse = 0) ; } ( rulename = "/urs directories", severity = $(SIG_HI) ) { /usr -> $(SEC_CRIT) (recurse = -1) ; } ( rulename = "/var directories", severity = $(SIG_HI) ) { /var -> $(SEC_CONFIG) (recurse = -1) ; /var/lib -> $(SEC_CONFIG) (recurse = -1) ; /var/spool -> $(SEC_CONFIG) (recurse = -1) ; !/var/spool/mail ; !/var/spool/mqueue ; } ( rulename = "SUID SGID binaries", severity = $(SIG_HI) ) { /usr/bin/man -> $(SEC_SUID) (recurse = 0) ; /usr/bin/slocate -> $(SEC_SUID) (recurse = 0) ; /usr/bin/passwd -> $(SEC_SUID) (recurse = 0) ; /usr/bin/crontab -> $(SEC_SUID) (recurse = 0) ; /usr/bin/sudo -> $(SEC_SUID) (recurse = 0) ; /usr/sbin/utempter -> $(SEC_SUID) (recurse = 0) ; /usr/sbin/exim -> $(SEC_SUID) (recurse = 0) ; /bin/su -> $(SEC_SUID) (recurse = 0) ; }
- 522. Tripwire 2 CHAPTER 3 522 ( rulename= "/chroot directory", severity = $(SIG_HI) ) { /chroot -> $(SEC_CRIT) (recurse = -1) ; } Step2 Now, set the permissions of the twpol.txt file to be (0640/-rw-r-----) and owned by the super-user ‘root’ for security reasons. • To change the permission mode and ownership of the twpol.txt file, use: [root@deep /]# chmod 640 /etc/tripwire/twpol.txt [root@deep /]# chown 0.0 /etc/tripwire/twpol.txt NOTE: Please, add to the above policy file all files, binaries, directories that you want the software to monitor for you. Remove any files, binaries, directories that you don’t want the software to monitor for you and don’t send emails to the mailing list if you receive error messages about the fact that some files, binaries, directories don’t exist on your system. Instead, review your policy file and make the changes related to the error that you received, because in most cases, this is why you have this kind of errors messages. Finally, reads the twpolicy manual page for more information on the parameters of this policy file. /etc/tripwire/twinstall.sh: The Tripwire Cryptographic File The twinstall.sh file is a script file used by Tripwire to configure, install and generate cryptographic keys used by Tripwire during operation and verification. Step 1 This script file asks you the passpharse that you want to run with Tripwire as well as other operations related to the Tripwire database generation and policies. By default, the twinstall.sh file does not exist after the installation; we have to create it as follows. • Create the twinstall.sh file (touch /etc/tripwire/twinstall.sh) and add the following lines inside the file. #!/bin/sh HOST_NAME='localhost' if uname -n > /dev/null 2> /dev/null ; then HOST_NAME=`uname -n` fi # Site Passphrase variable TW_SITE_PASS="" # Complete path to site key SITE_KEY="/etc/tripwire/site.key" # Local Passphrase variable TW_LOCAL_PASS="" # Complete path to local key
- 523. Tripwire 2 CHAPTER 3 523 LOCAL_KEY="/etc/tripwire/${HOST_NAME}-local.key" #If clobber==true, overwrite files; if false, do not overwrite files. CLOBBER="false" # If prompt==true, ask for confirmation before continuing with install. PROMPT="true" # Name of twadmin executeable TWADMIN="twadmin" # Path to twadmin executeable TWADMPATH=@sbindir@ # Path to configuration directory CONF_PATH="/etc/tripwire" # Name of clear text policy file TXT_POL=$CONF_PATH/twpol.txt # Name of clear text configuration file TXT_CFG=$CONF_PATH/twcfg.txt # Name of encrypted configuration file CONFIG_FILE=$CONF_PATH/tw.cfg # Path of the final Tripwire policy file (signed) SIGNED_POL=`grep POLFILE $TXT_CFG | sed -e 's/^.*=(.*)/1/'` if [ -z "$TW_SITE_PASS" ] || [ -z "$TW_LOCAL_PASS" ]; then cat << END_OF_TEXT ---------------------------------------------- The Tripwire site and local passphrases are used to sign a variety of files, such as the configuration, policy, and database files. Passphrases should be at least 8 characters in length and contain both letters and numbers. See the Tripwire manual for more information. END_OF_TEXT fi echo echo "----------------------------------------------" echo "Creating key files..." if [ "$CLOBBER" = "true" ] && [ "$PROMPT" = "false" ] && [ -f "$SITE_KEY" ] ; then rm -f "$SITE_KEY" fi if [ -f "$SITE_KEY" ] && [ "$CLOBBER" = "false" ] ; then echo "The site key file "$SITE_KEY"" echo 'exists and will not be overwritten.' else cmdargs="--generate-keys --site-keyfile "$SITE_KEY"" if [ -n "$TW_SITE_PASS" ] ; then cmdargs="$cmdargs --site-passphrase "$TW_SITE_PASS"" fi eval ""$TWADMPATH/$TWADMIN" $cmdargs" if [ $? -ne 0 ] ; then
- 524. Tripwire 2 CHAPTER 3 524 echo"Error: site key generation failed" exit 1 else chmod 640 "$SITE_KEY" fi fi if [ "$CLOBBER" = "true" ] && [ "$PROMPT" = "false" ] && [ -f "$LOCAL_KEY" ] ; then rm -f "$LOCAL_KEY" fi if [ -f "$LOCAL_KEY" ] && [ "$CLOBBER" = "false" ] ; then echo "The site key file "$LOCAL_KEY"" echo 'exists and will not be overwritten.' else cmdargs="--generate-keys --local-keyfile "$LOCAL_KEY"" if [ -n "$TW_LOCAL_PASS" ] ; then cmdargs="$cmdargs --local-passphrase "$TW_LOCAL_PASS"" fi eval ""$TWADMPATH/$TWADMIN" $cmdargs" if [ $? -ne 0 ] ; then echo "Error: local key generation failed" exit 1 else chmod 640 "$LOCAL_KEY" fi fi echo echo "----------------------------------------------" echo "Signing configuration file..." if [ "$CLOBBER" = "false" ] && [ -s "$CONFIG_FILE" ] ; then backup="${CONFIG_FILE}.$$.bak" echo "Backing up $CONFIG_FILE" echo " to $backup" `mv "$CONFIG_FILE" "$backup"` if [ $? -ne 0 ] ; then echo "Error: backup of configuration file failed." exit 1 fi fi cmdargs="--create-cfgfile" cmdargs="$cmdargs --cfgfile "$CONFIG_FILE"" cmdargs="$cmdargs --site-keyfile "$SITE_KEY"" if [ -n "$TW_SITE_PASS" ] ; then cmdargs="$cmdargs --site-passphrase "$TW_SITE_PASS"" fi eval ""$TWADMPATH/$TWADMIN" $cmdargs "$TXT_CFG"" if [ $? -ne 0 ] ; then echo "Error: signing of configuration file failed." exit 1 fi # Set the rights properly chmod 640 "$CONFIG_FILE" cat << END_OF_TEXT A clear-text version of the Tripwire configuration file $TXT_CFG has been preserved for your inspection. It is recommended
- 525. Tripwire 2 CHAPTER 3 525 thatyou delete this file manually after you have examined it. END_OF_TEXT echo echo "----------------------------------------------" echo "Signing policy file..." if [ "$CLOBBER" = "false" ] && [ -s "$POLICY_FILE" ] ; then backup="${POLICY_FILE}.$$.bak" echo "Backing up $POLICY_FILE" echo " to $backup" mv "$POLICY_FILE" "$backup" if [ $? -ne 0 ] ; then echo "Error: backup of policy file failed." exit 1 fi fi cmdargs="--create-polfile" cmdargs="$cmdargs --cfgfile "$CONFIG_FILE"" cmdargs="$cmdargs --site-keyfile "$SITE_KEY"" if [ -n "$TW_SITE_PASS" ] ; then cmdargs="$cmdargs --site-passphrase "$TW_SITE_PASS"" fi eval ""$TWADMPATH/$TWADMIN" $cmdargs "$TXT_POL"" if [ $? -ne 0 ] ; then echo "Error: signing of policy file failed." exit 1 fi # Set the proper rights on the newly signed policy file. chmod 0640 "$SIGNED_POL" cat << END_OF_TEXT A clear-text version of the Tripwire policy file $TXT_POL has been preserved for your inspection. This implements a minimal policy, intended only to test essential Tripwire functionality. You should edit the policy file to describe your system, and then use twadmin to generate a new signed copy of the Tripwire policy. END_OF_TEXT Step 2 Now, set the permissions of the twinstall.sh file to be (0500/---x------) and owned by the super-user ‘root’ for security reasons. • This procedure can be accomplished with the following command: [root@deep /]# chmod 500 /etc/tripwire/twinstall.sh [root@deep /]# chown 0.0 /etc/tripwire/twinstall.sh NOTE: The above script file can also be retrieved from the following URL: http://www.openna.com/products/books/securing-optimizing-linux/3rdedition/index.htm
- 526. Tripwire 2 CHAPTER 3 526 /etc/cron.weekly/tripwire.cron:The Tripwire Cron File The tripwire.cron file is a small script executed automatically by the crond program each week to scan your hard disk for possible changes to files or directories and mail the results to the system administrator. Step 1 This script will automate the procedure of integrity checking for you. If you want to automate this task, follow the simple steps below. • Create the tripwire.cron script file (touch /etc/cron.weekly/tripwire.cron) and add the following lines: #!/bin/sh HOST_NAME=`uname -n` if [ ! -e /var/lib/tripwire/${HOST_NAME}.twd ] ; then echo "**** Error: Tripwire database for ${HOST_NAME} not found. ****" echo "**** Run "/etc/tripwire/twinstall.sh" and/or "tripwire --init". ****" else test -f /etc/tripwire/tw.cfg && /usr/sbin/tripwire --check fi Step 2 Now, set the permissions of the tripwire.cron file to be (0500/---x------) and owned by the super-user ‘root’ for security reasons. • This procedure can be accomplished with the following command: [root@deep /]# chmod 500 /etc/cron.weekly/tripwire.cron [root@deep /]# chown 0.0 /etc/cron.weekly/tripwire.cron Running Tripwire for the first time Once all the files are installed and configured on your server, it’s time to run Tripwire to generate the cryptography key, passphrases and database files. These procedures should be made the first time you install the software and ONLY the first time. Step 1 Here we begin by running the twinstall.sh script file which will generate the cryptography keys and will ask us to enter our passphrase (password) which is required each time we want to update and accept Tripwire integrity reports. • To run the twinstall.sh file use the following command: [root@deep /]# /etc/tripwire/twinstall.sh ---------------------------------------------- The Tripwire site and local passphrases are used to sign a variety of files, such as the configuration, policy, and database files. Passphrases should be at least 8 characters in length and contain both letters and numbers. See the Tripwire manual for more information. ----------------------------------------------
- 527. Tripwire 2 CHAPTER 3 527 Creatingkey files... (When selecting a passphrase, keep in mind that good passphrases typically have upper and lower case letters, digits and punctuation marks, and are at least 8 characters in length.) Enter the site keyfile passphrase: Your site keyfile passphrase Verify the site keyfile passphrase: Your site keyfile passphrse again Generating key (this may take several minutes)...Key generation complete. (When selecting a passphrase, keep in mind that good passphrases typically have upper and lower case letters, digits and punctuation marks, and are at least 8 characters in length.) Enter the local keyfile passphrase: Your local keyfile passphrase Verify the local keyfile passphrase: Your local keyfile passphrase again Generating key (this may take several minutes)...Key generation complete. ---------------------------------------------- Signing configuration file... Please enter your site passphrase: Your site passphrase Wrote configuration file: /etc/tripwire/tw.cfg A clear-text version of the Tripwire configuration file /etc/tripwire/twcfg.txt has been preserved for your inspection. It is recommended that you delete this file manually after you have examined it. ---------------------------------------------- Signing policy file... Please enter your site passphrase: Your site passphrase Wrote policy file: /etc/tripwire/tw.pol A clear-text version of the Tripwire policy file /etc/tripwire/twpol.txt has been preserved for your inspection. This implements a minimal policy, intended only to test essential Tripwire functionality. You should edit the policy file to describe your system, and then use twadmin to generate a new signed copy of the Tripwire policy Step 2 Once our passphrase keyfiles have been generated, it’s time to run Tripwire in its’ initialization mode. The initialization mode will create the initial Tripwire database files based on what information has been provided inside the twpol.txt file. Tripwire must have a database to compare against, so we first create the file information database. This action will create a file called “tw.db_[hostname]” in the directory you specified to hold your databases (where [hostname] will be replaced with your machine hostname). • To run the Tripwire in initialization mode, use the following command: [root@deep /]# tripwire --init Parsing policy file: /etc/tripwire/tw.pol Generating the database... *** Processing Unix File System *** Please enter your local passphrase: Wrote database file: /var/lib/tripwire/deep.twd The database was successfully generated.
- 528. Tripwire 2 CHAPTER 3 528 NOTE:Initialization of the database Tripwire uses should be done manually because the key used to sign the database should be different for each system. Step 3 Finally, if you have not received any kind of error message, then you can safety remove the twcfg.txt and twpol.txt files from your system since they are no longer needed and it would be a security risk to keep these files on your server. • To remove the files from your system, use the following commands: [root@deep /]# rm -f /etc/tripwire/twcfg.txt [root@deep /]# rm -f /etc/tripwire/twpol.txt NOTE: You have to remove the files from your server ONLY if you are sure that the initialization of the databases has been completed without any errors. Otherwise you should keep these files and regenerate a new database once all the errors have been fixed inside the twpol.txt file, since in many cases errors come from twpol.txt file having some lines referring to files or directories that do not exist in your system. Securing Tripwire It is highly recommended that the database (tw.db_[hostname]) file of Tripwire be moved someplace (e.g. floppy) where it cannot be modified. This is important because data from Tripwire is only as trustworthy as its database. It is also recommended that you make a hardcopy printout of the database contents right away. In the event that you become suspicious of the integrity of the database, you will be able to manually compare information against this hardcopy. Tripwire Administrative Tools The commands listed below are some of the most used of this software, but many more exist. Check the Tripwire manual pages for more details. Running Tripwire in Interactive Checking Mode: In “Interactive Checking Mode” feature, Tripwire verifies files or directories that have been added, deleted, or changed from the original database and ask the user whether the database entry should be updated. This mode is the most convenient way of keeping your database up-to- date, but it requires that the user be "at the console". If you want to use this mode, then follow the simple step below. Once the file information database of Tripwire has been created, we can now run Tripwire in “Interactive Checking Mode”. This mode will prompt the user for whether or not each changed entry on the system should be updated to reflect the current state of the file. • To run Tripwire in Interactive Checking Mode, use the following command: [root@deep /]# tripwire --check --interactive Parsing policy file: /etc/tripwire/tw.pol *** Processing Unix File System *** Performing integrity check...
- 529. Tripwire 2 CHAPTER 3 529 NOTE:In interactive mode, Tripwire first reports all added, deleted, or changed files, and then allows the user to update the entry in the database. Further documentation For more details, there are several manual pages about Tripwire that you can read: $ man siggen (8) - Signature gathering routine for Tripwire. $ man tripwire (8) - A file integrity checker for UNIX systems. $ man twadmin (8) - Tripwire administrative and utility tool. $ man twintro (8) - Introduction to Tripwire software. $ man twprint (8) - Tripwire database and report printer. $ man twconfig (4) - Tripwire configuration file reference. $ man twpolicy (4) - Tripwire policy file reference. $ man twfiles (5) - Overview of files used by Tripwire and file backup process. Some possible uses of Tripwire software Tripwire can be used to: 1. Check the integrity of your files system. 2. Get a list of new installed or removed files on your system.
- 533. ucspi-tcp IN THIS CHAPTER 1.Compiling - Optimizing & Installing ucsip-tcp 2. Using ucsip-tcp
- 535. ucspi-tcp 2 CHAPTER 4 535 Linuxucspi-tcp Abstract UCSPI stand for (UNIX Client-Server Program Interface) and it's a command-line interface to client-server communications tools that provides several small programs like tcpserver or tcpclient, which are easy-to-use command-line tools for building TCP client-server applications. Some may ask why we would need to run this kind of program on our server. Well, in the UNIX world, there is some software that cannot run as a daemon and need the help of other software like ucspi-tcp to work. This is where ucspi-tcp is required. This small piece of software from D. J. Bernstein provides two important binary programs to achieve this. The first is called “tcpserver”, which waits for incoming connections and, for each connection, runs a program of your choice, the second is called “tcpclient”, which makes a TCP connection and runs a program of your choice. Other tools exist in this ucspi-tcp package but the most frequently used are tcpserver and tcpclient. In general, we use these programs to replace software like inetd or Xinetd, which perform the same functions as tcpserver and tcpclient. The main difference is that ucsip-tcp is really the most secure and faster software in this group. Personally, and each time you need to run third party software like IMAP, POP3, Qmail, vsFTPd, etc that depends on a super-server to work, I highly recommend you use ucspi-tcp instead of inet or Xinetd. That’s said; let’s go to the most interesting part now. These installation instructions assume Commands are Unix-compatible. The source path is /var/tmp (note that other paths are possible, as personal discretion). Installations were tested on OpenNA Linux & Red Hat Linux 7.3. All steps in the installation will happen using the super-user account “root”. Whether kernel recompilation may be required: No Latest ucsip-tcp version number is 0.88 The procedures given in this chapter are likely to work on all Linux platforms, but we have only tested it on OpenNA Linux and Red Hat Linux. Packages The following is based on information listed by ucsip-tcp as of 2002/04/19. Please regularly check at http://cr.yp.to/ucspi-tcp/install.html for the latest status. We chose to install the required components from source because it provides the facility to fine tune the installation. Source code is available from: ucsip-tcp Homepage: http://cr.yp.to/ucspi-tcp.html You must be sure to download: ucspi-tcp-0.88.tar.gz
- 536. ucspi-tcp 2 CHAPTER 4 536 Pristinesource If you don’t use the RPM package to install this program, it will be difficult for you to locate all the installed files in the system in the event of an update in the future. To solve the problem, it is a good idea to make a list of files on the system before you install ucsip-tcp, and one afterwards, and then compares them using the diff utility to find out what files were placed where. • Simply run the following command before installing the software: [root@deep root]# find /* > ucsip-tcp1 • And the following one after you install the software: [root@deep root]# find /* > ucsip-tcp2 • Then use the following command to get a list of what changed: [root@deep root]# diff ucsip-tcp1 ucsip-tcp2 > ucsip-tcp-Installed With this procedure, if any upgrade appears, all you have to do is to read the generated list of what files were