Symantec NetBackup™ for
Microsoft SQL Server
Administrator's Guide
for Windows
Release 7.5
Symantec NetBackup™ NetBackup for Microsoft SQL
Server Administrator's Guide
The software described in this book is furnished under a license agreement and may be used
only in accordance with the terms of the agreement.
Documentation version: 7.5
Legal Notice
Copyright © 2012 Symantec Corporation. All rights reserved.
Symantec and the Symantec Logo and NetBackup are trademarks or registered trademarks
of Symantec Corporation or its affiliates in the U.S. and other countries. Other names may
be trademarks of their respective owners.
This Symantec product may contain third party software for which Symantec is required
to provide attribution to the third party (“Third Party Programs”). Some of the Third Party
Programs are available under open source or free software licenses. The License Agreement
accompanying the Software does not alter any rights or obligations you may have under
those open source or free software licenses. Please see the Third Party Legal Notice Appendix
to this Documentation or TPIP ReadMe File accompanying this Symantec product for more
information on the Third Party Programs.
The product described in this document is distributed under licenses restricting its use,
copying, distribution, and decompilation/reverse engineering. No part of this document
may be reproduced in any form by any means without prior written authorization of
Symantec Corporation and its licensors, if any.
THE DOCUMENTATION IS PROVIDED "AS IS" AND ALL EXPRESS OR IMPLIED CONDITIONS,
REPRESENTATIONS AND WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT,
ARE DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD TO
BE LEGALLY INVALID. SYMANTEC CORPORATION SHALL NOT BE LIABLE FOR INCIDENTAL
OR CONSEQUENTIAL DAMAGES IN CONNECTION WITH THE FURNISHING,
PERFORMANCE, OR USE OF THIS DOCUMENTATION. THE INFORMATION CONTAINED
IN THIS DOCUMENTATION IS SUBJECT TO CHANGE WITHOUT NOTICE.
The Licensed Software and Documentation are deemed to be commercial computer software
as defined in FAR 12.212 and subject to restricted rights as defined in FAR Section 52.227-19
"Commercial Computer Software - Restricted Rights" and DFARS 227.7202, "Rights in
Commercial Computer Software or Commercial Computer Software Documentation", as
applicable, and any successor regulations. Any use, modification, reproduction release,
performance, display or disclosure of the Licensed Software and Documentation by the U.S.
Government shall be solely in accordance with the terms of this Agreement.
Symantec Corporation
350 Ellis Street
Mountain View, CA 94043
http://www.symantec.com
Technical Support
Symantec Technical Support maintains support centers globally. Technical
Support’s primary role is to respond to specific queries about product features
and functionality. The Technical Support group also creates content for our online
Knowledge Base. The Technical Support group works collaboratively with the
other functional areas within Symantec to answer your questions in a timely
fashion. For example, the Technical Support group works with Product Engineering
and Symantec Security Response to provide alerting services and virus definition
updates.
Symantec’s support offerings include the following:
■ A range of support options that give you the flexibility to select the right
amount of service for any size organization
■ Telephone and/or Web-based support that provides rapid response and
up-to-the-minute information
■ Upgrade assurance that delivers software upgrades
■ Global support purchased on a regional business hours or 24 hours a day, 7
days a week basis
■ Premium service offerings that include Account Management Services
For information about Symantec’s support offerings, you can visit our Web site
at the following URL:
www.symantec.com/business/support/
All support services will be delivered in accordance with your support agreement
and the then-current enterprise technical support policy.
Contacting Technical Support
Customers with a current support agreement may access Technical Support
information at the following URL:
www.symantec.com/business/support/
Before contacting Technical Support, make sure you have satisfied the system
requirements that are listed in your product documentation. Also, you should be
at the computer on which the problem occurred, in case it is necessary to replicate
the problem.
When you contact Technical Support, please have the following information
available:
■ Product release level
■ Hardware information
■ Available memory, disk space, and NIC information
■ Operating system
■ Version and patch level
■ Network topology
■ Router, gateway, and IP address information
■ Problem description:
■ Error messages and log files
■ Troubleshooting that was performed before contacting Symantec
■ Recent software configuration changes and network changes
Licensing and registration
If your Symantec product requires registration or a license key, access our technical
support Web page at the following URL:
www.symantec.com/business/support/
Customer service
Customer service information is available at the following URL:
www.symantec.com/business/support/
Customer Service is available to assist with non-technical questions, such as the
following types of issues:
■ Questions regarding product licensing or serialization
■ Product registration updates, such as address or name changes
■ General product information (features, language availability, local dealers)
■ Latest information about product updates and upgrades
■ Information about upgrade assurance and support contracts
■ Information about the Symantec Buying Programs
■ Advice about Symantec's technical support options
■ Nontechnical presales questions
■ Issues that are related to CD-ROMs, DVDs, or manuals
Support agreement resources
If you want to contact Symantec regarding an existing support agreement, please
contact the support agreement administration team for your region as follows:
Asia-Pacific and Japan customercare_apac@symantec.com
Europe, Middle-East, and Africa semea@symantec.com
North America and Latin America supportsolutions@symantec.com
Contents
Technical Support ............................................................................................... 4
Chapter 1 Introduction to NetBackup for SQL Server .................... 15
Overview of NetBackup for SQL Server ............................................. 15
About NetBackup operations .......................................................... 16
About the NetBackup for SQL Server graphical user interface .............. 17
NetBackup for SQL Server terminology ............................................ 18
Technical overview ....................................................................... 19
What are the components of NetBackup for SQL Server? ..................... 20
How does NetBackup for SQL Server back up a database? .................... 22
How does NetBackup for SQL Server recover a database? ..................... 22
What factors affect the data transfer rate during a SQL Server backup
or restore operation? .............................................................. 23
How does NetBackup resolve SQL Server host and instance
names? ................................................................................. 24
About SQL Server backups and restores in a VMware
environment ......................................................................... 26
NetBackup documentation ............................................................. 26
Chapter 2 Installing NetBackup for SQL Server ............................... 27
Planning the installation of NetBackup for SQL Server ....................... 27
Verifying the operating system and platform compatibility for
NetBackup for SQL Server ....................................................... 28
NetBackup server and client requirements for NetBackup for SQL
Server .................................................................................. 29
SQL Server server software requirements for NetBackup for SQL
Server .................................................................................. 29
Requirements for installing NetBackup for SQL Server in a NetBackup
cluster ................................................................................. 30
About support for Microsoft Cluster server (MSCS) clusters with
NetBackup for SQL ................................................................. 30
About support for Veritas Cluster Server (VCS) clusters with
NetBackup for SQL ................................................................. 31
Configuration and licensing requirements for NetBackup for SQL
Server with Snapshot Client ..................................................... 31
8 Contents
Adding license keys for NetBackup for SQL Server backups .................. 31
Chapter 3 Configuring NetBackup for SQL Server .......................... 33
About user interface terminology .................................................... 33
Configuring the Maximum jobs per client ......................................... 34
About configuring a backup policy for an SQL Server database ............. 35
Adding a new NetBackup for SQL Server policy ............................ 35
About policy attributes for NetBackup for SQL Server ................... 36
About adding schedules to a NetBackup for SQL Server
policy ............................................................................. 37
About NetBackup for SQL Server schedule properties ................... 37
About the types of NetBackup for SQL Server backups .................. 38
Adding clients to a NetBackup for SQL Server policy ..................... 39
Adding backup selections to a NetBackup for SQL Server
policy ............................................................................. 40
Adding batch files to the backup selections list ............................ 40
Testing configuration settings for NetBackup for SQL Server ............... 42
Configuring multistream operations ................................................ 42
Configuring multistriped backups ............................................. 43
Configuring multiplexed backups .............................................. 43
About configuration for a multi-interface network connection
(multi-NIC) ........................................................................... 44
Configuring the master server for a multi-NIC ............................. 44
Configuring a policy for a multi-NIC .......................................... 44
Adding the permissions that allow for browsing of backups across
the private interface ......................................................... 45
Configuring the client for a multi-NIC ........................................ 45
About SQL Server privileges ........................................................... 46
About setting the SQL Server logon for scheduled operations ............... 47
Authorizing scheduled operations (for sites with SQL Server security
restrictions) .......................................................................... 48
Mapping a Windows account to a standard SQL Server user ID ............. 49
About NetBackup for SQL performance factors .................................. 49
NetBackup for SQL buffer space parameters ................................ 49
About backup stripes .............................................................. 51
About shared memory usage .................................................... 51
About alternate buffer method ................................................. 51
About page verification ........................................................... 52
About instant data file initialization .......................................... 52
Using read-write and read-only filegroups .................................. 53
Backing up SQL Server in an environment with log shipping ................ 53
Contents 9
Chapter 4 Backup and recovery concepts ......................................... 55
Overview of backup and recovery concepts ....................................... 55
Overview of backup and recovery concepts ....................................... 55
About SQL Server system database types .................................... 56
About database backups .......................................................... 56
About filegroup backups .......................................................... 57
About differential backups ....................................................... 58
Protecting files and filegroups ........................................................ 58
About recovery considerations for files and filegroups ........................ 59
Reducing backup size and time by using read-only filegroups (SQL
Server 2005 or later) ............................................................... 60
About recovery factors for SQL Server ............................................. 61
About SQL Server transaction logs ............................................ 61
About recovery strategies ........................................................ 62
Backing up the transaction log .................................................. 62
About differential backups ....................................................... 63
About file and filegroup backups ............................................... 63
About database recovery ......................................................... 64
About staging recovery ........................................................... 64
Chapter 5 Performing backups and restores of SQL
Server ............................................................................... 67
Overview of using NetBackup for SQL Server .................................... 68
Using dbbackex to perform user-directed operations for SQL
Server .................................................................................. 69
Using client-based schedulers with dbbackex .............................. 70
Selecting the SQL host and instance ................................................ 71
Using the NetBackup for SQL Client interface .................................... 71
Starting the NetBackup SQL Client for the first time ........................... 71
Backup Microsoft SQL Server Objects field descriptions ...................... 72
Backing up databases .................................................................... 75
Backing up transaction logs ........................................................... 76
Backing up database filegroups ....................................................... 77
Backing up read-only filegroups (SQL Server 2005 or later) .................. 77
Viewing read-only backup sets ................................................. 77
Backing up read-write filegroups (SQL Server 2005 or later) ................. 78
Backing up database files ............................................................... 79
Performing partial backups (SQL Server 2005 or later) ........................ 79
Browsing for backup images ........................................................... 81
Restore Microsoft SQL Server Objects field descriptions ...................... 82
Restoring a database backup .......................................................... 86
Staging a full database recovery ...................................................... 87
10 Contents
Restoring filegroup backups ........................................................... 88
Recovering a database from read-write backups ................................. 89
Restoring read-only filegroups (SQL Server 2005 or later) .................... 90
Restoring database files ................................................................ 90
Restoring a transaction log image without staging a full
recovery ............................................................................... 91
Performing a database move .......................................................... 91
Restoring partial databases (SQL Server 2000) ................................... 94
Performing page-level restores (SQL Server 2005 or later) ................... 96
About performing page-level restores (SQL Server 2005 or
later) ............................................................................. 96
About page-level restore requirements and limitations ................. 96
Performing SQL Server page-level restores ................................. 97
Redirecting a restore to a different host ........................................... 98
Redirecting a database to a different location on a different host .......... 99
About selecting a server ......................................................... 100
Performing restores with a multi-NIC connection ............................. 101
Backing up and restoring the databases that contain full-text search
catalogs (SQL Server 2005) ..................................................... 101
About full-text catalog directory structure ................................ 101
Backing up and restoring the databases that contains full-text
catalogs ........................................................................ 102
Moving a database that contains full-text catalogs ...................... 103
About using batch files ................................................................ 103
About text format used in batch files ........................................ 104
Guidelines for creating and using batch files ............................. 104
Notes when creating and using batch files ................................. 104
About keywords and values used in batch files ........................... 105
Creating a batch file .............................................................. 116
Running batch files ............................................................... 117
About file checkpointing .............................................................. 117
About automatic retry of unsuccessful backups ................................ 118
Performing backups and restores of remote SQL Server
installations ........................................................................ 120
Restoring multistreamed backups ................................................. 120
About conventional backups using multiple streams ................... 121
About advanced backup methods using multiple streams ............. 121
Restoring a multiplexed SQL Server backup .............................. 121
Restoring a multistreamed SQL Server backup with fewer devices
than it was backed up with ............................................... 122
About using bplist to retrieve SQL Server backups ............................ 122
About NetBackup for SQL Server backup names ......................... 123
Contents 11
Chapter 6 Protecting SQL Server data with VMware
backups ......................................................................... 125
About Protecting SQL Server data with VMware backups ................... 125
About VMware SQL Server systems supported ................................ 126
About VMware SQL Server required applications and licenses ............ 127
Adding a new VMware backup policy to protect SQL Server ................ 127
Restoring SQL Server databases from a VMware backup .................... 128
Limitations of using a VMware policy to protect SQL Server ............... 129
Chapter 7 Disaster recovery ............................................................... 131
About disaster recovery ............................................................... 131
Preparing for disaster recovery of SQL Server .................................. 131
About disaster recovery of SQL Server ............................................ 132
Restoring an existing installation of SQL Server with other active
databases ............................................................................ 132
About disaster recovery of SQL Server databases .............................. 133
Rebuilding the master database .............................................. 133
Starting SQL Server in single-user mode ................................... 133
Restoring the SQL Server databases ......................................... 134
Chapter 8 Using NetBackup for SQL Server with clustering
solutions ........................................................................ 135
Overview of NetBackup for SQL Server with clustering solutions ......... 136
About using NetBackup for SQL in a cluster ..................................... 136
Configuring the NetBackup server to be aware of clustered SQL Server
instances ............................................................................ 136
Performing a backup on a virtual SQL Server instance ....................... 137
Performing a restore on a virtual SQL Server instance ....................... 138
About using SQL Server in a cluster with a multi-interface network
connection (multi-NIC) .......................................................... 138
About master server configuration for a SQL Server in a cluster with
a multi-NIC ......................................................................... 139
Configuring a policy for a SQL Server in a cluster with a
multi-NIC ............................................................................ 139
Adding the permissions that allow for browsing of backups across
the private interface in a cluster .............................................. 139
Configuring clients for a SQL Server in a cluster with a
multi-NIC ............................................................................ 140
Specifying the private name of the client for a SQL Server in a
cluster with a multi-NIC ................................................... 140
Performing backups in a cluster with a multi-NIC connection ............. 140
12 Contents
Performing restores in a cluster with a multi-NIC connection ............. 142
Chapter 9 Using NetBackup for SQL Server with Snapshot
Client .............................................................................. 145
About NetBackup for SQL Server with Snapshot Client ...................... 145
About NetBackup Snapshot Client for SQL Server overview ................ 145
About Snapshot Client features ............................................... 146
How SQL Server operations use Snapshot Client .............................. 147
About selection of backup method ........................................... 147
About SQL Server limitations ................................................. 147
What is backed up by NetBackup for SQL Server ........................ 147
About Snapshot Client and SQL Server performance
considerations ............................................................... 148
Performing SQL Server snapshot backups ................................. 148
Performing SQL Server snapshot restores ................................. 149
About SQL Server agent grouped backups ................................. 149
Restoring a database backed up in a group ................................ 151
Using copy-only or cloaked snapshot backups to affect how
differentials are based ..................................................... 152
Configuring a snapshot backup ..................................................... 156
Configuration requirements ................................................... 156
Configuration steps .............................................................. 157
Configuring Snapshot Client policies for NetBackup for SQL
Server .......................................................................... 157
Chapter 10 Backups and restores in an SAP environment ............ 161
About backups and restores in an SAP environment ......................... 161
About NetBackup manual backups ........................................... 161
About policy configuration for SQL Server in an SAP
environment ................................................................. 162
Creating batch files for automatic backups in for SQL Server in
an SAP environment ....................................................... 162
Monitoring backups on SQL Server .......................................... 163
Restoring the R/3 database ..................................................... 164
Chapter 11 Backup Exec restore topics ............................................. 167
Using NetBackup to restore SQL Server from Backup Exec
images ................................................................................ 167
Requirements for restoring SQL Server from Backup Exec
images .......................................................................... 167
Contents 13
Limitations when restoring SQL Server from Backup Exec
images .......................................................................... 168
Specifying the server, client, and policy type ................................... 168
Restoring from Backup Exec images ............................................... 169
Restore options for restoring SQL backups from Backup Exec
images ................................................................................ 169
Restoring Backup Exec SQL transaction logs up to a named
transaction .......................................................................... 174
Redirecting a restore that was backed up with Backup Exec ................ 175
Restoring Backup Exec SQL transaction logs backups up to a point
in time ................................................................................ 175
Restoring the SQL master database from a Backup Exec image ............ 176
Restarting SQL using database copies ............................................. 177
Restoring Backup Exec filegroup backups ....................................... 179
Restoring Backup Exec Database backups ....................................... 180
Chapter 12 Using NetBackup for SQL Server with database
mirroring ....................................................................... 181
About NetBackup for SQL Server with database mirroring ................. 181
Configuring NetBackup to support database mirroring ...................... 182
Initiating simultaneous backups for mirrored partners ..................... 182
Browsing for mirrored database backup images ............................... 183
Chapter 13 Troubleshooting ................................................................. 185
About progress reports created for NetBackup for SQL Server on the
client ................................................................................. 185
About sample progress report for NetBackup for SQL Server
backup ............................................................................... 186
About debug logging for SQL Server troubleshooting ........................ 188
Creating all NetBackup debug logs for SQL Server
troubleshooting ................................................................... 189
About backup operation debug logging for SQL Server ...................... 189
About restore operation debug logging for SQL Server ...................... 189
Setting the maximum trace level for NetBackup for SQL Server .......... 190
About NetBackup reports for SQL Server troubleshooting .................. 190
About minimizing timeout failures on large SQL Server database
restores .............................................................................. 190
Troubleshooting VMware SQL Server jobs ....................................... 191
14 Contents
Chapter 14 Sample batch files ............................................................. 193
About sample batch files .............................................................. 193
About NetBackup for SQL Server–Simple script to back up a
database named BUSINESS .............................................. 194
About NetBackup for SQL Server–Simple script to restore a
database named pubs ...................................................... 194
About NetBackup for SQL Server–Perform a striped database
backup and allow multiple internal buffers per stripe ............ 195
About NetBackup for SQL Server–Restore a database from
multiple stripes .............................................................. 195
About NetBackup for SQL Server–Restore a database transaction
log up to a point in time ................................................... 196
About NetBackup for SQL Server–Perform an operation and
specify the user ID and password to use to SQL Server ........... 196
About NetBackup for SQL Server–Perform multiple operations
in sequence ................................................................... 197
About NetBackup for SQL Server–Perform a set of operations in
parallel ......................................................................... 198
About NetBackup for SQL Server–Specify the maximum transfer
size and block size for a backup ......................................... 200
About NetBackup for SQL Server–Stage a database restore from
a database backup, a differential backup, and a series of
transaction backups ........................................................ 200
About NetBackup for SQL Server–Stage a database restore from
a filegroup backup, several file backups, and transaction log
backups ........................................................................ 202
About NetBackup for SQL Server–Using environment variables
to exclude instances and databases from backup .................. 205
Index ................................................................................................................... 207
Chapter 1
Introduction to NetBackup
for SQL Server
This chapter includes the following topics:
■ Overview of NetBackup for SQL Server
■ About NetBackup operations
■ About the NetBackup for SQL Server graphical user interface
■ NetBackup for SQL Server terminology
■ Technical overview
■ What are the components of NetBackup for SQL Server?
■ How does NetBackup for SQL Server back up a database?
■ How does NetBackup for SQL Server recover a database?
■ What factors affect the data transfer rate during a SQL Server backup or restore
operation?
■ How does NetBackup resolve SQL Server host and instance names?
■ About SQL Server backups and restores in a VMware environment
■ NetBackup documentation
Overview of NetBackup for SQL Server
NetBackup for SQL Server extends the capabilities of NetBackup for Windows to
include backing up and restoring SQL Server databases. These capabilities are
16 Introduction to NetBackup for SQL Server
About NetBackup operations
provided for a Windows client using either a UNIX or Windows NetBackup master
server.
NetBackup for SQL Server includes a client-based graphical user interface (GUI)
program to perform various activities on SQL Server. These activities include the
following:
■ Configuration of options for NetBackup for SQL Server operations.
■ Backups and restores of databases and database components, which include
transaction logs, differentials, files, and filegroups.
■ Starting NetBackup for SQL Server operations from the batch files which you
have created.
■ Monitoring NetBackup for SQL Server operations.
Microsoft SQL Server is referred to as SQL Server. NetBackup for Microsoft SQL
Server is referred to as NetBackup for SQL Server.
About NetBackup operations
NetBackup for SQL Server includes the following operation features.
■ Full integration with the NetBackup master server and Media Manager.
■ Stream-based backup and restore of SQL Server objects to tape or disk with
SQL Server's high-speed virtual device interface.
■ Snapshot-based backup and restore of SQL Server objects with NetBackup
Snapshot Client methods.
■ Backup and recovery of databases, differentials, files, filegroups, and
transaction logs.
■ Browse capability for SQL Server objects on the local nodes and remote nodes.
■ Support for redirection of SQL Server restores to different locations.
■ Support for multiple SQL Server instances.
■ Client operation monitoring through the NetBackup Client job monitor. Server
monitoring is also available through the NetBackup master.
■ Performance tuning through user control of backup stripes, transfer size, and
buffer usage.
■ Job launch is supported through the following options:
■ Immediate launch through the NetBackup Database Extension GUI
■ Scheduled backup in a backup policy
Introduction to NetBackup for SQL Server 17
About the NetBackup for SQL Server graphical user interface
■ Command line launch
■ Support for instances of SQL Server that are clustered with Microsoft Cluster
Server or Veritas Cluster Server.
■ Recovery of the Microsoft SQL Server images that were backed up with Backup
Exec, through the Backup, Archive, and Restore (NetBackup Client) interface.
■ Support for SQL Server 2005 enhancements: page level validation, page level
recovery, database mirroring, on-line restore, and partial database backup and
restore.
■ Support for SQL Server 2007 enhancements: XX.
■ Partial recovery
■ Optimization for read-only filegroups
■ Verify-only restore
■ Page verification during backup and restore
■ Page-level restore
■ Backup and restore of the full text search catalog
■ Extended object information in the SQL Server catalog
■ Progress statistics during backup and restore
■ Copy-only back up
■ Compression and encryption of backups
■ Option to retry unsuccessful backups automatically
■ Ability to restore a multistream backup with use of fewer devices than it was
backed up with
■ Support for application consistent full backups of VMware machines running
SQL Server
About the NetBackup for SQL Server graphical user
interface
NetBackup for SQL Server includes the following graphical user interface features:
■ GUI capability to browse SQL Server databases, filegroups, and files.
■ Stage a complete database recovery. This recovery can be made from the
backup images that were created for databases, filegroups, files, database
18 Introduction to NetBackup for SQL Server
NetBackup for SQL Server terminology
differentials, and transaction logs. Recovery can also be made from read-write
filegroup backups and partial database backups.
■ GUI assistance for page level restore.
■ Assistance to minimize backup volume thorough the use of the read-only
versus read-write properties of SQL Server data.
■ Restore any database objects that are backed up on one SQL Server client to
another SQL Server client.
■ If you do not want to perform an immediate job launch, you can create and
save a backup script.
■ Properties display for SQL Server databases, filegroups, and files.
■ Properties display for NetBackup for SQL Server backup images.
■ Online Help provided through the NetBackup Database Extension GUI.
NetBackup for SQL Server terminology
Table 1-1 shows the important terms that might be new to a SQL Server database
administrator or a NetBackup administrator.
Table 1-1 NetBackup for SQL Server terminology
Term Definition
batch file The script that is used to back up or to restore SQL Server objects. The
database agent performs all operations through a batch file. Batch
files are typically stored in the install_path\dbext\mssql\
directory. For operations executed immediately from the NetBackup
Microsoft SQL Client, a temporary batch file is placed in the following
directory:
\Veritas\Netbackup\dbext\mssql\temp directory
full backup A complete backup of the database that contains all of the data files
and the log file. (Note that a full backup does not truncate the
transaction log.)
differential backup A backup of the changed blocks since the last full backup.
transaction log An ongoing record of updates that were made to a database.
transaction log A backup of the inactive portion of the transaction log. Typically, this
backup portion of the transaction log is truncated after it has been backed up
successfully.
Introduction to NetBackup for SQL Server 19
Technical overview
Table 1-1 NetBackup for SQL Server terminology (continued)
Term Definition
restore To copy data back to a SQL Server object (see "recovery").
recovery To bring a database online as a result of a restore.
SQL host The host machine on which SQL Server resides. It may also refer to
the virtual name of a cluster that supports a SQL Server installation.
SQL instance A SQL Server installation. If an instance is not specified, it is
considered the default SQL instance for the SQL host.
source client A NetBackup term that identifies a host machine. The source client
is commonly the network name of the host. It can also be an IP address
or a cluster name, depending on how it is identified in the client
configuration.
backup stripes A data stream that is used for a backup or a restore of SQL Server
objects. The user specifies the number of stripes for the backup.
NetBackup performs a separate job each stripe that is specified.
multiplex When more than one backup stripe is written simultaneously to the
same tape.
multistream multistreaming is the generic method in which NetBackup manages
a backup or restore that includes multiple backup stripes. Multiplexing
is an example of multistreaming. NetBackup can also perform a
multistreamed backup by writing individual streams to individual
drives.
ODBC An open interface protocol that NetBackup for SQL Server uses to
interact with SQL Server.
VDI Virtual device interface. A proprietary interface that SQL Server
provides for backup and for restore. The interface is used both for
snapshot and for streamed operations. A VDI connection is managed
as a COM object.
Technical overview
This topic describes the NetBackup for SQL Server database agent in more detail.
These details include the following:
■ See “What are the components of NetBackup for SQL Server?” on page 20.
■ See “How does NetBackup for SQL Server back up a database?” on page 22.
20 Introduction to NetBackup for SQL Server
What are the components of NetBackup for SQL Server?
■ See “How does NetBackup for SQL Server recover a database?” on page 22.
■ See “What factors affect the data transfer rate during a SQL Server backup or
restore operation?” on page 23.
■ See “How does NetBackup resolve SQL Server host and instance names?”
on page 24.
What are the components of NetBackup for SQL
Server?
Table 1-2 describes the components of NetBackup for SQL Server.
Table 1-2 Components of NetBackup for SQL Server
Component Filename Description
graphical user dbbackup.exe You use this interface to browse database objects
interface (GUI) and backup images, create backup and restore
scripts, and launch backup and restore operations.
driver dbbackex.exe Launches backup and restore operations.
library dbbackmain.dll Facilitates backup and restore activities, access
to SQL Server, and other operations that
NetBackup for SQL Server performs.
These components also interface with VxBSA.dll, which is a common NetBackup
client module that connects NetBackup for SQL Server to the NetBackup server.
Figure 1-1 shows the relationships of NetBackup for SQL Server with other
software components.
Introduction to NetBackup for SQL Server 21
What are the components of NetBackup for SQL Server?
Figure 1-1 NetBackup for SQL Server components
NetBackup client NetBackup server
2 bphdb.exe
1
dbbackex.exe NetBackup
server
4 dbbackmain.dll
6
3 8 9
5 NetBackup
VxBSA.dll
Media Manager
dbbackup.exe
SQL Server
DBMS
The following interactions occur between NetBackup for SQL Server and other
software components:
■ Every backup or restore operation is initiated through dbbackex.exe, in one
of the following ways:
■ Scheduled backups
The NetBackup scheduler calls bphdb (1), which calls dbbackex (2).
■ GUI-initiated backups
dbbackup.exe invokes dbbackex.exe (3).
■ Command line
dbbackex.exe is invoked directly from a command line or third-party tool.
■ Dbbackex.exe makes function calls to dbbackmain.dll (4) to facilitate a backup
or a restore operation. The operation is carried out as dbbackmain.dll facilitates
one or more data streams between SQL Server and NetBackup server. The data
stream (7) is established through VDI (5) and the XBSA interface (6). VDI
interacts with SQL Server whereas XBSA interacts with the NetBackup database
client.
22 Introduction to NetBackup for SQL Server
How does NetBackup for SQL Server back up a database?
■ The NetBackup for SQL Server GUI (dbbackup.exe) lets you browse for SQL
Server objects, normally, databases, filegroups, and database files. dbbackup.exe
invokes dbbackmain.dll (8) for accessing the SQL Server master database.
NetBackup for SQL Server accesses information about SQL Server through
ODBC.
■ The NetBackup for SQL Server GUI (dbbackup.exe) also lets you browse for
SQL backup images. The NetBackup catalog contains the images you can
browse. To access the contents of the catalog the GUI invokes dbbackmain.dll,
which uses VxBSA function calls to access the NetBackup server database
manager.
How does NetBackup for SQL Server back up a
database?
From the NetBackup Microsoft SQL Client you can select the databases or the
database components that you want to back up. When you select an object for
backup, the GUI creates a batch file that contains a script. The SQL Server database
agent interprets this script. You can either execute the script immediately or save
it to run later. You can run a script later by including it in the backup selections
list of a backup policy.
When a backup script is executed, NetBackup for SQL Server does the following:
generates an SQL backup statement, logs into SQL Server, and delivers the SQL
statement to SQL Server through ODBC. Next, the database agent connects to SQL
Server through one or more VDI objects. One virtual device is created per backup
stripe. In addition, a VxBSA session is initiated for each stripe. These separate
sessions allow NetBackup to start a backup job for each stream that is generated
from SQL Server.
When the backup completes, the database agent obtains detailed properties of
the object that was backed up, including its relationships to other objects. The
agent writes this information to the NetBackup catalog and associates it with the
backup image. If there are multiple stripes, then the metadata is associated with
the first backup image. The adjunct stripes are associated with one another based
upon a common naming convention.
How does NetBackup for SQL Server recover a
database?
The NetBackup Microsoft SQL client displays backup images in a logical hierarchy
that mirrors the composition of the database. If you select a transaction log or
Introduction to NetBackup for SQL Server 23
What factors affect the data transfer rate during a SQL Server backup or restore operation?
differential image, then NetBackup examines the metadata that is stored with the
images for the selected database. It then determines the most efficient recovery
set. Then the agent generates a batch file that includes a sequence of scripted
restores. When the scripts are executed, the database is recovered.
The individual restore operations work in a similar manner to backups. An SQL
restore statement is generated and provided to SQL Server by ODBC. A VDI
connection is made. Then a VxBSA session is initiated that starts the data flow
between the media manager and SQL Server. NetBackup determines the number
of streams (and the corresponding virtual devices and VxBSA sessions) by the
number of stripes that were generated during backup.
After all of the recovery operations have completed, the NetBackup agent takes
the final step that sets the database into the recovered state. The database goes
back online and becomes available for use.
What factors affect the data transfer rate during a
SQL Server backup or restore operation?
To optimize your system for SQL Server backups the first step is to tune your
environment for standard backup operations. Details are provided in the NetBackup
Backup Planning and Performance Tuning Guide. With this information as a
background, you still must adjust several things specific to SQL Server. Note that
some of the following factors are only applicable to SQL Server stream-based
operations and have no affect on snapshot backups or restores. A detailed
description of how to make these adjustments is available.
See “About NetBackup for SQL performance factors” on page 49.
The following factors affect the data transfer rate during a SQL Server backup or
restore operation.
MAXTRANSFERSIZE (stream-based only) The size that SQL Server uses for
transferring data. Among the SQL Server-specific factors,
MAXTRANSFERSIZE has the greatest overall potential for
optimizing transfer rates. In general, improved performance
correlates directly with an increased MAXTRANSFERSIZE
value.
24 Introduction to NetBackup for SQL Server
How does NetBackup resolve SQL Server host and instance names?
NUMBER OF BUFFERS PER (stream-based only) Setting this factor to a value greater
STRIPE than one enables multi-buffer during data transfer.
Multi-buffer prevents short-term producer-consumer
imbalances during a backup or restore operation. Although
you can set the number of buffers as high as 32, normally
a value of 2 or 3 is sufficient.
This setting corresponds to the NUMBUFS keyword. You
can also use the BUFFERS keyword.
NUMBER OF STRIPES You can improve performance if you increase the number
of stripes that you use to back up a SQL Server object. You
must meet the following requirements:
■ Each stripe is backed up to a separate tape unit (the
stripes are not multiplexed to the same drive).
■ You have reached the maximum transfer rate of any
single drive.
Performance varies depending on your environment. Since
each stripe uses individual threads, multiple stripes may
lead to contention when they access internal SQL Server
structures. A cost that is associated with management of
the threads. However, you may see improvement on a large
system if the cost to start starting and stop I/O is greater
than using additional stripes.
SHARED MEMORY Shared memory should always be used if it is available in
your backup environment.
ALTERNATE BACKUP (stream-based only) A transfer method in which SQL Server
METHOD and NetBackup are allowed to share the same data transfer
buffers. This method may not be faster than ordinary shared
memory data transfer. But this method results in decreased
CPU utilization because it avoids a data copy step. If you
use ALTERNATE BACKUP METHOD, Symantec
recommends that you have MAXTRANSFERSIZE set to the
maximum value.
How does NetBackup resolve SQL Server host and
instance names?
Normally SQL Server identifies its installations with a combination that includes
the name of the host on which the installation resides plus an instance name. If
you omit the instance name then NetBackup assumes that the installation is the
default installation on the host. For example, a single host may contain several
Introduction to NetBackup for SQL Server 25
How does NetBackup resolve SQL Server host and instance names?
SQL Server installations, such as, TIGER, TIGER\ACCOUNTING, and
TIGER\WAREHOUSE. However, a clustered instance of SQL Server resides jointly
on multiple hosts and is identified with a virtual name.
Backups for a host are cataloged under the client name that is specified in the
client properties for that host. Ordinarily, the client name is the same as the host
name on which SQL Server resides. Normally, this name is the host's NetBIOS
name (such as TIGER). However, some backup administrators prefer to identify
the client to NetBackup in other ways. For example, an administrator might use
the IP name (for example, 20.81.74.123) or the fully qualified domain name
(tiger.apexworks.com). Or, an administrator might use an alternate network
interface that is attached to the host (tiger1.apexworks.com). For clustered SQL
instances, the client name is the virtual SQL Server name that is used for the
virtual SQL Server installation.
In most cases when you browse for backup images using the NetBackup Microsoft
SQL Client, you only need to specify the SQL Server host name. Specify this name
in the Backup History Options dialog box.
NetBackup then displays the backup images for all of the instances on that host.
However, to ensure that NetBackup displays the backup images you want, consider
the following special cases:
■ Retrieval of backups on a network interface not having the same name as the
host name (such as tiger1 or tiger.apexworks.com).
In this case, the backup images are stored under the network interface name
and not the NetBIOS name. To retrieve these images, you should specify the
host name in the SQL Host field and the network interface name in the Source
Client field.
SQL Host: TIGER
Source Client: TIGER1
■ Retrieving backups from a UNIX (or Linux) server.
This scenario may present a problem because UNIX names are case sensitive,
whereas Windows names are not. In this case, NetBackup tries to retrieve the
backup images by specifying the client name with all upper case characters
or all lower case characters. If the UNIX client name has mixed uppercase and
lowercase characters, you must provide the client name in the Source Client
field.
SQL Host: TIGER
Source Client: Tiger
■ The NetBackup client name is a qualified domain name.
To retrieve backup images specify the SQL Host as the NetBIOS name and the
Source Client as the fully qualified domain name.
SQL Host: Tiger
26 Introduction to NetBackup for SQL Server
About SQL Server backups and restores in a VMware environment
Source Client: tiger.apexworks.com
■ The NetBackup client name is an IP address.
To retrieve backup images specify the SQL Host as the NetBIOS name and the
Source Client as the IP address:
SQL Host: Tiger
Source Client: 10.80.136.68
■ Retrieval of the backups that were backed up from a SQL Server cluster.
This scenario does not present any special problem because the images are
stored under the cluster name. Provide the SQL Host name as the virtual name
and use the default value for the Source Client.
About SQL Server backups and restores in a VMware
environment
In NetBackup 7.5, application-aware VMware backups provide a single pass backup
that protects SQL Server. From one VMware backup the following restore options
are available: disk level restore, file-level recovery, or database restore. You can
also choose whether or not to truncate logs.
NetBackup documentation
NetBackup documentation is part of the NetBackup media kit and can be found
in a directory on the installation DVD for each release platform.
The documents are in Adobe® Portable Document Format (PDF), viewable with
the Adobe Acrobat Reader. Download the reader from the Adobe web site.
Symantec assumes no responsibility for the installation and use of the reader.
For a complete list of NetBackup technical documents, see the Related Documents
appendix in the NetBackup Release Notes.
The Symantec support Web site contains links to the most recent documentation
plus a list of helpful how-to and product alert topics.
Chapter 2
Installing NetBackup for
SQL Server
This chapter includes the following topics:
■ Planning the installation of NetBackup for SQL Server
■ Verifying the operating system and platform compatibility for NetBackup for
SQL Server
■ NetBackup server and client requirements for NetBackup for SQL Server
■ SQL Server server software requirements for NetBackup for SQL Server
■ Requirements for installing NetBackup for SQL Server in a NetBackup cluster
■ About support for Microsoft Cluster server (MSCS) clusters with NetBackup
for SQL
■ About support for Veritas Cluster Server (VCS) clusters with NetBackup for
SQL
■ Configuration and licensing requirements for NetBackup for SQL Server with
Snapshot Client
■ Adding license keys for NetBackup for SQL Server backups
Planning the installation of NetBackup for SQL Server
Table 2-1 shows the major installation steps needed to run NetBackup for SQL
Server. Each step contains one or more links to pertinent procedures and concepts.
28 Installing NetBackup for SQL Server
Verifying the operating system and platform compatibility for NetBackup for SQL Server
Table 2-1 Installation steps for NetBackup for SQL Server
Step Action Description
Step 1 Verify the installation prerequisites. See “Verifying the operating system and platform
compatibility for NetBackup for SQL Server” on page 28.
See “NetBackup server and client requirements for
NetBackup for SQL Server” on page 29.
See “SQL Server server software requirements for
NetBackup for SQL Server” on page 29.
See “Requirements for installing NetBackup for SQL Server
in a NetBackup cluster” on page 30.
See “Configuration and licensing requirements for
NetBackup for SQL Server with Snapshot Client” on page 31.
For VMware backups, follow the instructions for VMware
installation and configuration in the NetBackup for VMware
Administrator's Guide.
Step 2 Add the license key for NetBackup for See “Adding license keys for NetBackup for SQL Server
SQL Server. backups” on page 31.
Verifying the operating system and platform
compatibility for NetBackup for SQL Server
Verify that the NetBackup for SQL Server agent is supported on your operating
system or platform.
To verify operating system and compatibility
1 Go to the Symantec Support Web page:
http://www.symantec.com/business/support/index.jsp
2 Click on the link for NetBackup Enterprise Server.
3 In the list of documents, click on the following document:
NetBackup Database Agent Compatibility List
4 For information on supported cluster environments for NetBackup for SQL
Server, see the following document:
NetBackup Cluster Compatibility List
5 For information on support for Snapshot Client, see the following document:
NetBackup Snapshot Client Compatibility List
Installing NetBackup for SQL Server 29
NetBackup server and client requirements for NetBackup for SQL Server
NetBackup server and client requirements for
NetBackup for SQL Server
Verify that the following requirements are met for the NetBackup server:
■ The NetBackup server software is installed and operational on the NetBackup
server. The NetBackup server platform can be any that NetBackup supports.
See the NetBackup Installation Guide.
■ Make sure that you configure any backup media that the storage unit uses.
The number of media volumes that are required depends on several things:
■ The devices used
■ The sizes of the databases that you want to back up
■ The amount of data that you want to archive
■ The size of your backups
■ The frequency of backups or archives
See the NetBackup Administrator’s Guide, Volume I.
Verify that the following requirements are met for the NetBackup clients:
■ The NetBackup client software is installed on the computer that has the
databases you want to back up.
In a VMware environment, the NetBackup client software must be installed
on the virtual machines that have SQL Server running.
■ If the SQL client is on a different host than the master server or media server,
then install the NetBackup client on that host.
SQL Server server software requirements for
NetBackup for SQL Server
Verify the following regarding the SQL Server server software on the NetBackup
server or client:
■ SQL Server server software must be installed and operational.
30 Installing NetBackup for SQL Server
Requirements for installing NetBackup for SQL Server in a NetBackup cluster
Requirements for installing NetBackup for SQL Server
in a NetBackup cluster
If you plan to use the database agent software on a NetBackup server configured
in a NetBackup cluster, verify the following requirements:
■ Your cluster environment is supported by NetBackup:
■ See “Verifying the operating system and platform compatibility for
NetBackup for SQL Server” on page 28.
■ See “About support for Microsoft Cluster server (MSCS) clusters with
NetBackup for SQL” on page 30.
■ See “About support for Veritas Cluster Server (VCS) clusters with NetBackup
for SQL” on page 31.
■ The NetBackup server software is installed and configured to work in a
NetBackup cluster.
See the NetBackup Installation Guide.
See the NetBackup Clustered Master Server Administrator's Guide .
■ The SQL Server server software is installed and operational on each node to
which NetBackup can failover.
■ On each node where NetBackup server resides, add the license key for the
database agent.
About support for Microsoft Cluster server (MSCS)
clusters with NetBackup for SQL
NetBackup supports MSCS on 32-bit systems. On SQL Server 2000 a maximum of
four nodes is supported, on SQL Server 2005, 8 nodes, on SQL Server 2008, XX
nodes. Symantec recommends that the NetBackup server and the SQL Server be
in different cluster groups.
Table 2-2 includes the maximum failover nodes and the virtual servers that are
used in an MSCS.
Table 2-2 Maximum failover nodes and virtual servers allowed in an MSCS
SQL Server 2000 SQL Server 2005 SQL Server 2008
maximum failover 4 8 XX
nodes allowed,
32-bit systems
Installing NetBackup for SQL Server 31
About support for Veritas Cluster Server (VCS) clusters with NetBackup for SQL
Table 2-2 Maximum failover nodes and virtual servers allowed in an MSCS
(continued)
SQL Server 2000 SQL Server 2005 SQL Server 2008
maximum failover - 4 XX
nodes allowed,
64-bit systems
maximum virtual 24 24 XX
servers allowed
If SQL Server shares a cluster with a virtual NetBackup server, then SQL Server
and the NetBackup server may be placed in the same group. The SQL Server and
the NetBackup server must have different IP and network names.
About support for Veritas Cluster Server (VCS)
clusters with NetBackup for SQL
NetBackup for SQL Server supports VCS clusters with active SQL Server instances.
Configuration and licensing requirements for
NetBackup for SQL Server with Snapshot Client
To use NetBackup for SQL Server with Snapshot Client, you must have a license
for NetBackup Snapshot Client.
Adding license keys for NetBackup for SQL Server
backups
To use NetBackup for SQL Server, add a valid license key for it on the master
server. More information is available on how to add license keys.
See the NetBackup Administrator’s Guide, Volume I.
In a NetBackup cluster, add the key on each node where the NetBackup server is
installed.
To add a NetBackup for SQL Server license key in the NetBackup Administration
Console
1 On the master server, open the NetBackup Administration Console.
2 Choose Help > License Keys.
32 Installing NetBackup for SQL Server
Adding license keys for NetBackup for SQL Server backups
3 Click the New icon.
4 Type the license key and click Add.
Chapter 3
Configuring NetBackup for
SQL Server
This chapter includes the following topics:
■ About user interface terminology
■ Configuring the Maximum jobs per client
■ About configuring a backup policy for an SQL Server database
■ Testing configuration settings for NetBackup for SQL Server
■ Configuring multistream operations
■ About configuration for a multi-interface network connection (multi-NIC)
■ About SQL Server privileges
■ About setting the SQL Server logon for scheduled operations
■ Authorizing scheduled operations (for sites with SQL Server security
restrictions)
■ Mapping a Windows account to a standard SQL Server user ID
■ About NetBackup for SQL performance factors
■ Backing up SQL Server in an environment with log shipping
About user interface terminology
You perform many configuration steps from the NetBackup Administration
Console on the master server. The type of console available depends on your
master server platform. NetBackup supports a Java interface for both Windows
34 Configuring NetBackup for SQL Server
Configuring the Maximum jobs per client
and UNIX master servers. In addition, NetBackup supports a Windows interface
for Windows master servers.
The Java and Windows interfaces are nearly identical. If interface differences
exist in the configuration procedures, the term “Windows” or “Java” appears in
the procedure to differentiate between the two interfaces.
Configuring the Maximum jobs per client
The following procedure shows how to set the Maximum jobs per client attribute.
To configure the maximum jobs per client
1 In the left pane of the NetBackup Administration Console, expand NetBackup
Management > Host Properties.
2 Select Master Server.
3 In the right pane, double-click the server icon.
4 Click Global Attributes.
5 Change the Maximum jobs per client value to 99.
The Maximum jobs per client specifies the maximum number of concurrent
backups that are allowed per client. The default is 1.
You can use the following formula to calculate a smaller value for the Maximum
jobs per client setting:
Maximum jobs per client = number_of_streams X number_of_policies
Refer to the following definitions:
number_of_streams The number of backup streams between the database server and
NetBackup. If striping is not used, each separate stream starts a new
backup job on the client. If striping is used, each new job uses one
stream per stripe.
number_of_policies The number of policies of any type that can back up this client at the
same time. This number can be greater than one. For example, a client
can be in two policies to back up two different databases. These backup
windows can overlap.
Configuring NetBackup for SQL Server 35
About configuring a backup policy for an SQL Server database
About configuring a backup policy for an SQL Server
database
A backup policy for a database defines the backup criteria for a specific group of
one or more clients.
These criteria include the following:
■ Storage unit and media to use
■ Policy attributes
■ Backup schedules
■ Clients to be backed up
■ The batch files to run on the clients
To back up the database environment, define at least one MS-SQL-Server policy
with the appropriate schedules. A configuration can have a single policy that
includes all clients, or there can be many policies, some of which include only one
client.
Most requirements for database policies are the same as for file system backups.
In addition to the policy attributes for this database agent, other attributes are
available that you should consider.
See the NetBackup Administrator’s Guide, Volume I.
To add and configure a policy, see the following topics:
■ See “Adding a new NetBackup for SQL Server policy” on page 35.
■ See “About policy attributes for NetBackup for SQL Server” on page 36.
■ See “About adding schedules to a NetBackup for SQL Server policy” on page 37.
■ See “Adding clients to a NetBackup for SQL Server policy” on page 39.
■ See “Adding backup selections to a NetBackup for SQL Server policy”
on page 40.
Adding a new NetBackup for SQL Server policy
This topic describes how to add a new backup policy for a database.
Note: To perform multistreamed backups and restores, or if you have multiple
network interfaces, you need to perform other configuration.
36 Configuring NetBackup for SQL Server
About configuring a backup policy for an SQL Server database
To add a new NetBackup for SQL Server policy
1 Log on to the master server as administrator (Windows) or root (UNIX).
2 Start the NetBackup Administration Console.
3 If your site has more than one master server, choose the one on which you
want to add the policy.
4 Do one of the following:
■ Windows: In the left pane, right-click Policies and choose New Policy.
■ Java interface: In the left pane, click Policies. In the All Policies pane,
right-click the master server, and click New Policy.
5 In the Add a New Policy dialog box, in the Policy name box, type a unique
name for the new policy.
6 Click OK.
7 In the Add New Policy dialog box, in the Policy type list, select
MS-SQL-Server.
The database agent policy type does not appear in the drop-down list unless
your master server has a license key for the database agent.
8 Complete the entries on the Attributes tab.
See “About policy attributes for NetBackup for SQL Server” on page 36.
9 Add other policy information as follows:
■ Add schedules.
See “About adding schedules to a NetBackup for SQL Server policy”
on page 37.
■ Add clients.
See “Adding clients to a NetBackup for SQL Server policy” on page 39.
■ Add batch files to the backup selections list.
10 When you have added all the schedules, clients, and backup selections you
need, click OK.
About policy attributes for NetBackup for SQL Server
With a few exceptions, NetBackup manages a database backup like a file system
backup. Other policy attributes vary according to your specific backup strategy
and system configuration.
For more information on policy attributes, see the NetBackup Administrator’s
Guide, Volume I.
Configuring NetBackup for SQL Server 37
About configuring a backup policy for an SQL Server database
Table 3-1 Policy attribute descriptions for NetBackup for SQL Server policies
Attribute Description
Policy type Determines the types of clients that can be in the policy. In some cases the policy type
determines the types of backups that NetBackup can perform on those clients. To use the
SQL Server database agent, you must define at least one policy of type that is MS-SQL-Server.
Keyword phrase Although you can create a keyword phrase for MS-SQL-Server policies, NetBackup for SQL
Server does not record this information with the backup image.
Snapshot Client This option enables backups with Snapshot Client.
Perform snapshot This option enables backups with Snapshot Client.
backups
See “About NetBackup for SQL Server with Snapshot Client” on page 145.
About adding schedules to a NetBackup for SQL Server policy
Each policy has its own set of schedules. These schedules initiate automatic
backups and specify when a user can initiate operations. A database backup has
these types of schedules: application backup and automatic backup.
About NetBackup for SQL Server schedule properties
This topic describes the schedule properties that have a different meaning for
database backups than for file system backups. Other schedule properties vary
according to your specific backup strategy and system configuration. Additional
information about other schedule properties is available.
See the NetBackup Administrator’s Guide, Volume I.
Table 3-2 Description of schedule properties
Property Description
Type of backup Specifies the type of backup that this schedule controls. The selection list shows only the
backup types that apply to the policy you want to configure.
See “About the types of NetBackup for SQL Server backups” on page 38.
38 Configuring NetBackup for SQL Server
About configuring a backup policy for an SQL Server database
Table 3-2 Description of schedule properties (continued)
Property Description
Schedule type You can schedule a backup in one of the following ways:
■ Frequency
This setting is used only for scheduled backups. It is not used for user-directed backups.
Frequency specifies the period of time that can elapse until the next backup or archive
operation begins on this schedule. For example, assume that the frequency is seven
days and a successful backup occurs on Wednesday. The next full backup does not
occur until the following Wednesday. Typically, incremental backups have a shorter
frequency than full backups.
■ Calendar
This setting is used only for scheduled backups. It is not used for user-directed backups.
The Calendar option allows you to schedule backup operations that are based on specific
dates, recurring week days, or recurring days of the month.
Retention The retention period for an application backup schedule refers to the length of time that
NetBackup keeps backup images (stream-based backups). The retention period for an
automatic schedule controls how long NetBackup keeps records of when scheduled backups
occurred (proxy backups). For example, if your database is backed up once every Sunday
morning, you should select a retention period of at least 2 weeks.
The retention period is affected by the type of schedule you select, as follows:
■ Frequency-based scheduling
Set a retention period that is longer than the frequency setting for the schedule. For
example, if the frequency setting is set to one week, set the retention period to be more
than one week. The NetBackup scheduler compares the latest record of the automatic
backup schedule to the frequency of that automatic backup schedule to determine
whether a backup is due. This means that if you set the retention period to expire the
record too early, the scheduled backup frequency is unpredictable. However, if you set
the retention period to be longer than necessary, the NetBackup catalog accumulates
unnecessary records.
■ Calendar-based scheduling
The retention period setting is not significant for calendar-based scheduling.
Multiple copies If you want to specify multiple copies of a backup for the policy, configure Multiple copies
on the application backup schedule.
About the types of NetBackup for SQL Server backups
Table 3-3 shows the backup schedules you can specify.
Configuring NetBackup for SQL Server 39
About configuring a backup policy for an SQL Server database
Table 3-3 NetBackup for SQL Server backup types
Backup type Description
Application backup The application backup schedule enables user-controlled
NetBackup operations from the client. These operations include
those initiated from the client and those initiated by an automatic
schedule on the master server. NetBackup uses the application
backup schedule when the user starts a backup manually.
Configure at least one application backup schedule for each
database policy. The Default-Application-Backup schedule is
configured automatically as an application backup schedule.
Automatic backup An automatic backup schedule specifies the dates and times for
NetBackup to automatically start backups. NetBackup runs the
scripts in the order that they appear in the file list. If there is more
than one client in the policy, the scripts are run on each client.
Adding clients to a NetBackup for SQL Server policy
The client list is the list of hosts on which your batch files are run during an
automatic backup. A NetBackup client must be in at least one policy but can be in
more than one.
For a NetBackup for SQL Server policy, clients you want to add must have the
following software installed:
■ SQL Server
■ NetBackup client or server
■ The backup or restore batch file(s)
Note: Each batch file must be present on each client.
To add clients to a NetBackup for SQL Server policy
1 In the Policy dialog box, click the Clients tab.
To access the Policy dialog box, double-click the policy name in the Policies
list in the NetBackup Administration Console.
2 Click New.
3 Type the name of the client and press Enter (Windows) or click Add (Java).
If SQL Server is installed in a NetBackup cluster, specify the virtual SQL
Server name as the client name.
40 Configuring NetBackup for SQL Server
About configuring a backup policy for an SQL Server database
4 (Windows) To add another client, repeat step 2 and step 3.
5 (Java) To add another client, repeat step 3.
6 If this client is the last client you want to add, click OK.
7 (Java) In the Policy dialog box, click Close.
Adding backup selections to a NetBackup for SQL Server policy
The backup selections list in a database policy has a different meaning than for
non-database policies. For example, in a Standard or Microsoft Windows policy,
the list contains files and directories to be backed up. In a database policy, you
can specify batch files to be run. (For NetBackup for SQL Server, the scripts are
called batch files is identified by the .bch extension.) Batch files describe the
backup operations you want to start. They can be started through manual or
scheduled operations initiated from the NetBackup server. These files reside on
the client and direct the operation of NetBackup for SQL Server and SQL Server.
Add batch files to set up a policy for automatic scheduling. All batch files that are
listed in the backup selections list are run for manual backups and for Automatic
Backup schedules. The schedules are specified under the Schedules tab. NetBackup
runs the batch files in the order that the batch files appear in the backup selections
list.
Adding batch files to the backup selections list
The following procedures describe how to add batch files to the backup selections
list using the Java interface and using the Windows interface.
Note: Specify the correct batch file names in the backup selections list to prevent
an error or possibly a wrong operation. Make sure that the batch file resides on
the client before you try to add it to the backup selections list.
To add batch files to the backup selections list from the Java interface
1 Open the Policy dialog box.
To access the Policy dialog box, double-click the policy name in the Policies
list in the NetBackup Administration Console.
2 Click the Backup Selections tab.
3 Click New.
4 Specify the names of the batch files that you want NetBackup to use by doing
the following:
Configuring NetBackup for SQL Server 41
About configuring a backup policy for an SQL Server database
■ In the Script box, type the full path name of a batch file on the client.
For example, a path name for a batch file that backs up a database may
be:
install_path\NetBackup\DbExt\Mssql\bkup.bch
install_path refers to the directory where you installed the NetBackup
software. By default, this directory is C:\Program Files\VERITAS\.
You can also specify only the file name if the batch file resides in the
install_path\NetBackup\Dbext\Mssql\ directory.
■ Click Add to add the batch file to the list.
■ Add any other batch files.
5 Click OK.
To add batch files to the backup selections list from the Windows interface
1 In the Policy dialog box, click the Backup Selections tab.
To access the Policy dialog box, double-click the policy name in the Policies
list in the NetBackup Administration Console.
2 Click New.
3 Specify the names of the batch files that you want NetBackup to use.
Use one of the following methods:
■ Type the full path name of the batch file on the client.
For example, a path name for a batch file that backs up a database may
be:
install_path\NetBackup\DbExt\Mssql\bkup.bch
install_path refers to the directory where you installed the NetBackup
software. By default, this directory is C:\Program Files\VERITAS\.
You can also specify only the file name if the batch file resides in the
install_path\NetBackup\Dbext\Mssql\ directory.
■ Click the Remote Folder option.
Navigate to and select the batch file and click OK.
4 Click OK.
42 Configuring NetBackup for SQL Server
Testing configuration settings for NetBackup for SQL Server
Testing configuration settings for NetBackup for SQL
Server
After you configure the master server for NetBackup for SQL Server, test the
configuration settings. Perform a manual backup (or backups) with the automatic
backup schedules you created. A description of status codes and other
troubleshooting information is available.
See the NetBackup Status Codes Reference Guide.
See the NetBackup Troubleshooting Guide.
To test the configuration settings
1 Log onto the master server as administrator (Windows) or root (UNIX).
2 Start the NetBackup Administration Console.
3 In the left pane, click Policies.
4 Click the policy you want to test.
5 Select Actions > Manual Backup.
The Schedules pane contains the name of an automatic schedule (or schedules)
configured for the policy that you want to test. The Clients pane contains the
name of the client(s) listed in the policy that you want to test.
6 Follow the directions in the Manual Backup dialog box.
7 To check the status of the backup, click Activity Monitor in the NetBackup
Administration Console.
The Activity Monitor and the script output indicate the status of the backup
operation.
For more information, see the Troubleshooting chapter.
8 To test compression, perform a backup with and without compression. Then
compare the image sizes. The compressed image should be smaller.
Configuring multistream operations
NetBackup supports multistriped and multiplexed backups of SQL Server. Backup
streams can each be written to separate storage units or multiplexed to a single
storage unit.
See “Configuring multistriped backups” on page 43.
See “Configuring multiplexed backups” on page 43.
Configuring NetBackup for SQL Server 43
Configuring multistream operations
Configuring multistriped backups
SQL Server supports backups of databases through multiple data streams, which
are called stripes. NetBackup stores each stripe as a separate image. The purpose
of this feature is to speed up the rate of data transmission with the use of multiple
tape devices. This configuration is useful if the tape device is a bottleneck.
Symantec does not recommend that you enable multiplexing for a schedule that
is used for backing up multiple stripes. Restores will not work with MPX enabled
with multistripe backup configured (you must set STRIPES N to STRIPES 1).
Backup images can be written to more tapes than available drives. If this situation,
in the batch file for the restore operation, indicate the number of drives that are
available.
See “Restoring multistreamed backups” on page 120.
To configure a multistriped backup
1 In your "Application Backup" policy, set Media multiplexing to 1 in the
schedule you want to use for striped backups or restores.
Then multiplexing is disabled for this schedule. All streams are made
simultaneously available so restore operations are successful.
2 Ensure that the storage unit has as many drives as you want to have stripes.
3 Configure schedules for backups and restores so that enough drives are
available at the time you want to perform striped backups or restores.
4 In the Backup Microsoft SQL Server Objects dialog box, select a Stripes
setting greater than 1.
Configuring multiplexed backups
NetBackup provides a multiplexing feature that lets you interleave multiple
backups to the same tape. Multiplexing is useful if you have many simultaneous
backups using the same tape drive. However, it can interfere with SQL Server
recovery due to how SQL Server requests streams during restore.
If you use multiplexing when you perform multistreamed backups, see the
information on how to perform restores.
See “Restoring multistreamed backups” on page 120.
To configure multiplexed backups
1 In the schedule for your "Application Backup" policy, set Media multiplexing
to the number of backup stripes that you plan to use.
2 In the storage units that are associated with this schedule, select Enable
Multiplexing.
44 Configuring NetBackup for SQL Server
About configuration for a multi-interface network connection (multi-NIC)
3 Set Maximum streams per drive to the number of stripes that you plan to
use.
4 In the Backup Microsoft SQL Server Objects dialog box, select a Stripes
setting equal to the number of stripes that you want to use.
About configuration for a multi-interface network
connection (multi-NIC)
Many administrators want to reserve a separate network interface for their SQL
Server host machines that are used for routing backup traffic. This type of
configuration requires some special considerations when you configure both the
NetBackup master server and the NetBackup client that backs up SQL Server. The
process to browse backup images across a required network interface is also
different.
For information on how to perform restores with a multi-NIC connection, see the
following topic.
See “Performing restores with a multi-NIC connection” on page 101.
If you have installed SQL Server in a cluster, refer to the following topic.
See “About using SQL Server in a cluster with a multi-interface network connection
(multi-NIC)” on page 138.
Configuring the master server for a multi-NIC
The following configuration changes must be made on the master server to allow
for backups and restores over a private interface:
■ The backup policies must include the private interface name in the Clients list
of the policy.
See “Configuring a policy for a multi-NIC” on page 44.
■ Permissions must be added to allow for browsing of backups across the private
interface.
See “Adding the permissions that allow for browsing of backups across the
private interface” on page 45.
Configuring a policy for a multi-NIC
The private name of the client must be added to the Clients list of the policy. The
NetBIOS or public name of the client should not be used.
Configuring NetBackup for SQL Server 45
About configuration for a multi-interface network connection (multi-NIC)
In the following procedure, the NetBackup client machine that backs up SQL
Server is SQLHOST. The private interface that is used to back up SQLHOST is
SQLHOST-NB.
To configure a policy for a multi-NIC
1 Open the NetBackup Administration Console.
2 Create a new policy or open an existing policy.
3 In the Clients list for the policy, add a new client.
Specify the Client name as the private interface name. For example,
SQLHOST-NB.
Adding the permissions that allow for browsing of backups across the
private interface
The administrator can enable all clients or single clients to browse and restore
backups that are performed over the multi-nic connection.
To allow all clients to browse for backups and perform restores across the private
interface
◆ Add the empty file NB_INSTALL\db\altnames\No.Restrictions on the master
server.
This option allows any NetBackup client to browse the backups the SQL Server
host machine makes.
To allow a single client to browse for backups and perform restores across the
private interface
◆ Add the empty file, NB_INSTALL\db\altnames\SQLHOST on the master server.
This configuration only allows the SQL Server host machine, SQLHOST, to
access the backups the master server manages.
Configuring the client for a multi-NIC
To configure backups of a client in a multi-NIC environment where SQL Server
is clustered, you must provide the following for NetBackup:
■ The private interface name
You can set client name during the installation of the NetBackup client. Or
you can change it in the NetBackup Client Properties dialog box in the Backup,
Archive, and Restore interface.
■ Public interface name
46 Configuring NetBackup for SQL Server
About SQL Server privileges
Set the public interface name the NetBackup Microsoft SQL Client interface,
in the "SQL Server connection properties" dialog box.
Specifying the private name of the client for a multi-NIC
This topic describes how to specify the private name of the client when you want
to configure backups of a client in a multi-NIC environment.
To specify the private name of the client for a multi-NIC
1 Open the Backup, Archive, and Restore interface.
2 Select File > NetBackup Client Properties.
3 Click the General tab.
4 In the Client name box, specify the private name of the client.
For example, the private name for the machine SQLHOST would be
SQLHOST-NB.
Specifying the public name of the client for a multi-NIC
When SQL Server is clustered, specify the public name of the client for configuring
client backups in a multi-NIC environment..
To specify the public name of the client for a multi-NIC
1 Open the NetBackup for SQL Server interface.
2 Select File > Set SQL Server connection properties.
3 In the Host box, specify the public name of the SQL Server host.
4 Click Apply and Close.
About SQL Server privileges
When you install SQL Server, you can elect to use either integrated security or
standard security. Integrated security refers to the usage of Windows
authentication in lieu of standard SQL Server-based logons.
Note: Microsoft recommends using integrated security. Windows logons, unlike
SQL Server-based logons, can be traced with standard Windows security tools.
NetBackup for SQL Server supports both integrated security and standard security
for any level of SQL Server.
If you select integrated security, then SQL Server ignores the user ID and password
that you provided in the SQL Server connection properties dialog box of the
Configuring NetBackup for SQL Server 47
About setting the SQL Server logon for scheduled operations
NetBackup Database Extension GUI. Instead, use security validation to perform
backup and restore operations. Also use security validation for querying the
master database are based upon the SQL Server privileges that are granted to the
Windows account. The batch file user ID and password keywords are also ignored.
When integrated security is used SQL Server operations generated by NetBackup
for SQL Server are always validated against the Windows user ID.
If you select standard security, then you must supply NetBackup for SQL Server
with a SQL Server-based user ID and password. This user ID and encrypted
password are stored in the registry under the following key:
HKEY_CURRENT_USER\SOFTWARE\VERITAS\NETBACKUP\NetBackup for
Microsoft SQL Server\
To perform an operation with standard SQL Server security, NetBackup for SQL
Server looks up the SQL Server user ID and password from the registry.
Note: Since NetBackup for SQL Server does back up and restore commands and
issues select statements against the master database, review the SQL Server user
documentation to determine exactly what privileges are required by the user
accounts that you establish for NetBackup for SQL Server.
About setting the SQL Server logon for scheduled
operations
When using the NetBackup Database Extension GUI, SQL Server security is
satisfied. This is based upon either your current Windows account (if you use
integrated security) or a SQL Server-based user ID and password.
See “About SQL Server privileges” on page 46.
When the NetBackup scheduler is used, backup operations that are launched from
a daemon process called the NetBackup Client Service. You must ensure that the
Client Service has SQL Server privileges to perform backup and restore operations.
If you use integrated security, you must ensure that the NetBackup Client Service
specifies a Windows account that has been granted SQL Server administrative
privileges.
Note: If your site does not place any SQL Server security restrictions on the right
to back up databases then you do not need to authorize schedule operations or
configure standard SQL Server security.
48 Configuring NetBackup for SQL Server
Authorizing scheduled operations (for sites with SQL Server security restrictions)
Authorizing scheduled operations (for sites with SQL
Server security restrictions)
You must set up an account that the NetBackup Client Service can use to log in
to SQL Server. The account must have sufficient SQL Server privileges to perform
a backup. You can then use NetBackup to perform automatic SQL Server backups
of a site that places SQL Server security restrictions on the right to back up
databases.
Note: If your site does not place SQL Server security restrictions on the right to
back up databases, authorization of scheduled operations is not needed.
To authorize scheduled operations for a site with SQL Server security restrictions
1 If you use standard SQL Server security, first map a Windows account to a
standard SQL Server user ID.
See “Mapping a Windows account to a standard SQL Server user ID”
on page 49.
2 Open the Windows Services application.
3 Double-click the NetBackup Client Service entry.
4 Click the Log On tab.
5 Select This account.
6 Type the name of the account, or browse for and select an account.
7 Type the password and the password confirmation.
If you later change the password, you must also indicate the new password
for the NetBackup Client Service, then restart the NetBackup Client Service.
If you do not indicate the new password, NetBackup scheduled backup
operations fail.
8 Log on to Windows with the Windows account selected in step 6.
9 Select the SQL host and instance.
See “Selecting the SQL host and instance” on page 71.
The database logon parameters are used for the SQL Server backups that the
NetBackup scheduler launches.
10 Stop and restart the NetBackup Client Service from the service dialog box.
Configuring NetBackup for SQL Server 49
Mapping a Windows account to a standard SQL Server user ID
Mapping a Windows account to a standard SQL Server
user ID
If you use standard SQL Server security, the NetBackup Client Service must be
started with a Windows account. That account needs to be associated with a
standard SQL Server user ID. The account also requires sufficient administrative
privileges to perform backups and restores.
Note: If your site does not place SQL Server security restrictions on the right to
back up databases, you do not need to perform this procedure.
To associate a Windows user ID with a SQL Server user ID
1 Log on to your Windows host with the Windows account that you intend to
use for the NetBackup Client Service.
2 Start the NetBackup Database Extension GUI and go to the DBMS login
parameters window. Type the standard SQL Server logon name and password
into the edit boxes provided. The values that you supply are encrypted in a
registry entry that is associated with the current Windows account.
3 To set the logon account for the NetBackup Client Service, follow the
instructions to authorize scheduled operations.
See “Authorizing scheduled operations (for sites with SQL Server security
restrictions)” on page 48.
About NetBackup for SQL performance factors
Your hardware environment and the settings in SQL Server, the NetBackup master
server, and NetBackup for SQL Server, can all influence the backup performance.
The availability of buffer space has a definite influence on how fast backups run.
NetBackup for SQL buffer space parameters
Use the following parameters to increase buffer space in SQL Server:
■ See “Setting MAXTRANSFERSIZE for NetBackup for SQL Server” on page 50.
■ See “Setting BLOCKSIZE for NetBackup for SQL Server” on page 50.
■ See “Setting NUMBER OF BUFFERS PER STRIPE for NetBackup for SQL Server”
on page 50.
50 Configuring NetBackup for SQL Server
About NetBackup for SQL performance factors
Setting MAXTRANSFERSIZE for NetBackup for SQL Server
The MAXTRANSFERSIZE parameter can be set for each backup or restore
operation. It ranges in size from 64 kilobytes to 4 MB. The default is 64 kilobytes.
MAXTRANSFERSIZE is the buffer size used by SQL Server for reading and writing
backup images. Generally, you can get better SQL Server performance by using a
larger value.
Set the MAXTRANSFERSIZE for your NetBackup for SQL Server session by
choosing File > Set NetBackup client properties or add this parameter to batch
file.
See “About using batch files” on page 103.
Setting BLOCKSIZE for NetBackup for SQL Server
The BLOCKSIZE parameter can be set for each backup operation. For restores,
NetBackup for SQL Server automatically chooses the same size that you backed
up with. BLOCKSIZE ranges from 512 bytes to 64 kilobytes. The default is 512
bytes. BLOCKSIZE is the incremental size that SQL Server uses for reading and
writing backup images.
Set the BLOCKSIZE for your NetBackup for SQL Server session using the NetBackup
client properties dialog box. Otherwise, you can use the BLOCKSIZE parameter
to set this value when you create a batch file manually.
See “About using batch files” on page 103.
Setting NUMBER OF BUFFERS PER STRIPE for NetBackup for
SQL Server
The NUMBER OF BUFFERS PER STRIPE parameter affects buffer space availability.
This parameter is set in the NetBackup MS SQL Client interface for each operation.
NetBackup uses this parameter to decide how many buffers to allocate for reading
or writing each data stream during a backup or restore operation. By allocating
a greater number of buffers, you can affect how quickly NetBackup can send data
to the NetBackup master server.
The default value for the NUMBER OF BUFFERS PER STRIPE is 2, which allows
double buffering. You may get slightly better performance by increasing this value
to a higher value.
In the Microsoft SQL Client interface, you can change this value through the
NetBackup client properties dialog box. (File > Set NetBackup client properties).
Otherwise, you can use the NUMBUFS parameter to set this value when you create
a batch file manually.
See “About using batch files” on page 103.
Configuring NetBackup for SQL Server 51
About NetBackup for SQL performance factors
About backup stripes
You can divide the backup operation into multiple concurrent streams. This
configuration applies in any situation in which SQL Server dumps data faster
than your tape drive is capable of writing. NetBackup does supports multiplexing
many stripes to the same drive. Though you generally realize performance gains
only if you use the same number of tape drives as streams.
To divide a backup operation into multiple stripes, modify the Stripes value. This
value is in the Backup Microsoft SQL Server objects dialog box of the Microsoft
SQL Client interface.
About shared memory usage
If NetBackup server is installed on the same host as NetBackup for SQL Server,
then you see optimal performance. The optimal performance is seen if you use
shared memory for data transfer instead of sockets. Shared memory is the default
for this configuration and is used unless you create a
install_path\NetBackup\NOSHM file.
About alternate buffer method
NetBackup for SQL Server supports an alternate buffer method. It optimizes CPU
usage by allowing NetBackup and SQL Server to share the same memory buffers
without transferring data between them.
The alternate buffer method for backup and restore typically does not improve
data transfer rate, only CPU utilization. A situation may occur in which the transfer
rate is significantly degraded when alternate buffer method is in use. To improve
the transfer rate set the MAXTRANSFERSIZE for the backup to the maximum
allowed, which is 4 megabytes. Set this parameter by choosing File > Set
NetBackup client properties or add this parameter to a batch file.
See “About using batch files” on page 103.
About alternate buffer method with backup operations
This method is chosen automatically for backups if all of the following conditions
apply:
■ NetBackup shared memory is in use.
■ The backup is non-multiplexed.
■ The backup policy does not specify either compression or encryption.
■ The NetBackup buffer size equals the SQL Server blocksize.
52 Configuring NetBackup for SQL Server
About NetBackup for SQL performance factors
The default NetBackup buffer size is 64 kilobytes, but this value can be
overridden in the following settings:
install_path\Netbackup\db\config\SIZE_DATA_BUFFERS (for tape backups),
or,
install_path\Netbackup\db\config\SIZE_DATA_BUFFERS_DISK (for disk
backups)
You can set the SQL Server blocksize either in the NetBackup client properties
dialog box of the NetBackup Database Extension GUI. Or you can adjust it
directly using the BLOCKSIZE option in your batch file.
■ NetBackup for SQL Server must be started with the same account as the
NetBackup Client Service.
The backups that are initiated from an automatic backup policy are started
with the NetBackup Client Service so the same account is already in use.
However, you can start an SQL Server backup through NetBackup for SQL
Server or through dbbackex. Then your logon account must be the same as
the NetBackup Client Service account. Then your backups can be candidates
for the alternate buffer method.
About alternate buffer method with restore operations
Conditions for backups require that you use the alternate buffer method. Restores
also require that backups have been made with the alternate buffer method. You
can verify that the alternate buffer method was used. Look for the words Using
alternate buffer method, which appear in the dbclient log and the progress
report.
About page verification
With SQL Server 2005 or later, you can choose to perform page verification before
you perform a backup or restore. When this option is enabled, it imposes a
performance penalty on a backup or restore operation.
About instant data file initialization
When you restore a database, filegroup, or database file, SQL Server zeroes the
file space before it begins the restore operation. This action can slow the total
recovery time by as much as a factor of 2. To eliminate file initialization, run the
MSSQLSERVER service under a Windows account that has been assigned the
SE_MANAGE_VOLUME_NAME. For more information about eliminating the file
initialization, see the SQL Server and Windows 2003 documentation.
Configuring NetBackup for SQL Server 53
Backing up SQL Server in an environment with log shipping
Using read-write and read-only filegroups
You can significantly reduce backup time and the storage media that is needed if
you periodically back up only read-write filegroups. Then keep a single backup of
read-only filegroups, which is retained infinitely. You can set the retention level
in the schedule.
Backing up SQL Server in an environment with log
shipping
Log shipping is a SQL Server feature that may be employed to enhance the overall
availability of your installation. It uses a primary server, which contains the active
database, a monitor, and one or more secondary servers. Under log shipping,
copies of the transaction log are supplied to the secondary servers on an on-going
basis to the secondary servers. This configuration allows each secondary server
to be in a standby state in case the primary goes offline.
Many sites also use the secondary server to off-load certain activities from the
primary to minimize its load. However, a backup must NOT be performed on a
secondary (or standby) server. Databases must always be backed up on the primary
server and restored on the primary server. This requirement is based on the
Microsoft SQL Server restriction that is outlined in Microsoft knowledge base
article 311115. If you try to perform a backup on the secondary server, the result
cannot be predicted. However, you may see a message in the dbclient log similar
to the following:
16:33:26 [1208,2348] <16> CODBCaccess::LogODBCerr: DBMS MSG - ODBC message. ODBC return
code <-1>, SQL State <37000>, Message Text <[Microsoft][ODBC SQL Server Driver][SQL
Server]Database 'Mumbo' is in warm-standby state (set by executing RESTORE WITH
STANDBY) and cannot be backed up until the entire load sequence is completed.>
54 Configuring NetBackup for SQL Server
Backing up SQL Server in an environment with log shipping
Chapter 4
Backup and recovery
concepts
This chapter includes the following topics:
■ Overview of backup and recovery concepts
■ Overview of backup and recovery concepts
■ Protecting files and filegroups
■ About recovery considerations for files and filegroups
■ Reducing backup size and time by using read-only filegroups (SQL Server 2005
or later)
■ About recovery factors for SQL Server
Overview of backup and recovery concepts
These topics describe the concepts of backup and recovery of SQL Server and how
NetBackup for SQL Server performs these operations.
Overview of backup and recovery concepts
A SQL Server instance is created on a Windows host by installing Microsoft SQL
Server. You can install multiple instances on a single host including a default
instance and multiple named instances. You can select the database instance that
you want to browse and back up. Select File > Set SQL Server connection
properties.
See “About differential backups” on page 58.
56 Backup and recovery concepts
Overview of backup and recovery concepts
About SQL Server system database types
A SQL Server instance cannot be backed up as a single entity. The largest
granularity of a SQL Server backup is the database. SQL Server has system and
user databases.
The system databases are as follows:
Master This database is the "brains" of your installation. It contains a great
deal of the metadata that describes your instance. Be sure to retain
an up-to-date backup of the master database. Back up the master any
time you have made changes to your SQL Server installation, including
when you have created or modified other databases. Note that you
can only do full database backups on the master. You cannot back up
its component files, perform differentials, or backup up its transaction
log. Recovery of the master database requires special considerations.
See “About disaster recovery” on page 131.
msdb The Microsoft SQL Agent uses the msdb for schedules, alerts, and for
recording the backup history. All types of backups can be performed
on it, providing that it has the full or bulk-load recovery option set.
Model The model database serves as a template for new databases when the
Create Database statement is executed. All types of backups can be
performed on it, providing that it has the full or bulk-load recovery
option set.
tempdb The tempdb is for the temporary databases that applications use. It
cannot be backed up and does not appear in the NetBackup for SQL
Server backup browser.
About database backups
The following types of backup operations can be performed on databases:
Full The database, including all of its component files are backed up as a
single image. The log file is included in a full database backup.
Note: The transaction log is not automatically truncated following a
full backup. Thus a common practice to preserve disk space is to
manually truncate the transaction log following a successful full
backup.
Differential All of the changes since the last full are backed up to a single image.
Backup and recovery concepts 57
Overview of backup and recovery concepts
Transaction log Transaction log backups are only available for the full and bulk-load
recovery options. In this operation, the inactive portion of the
transaction log is backed up. Four options are available when you
select transaction log backup:
■ Back up and truncate transaction log.
■ Back up transaction log, but do not truncate it.
■ Truncate the transaction log, but do not back it up. (SQL Server
2000 only)
■ Back up and restore tail log. (SQL Server 2005 only)
The third option is not a backup at all. It only provides a convenient
way to truncate the transaction log for you. It is typically used after
a full database backup. The last option is a backup but it does not
create a permanent backup image. A typical use of this type of
transaction log backup is: after a page-level restore when the database
is recovered, but all of the filegroups have not been brought back on
line. By backing up and recovering the tail end of the log, SQL Server
is able to bring the database back to a usable state.
About filegroup backups
In addition to database backups, you can use separate images to back up the logical
filegroups and files that comprise databases.
Filegroups can be backed up in the following ways:
Filegroup Backups A backup can be created from a single filegroup. Scripts for filegroup
backups are created when you select individual filegroups in the object
browser of the backup database dialog box.
Read-Write A backup that contains only the read-write filegroups in a database.
Filegroups If all of the filegroups in a database are set to read-write, then the
Backups (SQL read-write filegroup backup has the same content as a full database
Server 2005 only) backup. You can create scripts for read-write filegroup backups when
you select individual databases and select the "read/write filegroups"
type of backup.
58 Backup and recovery concepts
Protecting files and filegroups
Partial Database A database backup contains an improvised selection of filegroups that
Backup (SQL can be selected by the user. You can create a template for partial
Server 2005 only) database backups when you select individual databases and select the
"Create a partial database template" type of backup. The template is
created with all of the filegroups commented-out. You can choose the
filegroups to include in the partial backup by removing the comments
from the filegroups.
Caution: Since the contents of a partial database backup are defined
by the user, NetBackup for SQL Server does not use them for staging
recovered backups. So if you rely on NetBackup to stage database
recovery for you, the partial backup may not be a good choice.
Backup of all a You effectively back up a filegroup when you back up all of the
filegroup’s database files in the filegroup.
database files
About differential backups
Differential backups can be created on the full database as well as all three types
of filegroup backup units as described in the previous section. A differential backup
contains the changes to the contents of the object since the last time that it was
captured in a full backup.
Note: Note that SQL Server does not let you create a differential on a database
file.
Protecting files and filegroups
If your plan to protect SQL Server includes backups of files and filegroups, then
the database must use the full or bulk-logged recovery model. In addition, you
must maintain the unbroken sequence of transaction log backups. You must create
the files and filegroups for your databases and place individual database
components into them. NetBackup places a restriction on the layout of your
database so it can successfully perform backups and restores of database files and
filegroups.
Before you try a file backup or filegroup backup, you must ensure that no table is
placed into a filegroup. That filegroup is different than any one of its indices.
For example, the layout as indicated by the following Transact SQL statements
should not be used:
Backup and recovery concepts 59
About recovery considerations for files and filegroups
use master
CREATE DATABASE MultiFileDB
ON
PRIMARY ( NAME = FileX,
FILENAME = 'd:\mssql\data\FileX.mdf'),
FILEGROUP AltGroup
( NAME = AltGroupFil,
FILENAME = 'd:\mssql\data\AltGroupFil.ndf')
GO
use MultiFileDB
CREATE TABLE Table1 (col1 char(10),col2 char(10), col3 char(10)) on AltGroup
go
create unique clustered index index4 on Table1 (col2)
go
Notice in this example, Table1 has been placed in filegroup AltGroup but its index
is placed (by default) in the primary filegroup.
If you do place a table into a different filegroup than one of its indices and use
NetBackup for SQL Server to back it up, you may fail. The following SQL Server
error message is displayed:
Database file <file name> is subject to logical recovery and
must be among the files to be backed up as part of the file
or filegroup backup.
About recovery considerations for files and filegroups
Always have backups of a full set of files and filegroups that constitute the entire
database. You also need transaction log backups that span the entire period of
time over which the backups were taken. When you have both types of backups,
it ensures that you can successfully restore a database from file and filegroup
backups. To maintain an unbroken sequence of transaction log backups, it is
essential to perform a transaction log backup following every file backup or
filegroup backup. If you back up several files or filegroups immediately, then you
only need to back up the transaction log after the last such backup. If the
transaction log is not backed up, SQL Server does not let you restore a file or
filegroup.
SQL Server does not keep a record in the transaction log of new files or filegroups
that are created. Therefore, after you add either a file or a filegroup to the database,
you must immediately back it up. Then perform a backup of all the filegroups in
the database so NetBackup selects the correct recovery set when subsequent
60 Backup and recovery concepts
Reducing backup size and time by using read-only filegroups (SQL Server 2005 or later)
backups are performed. Similarly, after you create a database file, you should
back up all of the files in the filegroup to which it belongs.
Reducing backup size and time by using read-only
filegroups (SQL Server 2005 or later)
Many applications contain a substantial amount of data that does not change. For
example, under time-based partitioning, historical data may be preserved
indefinitely. Only a fraction of the database is subject to change at any given time.
With SQL Server 2005 or later, static filegroups can be classified as read-only.
NetBackup uses the filegroup read-only designation to optimize the total backup
volume speed of recovery.
For backups, the advantage in using read-only filegroups is that you can reduce
total media usage. You back up the read-only filegroups one time and retain the
backup image indefinitely. This strategy reduces the total time you spend on
backup operations because only read-write data is backed up periodically.
For recovery, the advantage is that you can bring your database on-line more
quickly. Read-only filegroups do not need to be restored from backup media unless
they are corrupted due to disk error or other hardware failure.
To reduce backup size and time by with read-only filegroups
1 Create a policy for read-only filegroups.
2 Create an Application Backup schedule with Retention set to infinite.
3 Create a policy for read-write filegroups.
4 Create an Automatic Backup schedule with the wanted Retention period and
add the read-write filegroups batch file to the Backup Selections list.
5 Back up read-only filegroups.
6 In the NetBackup for SQL Server interface, select the policy you created in
step 1 and perform this backup one time.
7 All read-only filegroups must be included in some combination of full, partial,
or individual filegroup and file backups.
Note: Immediately back up any filegroup when you change it from read-write
to read-only.
8 Back up read-write filegroups.
Backup and recovery concepts 61
About recovery factors for SQL Server
9 This backup is done automatically through the schedule you created in step
3.
10 Or you can manually back up the read-write filegroups.
11 View the read-only backup set.
If necessary, confirm all read-only groups are backed up.
About recovery factors for SQL Server
There are several factors that you should take into consideration when you define
a recovery plan for your application environment. Much of this information is
based on Microsoft’s SQL Server Books Online. See that resource for a more
inclusive discussion.
■ See “About SQL Server transaction logs” on page 61.
■ See “About recovery strategies” on page 62.
■ See “Backing up the transaction log” on page 62.
■ See “About differential backups” on page 63.
■ See “About file and filegroup backups” on page 63.
■ See “About database recovery” on page 64.
■ See “About staging recovery” on page 64.
About SQL Server transaction logs
SQL Server maintains a write-ahead transaction log for each database. This log
helps to maintain database updates in cache memory to ensure that data is not
written to disk before it has been committed. Database writes occur as a part of
the checkpoint procedure.
SQL Server determines the checkpoint frequency based upon the "recovery
interval." This interval is a configuration parameter that indicates the maximum
time interval that can be tolerated during a system restart. When checkpoint
occurs the portion of the transaction log that is no longer needed for system
restart becomes inactive and is optionally truncated. The recovery strategy
determines whether the transaction log is truncated or not.
See “About recovery strategies” on page 62.
If the checkpoint procedure does not truncate the transaction log, then it can be
backed up. Then it can be used for point-in-time recovery, failure from disk crash,
or move and copy operations.
62 Backup and recovery concepts
About recovery factors for SQL Server
About recovery strategies
SQL Server provides the following levels for database recovery. Each level has
different implications for both backup performance and for the granularity of
recovery.
These levels are as follows:
Simple With this method the inactive portion of the transaction log cannot
be retained beyond the database checkpoint. This method provides
for minimal usage of log space. However, the database can only be
restored to the last full backup. Transaction log restores, including
point in time recovery and named transaction recovery are not
supported. In addition, maximum performance is provided for bulk
operations, such as (Create Index, Select Into, and Bulk Copy) because
they are not logged.
Full With this method, the inactive portion of the transaction log is retained
until it is truncated, which normally occurs when it is backed up. The
transaction log can then be used to stage a recovery either to a point
in time or to a named transaction. The Full Recovery model provides
maximum recoverability but it uses the most log space and does not
provide maximum performance for bulk operations.
Bulk-Logged This method is identical to the Full Recovery model except that bulk
operations are not logged and thus cannot be recovered.
Backing up the transaction log
Transaction logs can be backed up in SQL Server 2000 or later versions. Also,
databases must be set in either full or bulk-logged mode. By default a transaction
log is truncated after it has been backed up. However, it is not truncated following
a full database or differential backup.
The main factors in deciding how frequently to back up a transaction log would
be the following:
■ Conservation of log space.
■ How close to the failure point you must be able to recover in case of a disk
crash.
During peak periods in a high transaction environment, it may not be unusual to
back up the transaction log on an hourly basis.
Backup and recovery concepts 63
About recovery factors for SQL Server
About differential backups
Unlike the transaction log backup, the differential backup is a backup of the
database. The differential includes all of the changes that were made since the
last full backup. If you made several differential backups since the last full backup,
you only need to restore the last full database, followed by the last differential.
You would not need to restore any of the intermediate differentials.
Differential backups include the following types of backups:
■ Databases differential
■ Individual filegroup differentials
■ Read-write filegroup differentials, i.e., any backups that include differentials
on all the read-write filegroups in a database.
■ Partial differential filegroups, i.e., any backups that include differentials of
only the filegroups that the user selects.
Caution: Microsoft recommends that you do not create more than one type of
differential backup for the same object.
Caution: NetBackup does not consider differential images when it determines
recovery staging strategies if more than one type of differential is found for the
same object.
See “About staging recovery” on page 64.
A typical backup procedure may use full database, differential, and transaction
log backups in ascending order of frequency. For example the full database backup
may be taken bi-weekly and the differential may be taken nightly. Then the
transaction log backup may be made as frequently as every hour for either mission
critical or high volume applications.
About file and filegroup backups
SQL Server also supports the backup of up individual filegroups and files as distinct
images. A filegroup is composed of one or more database files. A backup of the
constituent files of a filegroup is logically equivalent to a backup of the filegroup
itself.
Filegroup and file backups would commonly be used in a tightly architected
application in which physical disk locations were mapped to logical objects. For
example, tables and indexes.
64 Backup and recovery concepts
About recovery factors for SQL Server
The following factors may lead you to use file and filegroup backups in this type
of environment:
■ Some portions of the database, which may be volatile or mission critical, should
be backed up more frequently than other portions.
■ The database may be so large that the time that is required for a full database
backup cannot fit in the allocated time window. Thus it may be more viable to
do a full backup of one or more files or filegroups on a rotating basis.
■ You may want to optimize on backup volume and recovery speed by placing
some of your data into read-only filegroups.
See “Reducing backup size and time by using read-only filegroups (SQL Server
2005 or later)” on page 60.
In the event of disk failure, you can choose to recover only the failed unit from a
filegroup backup or file backup. You do not have to restore the entire database.
To use filegroup and file backups you must maintain backups of the transaction
log.
For example, to perform a full database restore using filegroups and files, you are
required to restore the following:
■ All of the constituent filegroups and files
■ All of the transaction log segments
These translation logs segments must start from the first component backup
until a point in time following the last component backup.
About database recovery
During the restore process, a database goes into "loading mode" until the restore
command is executed against the database using the "recovery" option. Before
you place the database into recovery mode all of the restore commands will have
been executed using the "Not recovered" option. This way it is possible to continue
to stage additional restore statements to bring the database up to the state you
want. The database becomes usable again after the last restore statement has
been applied the "Recovered" option.
You can choose the recovery option you want when you perform restores.
About staging recovery
NetBackup for SQL Server keeps track of the backups you have performed and
when you performed them. You can display the backup history by opening the
Restore Microsoft SQL Server Objects dialog box.
Backup and recovery concepts 65
About recovery factors for SQL Server
This dialog box depicts all of the SQL Server backup images within the parameters
that you specify. The images appear in a tree-form that is based on the following
backup types:
■ Database
■ Database differential
■ Filegroup
■ Filegroup differential
■ File
■ Read-only filegroups
■ Read-only filegroups differential
■ Partial database
■ Partial database differential
■ Transaction log
When you select a transaction log for restore, NetBackup for SQL Server
automatically searches for a set of images. The images are used to stage a full
database recovery. The recovery set consists of the selected transaction log image
plus additional images which can reconstitute the database to a recovered state.
A recovery set can include a full database image, filegroup and file images, and
differentials images of the database or filegroups. It can include filegroups
contained in partial and read-write images or filegroup differentials contained in
these images. If the recovery set contains filegroup, file, partial, or read-write
images, then it also contains one or more transaction log images.
A read-write filegroup is also a full recovery set but it only contains backups
(including differential backups) of those filegroups that are writable. Read-only
filegroup images are not required because they are assumed not to have changed.
A read-write filegroup recovery set also contains one or more transaction log
images.
If a full recovery set is found, then all of the composite images are checked. In
addition, Stage full recovery is enabled. To view the full recovery set, right-click
the transaction log, select Properties, and click the Recovery Set tab.
66 Backup and recovery concepts
About recovery factors for SQL Server
Chapter 5
Performing backups and
restores of SQL Server
This chapter includes the following topics:
■ Overview of using NetBackup for SQL Server
■ Using dbbackex to perform user-directed operations for SQL Server
■ Selecting the SQL host and instance
■ Using the NetBackup for SQL Client interface
■ Starting the NetBackup SQL Client for the first time
■ Backup Microsoft SQL Server Objects field descriptions
■ Backing up databases
■ Backing up transaction logs
■ Backing up database filegroups
■ Backing up read-only filegroups (SQL Server 2005 or later)
■ Backing up read-write filegroups (SQL Server 2005 or later)
■ Backing up database files
■ Performing partial backups (SQL Server 2005 or later)
■ Browsing for backup images
■ Restore Microsoft SQL Server Objects field descriptions
■ Restoring a database backup
68 Performing backups and restores of SQL Server
Overview of using NetBackup for SQL Server
■ Staging a full database recovery
■ Restoring filegroup backups
■ Recovering a database from read-write backups
■ Restoring read-only filegroups (SQL Server 2005 or later)
■ Restoring database files
■ Restoring a transaction log image without staging a full recovery
■ Performing a database move
■ Restoring partial databases (SQL Server 2000)
■ Performing page-level restores (SQL Server 2005 or later)
■ Redirecting a restore to a different host
■ Redirecting a database to a different location on a different host
■ Performing restores with a multi-NIC connection
■ Backing up and restoring the databases that contain full-text search catalogs
(SQL Server 2005)
■ About using batch files
■ About file checkpointing
■ About automatic retry of unsuccessful backups
■ Performing backups and restores of remote SQL Server installations
■ Restoring multistreamed backups
■ About using bplist to retrieve SQL Server backups
Overview of using NetBackup for SQL Server
You can use the NetBackup Administration Console to start SQL Server backups
or use the NetBackup for SQL Server interface to initiate backups and restores.
You can perform user-directed operations for SQL Server on a Windows client in
the following ways:
■ Using the command line interface program, dbbackex.
See “Using dbbackex to perform user-directed operations for SQL Server”
on page 69.
Performing backups and restores of SQL Server 69
Using dbbackex to perform user-directed operations for SQL Server
■ Using the NetBackup for SQL Server interface.
See “Using the NetBackup for SQL Client interface” on page 71.
Using dbbackex to perform user-directed operations
for SQL Server
Execute dbbackex from a Windows MS-DOS with the following command:
install_path\NetBackup\bin\dbbackex -f file [-p policy][-u userid][-pw password]
[-s server][-np]
Refer to the description of the following parameters.
file The name of the batch file, which describes the operations you want
to start.
See “Running batch files” on page 117.
policy The MS-SQL-Server policy type that is used by NetBackup for the
operations that are specified in the batch file.
This parameter is ignored for restore operations. The NetBackup
server can retrieve the dump file based entirely on the image names
that are specified in the batch file for each restore. The policy name
is used for databases backups. If it is omitted, then the NetBackup
server uses the first active SQL Server policy that it finds in its policy
list. This policy name is used for all of the backup operations that are
specified in the batch file.
userid is the SQL Server user ID for logging into the database management
system.
password is the SQL Server password for logging into the database management
system.
server is the name of the host for the NetBackup master server that you want
to back up to or restore from.
If this parameter is omitted, then the client uses the default server
according to the Windows NetBackup client configuration. See the
NetBackup Backup, Archive, and Restore Getting Started Guide for
more information.
70 Performing backups and restores of SQL Server
Using dbbackex to perform user-directed operations for SQL Server
-np tells dbbackex not to create a message box to indicate the operation
status when it has completed.
Otherwise, a message appears when dbbackex completes. That
message tells you how many operations in the batch file were
successful and how many failed.
Note: Any of the options can be delimited with double quotation marks. For
example, use delimiters if the file name contains spaces.
Note: To protect logon passwords for SQL Server, do not use the -u or -pw
parameters. By omitting these parameters, you can force NetBackup for SQL
Server to read the default SQL Server logon data from an encrypted file.
See “Starting the NetBackup SQL Client for the first time” on page 71.
Using client-based schedulers with dbbackex
dbbackex lets you employ your choice of client-based schedulers to automatically
initiate NetBackup for SQL Server operations.
The following schedulers are available:
■ The Windows Scheduler. This scheduler is initiated through the MS-DOS at
command. Instructions for using the Windows Scheduler are provided in the
Microsoft Windows online documentation.
■ The SQL Server Scheduler. This scheduler is closely integrated with SQL Server.
It can be accessed through the Microsoft SQL Server Enterprise Manager.
One distinct advantage of the SQL Server Scheduler is that you can create scripts
for database maintenance operations. These operations are initiated as a result
of database events that you define. For example, you can create a script that
initiates dbbackex and tells it to back up a particular transaction log. You can also
create an alert which invokes that script when the transaction log for this database
becomes full.
Note: If you use dbbackex through a client-based scheduler, specify the -np option
to ensure that a message box is not generated. However, before you use the
scheduler with dbbackex consider the following. Try the dbbackex syntax on the
console WITHOUT the -np option. Do this to test for the successful completion
of the batch file that you have created for your operation.
Performing backups and restores of SQL Server 71
Selecting the SQL host and instance
Selecting the SQL host and instance
Use this procedure to set which SQL Server host and the instance that you want
the NetBackup MS SQL Client to access. It accesses the SQL Server host when it
logs into SQL Server.
To select the SQL host and instance
1 From the Windows Start menu, select Programs > Symantec NetBackup >
NetBackup Agents > NetBackup MS SQL Client.
2 Select File > Set SQL Server connection properties.
3 In the SQL Server connection properties dialog box, from the Host drop-down
list, select the SQL Server host.
You can type a host name if it does not appear in the drop-down list. If you
select a remote host and click Apply, the Host type is shown as "remote".
4 From the Instance drop-down list, select the SQL Server instance.
You can type an instance name if it does not appear in the drop-down list.
You can designate the default instance either by setting the Instance box to
<default> or to empty (not spaces).
5 If the SQL Server host and instance use standard or mixed security, you also
need to set the user ID and password. These items are set in the user ID,
Password, and Reenter password boxes.
6 Click Apply to save your changes.
7 Click Close.
Using the NetBackup for SQL Client interface
You can use the NetBackup for SQL Client interface to perform NetBackup
operations on the SQL Server. You can browse for available backup images, back
up databases and transaction logs, and back up database filegroups.
Starting the NetBackup SQL Client for the first time
This topic describes how to start the NetBackup SQL Client for the first time. For
subsequent sessions, the agent remembers the logon parameters you provided
for that Windows account.
72 Performing backups and restores of SQL Server
Backup Microsoft SQL Server Objects field descriptions
To start the NetBackup SQL Client for the first time
1 From the Windows Start menu, select Programs > Symantec NetBackup >
NetBackup Agents > NetBackup MS SQL Client.
2 When you are prompted to provide the logon parameters, click OK.
3 In the SQL Server Connection Properties dialog box, select the SQL Server
host and instance that you want to log into.
4 Click Apply.
5 Click Close.
Backup Microsoft SQL Server Objects field
descriptions
Table 5-1 describes the options that are available when you perform backups.
These options appear in the Backup Microsoft SQL Server Objects dialog box
after you select File > Backup SQL Server objects.
Table 5-1 Options for backup operations
Option Description
Expand database This pane lets you traverse live databases. You can expand the SQL
Server instance to view its databases. Expand each database to view
its filegroups or expand a filegroup to view its files. You can select
any object in this pane to view its constituent objects in the right-hand
pane.
Select database(s) Select the objects that you want to back up from this pane. This pane
for backup from displays the list of constituent database objects of the selected host
instance and instance in the left-hand pane. You can select one or more objects
host\instance (databases) in this pane. (Use Ctrl + click and Shift + click to select
multiple objects.)
Type of Backup ■ Full
Create a full database backup.
■ Full differential
Create a differential backup.
■ Transaction log
Create a transaction log backup. This type of backup is only
available for databases. When you select this type of backup, you
then need to select a backup option from the Transaction log
backup options list.
Performing backups and restores of SQL Server 73
Backup Microsoft SQL Server Objects field descriptions
Table 5-1 Options for backup operations (continued)
Option Description
The following backup types are only available for SQL Server 2005 or
later:
■ Read/write filegroups
Create a backup of read or write filegroups in a database.
■ Differential on read/write filegroups
Create a differential backup of read or write filegroups in a
database.
■ Create a template for partial backup
Create a backup of only the selected filegroups in a database.
■ Create a template for partial differential backup
Create a differential backup of only the selected filegroups in a
database.
Transaction log The following options are available when you have chosen a
backup options transaction log backup type:
■ Back up and truncate transaction log
Back up the transaction log and remove the inactive part of the
transaction log.
■ Back up transaction log, but do not truncate it
Back up a transaction log without truncating it.
■ Truncate transaction log, but don't back it up
(SQL Server 2000) Transaction log is truncated, but is not backed
up.
■ Back up and restore tail log
(SQL Server 2005 or later) Backup and recover the tail log from
disk.
Backup script ■ Launch Immediately
Start the backup operation immediately.
Launch immediately is disabled if you are logged into a SQL Server
instance that is not on the local host. If you generate a script for
a non-local host, then it must be executed on that host.
■ Save
Generate a script that can be started at a later time.
74 Performing backups and restores of SQL Server
Backup Microsoft SQL Server Objects field descriptions
Table 5-1 Options for backup operations (continued)
Option Description
Back up In the right-hand pane, choose one of the following backup options:
■ Selected
Back up only the objects selected.
■ All but selected
Back up all of the objects, except those selected.
■ All
Back up all of the objects.
Resume options ■ Do not resume unsuccessful backups
for this selection
■ Retry from the beginning
Restart failed backups after waiting 60 seconds.
■ Save work and restart at point of failure
Divide the backup into multiple streams and back up separately.
Any streams that fail are restarted after 60 seconds.
This option is available when the following conditions are met:
■ Exactly one object has been selected,
■ The object that is selected for backup is a database or filegroup
and the backup type is full,
■ The SQL Server object uses the “full” or “bulk-logged” recovery
method.
NetBackup policy If this host is the NetBackup master server, then this list includes all
active policies of type MS-SQL-Server. You can select one of these
policies or type the name of a policy.
The default is <any>. If you select the default, then NetBackup selects
which MS-SQL-Server policy to use.
Page verification (SQL Server 2005 or later) This option is enabled for objects have a
page verification type that is either torn page detection or checksum.
All of the objects in the right-hand pane must have the proper
verification type.
This indicates a performance penalty when you use page verification.
■ Do not perform verification
Do not perform page verification before you run the backup.
Performing backups and restores of SQL Server 75
Backing up databases
Table 5-1 Options for backup operations (continued)
Option Description
■ Perform verification
Perform page verification when you run the backup and stop the
backup if a verification error is encountered.
Stripes Set the number of backup stripes that you want SQL Server to create
for your backup. Type a number from 1 to 32.
See “Configuring multistream operations” on page 42.
Backup Start a database backup or generate a database backup script. This
option is enabled only when you select an object to back up.
Backing up databases
This procedure describes how to perform a database backup.
To back up a database
1 Change the host and instance you want to access.
See “Selecting the SQL host and instance” on page 71.
2 Select File > Backup SQL Server objects.
3 In the Backup Microsoft SQL Server Objects dialog box, in the left pane,
select the database instance.
4 In the right pane, select one or more databases that you want to back up.
5 Select the Type of Backup.
Select one of the following:
■ To perform a full backup, select Full backup.
■ To back up the database with the differential option, select Perform
differential backup.
6 Select the backup options.
See “Backup Microsoft SQL Server Objects field descriptions” on page 72.
7 Click Backup.
8 When you are prompted to start the backup, click Yes.
9 To view the progress of the backup, select File > View status.
76 Performing backups and restores of SQL Server
Backing up transaction logs
Backing up transaction logs
This procedure describes how to perform a transaction log backup.
To back up a transaction log
1 Set the Recovery Model setting to either Full or Bulk-logged.
Caution: You must ensure that the entire sequence of transaction logs
generated following any database backup are maintained on the same
NetBackup server. All transaction logs same facility and that none should be
allowed to expire before the others. NetBackup for SQL Server requires that
you follow these guidelines when you plan your backup strategy. Then you
have maximum assistance when you restore your databases.
2 Change the host and instance you want to access.
See “Selecting the SQL host and instance” on page 71.
3 Select File > Backup SQL Server objects.
4 In the Backup Microsoft SQL Server Objects dialog box, in the left pane,
select the database instance.
5 In the right pane, select one or more databases whose transaction logs you
want to back up.
6 In the Type of Backup list, select transaction log.
7 From the drop-down list, select the transaction log option. For more
information, see the following table.
Back up and truncate Back up the transaction log and remove the inactive
transaction log part of the transaction log.
Back up transaction log but (SQL Server 2000) Backup a transaction log without
don’t truncate it truncating it.
Truncate transaction log, but (SQL Server 2005 or later) Truncate the log without
don't back it up performing a backup.
Backup and restore tail log Back up and recover the tail log from disk.
8 Select the backup options.
9 Click Backup.
To view the progress of the backup, select File > View status.
Performing backups and restores of SQL Server 77
Backing up database filegroups
Backing up database filegroups
More information is available on how to use read-write and read-only filegroups
in your backup strategy.
See “Backing up read-write filegroups (SQL Server 2005 or later)” on page 78.
See “Backing up read-only filegroups (SQL Server 2005 or later)” on page 77.
To back up a database filegroup
1 Change the host and instance you want to access.
See “Selecting the SQL host and instance” on page 71.
2 Select File > Backup SQL Server objects.
3 In the Backup Microsoft SQL Server Objects dialog box, in the left pane,
expand the instance name.
4 Select a database whose filegroups you want to back up.
5 In the right pane, select one or more filegroups that you want to back up.
6 Select the backup options.
7 Click Backup.
To view the progress of the backup, select File > View status.
Backing up read-only filegroups (SQL Server 2005 or
later)
To back up read-only filegroups you must first create a separate policy for this
type of backup. You can also verify that all read-only filegroups are backed up.
See “Reducing backup size and time by using read-only filegroups (SQL Server
2005 or later)” on page 60.
See “Viewing read-only backup sets” on page 77.
Viewing read-only backup sets
If you perform periodic backups only on read-write filegroups, you can verify if
you have retained backups of the read-only filegroups.
78 Performing backups and restores of SQL Server
Backing up read-write filegroups (SQL Server 2005 or later)
To view read-only backup sets
1 Browse for the backup images that contain the read-only backup sets.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, in the left pane,
expand the instance name.
3 In the right pane, right-click the database and select Properties.
4 Click the "Read-only backup set" tab.
If the database does not contain read-only filegroups, then the message "This
database does not contain any read-only filegroups." is shown. If backups do
not exist for all of the read-only filegroups, then a list of the filegroups that
were not backed up is shown. Finally, if a backup of all of the read-only
filegroups is found, then the name appears of the latest image that contains
this backup.
5 If there are any read-only filegroups that are not backed up, back them up as
soon as possible. Do this to ensure you can perform a full recovery.
6 Click OK.
Backing up read-write filegroups (SQL Server 2005
or later)
This procedure describes how to back up read-write filegroups. This type of backup
can only be performed with SQL Server 2005 or later.
Note: Backups of read-write filegroups are not supported if you have SQL Express.
SQL Server Express Edition has limited support for recovering from this backup
type.
To back up read-write filegroups
1 Change the host and instance you want to access.
See “Selecting the SQL host and instance” on page 71.
2 Select File > Backup SQL Server objects.
3 In the Backup Microsoft SQL Server Objects dialog box, in the left pane,
select the database instance.
4 In the right pane, select one or more databases that you want to back up.
5 Select the Type of Backup.
Performing backups and restores of SQL Server 79
Backing up database files
Select one of the following:
■ To perform a full backup of the read-write filegroups, select Read/write
filegroups.
■ To perform a differential backup of the read-write filegroups, select
Differential on read/write filegroups.
6 Select the backup options.
7 Click Backup.
To view the progress of the backup, select File > View status.
Backing up database files
This procedure describes how to back up database files.
To back up a database file
1 Change the host and instance you want to access.
See “Selecting the SQL host and instance” on page 71.
2 Select File > Backup SQL Server objects.
3 In the Backup Microsoft SQL Server Objects dialog box, in the left pane,
expand the instance name and database.
4 In the left pane, select the filegroup that contains the files you want to back
up.
5 In the right pane, select one or more files that you want to back up.
6 Select the backup options.
7 Click Backup.
To view the progress of the backup, select File > View status.
Performing partial backups (SQL Server 2005 or later)
This procedure describes how to perform a partial backup. This type of backup
can only be performed with SQL Server 2005 or later.
Note: Partial backups are not supported if you have SQL Express. SQL Server
Express Edition has limited support for recovering from this backup type.
80 Performing backups and restores of SQL Server
Performing partial backups (SQL Server 2005 or later)
To perform a partial backup
1 Change the host and instance you want to access.
See “Selecting the SQL host and instance” on page 71.
2 Select File > Backup SQL Server objects.
3 In the Backup Microsoft SQL Server Objects dialog box, in the left pane,
select the database instance.
4 In the right pane, select a database that you want to back up.
5 Select the Type of Backup.
Select one of the following:
■ To perform a full backup, select Create a template for partial backup.
■ To perform a differential backup of the read-write filegroups, select Create
a template for partial differential backup.
6 Select the backup options.
7 Click Backup.
8 In the Save Script As dialog box, specify a file name and click OK.
9 When you are prompted to open the template, click Yes.
10 Edit the template by uncommenting the filegroups that you want to include
in the backup. You must uncomment at least one filegroup.
For example, replace:
#
# If you wish to include filegroup DBA_FG1 in the partial backup,
# then remove the hash mark that precedes the following line.
#FILEGROUP DBA_FG1
with:
#
# If you wish to include filegroup DBA_FG1 in the partial backup,
# then remove the hash mark that precedes the following line.
FILEGROUP DBA_FG1
11 When you are finished modifying the template, save it.
12 To run the backup, select File > Manage script files, select the script you
created, and click Start.
Performing backups and restores of SQL Server 81
Browsing for backup images
Browsing for backup images
This procedure describes how to browse for available backup images. When you
have displayed the backup images you want, then follow the instructions for
restoring a specific SQL object.
To browse for backup images
1 Change the host and instance you want to access.
See “Selecting the SQL host and instance” on page 71.
2 Select File > Restore SQL Server objects.
3 In the Backup History Options dialog box, select the SQL Host whose backup
images you want to browse, or type its name.
Also indicate the Source Client, if applicable. See the following descriptions.
SQL Host From this list, select the SQL Server host that you want to
generate a backup history for.
Source Client Specifies the NetBackup client name for the selected client. This
name is the one used in the policy to identify the client.
If you use a specific network interface for backup, then the
network interface name should be entered in the Source Client
box. The network interface name is defined in the Host Properties
for the server (open the properties for the server and click
Universal Settings).
4 Select the date range to search.
82 Performing backups and restores of SQL Server
Restore Microsoft SQL Server Objects field descriptions
5 Click OK.
6 Continue with the applicable instructions for how to restore the object(s).
See “Restoring a database backup” on page 86.
See “Staging a full database recovery” on page 87.
See “Restoring filegroup backups” on page 88.
See “Recovering a database from read-write backups” on page 89.
See “Restoring read-only filegroups (SQL Server 2005 or later)” on page 90.
See “Restoring database files” on page 90.
See “Restoring a transaction log image without staging a full recovery”
on page 91.
See “Performing a database move” on page 91.
See “Restoring partial databases (SQL Server 2000)” on page 94.
See “Performing page-level restores (SQL Server 2005 or later)” on page 96.
Restore Microsoft SQL Server Objects field
descriptions
Table 5-2 describes the options that are available when you perform restores.
Table 5-2 Options for restore operations
Option Description
Restore options ■ Use replace option
Restore with the SQL Server replace option.
■ Scripting
These scripting options are available for restoring from a
database image.
■ Restore selected object
Produce a script that performs a database restore. This script
is the default option.
■ Create a move template
Create a script template for moving the selected database.
Performing backups and restores of SQL Server 83
Restore Microsoft SQL Server Objects field descriptions
Table 5-2 Options for restore operations (continued)
Option Description
■ Create partial database restore template
(SQL Server 2000 only) Create a template for a partial database
restore.
■ Restore read-only filegroups
(SQL Server 2005 or later) Restore the most recent backup of
every read-only filegroup.
■ Create a page restore template
(SQL Server 2005 or later) Create a template for restoring a
database, filegroup, or file from the pages that are contained
in the selected backup image.
■ (Verify backup image, but don't restore
SQL Server 2005 or later) This option is only available if the
image was backed up with the page verification option.
NetBackup processes the image for errors, but does not perform
a restore.
■ Recovery
Specify one of the SQL Server recovery options.
■ Not recovered
Use this option during a restore if additional backup images
must be applied to the database following the current restore.
When you use this option, the database is left in a loading state.
■ Recovered
Select this option when restoring the last image in a restore
sequence. After the recovery operation, the database is ready
for use. If recover database is not performed, the database is
in an intermediate state, and is not usable. If Recovered is
selected when an intermediate backup is applied, you cannot
continue to restore backups; you must restart the restore
operation from the beginning.
■ Standby
Create and maintain a standby during a transaction log and
database restore. This option requires a standby undo log,
which by default is placed in
install_path\NetBackup\logs\SQLStandBy\. The
database is placed in "standby" state following the restore.
84 Performing backups and restores of SQL Server
Restore Microsoft SQL Server Objects field descriptions
Table 5-2 Options for restore operations (continued)
Option Description
■ Consistency Check
Select the consistency check to be performed after the restore.
Output from the consistency check is written to the SQL Server
client progress log. You cannot select consistency checking
unless the database is restored to the recovered state. If you
select consistency checking for a staged recovery, then the
check occurs following the last restore.
■ None
Do not perform consistency checking.
■ Full check, excluding indexes
Exclude indexes from the consistency check. If indexes are not
checked, the consistency check runs significantly faster but
is not as thorough. Only the data pages and clustered index
pages for each user table are included in the consistency check.
The consistency of the non-clustered index pages is not
checked.
■ Full check, including indexes
Include indexes in the consistency check. Any errors are logged.
■ Check catalog
Check for consistency in and between system tables in the
specified database.
■ Physical check only
(SQL Server 2000 or later) Select this item to perform a low
overhead check of the physical consistency of the SQL Server
2000 or later database. This option only checks the integrity
of the physical structure of the page and record headers. It
also checks the consistency between the pages’ object ID and
index ID and the allocation structures.
■ Page verification
(SQL Server 2005 or later) These options are available if the
source object was backed up with torn page detection or
checksum verification.
A performance penalty can happen when using page
verification.
■ Do not perform verification
Do not include page verification in the restore script.
Performing backups and restores of SQL Server 85
Restore Microsoft SQL Server Objects field descriptions
Table 5-2 Options for restore operations (continued)
Option Description
■ Perform verification
Include page verification in the restore script and stop the
restore if an error is encountered.
Transaction logs This group contains transaction log recovery options. It is enabled
if you select a transaction log backup image.
■ Stage full recovery
Select this option to recover the database by using the recovery
set that NetBackup found. If the transaction log that you select
does not belong to a recovery set, this option is disabled.
■ Restore selected transaction log
Select this option to restore only the selected transaction log.
If the transaction log that you select does not belong to a
recovery set, this option is disabled.
■ Transaction log recovery options
This list contains the controls for you to restore a transaction
log. You can restore the log to a point in time that precedes
the time when the transaction log was dumped. The individual
entries in this group are only enabled if you selected a
transaction log backup.
■ To point in time
Select this option to have the transaction log recovered to a
point in time.
■ To transaction log mark
Select this option to have the transaction log recovered to a
transaction log mark. With this option, you must enter a
transaction log mark name.
■ To transaction log mark but after
Select this option to have the transaction log recovered to a
transaction log mark but after a point in time. With this option,
you must enter a transaction log mark name.
■ Before transaction log mark
Select this option to have the transaction log recovered to a
point before the occurrence of a transaction log mark. With
this option, you must enter a transaction log mark name.
86 Performing backups and restores of SQL Server
Restoring a database backup
Table 5-2 Options for restore operations (continued)
Option Description
■ Before transaction log mark but after
Select this option recover the transaction log to a point before
the occurrence of a transaction log mark but after a point in
time. With this option, you must enter a transaction log mark
name.
■ MM, YYYY, DD, HH, MM, SS am, pm
Specify the time to which you want the transaction logs
restored. These fields are only enabled if you selected one of
the following transaction log recovery options:
■ To point in time
■ To transaction log mark but after
■ Before transaction log mark but after
■ Transaction log mark
This list is enabled if you selected a database transaction log
for restore. The transaction log contains one or more
transaction log marks, and you selected one of the following
transaction log recovery options:
■ To transaction log mark
■ To transaction log mark but after
■ Before transaction log mark
■ Before transaction log mark but after
Restore script ■ Launch immediately
Start the restore operation immediately.
Launch immediately is disabled if you are logged into a SQL
Server instance that is not on the local host. If you generate a
script for a non-local host, it must be executed on that host.
■ Save
Generate a script that can be started at a later time.
Restore Start the restore or generate a restore script. This is disabled if
you have not selected any objects to restore.
Restoring a database backup
This topic describes how to restore a database from a full database or differential
database backup.
Performing backups and restores of SQL Server 87
Staging a full database recovery
To restore a database backup
1 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance.
3 Expand the database.
4 Select the database image that you want to restore, as follows:
■ To restore a full backup, select the image of the database backup.
■ To restore a full backup and a differential database backup, click the "+"
and select a differential backup.
The full backup is automatically selected when you select a differential.
5 Select the restore options.
To place the database in recovery mode so that it is immediately usable
following the restore, select Recovered from the Recovery list. However, be
aware that after the database is placed in recovered mode, you cannot update
it with additional differential or transaction log backups.
6 Click Restore.
7 To view the progress of the restore, select File > View status.
Staging a full database recovery
This topic describes how to stage a full database recovery.
To stage a full database recovery
1 Browse for a backup image that contains the point in time to which you want
to recover.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance.
3 Click the "+" next to the database that contains the transaction log backup
you want to restore.
4 Select the transaction log image that includes the point in time from which
you want to recover.
88 Performing backups and restores of SQL Server
Restoring filegroup backups
5 Select Stage full recovery.
Stage full recovery is enabled if a set of images exists that includes the
transaction log image and that are adequate for staging a full database
recovery. When you are viewing the properties of the transaction log, a
Recovery Set tab appears.
The recovery set can include any combination of backup images that are
sufficient for staging the full recovery. These can include full database,
filegroup, and differentials.
6 Click Restore.
7 To view the progress of the restore, select File > View status.
Restoring filegroup backups
You can periodically back up only read-write filegroups as a part of your backup
strategy. If periodically is the method you use, see the instructions for that type
of restore.
See “Recovering a database from read-write backups” on page 89.
Information is also available for how to restore read-only filegroups.
See “Restoring read-only filegroups (SQL Server 2005 or later)” on page 90.
Note: If you attempt to restore a single differential backup without first restoring
the preceding database backup file, SQL Server halts the load process. An error
such as 4305 or 4306 is displayed. If you plan to restore a single differential, then
you are responsible for first restoring the database backup file. You can avoid this
problem by backing up the entire sequence of transaction logs. Also back up the
differential backup and the backup file to the same NetBackup server. Then you
can restore the entire sequence of backup objects.
Note: See “Staging a full database recovery” on page 87.
To restore a filegroup backup
1 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance and database.
3 Expand the filegroup and select a filegroup image to restore, as follows:
Performing backups and restores of SQL Server 89
Recovering a database from read-write backups
■ To restore a full backup, select the image of the filegroup backup.
■ To restore a differential filegroup backup, click the "+" next to the full
backup and select the differential backup.
4 Select the restore options.
5 Click Restore.
To view the progress of the restore, select File > View status.
Recovering a database from read-write backups
NetBackup for SQL Server automatically generates the most efficient recovery
path when you select a transaction log image for restore. The recovery path can
be based on read-write filegroups if you use them in your backup strategy. After
restoring the read-write filegroups, you can bring the database online without
having to restore the read-only filegroups provided they are not damaged.
To recover a database from read-write filegroups
1 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance.
3 Expand the database that contains the read-write filegroups you want to
restore.
4 Select the transaction log backup.
5 Right-click the transaction log backup and select Properties.
6 On the Recovery set tab, verify that a complete backup set is available.
The following screenshot shows an example of a recovery set that is based
on a read-write filegroups image.
7 Click OK.
8 To begin the database restore, click Restore.
After the restore completes the database is back online. However, you cannot
recover the read-only filegroups until they are restored.
See “Restoring read-only filegroups (SQL Server 2005 or later)” on page 90.
90 Performing backups and restores of SQL Server
Restoring read-only filegroups (SQL Server 2005 or later)
Restoring read-only filegroups (SQL Server 2005 or
later)
This topic describes how restore read-only filegroups. This type of restore
operation only applies to SQL Server 2005 or later.
To restore read-only filegroups
1 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
Be sure that the start date for the Time Filter is early enough to include the
timestamp of the earliest backup of the read-only filegroups.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance.
3 Select the database that contains the read-only filegroups you want to restore.
In the Scripting list, Restore read-only filegroups is selected.
The restore option is enabled if a full set of read-only filegroups is available.
4 Click Restore.
5 To view the progress of the restore, select File > View status.
Restoring database files
This topic describes how to restore database files.
To restore a database file
1 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance and the database.
3 Expand the filegroup that contains the file you want to restore.
4 Expand the file.
5 Select the database file image that you want to restore.
6 Select the restore options.
7 Click Restore.
To view the progress of the restore, select File > View status.
Performing backups and restores of SQL Server 91
Restoring a transaction log image without staging a full recovery
Restoring a transaction log image without staging a
full recovery
This topic describes how to restore a transaction log image without staging a full
recovery.
To restore a transaction log without staging a full recovery
1 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance.
3 Select the transaction log image that you want to restore.
If a set of images exists that include the transaction log image and that are
sufficient for staging a full database recovery, Stage full recovery is enabled.
When viewing the properties of the transaction log, a Recovery Set tab
appears.
4 Select Restore only the transaction log that you selected.
5 Click Restore.
To view the progress of the restore, select File > View status.
Performing a database move
A database move lets you use a full set of backup images to copy an existing
database to a location under a different name. Database move operations can only
be carried out when your selection includes a database image. This move can occur
either when you directly select the database backup image, or when NetBackup
finds a recovery set that contains a database backup image.
To perform a database move
1 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance.
3 Select the database backup image that you want to restore.
92 Performing backups and restores of SQL Server
Performing a database move
4 From the Scripting list, select Create a move template.
When you create a move or partial database restore script, the capability to
perform an immediate launch is disabled. You must edit the script to specify
certain destination parameters.
5 Select the restore options.
6 Click Restore.
7 Indicate a file name and click Save.
8 In the Save Script As dialog box, click Yes to open the template in Notepad.
9 Change the database name in the template to the name of the database to
restore to.
For example, replace:
# Replace the database name in the following line with the name of the database that
you
# want to move to. Also remove the hash mark <#> which precedes the keyword
<DATABASE>.
#
# DATABASE "DatabaseA"
with:
# Replace the database name in the following line with the name of the database that
you
# want to move to. Also remove the hash mark <#> which precedes the keyword
<DATABASE>.
#
DATABASE "DatabaseB"
Performing backups and restores of SQL Server 93
Performing a database move
10 Change the path for the database files that you want to restore.
You must uncomment at least one file. For example, replace:
# Replace the file path <C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DBA_FG1_File1.ndf>
# with a new file path. Also remove the hash mark <#> which precedes the keyword <TO>.
# The target of the MOVE keyword must be "DBA_FG1_File1".
MOVE "DBA_FG1_File1"
#TO "C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DBA_FG1_File1.ndf"
with:
# Replace the file path <C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DBA_FG1_File1.ndf>
# with a new file path. Also remove the hash mark <#> which precedes the keyword <TO>.
# The target of the MOVE keyword must be "DBA_FG1_File1".
MOVE "DBA_FG1_File1"
TO "C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DBB_FG1_File1.ndf"
11 Change the database file path.
For example, replace:
# Replace the file path <C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DatabaseA.mdf>
# with a new file path. Also remove the hash mark <#> which precedes the keyword <TO>.
# The target of the MOVE keyword must be "DatabaseA".
MOVE "DatabaseA"
#TO "C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DatabaseA.mdf"
with:
# Replace the file path <C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DatabaseA.mdf>
# with a new file path. Also remove the hash mark <#> which precedes the keyword <TO>.
# The target of the MOVE keyword must be "DatabaseA".
MOVE "DatabaseA"
TO "C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DatabaseB.mdf"
12 Make similar changes to the template for any differential backups or
transaction log backups you want to move.
13 When you finish modifying the template, save it.
14 To run the restore, select File > Manage script files, select the script you
created, and click Start.
15 Click Yes to launch the restore.
To view the progress of the restore, select File > View status.
94 Performing backups and restores of SQL Server
Restoring partial databases (SQL Server 2000)
Restoring partial databases (SQL Server 2000)
This topic shows you how to perform a partial database restore on a SQL Server
2000 database. A partial database restore lets you select individual filegroup
components of a database and restore them to a new database having a different
name.
Note: When you prepare your database design, map logically related database
components such as tables, indexes, and keys to the same filegroup. This design
enables you to use the partial database restore procedure to recover these logically
related components as a single unit.
To perform a partial database restore
1 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance.
3 Expand the database.
4 Select the database image that you want to restore.
5 From the Scripting list, select Create a partial database restore template.
When you create a move or partial database restore script, the capability to
perform an immediate launch is disabled. You must edit the script to specify
certain destination parameters.
6 Click Restore.
7 In the Save Script As dialog box, type the name of a file to which you want
NetBackup to write the template and click Save.
8 Click Yes to open the template in Notepad.
Performing backups and restores of SQL Server 95
Restoring partial databases (SQL Server 2000)
9 Replace the database name in the template with the name of the target
database of the restore.
For example, replace:
# Replace the database name in the following line with the name of the database that
you
# want as the target of the partial restore. Also remove the hash mark <#> which
precedes
# the keyword <DATABASE>.
#DATABASE "DatabaseA"
with:
# Replace the database name in the following line with the name of the database that
you
# want as the target of the partial restore. Also remove the hash mark <#> which
precedes
# the keyword <DATABASE>.
DATABASE "DatabaseB"
10 To restore a file, indicate a new file path and uncomment the MOVE and TO
lines.
For example, replace:
# If you wish to restore file <C:\Program Files\Microsoft SQL
Server\MSSQL\data\Sample2.mdf>,
# then replace this path with a new file path. Also remove the hash marks <#>
# which precede the keywords <MOVE> and <TO>.
#MOVE "DBA_FG1_File1"
#TO "C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DBA_FG1_File1.ndf"
with:
# If you wish to restore file <C:\Program Files\Microsoft SQL
Server\MSSQL\data\Sample2.mdf>,
# then replace this path with a new file path. Also remove the hash marks <#>
# which precede the keywords <MOVE> and <TO>.
MOVE "DBA_FG1_File1"
TO "C:\Microsoft SQL Server\MSSQL.3\MSSQL\DATA\DBB_FG1_File1.ndf"
11 Make similar changes to the template for any differential backups or
transaction log backups you want to restore.
96 Performing backups and restores of SQL Server
Performing page-level restores (SQL Server 2005 or later)
12 When you finish modifying the template, save it.
13 To run the restore, select File > Manage script files, select the script you
created, and click Start.
Performing page-level restores (SQL Server 2005 or
later)
This topic describes page-level restores. This type of restore only applies to SQL
Server 2005 or later.
About performing page-level restores (SQL Server 2005 or later)
If a portion of a SQL Server 2005 or later database is corrupted due to hardware
failure, you may be able to use page-level restore. Use page-level restore to recover
only the pages that were corrupted. Page-level restore can reduce the total
downtime if you only need to restore a relatively small number of pages. If many
pages are corrupt, then a full database recovery may be faster.
When you select the page restore option, NetBackup for SQL Server creates a page
restore template.
This template includes the following parts:
■ A page restore operation that you can modify by inserting the IDs of the pages
that you want to restore.
■ A series of transaction log images for recovering the database to the current
point in time.
■ A tail-log backup and recovery operation, which is required to bring the
database online.
See “Redirecting a restore to a different host” on page 98.
See “Performing restores with a multi-NIC connection” on page 101.
See “Overview of using NetBackup for SQL Server” on page 68.
About page-level restore requirements and limitations
The following requirements and limitations exist when you perform SQL Server
page-level restores:
■ Pages can be restored from the following backup types: Database, filegroup,
file, read-write filegroups, partial database.
Performing backups and restores of SQL Server 97
Performing page-level restores (SQL Server 2005 or later)
■ Your SQL Server 2005 or later database must use either the full or bulk-logged
recovery model. Use this model in order for page-level restore to be available
to you.
■ SQL Server sometimes cannot recover the specific pages that you request if
they contain critical information about the definition of the database itself.
For example, you cannot use page-level restore for the first page in a database
file. When you detect that page-level restore does not work, you need to use
full database recovery.
■ A maximum of 1000 pages can be recovered from a backup image through a
page-level restore.
Performing SQL Server page-level restores
This topic describes how to perform page-level restores.
To perform a page-level restore
1 Obtain a list of corrupt pages in the database.
SQL Server 200x Books Online suggests several methods for obtaining a list
of corrupt pages. One of these methods is to run the command DBCC checkdb
from the SQL Server 200x Management Studio.
2 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
3 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance.
4 Expand the database.
5 Select the database backup image that contains pages you want to restore.
6 From the Scripting list, select Create a page restore template.
7 Click Restore.
8 In the Save Script As dialog box, type a file name for the page restore script
and click Save.
9 Click Yes to open the template in Notepad.
98 Performing backups and restores of SQL Server
Redirecting a restore to a different host
10 Edit the page first operation the page IDs that you want to replace.
For example, replace:
#
# Create one or more page restore requests. These use the following format
#PAGE file-id:page-id
with
#
# Create one or more page restore requests. These use the following format
PAGE 1:14
PAGE 1:20
11 When you finish modifying the template, save it.
12 To run the restore, select File > Manage script files, select the script you
created, and click Start.
13 Click Yes to start the restore.
Redirecting a restore to a different host
Multiple NetBackup database clients can be configured to use a single NetBackup
master server for backing up SQL Server databases. With this configuration, you
can back up SQL Server database objects from one client and restore them to
another. In the following procedure, HostA is the client that was originally backed
up and HostB is the other client to which the restore is redirected.
To redirect a restore to a different host
1 Establish permission settings on the master server.
Do one of the following:
■ Create a install_path\NetBackup\db\altnames\No.Restrictions file
to allow unrestricted redirected restore privileges.
■ Create a install_path\NetBackup\db\altnames\HostB file to allow HostB
to restore HostA's data. The client named HostB can then access HostA's
data on the master as well as any other client's data that was backed up
on the master.
2 Select File > Set SQL Server connection properties.
3 Log onto the target instance on HostB.
Performing backups and restores of SQL Server 99
Redirecting a database to a different location on a different host
4 Select File > Set NetBackup client properties and select the current master
server to be a master server that is common to both HostA and HostB.
5 Select File > Restore SQL Server objects.
6 From the Backup History Options dialog box, select HostA as the SQL host
from which the target objects were backed up.
When the restore window appears it contains a history of objects that were
backed up from HostA. Select those objects and restore them to HostB.
Redirecting a database to a different location on a
different host
The database move redirects the restore of a SQL Server database to a different
location. The new location can be a different instance on the same host, a different
host, or a different file path. The move operation also lets you restore the database
under a different name than the original one.
Note: The destination host and instance of a move or restore operation is the one
that you log into. The source (or browse) host and instance for move or restore
operations is designated by using the Backup History Options dialog box, which
comes up automatically when you select File > Restore SQL Server objects.
To redirect a database to another location on a different host
1 The server that backed up the database you want to restore must appear in
the server list of the destination host. If the server is not in the list, add it.
See “About selecting a server” on page 100.
2 Select File > Set SQL Server connection properties.
3 From the Host list, select the host you want to restore to.
4 From the Instance list, select the database instance.
To select the default instance, either select <default> or leave the field empty.
5 Click Apply and then Close.
6 Select File > Set NetBackup client properties.
7 In the NetBackup client properties dialog box, from the Current NetBackup
Server list, select the NetBackup master server that contains the SQL Server
backup images that you want to restore on the destination host.
8 Click OK.
100 Performing backups and restores of SQL Server
Redirecting a database to a different location on a different host
9 Select File > Restore SQL Server objects.
10 In the Backup History Options dialog box, in the SQL Host list, select the host
that has the database you want to restore.
11 Indicate the Source Client, if applicable.
This option specifies the NetBackup client name for the selected client. This
is the name used in the policy to identify the client.
If you use a specific network interface for backup, enter the network interface
name in the Source Client box. The network interface name is defined in the
Host Properties for the server (open the properties for the server and click
Universal Settings).
12 Click OK.
13 Browse for the database that you want to move.
14 From the Scripting list, select Create a move template.
15 Click Restore.
16 In the Save As dialog box, enter a file name and click Save.
17 Click Yes to open the template.
18 Edit the template to designate the name that you want to use for the
destination database as well as the file paths that you want to use for each
of the database files.
19 After you edit the template, launch it by selecting File > Manage script files.
About selecting a server
When you perform a move, the backup images must be available on the host
machine that acts as the NetBackup master server for the destination host. If this
server is contained in the server list of the destination host, you can select the
current master server by selecting File > Set NetBackup client properties.
If the server is not in the server list of the destination host—perhaps, because the
server machine is remote or has access limitations—you must duplicate the images
onto removable media (with a unique ID), transport that media to the master
server used by the destination host, and import the images to that server. After
the images are imported, continue with the instructions for performing a move.
See “Performing a database move” on page 91.
Performing backups and restores of SQL Server 101
Performing restores with a multi-NIC connection
Performing restores with a multi-NIC connection
To perform restores with a multi-NIC connection, you need to specify the public
name of the SQL Server host in the SQL Server connection properties dialog box.
You also need to provide both the SQL Server host name and the private interface
name in the Backup History Options dialog box.
To perform restores with a multi-NIC connection
1 Open the NetBackup for SQL Server interface.
2 Select File > Set SQL Server connection properties.
3 In the Host box, specify the public name of the SQL Server host (SQLHOST).
4 Click OK.
5 Select File > Restore SQL Server objects.
6 In the SQL Host box, specify the primary (or NetBIOS) name.
7 In the Source Client box, specify the private interface name.
8 Click OK.
A dialog box opens that shows the SQL Server backups that the SQL Host
made on the private network interface.
9 Continue with the restore as normal.
Backing up and restoring the databases that contain
full-text search catalogs (SQL Server 2005)
If your databases contain full-text search catalogs, then they are backed up with
the databases and are restored as a part of the recovery process. You can also find
additional information about full-text catalogs to help you understand how the
backup and restore processes work.
About full-text catalog directory structure
A full-text catalog consists of a root directory, subdirectories, and dozens of files.
SQL Server names the root full-text directory that is based on the directory name
the end-user gives. Typically, the directory name is the same as the index name,
but SQL Server may alter if there are name conflicts. For example, consider if
there is a full-text catalog named ‘cats’ for database ONE. Then a user creates a
second database named ‘cats’ for database TWO. SQL Server creates a root full-text
directory named ‘cats0000’ for the second full-text catalog.
102 Performing backups and restores of SQL Server
Backing up and restoring the databases that contain full-text search catalogs (SQL Server 2005)
SQL Server also creates a logical file to contain full-text catalogs. The logical file
is named as sysft_full-text catalog name. For example, in the previous
example, the "cats" catalog would be contained by sysft_cats. This name has
import for the database move operation.
SQL Server places the full-text catalog directory into the following path by default:
SQL-instance-install-path\FTData\full-text-catalog-root-directory.
For instance, in the previous example, the cats directory for database TWO is
placed into a path such as:
C:\Program Files\Microsoft SQL Server\MSSQL.1\FTData\cats0000\
However, when you perform a database move command, you can establish the
path at the location of your choosing.
One additional characteristic of the full-text catalog is its association with a
filegroup. By default, the association is made with the primary filegroup. However,
you can also select a different filegroup for which to make the association.
Backing up and restoring the databases that contains full-text catalogs
When you back up a database that contains a full-text catalog, NetBackup writes
the catalog name and directory path to NetBackup catalog. NetBackup for SQL
Server displays the full-text catalog information in the backup history properties
windows when you select the database backup image. If you choose to move the
database, then the saved full-text catalog information allows NetBackup for SQL
Server to generate specifications for the move template. In the move template,
you can specify a path for the catalog directory.
snapshot backups also include all of the files in any full-text catalog that is
associated with the object that is backed up. This affects backups database,
filegroup, and file backups as follows:
■ Database backup: The full-text catalog files are backed up with the full database
snapshot backup.
■ Filegroup backup: The full-text catalog files are backed up for any catalog that
is associated with the filegroup.
■ File backup: NetBackup for SQL Server backs up the full-text catalog files along
with a snapshot file backup
The backup is performed based on the following conditions:
■ The file belongs to a filegroup that is associated with a full-text catalog
■ The file has the lowest sort-order of any file that belongs to the filegroup.
Performing backups and restores of SQL Server 103
About using batch files
Backups and restores based on snapshots are affected based on the extra catalog
files that are included in the snapshot.
Moving a database that contains full-text catalogs
NetBackup for Microsoft SQL Server extends the move template when a database
contains a full-text catalog by including the following text:
#
# Replace the full text catalog directory path
# <C:\Program Files\Microsoft SQL Server\MSSQL.1\MSSQL\FTData\HisCatalog>
# with a new directory path. Also remove the hash mark <#> which precedes the keyword
<TO>.
# The target of the MOVE keyword must be "sysft_HisCatalog".
MOVE "sysft_HisCatalog"
#TO "C:\Program Files\Microsoft SQL Server\MSSQL.1\MSSQL\FTData\HisCatalog"
#
Modify the target directory path in the template. The name sysft_HisCatalog
is the logical name of the file that contains the full-text directory named
HisCatalog. NetBackup creates this name by prepending the catalog name with
"sysft_".
About using batch files
NetBackup for SQL Server uses scripts called batch files for initiating backup and
restore operations. A batch file uses the .bch extension and is typically executed
from the install_path\DbExt\MsSql\ directory.
See the following topics to understand the contents of a batch file and for
information on how to modify or create one on your own:
■ See “About text format used in batch files” on page 104.
■ See “Guidelines for creating and using batch files” on page 104.
■ See “Notes when creating and using batch files” on page 104.
■ See “About keywords and values used in batch files” on page 105.
■ See “Running batch files” on page 117.
You must create a batch file if you start up operations with the Manage Scripts
dialog box.
Also see the example batch files.
See “About sample batch files” on page 193.
104 Performing backups and restores of SQL Server
About using batch files
■ the dbbackex command line
■ the NetBackup scheduler
About text format used in batch files
Batch files created by the SQL server agent interface are Unicode text. This permits
inclusion of object names and image names that include non-English characters.
If you create a batch file manually, use notepad or some other tool that generates
plain text. If your batch file contains any non-English characters save it as Unicode.
Guidelines for creating and using batch files
Use the following guidelines when you create and use batch files:
■ Give the file a unique name with the extension .bch.
■ Place the batch file in the install_path\NetBackup\DbExt\MsSql\ folder.
■ Batch files must reside on the host from which they executed. If performing
actions on a remote host, the batch file must reside on that remote host.
Notes when creating and using batch files
Review the following information before you create and use batch files:
■ A batch file consists of a series of operations (backups and restores) which by
default are run in sequence.
■ Each operation consists of a series of <keyword value> pairs, which completely
define the total operation.
■ The keyword is not case sensitive but the value is. Generally, you can code
both the keyword and value in uppercase. The exception is that if you use the
NBIMAGE keyword option, the value must be specified exactly as it is stored
by NetBackup server .
■ Operations are not nested.
■ With the exception of the BATCHSIZE, GROUPSIZE, RESTARTTYPE,
NUMRESTARTS, and RESTARTWAITSECONDS parameters, <keyword value>
pairs are not global. If you use BATCHSIZE, GROUPSIZE, RESTARTTYPE,
NUMRESTARTS, or RESTARTWAITSECONDS then it must appear only once
in your batch file and it must appear in the first operation.
■ If SQLINSTANCE $ALL is used, then it must appear in the first operation of
the batch file. Each operation in the batch file is performed for all SQL Server
instances on the client where the batch file is executed. Also, it is not necessary
to specify an SQLHOST or SQLINSTANCE on any subsequent operations.
Performing backups and restores of SQL Server 105
About using batch files
■ Within an operation, the <keyword value##gt## pairs may appear in any order
except that each operation must be terminated by ENDOPER TRUE.
■ You can include comment lines in your batch file by placing a hash mark (‘#’)
in the first column.
■ STOPAT, RESTORETOMARK, RESTORETOMARKAFTERTIME, RESTOREBEFOREMARK,and
RESTOREBEFOREMARKAFTERTIME are mutually exclusive restore parameters. If
either RESTORETOMARKAFTERTIME or RESTOREBEFOREMARKAFTERTIME are used,
then the batch file must also specify a datetime string with the keyword
STOPAFTER.
About keywords and values used in batch files
Table 5-3 describes the keywords and values that can be used in batch files.
Table 5-3 Keywords and values used in batch files
Keyword Values Required? Default Description
ALTCLIENT (Same as string no none Restores images from a
BROWSECLIENT) host other than the
local host.
BACKUPMODEL BACKUPMODEL_ no BACKUPMODEL_ Valid only for restore.
CONVENTIONAL, CONVENTIONAL Indicates whether the
BACKUPMODEL_ SNAPSHOT backup was originated
from a snapshot method
BATCHSIZE integer no 1 Number of back up
operations to start up
simultaneously. Applies
to all of the operations
in the batch file. Must
appear before the end
of the first operation.
Range is 1-10.
BLOCKSIZE integer no 0 Applicable for backup
operations only. Block
size is calculated as 512
bytes * 2BLOCKSIZE.
Range is 1-7.
BROWSECLIENT (Same string no none Restores images from a
as ALTCLIENT) host other than the
local host.
106 Performing backups and restores of SQL Server
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
BUFFERS See NUMBUFS.
CLOAKEDBACKUP TRUE or FALSE no FALSE Applies only to
Snapshot Client
database backups of
SQL 2000. If TRUE, the
backup is not recorded
by SQL Server and
therefore cannot be the
basis for any
subsequent differential
backup.
See “Using copy-only or
cloaked snapshot
backups to affect how
differentials are based”
on page 152.
CONSISTENCYCHECK FULLINCLUDINGINDICES, no none Performs the specified
FULLEXCLUDINGINDICES, consistency check after
PHYSICALCHECKONLY, the restore has been
CHECKCATALOG completed.
Performing backups and restores of SQL Server 107
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
COPYONLY TRUE or no See description If TRUE, an out-of-band
backup is created by
FALSE
SQL Server so that it
does not interfere with
the normal backup
sequence. The default
value is FALSE except
for full database
persistent frozen image
backups. COPYONLY is
only available for SQL
Server 2005 or later
database, differential,
and transaction log
backups.
See “Using copy-only or
cloaked snapshot
backups to affect how
differentials are based”
on page 152.
DATABASE string yes none Name of database. For
backup operations,
specify value $ALL to
designate all databases
(except for tempdb.)
DBMS MSSQL no MSSQL You can specify MSSQL
only.
DSN string no saved from GUI ODBC data source
user session name. Deprecated.
DUMPOPTION INCREMENTAL no none Specifies INCREMENTAL
restoring from an
incremental backup.
ENDOPER TRUE yes none Terminates each
operation specified in
the batch file.
108 Performing backups and restores of SQL Server
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
EXCLUDE string no none Name of a database to
exclude when
DATABASE $ALL is
specified in a batch
operation
EXCLUDE can be used in
a batch file only if
DATABASE $ALL is
used.
GROUPSIZE integer between 1 and 32 no none The number of
databases that will be
snapped as a single SQL
Server backup image.
See “About NetBackup
for SQL Server with
Snapshot Client”
on page 145.
INHIBITALTBUFFER TRUE, FALSE no FALSE Tells NetBackup
METHOD whether to consider the
candidacy of alternate
buffer method.
MAXRESTARTSETS integer no none Use MAXRESTARTSETS
to enable file
checkpointing. The
valid range is 2 to 32.
This parameter
specifies the number of
separate streams into
which the backup
request is sub-divided.
MAXTRANSFERSIZE integer no 0 Maximum transfer size
is calculated as 64
kilobytes bytes *
2MAXTRANSFERSIZE. Range
is 1-6.
Performing backups and restores of SQL Server 109
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
MOVE file group no none Specifies a filegroup
name. Used for restore
types PARTIAL and
MOVE.
NBIMAGE string yes* none Specifies a NetBackup
image for the restore
operations. See note for
NBSERVER.
* Required for restore
operations.
NBSCHED string no none If the NetBackup policy
has several Application
Backup Policy
schedules, use NBSCHED
to select amongst them.
NBSERVER string no none Specifies which master
server to use for the
backup or restore
operation.
Note: If NBSERVER is
not specified in a batch
file operation, the
master server defaults
to the name specified at
HKEY_CURRENT_USER\
Software\VERITAS\
NetBackup\
NetBackup for
Microsoft SQL
Server\DEFAULT_
SQL_NB_MASTER_
SERVER.
NUMBUFS integer no 1 Number of buffers per
stripe. Range is 1-32.
NUMRETRIES See NUMRESTARTS.
110 Performing backups and restores of SQL Server
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
NUMRESTARTS 1-9 no 1 The number of times to
retry a backup if
RESTARTTYPE AUTO is
specified. Use this
keyword only once in
the batch file and in the
first operation of the
batch file.
OBJECTNAME string yes* none Specifies a file or
filegroup name for
file/filegroup backups
and restores,
* If OBJECTTYPE=
FILE or FILEGROUP.
OBJECTTYPE DATABASE, TRXLOG, no DATABASE Specifies whether you
FILEGROUP, FILE are backing up or
restoring a database,
transaction log,
filegroup, or file.
OPERATION BACKUP, RESTORE no BACKUP Type of operation,
either backup or
restore.
PAGE Page ID no none SQL Server 2005 or
later. Specifies a page
ID for a page restore
operation.
PARTIAL TRUE, FALSE no FALSE SQL Server 2005 or
later. Specifies
NetBackup perform a
partial backup or
restore.
PASSWORD string no null Password for logging
into SQL Server.
Performing backups and restores of SQL Server 111
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
RECOVERED STATE RECOVERED, STANDBY, no RECOVERED RECOVERED means that
NOTRECOVERED, TRUE, the database should be
FALSE restored to the
recovered state.
NOTRECOVERED means
that it should remain in
the loading state
following the restore.
STANDBY means that
the database should be
restored to standby
state. If STANDBY is
used, then the
STANDBYPATH keyword
is also required. TRUE
and FALSE, when used
as values for
RECOVEREDSTATE, are
synonyms for
RECOVERED and
NOTRECOVERED.
RESTOREBEFOREMARK string no none Specify transaction log
mark.
RESTOREBEFOREMARK string no none Specify transaction log
AFTERTIME mark.
RESTOREOPTION REPLACE no none Tells NetBackup to use
the SQL Server replace
option on a restore.
RESTOREPAGES TRUE, FALSE no FALSE SQL Server 2005 or
later. Specifies that
NetBackup perform a
page restore operation.
RESTORETOMARK string no none Specify transaction log
mark.
RESTORETOMARK string no none Specify transaction log
AFTERTIME mark.
112 Performing backups and restores of SQL Server
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
RESTORETYPE FULL, PARTIAL, MOVE no FULL Full = Full database
restore, Partial =
Partial database restore,
Move = Database move
RESTORETYPE is
applicable only to
RESTORE database
operations. If MOVE is
used, then the batch file
should contain a series
of one or more
<MOVE><filegroup>
and <TO><file path>
sequences. If PARTIAL
is used, the sequence for
PARTIAL must specify
all of the filegroups in
the database whose
backup image is
referenced by the
NBIMAGE keyword.
RESTARTTYPE AUTO, MANUAL no none Available only for
backups. Use AUTO to
automatically retry
backup of failed objects.
Use MANUAL to create a
batch file for backing up
any of the objects that
were not successfully
backed up. Use this
keyword only once in
the batch file and in the
first operation of the
batch file.
RETRYTYPE See RESTARTTYPE.
Performing backups and restores of SQL Server 113
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
RESTARTWAITSECONDS integer number no 60 The time to following a
backup failure before
making a second
attempt. Use this
keyword only once in
the batch file and in the
first operation of the
batch file.
RETRYWAITSECONDS See
RESTARTWAITSECONDS.
ROLLBACKVOLUME TRUE,FALSE no FALSE Tells NetBackup to do
the recovery of a
persistent frozen image
backup using the
volume rollback
method.
SQLHOST string no Name of SQL Server
host.
If SQLHOST is not
specified in a batch file
operation, then the SQL
Server host is obtained
from HKEY_CURRENT_
USER\Software\
VERITAS\NetBackup\
NetBackup for
Microsoft SQL
Server\
DEFAULT_SQL_HOST. If
SQLINSTANCE is
missing, then the
default SQL Server
instance is assumed for
the SQL Host.
114 Performing backups and restores of SQL Server
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
SQLINSTANCE string no Name of the SQL Server
instance. Or for backup
operations specify $ALL
to designate all SQL
Server instances
including the default
instance.
If SQLINSTANCE $ALL
is used, then it must
appear in the first
operation of the batch
file. Each operation in
the batch file is
performed for all SQL
Server instances on the
client where the batch
file is executed. Also, it
is not necessary to
specify an SQLHOST or
SQLINSTANCE on any
subsequent operations.
STANDBYPATH string no none Specify a fully- qualified
file path to use for the
standby redo log.
STOPAFTER datetime string no none Specifies datetime for
RESTORETOMARK
options. The datetime
string is formatted as
YYYY/MMDDHH:MM:SS.
STOPAT datetime string no none Specifies point-in-time
recovery of a
transaction log. The
datetime string is
formatted as
YYYY/MMDDHH:MM:SS.
Performing backups and restores of SQL Server 115
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
STORAGEIMAGE string no none Used for restoring a
database that was
backed up using a
grouped Snapshot
Client snapshot.
STORAGEIMAGE
identifies the image
with which the physical
files are associated.
STRIPES integer no 1 Number of stripes.
Range is 1-32.
TO file path no none Specifies a filegroup
destination path.
Required for each MOVE
keyword. Also must
sequentially follow each
MOVE entry. The value
may be delimited with
single quotes.
TRACELEVEL MIN, MID, MAX no MIN Trace level.
TRXOPTION NOTRUNC, TRUNCONLY, no none SQL Server transaction
TAILLOG log backup options.
Note: TRUNCONLY
applies only to SQL
Server 2000. TAILLOG
only applies to SQL
Server 2005 or later.
If neither NOTRUNC nor
TRUNCONLY is selected,
then the transaction log
can be backed up and
truncated. If TAILLOG
is selected, the tail log
is backed up and
restored.
USERID string no sa User ID for logging into
SQL Server.
116 Performing backups and restores of SQL Server
About using batch files
Table 5-3 Keywords and values used in batch files (continued)
Keyword Values Required? Default Description
VDITIMEOUTSECONDS integer no 300 Timeout interval for
SQL Server Virtual
Device Interface
VERIFYONLY TRUE,FALSE no FALSE (SQL Server 2005 or
later) Tells SQL Server
to verify a backup image
but not to restore it.
VERIFYOPTION NONE, STOPONERROR no NONE (SQL Server 2005 or
CONTINUEAFTERERROR later) This option is
only valid for databases
that have an active
page. STOPONERROR
performs verification
and stops if a
verification error
occurs.
CONTINUEAFTERERROR
performs verification
but continues if a
verification error
occurs.
Creating a batch file
You can use any of the backup or restore dialog boxes to create a batch file that
contains a NetBackup for SQL Server script. This script can be executed at a later
time from the Manage Scripts dialog box.
Or you can launch the script from the dbbackex command line program or through
the NetBackup scheduler. See the example batch files.
See “About sample batch files” on page 193.
To create a batch file
1 Select File > Backup SQL Server objects or File > Restore SQL Server objects.
2 Select the object you want to back up or restore.
3 Select the backup or restore options.
4 In the Backup script or Restore script group, click Save.
5 Click Backup or Restore.
Performing backups and restores of SQL Server 117
About file checkpointing
6 Specify a file name and click Save.
Alternatively, you can select the name of an existing file and NetBackup
appends the new script to it.
7 Click Yes to open and edit the batch file.
See “About sample batch files” on page 193.
Running batch files
Once you have created a batch file, you manually run it from the NetBackup for
SQL Server interface.
To run a batch file
1 Log on to the host and instance you want to access.
See “Selecting the SQL host and instance” on page 71.
2 Select File > Manage script files.
3 Double-click the batch file.
4 Click Start.
5 To monitor the operation, select File > View status.
About file checkpointing
Use file checkpointing if you need to perform a large backup and want to save
completed work in case the operation fails before it completes. When file
checkpointing is enabled, the database or filegroup is divided into file sets that
are backed up as separate units. The following batch file command initiates file
checkpointing:
MAXRESTARTS integer
The backup operation is split into the number of operations specified by the integer
value, unless the total number of files comprising the database or filegroup
operation is less than that value. In this case, the number of separate operations
is the number of files.
File checkpointing is available for databases and filegroups that are backed up as
streams or with the snapshot option. However, the following restrictions exist:
■ The backup object must contain at least two files.
■ The recovery model of the database cannot be "simple".
118 Performing backups and restores of SQL Server
About automatic retry of unsuccessful backups
■ If the snapshot option is used for backup, then the method cannot be instant
recovery. However, file checkpointing that uses instant recovery to a storage
unit is supported.
■ The batch file that you use for a file checkpoint backup can specify only one
database or filegroup. You cannot use the "'backup database $ALL" option.
When you use file checkpointing for backing up a full database, NetBackup for
SQL Server automatically splits the database into fileset components. Recovering
the database from components requires a restore of the transaction log. NetBackup
for SQL Server automatically includes a backup log directive in the generated
batch file when you choose file checkpointing from the backup dialog box.
About automatic retry of unsuccessful backups
NetBackup for SQL Server provides the following options to retry unsuccessful
backup attempts.
Automatic retry NetBackup for SQL Server keeps track of the unsuccessful backups
that may have resulted from the execution of a batch file. When the
initial backup attempt is complete, the agent rewrites the batch file,
including only those operations that failed. The rewritten batch file
is launched automatically.
Manual retry A manual retry is similar to an automatic retry except that NetBackup
does not launch the rewritten batch file. Instead it is written to the
install_path\dbext\mssql\retry\ directory. The user can then
choose when to run the new batch file.
To use automatic retry, add the following line to your batch file.
RESTARTTYPE AUTO
By default, the unsuccessful backups are retried one time automatically after 60
seconds. To change the delay following the unsuccessful attempt, then add the
following to your batch file.
RESTARTWAITSECONDS <integer>
You can also specify the number of retries. Add the following to your batch file.
NUMRESTARTS <1 to 9>.
To use manual retry, add the following line to your batch file.
RESTARTTYPE MANUAL
Performing backups and restores of SQL Server 119
About automatic retry of unsuccessful backups
Retry may also be used with file checkpoints. You can create the batch file so that
any parts of the operation that fail are written to a new batch file that can be
launched either automatically or manually.
See “About file checkpointing” on page 117.
You can easily enable file checkpointing with automatic retry in the backup dialog
in the NetBackup for SQL Server graphical user interface. Simply select a single
database (or filegroup), then from the "Resume options for this selection" list,
select Save work and restart at point of failure.
This action creates a batch file that contains the following scripting:
MAXRESTARTSETS 32
RESTARTWAITSECONDS 60
NUMRESTARTS 1
MAXRESTARTSETS 32 means that up to 32 pieces are backed up independently. The
keywords RESTARTWAITSECONDS and NUMRESTARTS are synonymous with the
following:
RETRYWAITSECONDS 60
NUMRETRIES 1
These keywords indicates the following things: first, that an automatic retry is
launched after 60 seconds for all of the pieces that failed to get backed up on the
first time. Second, the restart is attempted only one time. You can manually change
either of these parameters.
In addition, you can choose to not have the retry script automatically launched.
Replace the NUMRETRIES command with RETRYTYPE MANUAL.
NUMRETRIES 1
with
RETRYTYPE MANUAL.
When manual retry is enabled, the generated batch file for the components that
were unsuccessfully backed up is written to the following folder:
NB_INSTALL\dbext\mssql\temp folder
Note: All of the keyword-value pairs that are described in this topic are only
permitted in the first operation of the batch file.
120 Performing backups and restores of SQL Server
Performing backups and restores of remote SQL Server installations
Performing backups and restores of remote SQL
Server installations
You can use NetBackup for SQL Server to back up and restore databases on a
remote host. Generated batch files must be saved on the remote host. You can
launch the operation from the local installation of NetBackup for SQL Server,
from an automatic backup policy, or from a manual backup.
To perform backups and restores of remote SQL Server installations
1 Select the host and instance you want to access.
See “Selecting the SQL host and instance” on page 71.
2 Create the backup or restore script.
Select one of the following:
■ To create a backup script, select File > Backup SQL Server objects.
■ To create a restore script, select File > Restore SQL Server objects.
3 Select the options for the operation.
Save is enabled in the backup and restore dialog boxes. Launch immediately
is disabled because the generated script must be executed on the remote host
that you are logged on to.
4 Click Backup or Restore.
5 In the Save Script As dialog box, navigate to the
install_path\NetBackup\DbExt\MsSql\ folder on the remote host, and save
the batch file there.
6 Launch the backup or restore operation.
Do one of the following:
■ Run the operation from the local installation of NetBackup for SQL Server.
■ Create a new policy and add the batch file to the Backup Selections list
in the policy.
Restoring multistreamed backups
When you use the NetBackup SQL Client, backups using multiple stripes are
restored automatically. Select the object you want to restore and NetBackup finds
all of the related backups and restore them. Upon restore, all of the streams must
also be available at the same time.
Performing backups and restores of SQL Server 121
Restoring multistreamed backups
About conventional backups using multiple streams
If you specified multiple stripes for a non-snapshot backup, then the number of
backup streams that you specified was created. NetBackup names these streams,
for example:
juneberry.MSSQL7.COLE.db.pubs.~.7.001of003.20060908200234..C
juneberry.MSSQL7.COLE.db.pubs.~.7.002of003.20060908200234..C
juneberry.MSSQL7.COLE.db.pubs.~.7.003of003.20060908200234..C
To create your own batch file to restore a striped object, specify only the first
stripe name with the NBIMAGE keyword. NetBackup for SQL Server finds the
remaining ones automatically. More information is available about the backup
names that are used for SQL Server objects
See “About using bplist to retrieve SQL Server backups” on page 122.
About advanced backup methods using multiple streams
If you specified multiple stripes for any Snapshot Client backup, which streams
the frozen image to tape, then NetBackup divides the number of component files
equally among the number of stripes. If the number of files is less than the
specified number of stripes, then the agent performs the backup using only as
many stripes as there are files.
Note: Since Persistent Frozen Image (PFI) policies do not stream the component
files to tape, PFI backups ignore the multistream directive.
With SQL Server backups performed with Snapshot Client, NetBackup identifies
all of the backup streams by the same name, such as:
juneberry.MSSQL7.COLE.db.Northwind.~.7.001of003.20061012131132..C
and are differentiated by NetBackup by their backup IDs.
Restoring a multiplexed SQL Server backup
In most cases, Symantec does not recommend multiplexing multiple SQL Server
streams from the same backup to a single tape. However, you may want to do this
if you vault or export backup images. During the restore of this type of multiplexed
backup, NetBackup may time out while trying to synchronize access to data blocks
from the backup tape. To prevent this time out, change the stripes parameter in
the recovery batch file from STRIPES N to STRIPES 1.
122 Performing backups and restores of SQL Server
About using bplist to retrieve SQL Server backups
When you change this value it causes the restore to be performed in a
single-stream. NetBackup presents the N backup images to SQL Server one at a
time. The tape is rewound between the restore of each image.
Restoring a multistreamed SQL Server backup with fewer devices than
it was backed up with
In your recovery environment, you may have fewer drives available for restores
than you used for backups. In this situation, SQL Server times out while it waits
for the additional backup images to be mounted. To prevent this time out, modify
the recovery batch file to specify the number of drives that are available for restore.
Consider, for example, if you had performed a backup using 5 drives, and only 2
are available for recovery. In the recovery batch file, change the stripes parameter
from STRIPES 5 to STRIPES 2. This change causes SQL Server to request two
backup images at a time until all five images are restored.
About using bplist to retrieve SQL Server backups
You can use the bplist command to obtain restore images. Use this command if
you plan to manually create a restore script, rather than through the NetBackup
for SQL Server interface. See the NetBackup Commands guides for complete
information about bplist.
To extract all of the NetBackup for SQL Server backups from a specific server for
a specific client, execute the following command from the MS-DOS prompt.
install_path\NetBackup\bin\bplist -C client -t 15 -S server -R \
where client is the host machine on which NetBackup for SQL Server resides and
server is the host machine of NetBackup server.
The following example shows how to obtain the list of SQL Server backups that
were backed up from client juneberry to server Cole:
C:\Program Files\NetBackup\bin\bplist -C juneberry -t 15 -S cole -R \
juneberry.MSSQL7.JUNEBERRY.db.pubs.~.7.001of003.20060920101716..C:\
juneberry.MSSQL7.JUNEBERRY.db.pubs.~.7.002of003.20060920101716..C:\
juneberry.MSSQL7.JUNEBERRY.db.pubs.~.7.003of003.20060920101716..C:\
juneberry.MSSQL7.JUNEBERRY.fil.pubs.pubsnew.7.001of001.20060919175149..C:\
juneberry.MSSQL7.JUNEBERRY\NEWINSTANCE.trx.abc.~.7.001of001.20060902170920..C:\
juneberry.MSSQL7.JUNEBERRY\NEWINSTANCE.fg.abc.PRIMARY.7.001of001.20060902170824.C:\
juneberry.MSSQL7.JUNEBERRY\NEWINSTANCE.db.Howard's
Barbeque.~.7.001of001.20060901085255..C:\
juneberry.MSSQL7.JUNEBERRY\NEWINSTANCE.inc.Howard's
Performing backups and restores of SQL Server 123
About using bplist to retrieve SQL Server backups
Barbeque.~.7.001of001.20060903108552..C:\
juneberry.MSSQL7.COLE.db.pubs.~.7.001of001.20060907100101..C:\
juneberry.MSSQL7.COLE.db.pubs.~.7.001of001.20060908200234..C:\
Note: The colon and backslash that terminate each line are not part of the backup
name.
About NetBackup for SQL Server backup names
The backup name is a string that consists of the following components. These
components are separated by a delimiter that is specified by the character that
precedes the "C" at the end of the backup image name.
SQL Server
identifier
Instance 2 Version
Host where SQL
Server resides name1 Object type Database Indicator
name 3
cole.MSSQL7.cole\INSTANCE2.fg.abc.PRIMARY.7.001of001.20060902170824.C:
File or filegroup
name Timestamp 6
Blocksize 4 Stripes5
1 - Named instances are formatted as host\instance-name. The default instance
is the name of the host machine.
2 - The object types are as follows:
db database
inc database differential
trx transaction log
fg filegroup
fdg filegroup differential
fil file
3 - The name of the file or filegroup if the object type is a file or filegroup; otherwise
the symbol ~ is used.
4- The blocksize is specified as 256 kilobytes bytes * 2 blocksize.
124 Performing backups and restores of SQL Server
About using bplist to retrieve SQL Server backups
5 - Stripes are specified as <stripe number>of<total stripes>. non-striped backups
are always 001of001. For striped backups, <total stripes> is the total number of
stripes for the backup. <stripe number> is the count number of the backup for
that backup, starting with 001.
6 - The format of the timestamp is YYYYMMDDHHMMSS.
7 - The character that immediately precedes the version indicator is the delimiter,
which by default is a period. However, if a period is used in any of the fields, the
delimiter may be another character. "C" is for NetBackup 6.0 and higher.
Chapter 6
Protecting SQL Server data
with VMware backups
This chapter includes the following topics:
■ About Protecting SQL Server data with VMware backups
■ About VMware SQL Server systems supported
■ About VMware SQL Server required applications and licenses
■ Adding a new VMware backup policy to protect SQL Server
■ Restoring SQL Server databases from a VMware backup
■ Limitations of using a VMware policy to protect SQL Server
About Protecting SQL Server data with VMware
backups
The Application Aware VMware Backups for SQL Server capability is introduced
with the release of NetBackup 7.5. Single pass VMware backup protects the SQL
Server. What is added is the ability to do database level restores of SQL Server
databases from VM backups. Currently customers with virtualized environments
must do a second backup. To protect an application within virtual machines install
the NetBackup agent and run the backup from inside the guest OS. The restore
experience for the end user is the same as it is today when using the traditional
application agent.
Virtualization of applications is an increasing trend. Customers use a greater
number of smaller SQL Server databases in virtualized environments. This feature
provides the capability to more efficiently protect SQL Server virtualized
environments. In previous versions you had to protect the application by using
126 Protecting SQL Server data with VMware backups
About VMware SQL Server systems supported
the traditional agent. This feature replaces the need for having the traditional
agent in the guest OS.
Protecting SQL in virtualized environments is required using the traditional
NetBackup SQL Agent . In previous versions you had to protect the application
by using the traditional agent. Replaces need for agent backups in guest OS. This
feature replaces the need for traditional agent backups.
Through a VMware backup policy, NetBackup can create consistent full backups
of an SQL Server that resides on a virtual machine. To protect a supported
application with a VMware policy, there is a new job or phase during the backup.
An Application State Capture (ASC) job executes after the VMware discovery job
and before the snapshot job(s). This ASC job contacts the NetBackup client on the
guest virtual machine. The ASC job collects and catalogs application, the specific
data that is needed for application recovery and granular recovery (GRT)
functionality.
■ Customers can perform single pass VMware backups that can quiesce all
instances of SQL Server in that guest OS and their databases. This allows them
to be fully recoverable up to the point of time of the backups.
■ Customers can use the existing SQL Server restore process to restore and
recover data from VMware backups.
■ Customers can restore and recover databases from VMware backups to
alternate clients. It did not matter if the target destination client was a physical
machine or virtual machine.
About VMware SQL Server systems supported
The following VMware SQL Server systems are supported.
■ VMware only (No Hyper-V support)
■ SQL Server 2005 and SQL Server 2008
■ Windows Server 2003 and Windows Server 2008
■ x86 and x64 only (no IA64 support)
■ Virtual machine infrastructure. The ESX 3.5 Update 5, 4.0, 4.1 and 5.0 are
supported.
■ vSphere 4.0, 4.1 and 5.0
■ vCenter 2.5
Protecting SQL Server data with VMware backups 127
About VMware SQL Server required applications and licenses
About VMware SQL Server required applications and
licenses
The following VMware SQL Server applications and licenses are required.
■ NetBackup 7.5 on the master server and media server
■ NetBackup 7.5 Client needs to be installed in the guest OS
Note: In previous releases protecting virtual machines do not require having
a client in that virtual machine. To protect the applications a client in that OS
is needed.
■ Customers must have VMware Tools and the Symantec VSS Provider of the
VMware Snapshot Provider.
Note: If a customer does not have VMware Tools the VSS Provider is packaged
in the NetBackup Goodies directory. There you can find documentation
explaining how to install the VSS Provider.
■ The Enterprise Client and App/DB Package licenses are required. One license
per ESX Host is required.
Adding a new VMware backup policy to protect SQL
Server
The following steps describe how to add a new backup policy for an SQL Server
database that runs in a VMware environment.
To add a new policy
1 Log on to the master server as administrator (Windows) or root (UNIX).
2 Log on to the master server as administrator.
3 Start the NetBackup Administration Console.
4 If your site has more than one master server, choose the one on which you
want to add the policy.
5 (Windows) In the left pane, right-click Policies and choose New Policy.
6 (Java interface) In the left pane, click Policies. In the All Policies pane,
right-click the master server, and click New Policy.
128 Protecting SQL Server data with VMware backups
Restoring SQL Server databases from a VMware backup
7 In the left pane, right-click Policies and choose New Policy.
8 In the Add a New Policy dialog box, in the Policy name box, type a unique
name for the new policy.
9 Click OK.
10 In the Add New Policy dialog box, in the Policy type list, select VMware.
The VMware policy type does not appear in the drop-down list unless your
master server has a license key for VMware.
11 In the Policy storage dialog box, select a disk storage unit.
12 Create a backup schedule to run the full backup.
See “About adding schedules to a NetBackup for SQL Server policy”
on page 37.
13 In the Clients tab select the Select automatically through query menu item.
14 Select the Field on which to match and choose the drop-down item. For
example select Displayname.
15 Select the Operator on which to match and choose the drop-down item Equal.
16 Select the Value(s): on which to match by entering the virtual machine name
between quotation marks. For example, a virtual machine name test should
be entered as "testvm1".
17 Add the virtual machine name to the query by selecting the + sign.
The Query shows the Displayname Equal "testvm1".
18 Select the Backup Selections tab.
You can view the vmware/?filter=Displayname Equal '"testvm1"' entry.
19 Select the VMware tab.
20 Under the Application Protection item select the Enable SQL Server
Recovery.
21 Under the Primary VM identifier select the VM identifier for which they
want the backups to be cataloged under.
22 Click OK.
Restoring SQL Server databases from a VMware
backup
Thisfollowing steps describe how to restore an SQL Server database that runs on
VMware from a full database or differential database backup.
Protecting SQL Server data with VMware backups 129
Limitations of using a VMware policy to protect SQL Server
To restore a SQL Server database from a VMware backup
1 Browse for the backup images you want to restore.
See “Browsing for backup images” on page 81.
2 In the Restore Microsoft SQL Server Objects dialog box, expand the database
instance.
3 Expand the database.
4 Select the VMware database image that you want to restore, as follows:
■ To restore a full backup, select the image of the VMware database backup.
Note: Only full database restores are supported. Incremental restores are not
supported.
5 Select the restore option Recovered.
Note: The restore options are defaulted to Recovered. Users are unable to
choose any other option here.
6 Click Restore.
7 To view the progress of the restore, select File > View status.
Displayed is a database URI that is generated. The URI holds the information
that drives the restore. It indicates the logical objects, and the physical files
being restored.
Limitations of using a VMware policy to protect SQL
Server
The following limitations exist when you configure a VMware policy to protect
SQL Server:
■ VMware incremental backups of SQL Server are not supported with this version
of NetBackup.
■ The Application State Capture (ASC) job fails and the databases are not
protected if you do any of the following:
■ Disable the Virtual Machine quiesce option.
■ Select the Exclude data disks option.
130 Protecting SQL Server data with VMware backups
Limitations of using a VMware policy to protect SQL Server
Databases are cataloged and protected only if they exist in a configuration that
is supported for VMware backups. As long as there are any databases that can be
protected, the ASC job continues. If you select databases for backup that exist on
supported and on unsupported disks, the ASC job produces a status 1 (partially
successful). The ASC job detects these situations and the job details include the
result of the backup operation.
SQL Server databases are not cataloged and backed up if they exist on the following:
■ Mount point volumes.
■ Virtual hard disks (VHDs).
If NetBackup detects any database objects on a VHD, the ASC job fails and no
SQL Server content is cataloged. This exclusion includes any objects that do
not exist on the VHD.
■ The ASC job detects an excluded Windows boot disk and treats it like an
independent disk.
The VMware backup cannot exclude for any reason the disk on which
NetBackup is installed. For example, do not select the Exclude boot disk option
if NetBackup is installed on the boot drive (typically C:).
Chapter 7
Disaster recovery
This chapter includes the following topics:
■ About disaster recovery
■ Preparing for disaster recovery of SQL Server
■ About disaster recovery of SQL Server
■ Restoring an existing installation of SQL Server with other active databases
■ About disaster recovery of SQL Server databases
About disaster recovery
This topic contains information and instructions on restoring the Microsoft SQL
Server in a disaster recovery scenario.
Preparing for disaster recovery of SQL Server
When you are developing your SQL Server disaster recovery plan you need to plan
how to recover from corruption of the master database. You also need to plan for
loss of your host machine. If the master database has been corrupted, then SQL
Server does not start. When disaster happens you need to use the SQL Server
rebuild master (rebuildm.exe) utility to start the SQL Server service. This utility,
however, does not recreate the schema information of your application databases.
To recover your database schema use the NetBackup SQL Server agent to restore
your latest backup of the master database.
Disaster recovery of SQL Server assumes that you have already put in place a
strategy to recovery from other sorts of data loss. Data loss can include disk,
software, and human error. To prepare for disaster recovery you need to make
frequent backups of the master database, minimally. Do frequent backups after
132 Disaster recovery
About disaster recovery of SQL Server
you have added or dropped databases or carried out other operations that may
result in schema definitions.
About disaster recovery of SQL Server
SQL Server corrects itself automatically from temporary or minor problems.
However, most disasters are beyond the scope of the automatic recovery feature.
For example, if a database becomes severely corrupted, or there is a catastrophic
failure, recovery is initiated by the system administrator.
User-initiated recovery can entail either restoring the entire server, including the
SQL Server databases, from full system backups. Or recovery can include restoring
only the SQL Server databases to a newly-installed or other available SQL Server.
Restoring the entire server has the added benefit of recovering other applications
and data which may have resided on the server at the time of failure. Restoring
be accomplished using one of the following methods:
■ Manual recovery of the server. This method involves manually restoring the
server from full system backups.
See “Preparing for disaster recovery of SQL Server” on page 131.
■ NetBackup Bare Metal Restore. BMR automates system recovery by restoring
the operating system, system configuration, and all system files and data files.
See the NetBackup Bare Metal Restore System Administrator's Guide for more
information.
After recovery of the server is complete, or after the new server installation is
available, recovery of the SQL Server databases can begin.
Restoring an existing installation of SQL Server with
other active databases
For the purposes of disaster recovery, you should only restore to a new installation
of SQL Server. However, you can restore an existing installation of SQL Server
with other active databases. The server should be running the same version of
Windows on the same hardware platform. It also should be running the same
Service Pack level: the same version of SQL Server with the same service pack as
the original server.
See “About disaster recovery of SQL Server databases” on page 133.
Disaster recovery 133
About disaster recovery of SQL Server databases
About disaster recovery of SQL Server databases
If you want to restore to a new SQL Server installation, skip the steps for how to
rebuild the master database. If you installed SQL Server in a cluster, you need to
start SQL Server in single-user mode before restoring the databases.
To recover the SQL Server databases, you need to perform the following tasks:
■ See “Rebuilding the master database” on page 133.
■ See “Starting SQL Server in single-user mode” on page 133.
■ See “Restoring the SQL Server databases” on page 134.
Rebuilding the master database
If you want to restore to an existing SQL Server, you must first rebuild the master
database.
To rebuild the master database
1 Run the Rebuild Master utility (\SQL Server installation
directory\bin\Rebuildm.exe).
Note: See the Microsoft SQL Server official documentation for information
on how to use rebuildm.exe.
2 When the rebuild is complete, restart the SQL Server services if necessary.
Starting SQL Server in single-user mode
This procedure describes how to start SQL Server in single-user mode.
Note: The SQL Agent automatically puts the database in single-user mode for
restores of non-clustered databases that are the version SQL Server 2000 or higher.
You need to start SQL Server in single-user mode only if SQL Server is configured
in a cluster.
To start SQL Server in single-user mode
1 In the Windows Control Panel, open Services.
2 Select the MSSQLServer service.
134 Disaster recovery
About disaster recovery of SQL Server databases
3 Type the following in the Startup Parameters box:
/m
4 Click Start.
Restoring the SQL Server databases
This procedure describes how to restore the SQL Server databases after disaster
recovery.
To restore the SQL Server databases
1 If SQL Server is configured in a cluster, you must start SQL Server in
single-user mode before you restore the SQL databases.
See “Starting SQL Server in single-user mode” on page 133.
2 Open the NetBackup for SQL Client interface.
3 Locate all the media that is required to perform the restore operations.
4 Select File > Restore SQL Server objects.
5 Select the backup image that contains the copy of the master database you
want to restore.
Select only the master database at this time.
6 Click Restore.
7 If SQL Server is configured in a cluster, restart the SQL Server service after
the restore has completed.
8 Continue restoring the remaining SQL Server databases.
Follow the instructions for restoring SQL databases, differentials, transaction
logs, files, and filegroups.
See “Using the NetBackup for SQL Client interface” on page 71.
When all of the restore operations have completed successfully, then the
recovery of the SQL Server databases is complete.
After the recovery is complete, Symantec recommends that you perform a
full database backup as soon as possible.
Chapter 8
Using NetBackup for SQL
Server with clustering
solutions
This chapter includes the following topics:
■ Overview of NetBackup for SQL Server with clustering solutions
■ About using NetBackup for SQL in a cluster
■ Configuring the NetBackup server to be aware of clustered SQL Server instances
■ Performing a backup on a virtual SQL Server instance
■ Performing a restore on a virtual SQL Server instance
■ About using SQL Server in a cluster with a multi-interface network connection
(multi-NIC)
■ About master server configuration for a SQL Server in a cluster with a
multi-NIC
■ Configuring a policy for a SQL Server in a cluster with a multi-NIC
■ Adding the permissions that allow for browsing of backups across the private
interface in a cluster
■ Configuring clients for a SQL Server in a cluster with a multi-NIC
■ Performing backups in a cluster with a multi-NIC connection
■ Performing restores in a cluster with a multi-NIC connection
136 Using NetBackup for SQL Server with clustering solutions
Overview of NetBackup for SQL Server with clustering solutions
Overview of NetBackup for SQL Server with clustering
solutions
NetBackup for SQL Server can be used to back up and to restore any installations
that are clustered. NetBackup for SQL Server is supported with Microsoft Cluster
server (MSCS) or Veritas Cluster Server (VCS).
About using NetBackup for SQL in a cluster
Note the following when you use NetBackup for SQL in a cluster:
■ For VCS clusters, the VirtualName attribute under the Veritas Cluster Server
resource type, Lanman, is the name of the virtual SQL Server.
■ For MSCS clusters, the unique SQL Server instances are distinguished by the
virtual server name.
■ You perform a backup or restore with a clustered instance of SQL Server as
with a non-clustered instance with two exceptions. The client name or host
name is the SQL virtual name or VirtualName and the designated instance is
<default>.
Configuring the NetBackup server to be aware of
clustered SQL Server instances
Perform the following configuration steps after the installation of a virtual SQL
Server (VIRTUALSERVER) has been created. Also perform the installation of a
virtual NetBackup media server. The following actions must be performed on the
master server or on a NetBackup remote client console that acts for the master
server.
To configure NetBackup server to be aware of clustered SQL Server instances
1 Using the NetBackup Administration Console, create a MS-SQL-Server policy
(for example, VIRTSQLPOLICY), to specify the storage attributes of the backup.
Set up the following attributes:
■ Policy-storage unit: Specify a storage unit that belongs to the storage
group that is intended for backup. If you use a virtual media server, then
specify a storage unit that belongs to the virtual media server.
■ Create a backup policy schedule for VIRTSQLPOLICY.
Using NetBackup for SQL Server with clustering solutions 137
Performing a backup on a virtual SQL Server instance
■ Add the virtual SQL Server name (VIRTUALSERVER) to the client list.
2 To incorporate SQL Server backup scheduling, create an automatic schedule
in the VIRTSQLPOLICY. Create the automatic schedule as follows:
■ Add an automatic backup schedule to VIRTSQLPOLICY.
■ Add one or more script names (batch files) to the file list.
3 Create a standard backup policy (say, STDPOLICY). Add all physical names
in the cluster to the client list of STDPOLICY.
4 Establish the permissions settings as follows for a redirected restore to a
different client.
On the master server, do one of the following:
■ Create a install_path\NetBackup\db\altnames\No.Restrictions file.
■ Create each of the files, install_path\NetBackup\db\altnames\NODEA,
install_path\NetBackup\db\altnames\NODEB.
Note: The No.Restrictions file allows all clients to perform redirected
restores to different clients. This parameter may need to be added and
removed according to site policies.
Performing a backup on a virtual SQL Server instance
NetBackup for SQL Server supports browsing for databases and images on any of
the physical nodes. But a user backup or restore is not successful unless it is
launched from the NetBackup client that is active. After NetBackup has been
properly configured: a backup of a SQL Server database from a virtual instance
is similar to a backup of a database from a non-virtual one.
See “Configuring the master server for a multi-NIC” on page 44.
See “Configuring clients for a SQL Server in a cluster with a multi-NIC” on page 140.
To perform a backup on a virtual SQL Server instance
1 Open the NetBackup for SQL Server interface on the active platform node.
2 Select File > Backup SQL Server objects.
3 Select one or more databases.
138 Using NetBackup for SQL Server with clustering solutions
Performing a restore on a virtual SQL Server instance
4 In the NetBackup Policy field, enter the name of the MS-SQL Server policy
(VIRTSQLPOLICY).This policy was created for specifying storage attributes
for the virtual SQL Server backup.
5 Click OK.
Performing a restore on a virtual SQL Server instance
This procedure describes how to perform a restore on a virtual SQL Server instance.
To perform a restore on a virtual SQL Server instance
1 Open the NetBackup for SQL Server interface on the active platform node.
2 Select File > Restore SQL Server objects.
3 In the Backup History Options dialog box, in the SQL Host list; Select the
Virtual Server name (VIRTUALSERVER) as the SQL host.
4 From the SQL Server Instance drop-down list, select <default>.
5 Click OK.
6 In the Restore Microsoft SQL Server Objects dialog box, select a backup
image or staged image list.
7 Click OK.
About using SQL Server in a cluster with a
multi-interface network connection (multi-NIC)
To combine SQL Server clusters with the use of public-private interfaces to perform
backups, special configuration is required. Many administrators want to reserve
a separate network interface for their SQL Server host machines that is used for
routing backup traffic. This type of configuration requires special consideration
in terms of configuring both the NetBackup master server and the NetBackup
client which backs up SQL Server. In addition, a special consideration is required
in terms of how to browse for SQL Server backup images.
The following distinct network resources can be identified in a dual-interface SQL
Server cluster.:
■ The public name of each SQL Server host (for example, SQLHOST1 and
SQLHOST2),
■ The private interface name that is used to back up each of the SQL Server hosts
(for example, SQLHOST1-NB and SQLHOST2-NB),
■ The public virtual name of the SQL Server (for example, VIRTSQL), and
Using NetBackup for SQL Server with clustering solutions 139
About master server configuration for a SQL Server in a cluster with a multi-NIC
■ The private virtual name of the SQL Server (for example, VIRTSQL-NB).
About master server configuration for a SQL Server
in a cluster with a multi-NIC
The following configuration changes must be made on the master server to allow
for backups and restores over a private interface:
■ The backup policies must include the private interface name in the Clients list
of the policy.
■ Permissions must be added to allow for browsing of backups across the private
interface.
Configuring a policy for a SQL Server in a cluster with
a multi-NIC
The private name of the client must be added to the Clients list of the policy. The
NetBIOS or public name of the client should not be used.
To configure a policy for a SQL Server in a cluster with a multi-NIC
1 Open the NetBackup Administration Console.
2 Create a new policy or open an existing policy.
3 In the Clients list for the policy, add a new client.
Specify the Client name as the private SQL virtual instance name. For example,
VIRTSQL-NB.
Adding the permissions that allow for browsing of
backups across the private interface in a cluster
The administrator can allow all clients or allow single clients to browse and restore
backups that are performed over the multi-nic connection.
To allow all clients to browse for backups and perform restores across the private
interface in a cluster
◆ Add the empty file NB_INSTALL\db\altnames\No.Restrictions on the master
server.
This option allows any NetBackup client to browse the backups the SQL Server
host machine makes.
140 Using NetBackup for SQL Server with clustering solutions
Configuring clients for a SQL Server in a cluster with a multi-NIC
To allow a single client to browse for backups and perform restores across the
private interface in a cluster
◆ Add the empty file, NB_INSTALL\db\altnames\SQLHOST on the master server.
This configuration only allows the SQL Server host machine, SQLHOST, to
access the backups the master server manages.
Configuring clients for a SQL Server in a cluster with
a multi-NIC
To configure backups of a client in a multi-NIC environment, you must specify
the private interface name of the SQL Server host machine.
You can set the Client name during the installation of the NetBackup client. Or it
can be changed through the Backup, Archive, and Restore interface on the client,
in the NetBackup Client Properties dialog box.
Specifying the private name of the client for a SQL Server in a cluster
with a multi-NIC
This topic describes how to specify the private name of the client when you want
to configure backups of a client. Configuration is done in a multi-NIC environment
where SQL Server is clustered.
To specify the private name of the client for a SQL Server in a cluster with a
multi-NIC
1 Open the Backup, Archive, and Restore interface.
2 Select File > NetBackup Client Properties.
3 Click the General tab.
4 In the Client name box, specify the private name of the client.
For example, the private name for the machine SQLHOST1 would be
SQLHOST1-NB. The private name for the machine SQLHOST2 would be
SQLHOST2-NB.
Performing backups in a cluster with a multi-NIC
connection
To perform backups of SQL Server using a multi-NIC connection, you need to
specify the public name of the virtual SQL Server host. You also need to create
Using NetBackup for SQL Server with clustering solutions 141
Performing backups in a cluster with a multi-NIC connection
and edit a batch file for backups so that it includes the private virtual SQL Server
name.
To perform backups in a cluster with a multi-NIC connection
1 On either SQLHOST1 or SQLHOST2, open the NetBackup for SQL Server
interface.
2 Select File > Set SQL Server connection properties.
3 In the Host box, specify the public name of the virtual SQL Server host
(VIRTSQL).
4 Click Apply and Close.
5 Select File > Backup SQL Server objects.
6 Select the databases to back up.
7 Select the backup options.
Note: Do not attempt to perform an immediate backup from the backup dialog
box. The generated batch files must be modified before they can be run
successfully.
8 From the Backup script options, click Save.
9 Click Backup.
A batch file similar to the following is created:
OPERATION BACKUP
DATABASE "ACCOUNTING"
SQLHOST "VIRTSQL"
NBSERVER "THOR"
BROWSECLIENT "VIRTSQL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
ENDOPER TRUE
142 Using NetBackup for SQL Server with clustering solutions
Performing restores in a cluster with a multi-NIC connection
10 Change the line value associated with the BROWSECLIENT from the public
virtual SQL Server name to the private name.
OPERATION BACK
UPDATABASE "ACCOUNTING"
SQLHOST "VIRTSQL"
NBSERVER "THOR"
BROWSECLIENT "VIRTSQL-NB"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
ENDOPER TRUE
11 Place the modified batch file on both nodes in the cluster so that it is available
for scheduled backups. Backups are done regardless of which node is active
when a backup is initiated.
Performing restores in a cluster with a multi-NIC
connection
To perform restores of SQL Server in a multi-NIC, cluster environment, you need
to do the following:
■ Specify the public name of the virtual SQL Server host. (In the SQL Server
connection properties dialog box.)
■ Indicate both the virtual SQL Server host name and the private interface name
of the virtual SQL Server host. (In the Browse History Options dialog box.)
■ Create a batch file for restores and manually edit it to include the private name
of the virtual SQL Server.
To perform restores in a cluster with a multi-NIC connection
1 On either SQLHOST1 or SQLHOST2, open the NetBackup for SQL Server
interface.
2 Select File > Set SQL Server connection properties.
3 In the Host box, specify the public name of the virtual SQL Server host
(VIRTSQL).
4 Click Apply and Close.
5 Select File > Restore SQL Server objects.
Using NetBackup for SQL Server with clustering solutions 143
Performing restores in a cluster with a multi-NIC connection
6 In the Backup History Options dialog box, do the following.
SQL Host Specify the public name of the virtual SQL Server (VIRTSQL).
Source Client Specify the private name of the virtual SQL Server (VIRTSQL-NB).
7 Click OK.
8 Select the databases to restore.
Note: Do not try to perform an immediate restore from the restore dialog box.
The generated batch files must be modified before they can be run
successfully.
9 Select the restore options.
10 From the Restore script options, select Save.
11 Click Restore.
A batch file similar to the following is generated.
OPERATION RESTORE
OBJECTTYPE DATABASE
DATABASE "ACCOUNTING"
NBIMAGE "SQLHOST1.MSSQL7.VIRTSQL.db.ACCOUNTING.~.7.001of001.20040306111309..C"
SQLHOST "VIRTSQL"
NBSERVER "THOR"
BROWSECLIENT "VIRTSQL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE RECOVERED
ENDOPER TRUE
144 Using NetBackup for SQL Server with clustering solutions
Performing restores in a cluster with a multi-NIC connection
12 Change the line value that is associated with BROWSECLIENT from the public
virtual SQL Server name to the private name.
OPERATION RESTORE
OBJECTTYPE DATABASE
DATABASE "ACCOUNTING"
NBIMAGE "SQLHOST1.MSSQL7.VIRTSQL.db.ACCOUNTING.~.7.001of001.20040306111309..C"
SQLHOST "VIRTSQL"
NBSERVER "THOR"
BROWSECLIENT "VIRTSQL-NB"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE RECOVERED
ENDOPER TRUE
13 Select File > Manage script files.
14 Select the modified batch file and click Start.
Chapter 9
Using NetBackup for SQL
Server with Snapshot Client
This chapter includes the following topics:
■ About NetBackup for SQL Server with Snapshot Client
■ About NetBackup Snapshot Client for SQL Server overview
■ How SQL Server operations use Snapshot Client
■ Configuring a snapshot backup
About NetBackup for SQL Server with Snapshot Client
The NetBackup Snapshot Client feature can be used with the NetBackup SQL
Agent to employ snapshot technology to back up and restore SQL Server objects.
To use NetBackup Snapshot Client with NetBackup for SQL Server, you must have
a license key for both Snapshot Client and Microsoft SQL Server.
About NetBackup Snapshot Client for SQL Server
overview
Snapshot Client, when used with NetBackup for SQL Server, backs up and restores
SQL Server objects with snapshots of the component files. Then it backs up the
snapshot version to tape or commits it to another form of persistent storage. The
snapshot technology uses SQL Server VDI (virtual device interface) quiescence
to affect a momentary freeze on database activity. Then NetBackup takes a
snapshot of the identified files.
146 Using NetBackup for SQL Server with Snapshot Client
About NetBackup Snapshot Client for SQL Server overview
The term database freeze is interchangeable with database quiescence. It refers
to the momentary state in which all activity in the database has been halted while
a snapshot (or frozen image) copy is created. The state-change in which the freeze
is lifted is referred to as database thaw.
The role of NetBackup for management of the backup images that were taken
from snapshots is common for both Snapshot Client and standard database
backups. NetBackup provides an identical view for cataloging and browsing images
and for accessing them in restore operations. It is transparent whether standard
database backups or the Snapshot Client created the images. Operational
experience is similar whether the user backs up and restores standard SQL Server
database images or any images that were taken from snapshots. However, there
are fundamental differences in the technologies.
Snapshot Client backups, in contrast to standard backups, are file-based.
NetBackup determines the file list that constitutes the SQL Server object and
backs it up asynchronously with respect to SQL Server. On the other hand, standard
backups are stream-based, which means that SQL Server provides data to
NetBackup buffer-by-buffer that constitutes a backup stream. The key role of SQL
Server in file-based backups is to provide the mechanism to freeze database
activity. NetBackup can then invoke a so-called snapshot provider that creates
volume snapshots of the files.
The snapshot method is the software component or hardware component that
creates the snapshot image of the files. The method may be a Symantec product
such as Volume Manager or an operating system component such as VSS (Microsoft
Virtual Shadow Service). Ia may also be a third-party product. Snapshot methods
use a variety of technologies such as copy-on-write, split-mirror, or RAID. They
also provide several competing methodologies. These technologies may have
different consequences for recovery speed: effect of the backup on the client
processor, and the cost of specialized disks or processing units.
About Snapshot Client features
The following NetBackup Snapshot Client features are available for use with
NetBackup for SQL Server:
Snapshot backup A snapshot is a disk image of the client's data that is made
almost instantaneously. NetBackup backs up the data from the
snapshot image, not directly from the client’s primary data.
Instant Recovery An instant recovery backup is similar to a snapshot backup
(Persistent Frozen Image except that the snapshot is not transferred to tape. The restore
or PFI) may be made directly from snapshot copy.
Using NetBackup for SQL Server with Snapshot Client 147
How SQL Server operations use Snapshot Client
Persistent Frozen Image This feature is similar to the Instant recovery feature except
with Backup to Tape that a backup copy is made to a tape storage unit. If the image
is needed for a restore, then a disk copy is used unless it has
been swapped out due to storage contention. In this case, the
restore falls back to the tape copy of the backup.
Off-host alternate client An off-host alternate client backup shifts the burden of backup
backup processing to an alternate client. This action reduces the effect
on the client's resources ordinarily caused by a local backup.
The backup agent reads the data from the SharedDisk and writes
it to storage.
Although all of these features are provided through Snapshot Client support for
SQL Server, not all methods are supported. For information on how to select a
method, see the NetBackup Snapshot Client Administrator’s Guide. For a description
of snapshot methods available for use with NetBackup for SQL Server, see the
supported platforms matrix on the Symantec Support Web site.
How SQL Server operations use Snapshot Client
This section describes how SQL Server operations use the Snapshot Client.
About selection of backup method
The selection of a backup methodology, whether standard or Snapshot Client, is
dependent on what policy is used. If a policy configured for Snapshot Client is
selected, then additional attributes of policy determine the Snapshot Client
features. It also determines the specific snapshot methods that are used.
About SQL Server limitations
Due to SQL Server limitations certain objects cannot be backed up by snapshots.
These are database differentials, filegroup differentials, and transaction logs. If
a Snapshot Client policy is selected to back up one of these object types, then
NetBackup performs a stream-based backup. NetBackup uses the storage unit
that is provided in the policy configuration. If a storage unit is not provided, then
NetBackup uses the default storage unit for the server.
What is backed up by NetBackup for SQL Server
The database administrator works exclusively with logical objects, such as
databases and filegroups. However, it is useful to understand the differences
between file- and stream-based backups in terms of the data content that is
148 Using NetBackup for SQL Server with Snapshot Client
How SQL Server operations use Snapshot Client
archived. For stream-based backups, NetBackup captures the data stream content
that is provided by SQL Server. If the user has specified multiple streams, then
SQL Server opens multiple streams that NetBackup catalogs as separate images.
For file-based backups, NetBackup creates a file list that consists of all the physical
files that constitute the object. This file list is supplied to the Snapshot Client,
which is responsible for snapshot creation. If multiple streams are specified, then
NetBackup divides the file list into sub-lists. Each sub-list is backed up separately
and constitutes a separate image. Users may notice that if multiple streams are
specified for a file-based backup and if the number of streams exceeds the number
of component files, then the number of file-based streams does not exceed the
number of files. With stream-based SQL Server backups, SQL Server always creates
exactly the number of streams that the end user specifies.
The file list that is used to back up a SQL Server database consists of the physical
files that constitute the primary filegroup. The file list also consists of any
secondary filegroups, and the transaction log. Typically, these can be identified
respectively by their name extensions, which are .mdf, .ndf, and .ldf. The file
list for a filegroup backup consists of the physical files that belong to the filegroup.
And, finally, the file list for a file object backup consists of a single physical file.
This file is the file that maps to the SQL Server file object.
About Snapshot Client and SQL Server performance considerations
When a physical file is backed up with the Snapshot Client, the backup consists
of the entire extent. This backup contrasts with stream-based SQL Server backups
where only the actual data content of the objects are archived. If you intend to
use snapshot technology to back up SQL Server, you may want to use the SQL
Server dynamic file allocation. This configuration reduces the likelihood that any
of the component files contain large areas of empty space.
Another consideration for choosing between file- and stream-based backups
concerns how SQL Server zeroes the target disk area before a stream-based restore.
In some cases, the time almost equals the total disk-copy time for restore. For
Snapshot Client restores, however, disk zeroing is not done so the total recovery
time can be substantially less.
Review the other considerations for SQL Server disk initialization.
See “About instant data file initialization” on page 52.
Performing SQL Server snapshot backups
No special interfacing considerations exist when you perform Snapshot Client
backups of SQL Server. A snapshot backup is performed if the backup object is: a
database, a filegroup, or a file and a policy is selected and configured for Snapshot
Using NetBackup for SQL Server with Snapshot Client 149
How SQL Server operations use Snapshot Client
Client. If a differential backup or transaction log backup is tried with a Snapshot
Client backup, then the operation uses the selected policy. But a standard database
backup is performed with the configured storage unit.
Performing SQL Server snapshot restores
Any backup images that were created from snapshots display along with standard
backup images. That is, all backup items—without regard to method—display in
a time-sequenced ordering that respects the composition of the database hierarchy.
In addition, no weighting is given in to determine an optimal recovery that is
based on the backup method. To determine what backup method and policy were
used when a SQL Server backup was created, right-click the backup image and
select Properties.
Figure 9-1 Backup method that appears in the backup image properties
Indicates backup was
created with a frozen
image (snapshot)
method.
About SQL Server agent grouped backups
The SQL Server agent provides a method in which multiple databases can be
quiesced together and split-off to form a single snapshot. This method minimizes
150 Using NetBackup for SQL Server with Snapshot Client
How SQL Server operations use Snapshot Client
the usage of system resources if the databases exist on a single volume. This
happens because the aggregation of constituent files uses one snapshot volume
instead of one per database. The method for aggregating database Snapshot Client
backups is called backup "grouping".
When databases are backed up in a group, all of the databases are quiesced
simultaneously. The constituent files of all databases are backed up to a single
storage image under the same backup ID. This means that an "import and copy"
procedure would use only one image to export all of the database backups in the
group.
Requirements for a grouped backup
Certain requirements must be met for a grouped backup to be performed. If any
of the following requirements are not met, a standard backup is performed:
■ All backup operations must be full backups. Differential backups are not
supported.
■ The master database cannot be included in a grouped backup.
■ The same policy must be specified for each backup operation in the group.
■ The same NetBackup server must be specified for each backup operation in
the group.
The simplest way to use grouped backup is to select multiple databases using the
Backup Microsoft SQL Server Objects dialog box. If the conditions described apply,
then the selected databases are backed up as a group.
Viewing the progress of a grouped backup
You can determine that a grouped backup is underway from the progress report.
See Figure 9-2 on page 151.
The keyword GROUPSIZE appears at the beginning of the batch file. This keyword
indicates that NetBackup uses grouping to back up the selected SQL Server
databases. If the appropriate conditions apply all operations are full database
backups. Then all of the databases are snapped and backed up as a group. When
this action happens, the progress log displays the backup image name as well as
the storage image for each database in the group.
Using NetBackup for SQL Server with Snapshot Client 151
How SQL Server operations use Snapshot Client
Figure 9-2 Progress report for a grouped backup operation
Indicates that this set of
databases is a candidate to
be backed up as a group.
Backup image name and
storage image name is shown
in a grouped backup.
Restoring a database backed up in a group
A database that is backed up in a group can be restored like any other database.
See “Restoring a database backup” on page 86.
When you launch the restore operation, note that the batch file specifies the
storage image name and the backup image name.
See Figure 9-3 on page 152.
152 Using NetBackup for SQL Server with Snapshot Client
How SQL Server operations use Snapshot Client
Figure 9-3 Batch file shown in the progress report for the restore operation
Storage image name and backup
image name are shown when
restoring from a grouped
backup.
Using copy-only or cloaked snapshot backups to affect how differentials
are based
SQL Server records the history of successful database backups in the msdb system
database. It uses this history in to decide how to base differential backups. In
particular, SQL Server creates differential database backups as cumulative with
respect to the last full database backup that it has recorded in the msdb. This
action allows for a quick recovery in case a failure has been detected after the last
full database backup.
Assume that full backups are created every day at midnight; differentials are
created every day at 6AM, noon, and 6PM; and transaction log backups are created
every two hours. If a failure occurs at 7:50 P.M. on Tuesday, then a point in time
recovery could be achieved with a restore of: the full database from Tuesday at
12AM, followed by the differential at 6PM on Tuesday, and finally the transaction
log at 8PM (choose "to 7:50 P.M.").
Using NetBackup for SQL Server with Snapshot Client 153
How SQL Server operations use Snapshot Client
However, with persistent frozen image backups, it may not be feasible to retain
the daily full backup after the next full backup has been created. If a point in time
restore is required before the latest backup, the differentials can effectively be
based on the backups that no longer exist. The only alternative would be to recover
based upon the last full backup that had been retained. You would have to use a
potentially long sequence of transaction log images.
To resolve this issue, NetBackup lets you create the SQL Server Snapshot Client
backups that are not recorded in the msdb. To create these backups in SQL Server
2005 or later, NetBackup uses the copy-only backup feature, which allows the
backups to be created as out-of-band. In SQL Server 2000, since copy-only backups
are not available, NetBackup allows these backups to be cloaked. This action
creates the backup of which SQL Server is not aware.
In SQL Server 2000, NetBackup uses cloaking to provide SQL Server with an
"unsuccessful" status. This action inhibits SQL Server from recording that the
backup had occurred. Due to the unsuccessful status, the job line of the MSSQL
progress monitor shows a -1 status. However the NetBackup server job monitor
registers the job as successful. NetBackup places information in its catalog to
denote when a backup has been cloaked and uses this information to determine
full database recovery paths. In SQL Server 2005 or later, cloaking is not needed
because SQL Server does not record the backup, but declares it as copy-only.
Sample backup schedule using cloaked backups
To understand how recovery staging works with copy-only or cloaked backups,
consider a sample backup schedule with the following characteristics:
■ The transaction log is backed up frequently, e.g., every two hours
■ A full backup is saved to secondary tape storage once every several days
■ Differential database backups are created several times per day
■ A persistent frozen image backup is created several times per day and expires
when the next one is created. This backup is created as copy-only (SQL Server
2005 or later) or it is cloaked (SQL Server 2000).
Table 9-1 shows an excerpt from this schedule.
Table 9-1 Sample backup schedule using cloaked backups
Time A full backup Differential PFI Copy-Only Transaction
saved to backup or Cloaked log backup
secondary Backup
storage
Day 1
154 Using NetBackup for SQL Server with Snapshot Client
How SQL Server operations use Snapshot Client
Table 9-1 Sample backup schedule using cloaked backups (continued)
Time A full backup Differential PFI Copy-Only Transaction
saved to backup or Cloaked log backup
secondary Backup
storage
12:00 A.M. X X
2:00 A.M. X
4:00 A.M. X X
6:00 A.M. X X
8:00 A.M. X
10:00 A.M. X X
12:00 P.M. X X
2:00 P.M. X
4:00 P.M. X X
6:00 P.M. X X
8:00 P.M. X
10:00 P.M. X X
Day 2
12:00 A.M. X X
2:00 A.M. X
Under this schedule, full backups are performed every six hours. If a failure occurs,
and is detected immediately, then you can restore the last full backup. Then you
can replay, on average, three hours of transaction logs to achieve recovery.
However, if a failure is not detected until after the next full backup, then there
are not any full backups available. There are none available since 12:00 A.M. on
day 1. The persistent frozen image backups are cloaked. However, the differential
backups would each be cumulative with respect to the last full backup that is
non-cloaked.
In this example, suppose that an error occurs at 11:30 P.M. on day 1. But the error
is not detected until 12:30 A.M. on day 2, after the 12:00 A.M. full backup. Since
the 6:00 P.M. full backup no longer exists it would be necessary to begin the
recovery with the backup taken at 12:00 A.M. on day 1. However, since all of the
Using NetBackup for SQL Server with Snapshot Client 155
How SQL Server operations use Snapshot Client
full backups were cloaked since then, the differential backup from 10:00 P.M.
would be cumulative with respect to that backup. The recovery sequence would
be restore the 12:00 A.M. day 1 backup. Restore the 10:00 P.M. differential backup.
Restore the 1½ hours of transaction log backups.
If you use copy-only or cloaked backups, then the copy-only or cloaking attribute
appears in the properties for the snapshot backup image. Differential backups
are automatically associated with the correct full backup. The SQL Agent
recognizes these backups when it selects the recovery set for the full database
restore.
Caution: Microsoft SQL Server does not recognize the cloaked backup. Therefore,
if it is incorporated in your database protection strategy, it is essential that you
maintain a comprehensive set of transaction logs. These logs span the time
duration back to the last full backup that non-cloaked.
Creating a cloaked backup (SQL Server 2000)
This procedure describes how to create a batch file that performs a cloaked backup.
To create a cloaked backup
1 Open an existing batch file in a text editor.
2 Insert the following:
CLOAKEDBACKUP TRUE
3 Save the batch file.
Creating a copy-only backup (SQL Server 2005 or later)
Any backup can be created as copy-only in SQL Server 2005 or later. A persistent
frozen image snapshot is automatically created as copy-only. You can also create
a snapshot that is NOT copy-only.
See “Creating a cloaked backup (SQL Server 2000)” on page 155.
To create a copy-only backup
1 Open an existing batch file in a text editor.
2 Insert the following:
COPYONLY TRUE
3 Save the batch file.
156 Using NetBackup for SQL Server with Snapshot Client
Configuring a snapshot backup
Creating a PFI snapshot that is not copy-only
If a persistent frozen image snapshot is selected for backup, NetBackup
automatically creates the backup image as copy-only. You can choose NOT to have
the backup created as copy-only.
To create a PFI snapshot that is not copy-only
1 Open an existing batch file in a text editor.
2 Insert the following:
COPYONLY FALSE
3 Save the batch file.
Configuring a snapshot backup
Before you configure NetBackup for SQL Server for snapshot backups, review the
configuration requirements and configuration steps in the following sections:
■ See “Configuration requirements” on page 156.
■ See “Configuration steps” on page 157.
Once you have reviewed this information, continue with the instructions for how
to configure the policy. You can create a standard policy with snapshot methods
or a policy that uses instant recovery.
See the following topics:
■ See “Overview of NetBackup for SQL Server” on page 15.
■ See “Configuring a policy for instant recovery” on page 159.
■ See “Configuring Snapshot Client policies for NetBackup for SQL Server”
on page 157.
■ See “Configuring a Snapshot Client policy” on page 157.
Configuration requirements
Review the following requirements before you configure NetBackup for SQL Server
with snapshot backups:
■ See the NetBackup Snapshot Client Administrator’s Guide for details on the
hardware requirements and software requirements for the snapshot method
that you want to use.
■ Go to the Symantec Support Web site for details on the snapshot methods and
platforms that are supported for NetBackup for SQL Server.
Using NetBackup for SQL Server with Snapshot Client 157
Configuring a snapshot backup
■ The volume(s) which contains the SQL Server databases and log files should
be dedicated to SQL Server only. Other types of databases (e.g., Exchange)
should not reside on the volume(s).
■ NetBackup Snapshot Client is installed and configured correctly and the license
key for this option has been registered. See the NetBackup Snapshot Client
Administrator’s Guide for details.
■ SQL Server 2000 or later
Configuration steps
To perform a Snapshot Client for SQL Server backup, you must perform the
following configuration steps:
■ Create a backup script (.bch file) using the NetBackup for SQL Server interface.
See “About sample batch files” on page 193.
■ Configure an MS-SQL-Server policy that has the Snapshot Client attributes
selected.
See “Configuring Snapshot Client policies for NetBackup for SQL Server”
on page 157.
Configuring Snapshot Client policies for NetBackup for SQL Server
This topic only covers what is necessary to configure snapshot backups for a
MS-SQL-Server policy.
Also see the following information:
■ Information on other policy attributes, how to schedules, how to add clients,
and how to add backup selections is described in other topics.
See “Configuring multistream operations” on page 42."
■ For information on how a snapshot method is selected automatically and
details on the types of snapshot methods, See the NetBackup Snapshot Client
Administrator’s Guide.
Only one snapshot method can be configured per policy. If you want to select one
snapshot method for clients, one group of clients, and a different method for
another group, then you would need to create two policies for each group of clients.
Then select one method for each policy.
Configuring a Snapshot Client policy
This topic describes how to configure a Snapshot Client policy. Optionally you
can choose to perform an off-host backup.
158 Using NetBackup for SQL Server with Snapshot Client
Configuring a snapshot backup
To configure a Snapshot Client policy
1 Open the policy you want to configure.
2 Click the Attributes tab.
3 From the Policy type list, select MS-SQL-Server.
4 Select the Policy storage unit/lifecycle policy.
If database differentials, filegroup differentials, or transaction logs are
included in the Backup Selections list of a Snapshot Client policy, then
NetBackup performs a stream-based backup. The selected storage unit is
used. If a storage unit is not provided, then NetBackup uses the default storage
unit for the server.
5 Select Perform snapshot backups.
6 Choose to have NetBackup select the snapshot method or select the snapshot
method manually.
Perform one of the following:
■ By default, NetBackup chooses a snapshot method for you. If you have
changed this setting and want NetBackup to choose the method
automatically, click Snapshot Client Options. Then from the Snapshot
method list, select auto.
■ To use a specific snapshot method, click Snapshot Client Options. From
the Snapshot method list, select the method you want to use for this
policy.
See the NetBackup Snapshot Client Administrator’s Guide for details on how
to select the snapshot method and automatic snapshot selection.
7 To configure schedules, click the Schedules tab.
8 Configure an Application and an Automatic schedule.
See “Configuring multistriped backups” on page 43."
9 (Optional) To use an alternate client to reduce the processing load on the
client, perform the following steps:
■ The alternate client must be the client that shares the disk array. This
option may require additional configuration. See the NetBackup Snapshot
Client Administrator’s Guide.
■ Select Perform off-host backup.
■ Click Use alternate client and enter the name of the alternate client.
Note: Use data mover is not a supported option for NetBackup for SQL Server.
Using NetBackup for SQL Server with Snapshot Client 159
Configuring a snapshot backup
10 Use the Clients tab to specify clients to be backed up by this policy.
11 Use the backup selections tab to enter the batch files.
12 Configure other attributes, and add the schedules and backup selections.
13 Click OK to close the dialog box.
Configuring a policy for instant recovery
These instructions describe how to configure a policy for instant recovery.
Optionally you can choose to back up to disk only.
To configure a policy for instant recovery
1 Open the policy you want to configure.
2 Click the Attributes tab.
3 From the Policy type list, select MS-SQL-Server.
4 Select the Policy storage unit/lifecycle policy.
If you select an Instant Recovery option on the Schedules tab (see step 8),
the storage unit is not used.
NetBackup creates only a disk snapshot.
If database differentials, filegroup differentials, or transaction logs are
included in the Backup Selections list of a Snapshot Client policy, then
NetBackup performs a stream-based backup. This backup uses the selected
storage unit. If a storage unit is not provided, then NetBackup uses the default
storage unit for the server.
5 Click Perform snapshot backups.
6 Choose to have NetBackup select the snapshot method or select the snapshot
method manually.
Perform one of the following:
■ By default, NetBackup chooses a snapshot method for you. If you have
changed this setting and want NetBackup to choose the method
automatically, click Snapshot Client Options. In the Snapshot Client
Options dialog box, from the Snapshot method list, choose auto.
■ To use a specific snapshot method, click Snapshot Client Options. In the
Snapshot Client Options dialog box, from the Snapshot method list,
choose the method you want to use for this policy.
See the NetBackup Snapshot Client Administrator’s Guide for details on how
to select the snapshot method and automatic snapshot selection.
160 Using NetBackup for SQL Server with Snapshot Client
Configuring a snapshot backup
7 Select Retain snapshots for Instant Recovery.
NetBackup retains the snapshot on disk, so that instant recovery can be
performed from the snapshot.
A normal backup to storage is also performed, if you do not choose to create
a snapshot only (see step 8).
8 To configure schedules, click the Schedules tab.
9 Follow the instructions to configure an Application and Automatic schedule.
10 (Optional) To create a disk image only, open the Application schedule and
select an instant recovery option.
Select one of the following options:
■ If Snapshots and copy snapshots to a storage unit is selected, NetBackup
creates a disk snapshot. NetBackup also backs up the client’s data to the
storage unit that is specified for the policy.
■ If Snapshots only is selected, the image is not backed up to tape or to
other storage. NetBackup creates a disk snapshot only. Note that this disk
snapshot is not considered a replacement for traditional backup.
11 Use the Clients tab to specify clients to be backed up by this policy.
12 Use the backup selections tab to enter the batch files.
13 Configure other attributes, and add the schedules and backup selections.
14 Click OK to close the dialog box.
Chapter 10
Backups and restores in an
SAP environment
This chapter includes the following topics:
■ About backups and restores in an SAP environment
About backups and restores in an SAP environment
With NetBackup you can perform scheduled SAP backups, in accordance with a
predefined backup strategy, or manual backups. These backups may not be planned
and may be necessary in exceptional situations. The practices that are described
here are based on the practices that are recommended by SAP in SAP/MS SQL
Server 2000 DBA in CCMS.
The NetBackup backup and restore procedures for the SAP R/3 database are
identical to the NetBackup procedures with any other SQL Server database.
You can create scripts to perform full or differential backups of databases and
backups of transaction logs. In addition to the database backups and restores,
NetBackup also provides the capabilities to back up the SAP file systems.
About NetBackup manual backups
The administrator on the master server can use the NetBackup Administration
Console to manually execute an automatic backup schedule. This schedule can be
for an "MS-SQL-Server" policy, where the R/3 database is specified in the backup
script.
For more information, see the section on manual backups in the Netbackup System
Administrator Guide for Windows.
162 Backups and restores in an SAP environment
About backups and restores in an SAP environment
About policy configuration for SQL Server in an SAP environment
To automatically perform backups of an SAP environment, you need to create
backup policies. A backup policy with the "MS-SQL-Server" policy type that is
selected must be created for R/3 database backups. Batch files, which initiate the
backup of the database and transaction logs, must be added to the Backup
Selections list in the policy.
Information is available for how to create the batch files that are needed and how
to configure backup policies.
For backups of the executables disk (a file-system backup), a backup policy must
be created with the Windows policy type selected.
For information on Windows policies, see the NetBackup System Administrator
Guide for Windows, Volume 1.
Creating batch files for automatic backups in for SQL Server in an SAP
environment
NetBackup for SQL Server uses batch files to initiate database backup and restore
operations. A batch file must be created for database backups and for transaction
log backups. These batch files must then be added to the Selections list in the
backup policies that you created.
Creating a batch file for database backups
This topic describes how to create a batch file for database backups.
To create a script for database backups
1 From the Windows Start menu, select Programs > NetBackup > NetBackup
Agents > NetBackup MS SQL Client.
2 Select File > Backup SQL Server objects.
3 In the Backup Microsoft SQL Server Objects dialog box, in the left pane,
expand the database instance.
4 In the right pane, select the R/3 database.
5 From the Type of Backup list, select the type of backup you want to perform,
Full, or Full differential.
6 Under Backup Script, select Save.
7 Click Backup.
Backups and restores in an SAP environment 163
About backups and restores in an SAP environment
8 Specify a file name and click Save.
Alternatively, you can select the name of an existing file, and NetBackup
appends the new script to it.
9 Click Yes to open and edit the batch file.
Creating a batch file for transaction log backups
This topic describes how to create a batch file for transaction log backups.
To create a batch file for transaction log backups
1 Before starting a transaction log backup, the database administrator should
set the Transaction log backup options database option to off. This option
on the SQL Server interface applies to the databases.
The entire sequence of transaction logs generated following any database
dump must be maintained on the same NetBackup server. NetBackup for SQL
Server requires that you follow these guidelines in devising your backup
strategy to ensure success in restoring your database.
2 Select File > Backup SQL Server objects.
3 In the Backup Microsoft SQL Server Objects dialog box, in the left pane,
expand the database instance.
4 In the right pane, select the R/3 database.
5 For the Type of Backup, select transaction log.
6 Under Backup Script, select Save.
7 Click Backup.
8 Specify a file name and click Save.
Alternatively, you can select the name of an existing file, and NetBackup
appends the new script to it.
9 Click Yes to open and edit the batch file.
Monitoring backups on SQL Server
Check scheduled backups regularly to ensure that they completed successfully.
Always check the following:
■ That the most recent backup has run successfully.
See “About progress reports created for NetBackup for SQL Server on the
client” on page 185.
164 Backups and restores in an SAP environment
About backups and restores in an SAP environment
■ All the backups in the backup cycle are executed according to the schedule.
Gaps in a backup sequence can have serious consequences in a subsequent
attempt to restore the database.
Restoring the R/3 database
This topic describes how to restore the R/3 database.
Determine how to perform the restore based on the following scenarios:
■ If you have scheduled differential backups, review the information for that
type of restore.
See “About including differential backups in a restore operation” on page 164.
■ If the R/3 database disk system is damaged or the transaction log disk system
is damaged, follow the instructions for that scenario.
See “Restoring the R/3 database after a disk crash” on page 164.
■ To perform a regular restore of the R/3 database, follow the instructions for
that type of restore.
See “Restoring the database backups and transaction log backups” on page 165.
About including differential backups in a restore operation
If you incorporated differential backups in the backup strategy, the restore process
differs depending on the type of backups available.
Determine how to perform the restore based on which of the following differential
backups you have:
■ If differential backups were made after the last full database backup, restore
the last database backup that is followed by the most recent differential backup.
Then apply all subsequent transaction logs.
■ If no differential backups were made since the last full database backup, restore
the last full database backup and then apply all subsequent transaction logs.
■ If several differential backups are available but the latest one cannot be read,
restore the most recent full database backup. And restore the latest readable
differential backup and apply all subsequently created transaction logs.
Restoring the R/3 database after a disk crash
This topic describes how to restore the database when the R/3 database disk
system is damaged or the transaction log disk system is damaged. This process
is only applicable to a configuration with three disk systems: one system for the
R/3 database, one for the R/3 transaction logs and one for all others.
Backups and restores in an SAP environment 165
About backups and restores in an SAP environment
Note: The R3 database must not be in use when you are performing a restore
operation. Make sure that all SAP services are stopped before you attempt a restore
with NetBackup.
Warning: If the disk system on which the R/3 database resides is damaged, it is
vital to immediately back up the currently active transaction log. This log backup
is done to prevent loss of data. Without a backup of the current log, the database
can only be restored to the status at the time of the last transaction log backup.
If work has been carried out on the R/3 system since then, this work is lost.
To restore the R/3 database after a disk crash
1 Back up the current transaction log.
2 Replace damaged disks.
Replacing damaged disks in a RAID disk system is normally a straightforward
procedure. If you are uncertain how to proceed, see the documentation of
your hardware vendor to learn how to handle the disks. The new disks must
be formatted and assigned the same drive letter as the old disks.
3 Restore the database logs and transaction logs.
The central phase of a restore operation is the reloading of the database
backup and the application of the available transaction logs. When the
database backup is reloaded, the database files are automatically recreated.
The data is copied from the backup device to the newly created files. Once
this copy has been done, the transaction logs are applied in the same sequence
as they were originally made. In a final step, open transactions that were not
completed at the time of the database failure are rolled back.
Restoring the database backups and transaction log backups
NetBackup MS-SQL server agent GUI provides for automatic staging. By selecting
the latest transaction log backup, the GUI automatically restores the previous
full database backup. It also restores any optional differential backups and
subsequent transaction log backups. You can also use the option to specify a point
in time to which to restore to.
Note: The R3 database must not be in use when performing a restore operation.
Make sure that all SAP services are stopped before you attempt a restore with
NetBackup.
166 Backups and restores in an SAP environment
About backups and restores in an SAP environment
Warning: To restore the R/3 database you first restore the most recent database
backup and then the subsequent transaction logs. During the entire procedure,
do not execute any transactions and do not shut down the database server. A
server shutdown would write a checkpoint to the log and as a result you would
not be able to restore further transaction logs.
To restore the database backups and transaction log backups
1 Restore the most recent database backup.
2 Restore the latest differential database backup (if available).
3 Restore all succeeding transaction log backups.
4 Restore the latest transaction log backup.
Chapter 11
Backup Exec restore topics
This chapter includes the following topics:
■ Using NetBackup to restore SQL Server from Backup Exec images
■ Specifying the server, client, and policy type
■ Restoring from Backup Exec images
■ Restore options for restoring SQL backups from Backup Exec images
■ Restoring Backup Exec SQL transaction logs up to a named transaction
■ Redirecting a restore that was backed up with Backup Exec
■ Restoring Backup Exec SQL transaction logs backups up to a point in time
■ Restoring the SQL master database from a Backup Exec image
■ Restarting SQL using database copies
■ Restoring Backup Exec filegroup backups
■ Restoring Backup Exec Database backups
Using NetBackup to restore SQL Server from Backup
Exec images
NetBackup can restore SQL Server from Backup Exec images, using the NetBackup
Backup, Archive, and Restore (NetBackup Client) interface.
Requirements for restoring SQL Server from Backup Exec images
The following requirements apply when you restore SQL Server from Backup Exec
images:
168 Backup Exec restore topics
Specifying the server, client, and policy type
■ To restore Backup Exec images successfully, you need to do the following:
Import the Backup Exec media into NetBackup with the vmphyinv command,
run the bpimport command, Phase 1, and Phase 2. Refer to the instructions
for importing images from Backup Exec media.
See the NetBackup System Administrator's Guide for Windows, Volume I.
NetBackup can restore Backup Exec images of SQL 2000 and later.
■ The NetBackup Client Service should be running in a user account that has
been granted the system administrator role on the target SQL instance.
Details are available on how to configure the logon account for this service.
See “About SQL Server privileges” on page 46.
Limitations when restoring SQL Server from Backup Exec images
The following limitation exist when you restore SQL Server from Backup Exec
images:
■ Multiple databases cannot be restored in a single restore job. You can restore
full, differential, and transaction log backups of a same database in a single
restore job.
■ Table backups cannot be restored, though they are included in the backup
image and can be selected for restore. If you try to restore a table, the job fails
with the error "unsupported object was selected for restore".
Specifying the server, client, and policy type
To browse for backups of SQL databases that are backed up by Backup Exec, you
must first specify the server that contains the backup images. You must also
specify the client that performed the backups, and the type of policy that is
associated with the backups.
To specify the server, client, and policy type
1 Select File > Specify NetBackup Machines and Policy Type.
2 In the Specify NetBackup Machines and Policy Type dialog box, select the
desired server from the Server to use for backups and restores list.
3 From the Source client for restores list, select the desired client.
The source client is the SQL Server machine name whose backup images you
want to browse.
4 From the Policy type for restores list, select MS-SQL-Server.
Backup Exec restore topics 169
Restoring from Backup Exec images
5 To redirect a restore to a different client, select the client to which to redirect
the restore from the Destination clients for restores list.
See “Redirecting a restore that was backed up with Backup Exec” on page 175.
6 Click OK.
NetBackup browses for SQL Server backup images.
The NetBackup History pane displays SQL Server backup information. The
top split windows show individual image information. The bottom split gives
file and folder information and also allows the user to select what files are
to be restored.
Restoring from Backup Exec images
When you restore SQL Servers from Backup Exec images, there are several options
and procedures that you can use. For example, you can redirect a restore that you
backed up with Backup Exec. You can also restore Backup Exec SQL transaction
logs up to a point in time, and restart SQL using copies of the master databases.
You can restore the SQL master database from a Backup Exec image.
Restore options for restoring SQL backups from
Backup Exec images
Microsoft SQL Server Redirection tab field descriptions describe the restore options
that are available when you restore SQL backups from Backup Exec images.
Table 11-1 Restore options on the Microsoft SQL Server tab
Item Description
Recovery ■ Leave database operational. No additional transaction logs can
completion state be restored.
(With Recovery)
Select this option to restore the last database, differential, or log
backup in the restore sequence. This action causes the restore
operation to roll back all incompleted transactions. After the
recovery operation, the database is ready for use. If Leave database
operational is not performed, the database is left in an intermediate
state and is not usable.
If Leave database operational is selected when an intermediate
backup is applied, you cannot continue to restore backups.
You must restart the restore operation from the beginning.
170 Backup Exec restore topics
Restore options for restoring SQL backups from Backup Exec images
Table 11-1 Restore options on the Microsoft SQL Server tab (continued)
Item Description
■ Leave database nonoperational but able to restore additional
transaction logs
(No Recovery)
Select this option during a restore if you have additional differential
or transaction log backups to be restored in another restore job.
■ Leave database read-only and able to restore additional
transaction logs
(Standby)
Select this option during transaction log and database restore to
create and maintain a standby database. See your SQL
documentation for information on standby databases.
Replacedatabases Select this checkbox to replace a database or filegroup, even if another
or filegroups database or filegroup with the same name already exists on the server.
If Replace Databases or Filegroups is not specified for a restore, SQL
performs a safety check. This check ensures that a different database
or filegroup is not accidentally overwritten. Refer to your SQL
documentation for more information about the safety check that
occurs when the REPLACE option is not selected.
Automate master Enable NetBackup to stop SQL so that the master database can be
database restore restored. All existing users are logged off, and SQL Server is put into
single-user mode.
When this option is selected, only the master database can be restored;
if this option is selected for any other database, those jobs fail.
NetBackup must have access to the following SQL registry keys:
HKEY_LOCAL_MACHINE\Software\Microsoft\Microsoft SQL Server
HKEY_LOCAL_MACHINE\Software\Microsoft\MSSQLServer
Without this access, a restore to the default directory may not work.
Also, the option Automate master database restore does not work.
To ensure that NetBackup has access rights, verify that the user
account of the NetBackup Client Service has administrator rights to
the Windows server that the SQL instance is installed on.
Backup Exec restore topics 171
Restore options for restoring SQL backups from Backup Exec images
Table 11-1 Restore options on the Microsoft SQL Server tab (continued)
Item Description
Consistencycheck To recover the database after the restores are complete and when you
after restore selected the Leave database operational option, you need to select a
consistency check option.
The following consistency checks are available:
■ Full check, excluding indexes.
Select this option to exclude indexes from the consistency check.
If indexes are not checked, the consistency check runs significantly
faster but is not as thorough. Only the data pages and clustered
index pages for each user table are included in the consistency
check. The consistency of the nonclustered index pages is not
checked.
■ Full check, including indexes.
Select this option to include indexes in the consistency check. Any
errors are logged. This option is selected by default.
■ Physical check only (SQL 2000 only).
Select this option to perform a low overhead check of the physical
consistency of the SQL Server 2000 database. This option only
checks the integrity of the physical structure of the following: the
page and the record headers, the consistency between the pages'
object ID and index ID, and the allocation structures.
■ None.
Select this option to perform sequential restores. Do not run a
consistency check after a restore until all sequential restores are
done. If a consistency check is selected during a restore, the restore
completes but the consistency check is not performed. Check the
job log for this information.
172 Backup Exec restore topics
Restore options for restoring SQL backups from Backup Exec images
Table 11-1 Restore options on the Microsoft SQL Server tab (continued)
Item Description
Alternate drive Select a drive to which SQL database files can be restored. Use this
for restoring option if the drive where one or more of the database files previously
database files resided no longer exists.
Note: Do not select the Alternate drive option when you restore
filegroups. Filegroups must be restored to the same drive letter and
path that they were backed up from.
In the backup set, SQL Server stores the physical file names of the
files that make up the database. The physical file name includes the
directory path. For the logical file, "pubs," the physical file name is
stored as E:\MSSQL7\DATA\pubs.mdf. If the database is restored
later, SQL uses these same physical file names for the target of the
restore. During a restore, NetBackup automatically creates any
necessary subdirectories that do not exist.
However, the behavior is different if the drive where one or more of
the database files previously resided no longer exists. Then NetBackup
moves those files to their original directory path, but on the alternate
drive specified.
Using the same example, if drive C: is specified, then the file with the
original directory path of E:\MSSQL7\DATA\pubs.mdf is restored to
C:\MSSQL7\DATA\pubs.mdf.
If no alternate drive is specified in this situation, the job fails.
Restore to You can restore to an alternate drive in one of the following ways:
alternate drive
■ Only when original drive does not exist
Select this option to restore all database files to their original
directory path on the alternate drive. The restore is performed
only if the drive where they originally resided exists. To make this
option available, select a drive letter in Alternate drive for
restoring database files list.
■ Even when original drive does exist
Select this option to restore all database files to their original
directory path on the alternate drive. The restore is performed
even if the drive where they originally resided exists. To make this
option available, select a drive letter in Alternate drive for
restoring database files list.
Backup Exec restore topics 173
Restore options for restoring SQL backups from Backup Exec images
Table 11-1 Restore options on the Microsoft SQL Server tab (continued)
Item Description
Restore all Select this checkbox to restore files to the default data and log
database files to directories of the destination instance. For example, to restore a
the target database to a different instance of SQL. This option moves the database
instance’s data files to the correct location for the new instance.
location Note: Do not select the Restore all database files to the target
instance’s data location option when restoring filegroups. Filegroups
must be restored to the same drive letter and path that they were
backed up from.
If this option is not selected, then the files are restored to the directory
that the master database is in.
Point in time log Select this checkbox to restore transactions from a transaction log up
restore to and including a point in time in the transaction log. After the point
in time, recovery from the transaction log is stopped.
Select the part of the date you want to change. Then enter a new date
or click the arrow to display a calendar from which you can select a
date.
Select the part of the time you want to change, and then enter a new
time or click the arrows to select a new time.
Restore log up to Select this checkbox to restore transactions from a transaction log up
named to a named transaction (or named mark) in the transaction log. After
transaction (SQL the named transaction is restored,the recovery from the transaction
2000) log is stopped. The named transactions are case sensitive.
You can restore transactions in one of the following ways:
■ Include the named transaction
Select this checkbox to include the named transaction in the
restore; otherwise the restore stops immediately before the named
transaction is restored.
■ Found after
Select this checkbox to specify a date and time after which the
restore operation is to search for the named transaction. For
example, consider that you specified a restore from a log up to the
named transaction "AfternoonBreak." This transaction was found
after 6/02/2007, 12:01 P.M. In that case, the restore operation does
not search for "AfternoonBreak" until after that time.
174 Backup Exec restore topics
Restoring Backup Exec SQL transaction logs up to a named transaction
Restoring Backup Exec SQL transaction logs up to a
named transaction
When restoring SQL Server 2000 transaction logs, you can restore transactions
from a transaction log up to and including a named transaction (or mark). After
the named transaction is reached, recovery from the transaction log is stopped.
To restore from a Backup Exec image a SQL transaction log up to a named
transaction
1 Log on as Administrator.
2 Open the Backup, Archive, and Restore interface.
3 Specify the appropriate server, client, and policy type.
See “Specifying the server, client, and policy type” on page 168.
4 Select File > Select Files and Folders to Restore > from Backup Exec Backup.
5 In the Restore window, from the NetBackup History pane, select the backup
image that contains the objects you want to restore. In the All Folders pane,
select the most recent full database backup. Select also the most recent
differential database backup, if any. And select all the log backups you want
to restore.
6 Select Actions > Start Restore of Marked Files.
7 In the Restore Marked Files dialog box, from the Recovery completion state
group, select Leave database operational. No additional transaction logs
can be restored.
8 Select Restore log up to named transaction and then enter the name of the
transaction.
The names are case sensitive. Make sure that you enter the correct uppercase
and lowercase characters.
Select from one of the following options:
■ To include the named transaction in the restore, select Include the named
transaction.
■ To specify a particular named transaction in the log, select Found after
and select a date and time.
If a date and time are not entered, recovery from the transaction log is
stopped at the first transaction with the specified name.
See “Restoring the SQL master database from a Backup Exec image”
on page 176.
9 Click Start Restore.
Backup Exec restore topics 175
Redirecting a restore that was backed up with Backup Exec
Redirecting a restore that was backed up with Backup
Exec
You can redirect the following Backup Exec backups:
■ A database backup to a different server, database, or instance.
■ Differential and log backups to wherever the associated database is restored.
■ One or more filegroups in a backup to a different server or instance. Filegroups
can be redirected to a different server, but the database file paths cannot be
changed. For example, if the filegroup was backed up from G:\SQLDATA, it must
be restored to G:\SQLDATA even if it is redirected to another server. Filegroups
must be restored to the same drive letter and path that they were backed up
from.
To redirect a restore that was backed up with Backup Exec
1 Log on as Administrator.
2 Open the Backup, Archive, and Restore interface.
3 Specify the appropriate server, source client, destination client, and policy
type.
See “Specifying the server, client, and policy type” on page 168.
4 Select File > Select Files and Folders to Restore > from Backup Exec Backup.
5 In the Restore window, select the objects you want to restore.
6 Select Actions > Start Restore of Marked Files.
7 In the Restore Marked Files dialog box, click the Microsoft SQL Server
Redirection tab.
8 Select the other restore options that you want on the Microsoft SQL Server
tab.
9 Click OK.
Restoring Backup Exec SQL transaction logs backups
up to a point in time
You can restore transactions from a transaction log up to and including a point
in time in the transaction log. After the point in time is reached, recovery from
the transaction log stops.
176 Backup Exec restore topics
Restoring the SQL master database from a Backup Exec image
To restore SQL transaction logs up to a point in time from Backup Exec images
1 Log on as Administrator.
2 Open the Backup, Archive, and Restore interface.
3 Specify the appropriate server, client, and policy type.
See “Specifying the server, client, and policy type” on page 168.
4 Select File > Select Files and Folders to Restore > from Backup Exec Backup.
5 In the Restore window, from the NetBackup History pane, select the backup
image that contains the objects you want to restore.
6 In the All Folders pane, select the most recent full database backup. Select
also the most recent differential database backup, if any. And select all the
log backups you want to restore.
7 Select Actions > Start Restore of Marked Files.
8 In the Restore Marked Files dialog box, from the Recovery completion state
group, select Leave database operational. No additional transaction logs
can be restored.
9 Select Point in time log restore and select a date and time.
10 Click Start Restore.
Restoring the SQL master database from a Backup
Exec image
If the master database is damaged, symptoms can include the following:
■ An inability to start SQL.
■ Segmentation faults or input or output errors.
■ A report that is generated by SQL Database Consistency Checker utility (DBCC).
If you can start SQL, you can restore the latest copy of the master database backup
by using the Automate master database restore option. Then restore any other
databases, if needed.
If you cannot start SQL, you can use copies of the master database and model
databases.
See “Restarting SQL using database copies” on page 177.
When a backup is restored, all changes that are made to the master database after
the last backup was created are lost. You must reapply any changes. If you created
any user databases after the master database was backed up, you cannot access
Backup Exec restore topics 177
Restarting SQL using database copies
those databases. You cannot access the databases until the databases are restored
from backups or reattached to SQL.
To restore the SQL master database from a Backup Exec image
1 Log on as Administrator.
2 Open the Backup, Archive, and Restore interface.
3 Specify the appropriate server, client, and policy type.
See “Specifying the server, client, and policy type” on page 168.
4 Select File > Select Files and Folders to Restore > from Backup Exec Backup.
5 In the Restore window, from the NetBackup History pane, select the backup
image that contains the last master database backup.
6 In the All Folders pane, select the master database.
7 Select Actions > Start Restore of Marked Files.
8 In the Restore Marked Files dialog box, on the Microsoft SQL Server tab,
select Automate master database restore.
When this option is selected, only the master database can be restored; if this
option is selected for any other database, those jobs fail. When you start the
restore operation, all existing users are logged off, and SQL Server is put into
single-user mode.
If NetBackup does not have access to the SQL registry keys
HKEY_LOCAL_MACHINE\Software\Microsoft\Microsoft SQL Server and
HKEY_LOCAL_MACHINE\Software\Microsoft\MSSQLServer, then a restore to
the default directory may not work. And the option Automate master
database restore does not work. To ensure that NetBackup has access rights,
verify that the account that NetBackup uses has administrator rights to the
computer that is running SQL.
9 Select a consistency check to be run after the restore.
10 Click Start Restore.
After the restore, SQL is restarted in multi-user mode.
Restarting SQL using database copies
If the master database is critically damaged and SQL cannot be started, you can:
replace the corrupted or missing databases with the copies of the master databases
and model databases. Backup Exec automatically creates and updates these copies
whenever backups of those databases are run. (Alternatively, restart SQL by
running the Rebuild Master utility or reinstall SQL.)
178 Backup Exec restore topics
Restarting SQL using database copies
After SQL is running again, you can restore the latest copy of the master database
and restore any other databases, if needed.
See “Restoring the SQL master database from a Backup Exec image” on page 176.
If copies of the master databases and model databases were not made, then you
must use Microsoft’s rebuildm.exe utility to rebuild the master database and
start SQL.
To restart SQL using database copies
1 Verify that the database copies are present.
The database copies are named master$4idr, mastlog$4idr, model$4idr,
and modellog$4idr.
In a default installation of SQL Server 2000, the databases are in:
C:\Program Files\Microsoft SQL Server\MSSQL\Data\*.*.
In a named instance of SQL Server 2000, the databases are in:
C:\Program Files\Microsoft SQL
Server\MSSQL$Instance_Name\Data\*.*
If necessary, restore the master database and model database copies from a
backup set to the same directory. Restore them from the original master and
model databases are in.
2 Open a command prompt window and delete the original master and model
databases and their transaction logs.
For example:
C:\Program Files\Microsoft SQL Server\MSSQL\Data> del master.mdf mastlog.ldf model.mdf modellog.ld
3 Rename the copies of the databases back to their original names.
Type the following:
C:\Program Files\Microsoft SQL Server\MSSQL\Data>rename master$4idr master.mdf
C:\Program Files\Microsoft SQL Server\MSSQL\Data>ename mastlog$4idr mastlog.ldf
C:\Program Files\Microsoft SQL Server\MSSQL\Data>rename model$4idr model.mdf
C:\Program Files\Microsoft SQL Server\MSSQL\Data>rename modellog$4idr modellog.ldf
Backup Exec restore topics 179
Restoring Backup Exec filegroup backups
4 Use the SQL Service Control Manager to start SQL Server.
5 Continue with the procedure to restore the latest changes to the master
database.
See “Restoring the SQL master database from a Backup Exec image”
on page 176.
Restoring Backup Exec filegroup backups
With filegroup backups, you can restore the entire database, a primary filegroup,
a filegroup that contains a deleted or changed table, and a nonprimary filegroup.
Note: Use separate restore jobs to restore the primary filegroup, the rest of the
filegroup backup sets, and the transaction logs.
To restore a SQL filegroup from a Backup Exec image
1 Log on as Administrator.
2 Open the Backup, Archive, and Restore interface.
3 Specify the appropriate server, client, and policy type.
See “Specifying the server, client, and policy type” on page 168.
4 Select File > Select Files and Folders to Restore > from Backup Exec Backup.
5 In the Restore window, from the NetBackup History pane, select the backup
image that contains the objects you want to restore.
6 In the All Folders pane, select the full backup of the primary filegroup and
any differential backups.
7 Select Actions > Start Restore of Marked Files.
8 In the Restore Marked Files dialog box, from the Recovery completion state
group, select Leave database nonoperational but able to restore additional
transaction logs.
9 Click Start Restore.
10 After the primary filegroup is restored, select the latest full and differential
backups for the other filegroups.
11 On the Microsoft SQL Server tab, select the recovery completion state Leave
database nonoperational but able to restore additional transaction logs,
and then start the restore job.
12 When the other filegroups are restored, select the transaction logs.
180 Backup Exec restore topics
Restoring Backup Exec Database backups
13 On the Microsoft SQL Server tab, for the recovery completion state, select
Leave database operational and restore all of the transaction logs.
14 Optionally, you can also select Point in time log restore or Restore log up to
named transaction.
15 Click Start Restore.
Restoring Backup Exec Database backups
If the database you want to restore uses the simple recovery model, there are no
transaction log backups to restore. You only need to restore the most recent full
database backup and if you were running differential database backups, restore
the most recent differential database backup.
To restore a SQL database from a Backup Exec backup
1 Log on as Administrator.
2 Open the Backup, Archive, and Restore interface.
3 Specify the appropriate server, client, and policy type.
See “Specifying the server, client, and policy type” on page 168.
4 Select File > Select Files and Folders to Restore > from Backup Exec Backup.
5 In the Restore window, in the NetBackup History pane, select the backup
image that contains the objects you want to restore.
6 Select the most recent full database backup and the most recent differential
database backup to restore.
7 Select Actions > Start Restore of Marked Files.
8 In the Restore Marked Files dialog box, from the Recovery completion state
group, select Leave database operational. No additional transaction logs
can be restored.
9 Click Start Restore.
Chapter 12
Using NetBackup for SQL
Server with database
mirroring
This chapter includes the following topics:
■ About NetBackup for SQL Server with database mirroring
■ Configuring NetBackup to support database mirroring
■ Initiating simultaneous backups for mirrored partners
■ Browsing for mirrored database backup images
About NetBackup for SQL Server with database
mirroring
Database mirroring is a software solution introduced in SQL Server 2005, service
pack 1, to increase the availability of a SQL Server database. It uses two database
instances (normally on different hosts), which contain copies of the same SQL
Server database. These databases are identical in both name and content. The
copies are the principal and the mirror. The mirror serves as a hot standby to the
principal, where transactions take place. The mirror is very closely synchronized
with the principal through transaction log porting. It is immediately available in
case the principal fails.
The primary consideration when you establish your backup and restore procedures
for database mirroring is that these operations are only available on the principal
database.
182 Using NetBackup for SQL Server with database mirroring
Configuring NetBackup to support database mirroring
For a complete description of database mirroring refer to the SQL Server 2005
Books online.
Configuring NetBackup to support database mirroring
To use database mirroring with NetBackup, both the principal and the mirror
should be set up as clients of the same master server.
To configure NetBackup to support database mirroring
1 The hosts that contain both databases should specify the same master server
in their server lists.
2 Any policy that is used to back up the principal should also specify the host
that contains the mirror database.
3 Permission for a redirected restore must be established between both
mirroring partners.
To establish redirected restore permission between two hosts
To establish redirected restore permission between two hosts, do one of the
following:
1 On the master server, create a file called:
install_path\NetBackup\db\altnames\No.Restrictions
2 Create each of the files on the master server:
install_path\NetBackup\db\altnames\hostA
install_path\NetBackup\db\altnames\hostB
Note: When you create the No.Restrictions file it allows all clients to perform
redirected restores to different clients. This file may need to be added and removed
according to your site policies.
Initiating simultaneous backups for mirrored partners
Since backups can occur only on the principal, you must take steps to ensure that
you don't miss any scheduled backups due to failover. Establish a procedure to
simultaneously initiate backups for both partners, but suppress the operation on
the mirror.
To achieve this objective, first create a backup schedule for the principal.
Then take the following steps:
Using NetBackup for SQL Server with database mirroring 183
Browsing for mirrored database backup images
To simultaneously initiate backups for both partners
1 Add the host that contains the mirroring partner to the client list of the policy
that you created for the principal.
2 Create a batch file on the mirroring partner that has the same name as the
batch file specified in the backup selections policy.
The batch file on the mirroring partner should be identical to the one used
on the principal, with one exception. The value for SQLHOST and
SQLINSTANCE are different.
When you follow these steps, it ensures that any scheduled backup for a
mirrored database is launched on both the principal and the mirror. The
backup on the principal should run normally. The backup on the mirror is
inhibited. You may notice that the NetBackup for SQL Server progress log
for the inhibited operation specifies a message similar to the following:
Database ACCOUNTING is in a mirroring role on SQL Server
instance SUN. This operation must be performed on SQL Server
instance MOON.
Browsing for mirrored database backup images
When you restore a mirrored database, you must run the NetBackup SQL Client
from the host that contains the mirroring partner. The mirroring partner is in
the principal role. See SQL Server 2005 Books Online for information on how to
determine which partner is the principal.
Backup images may have been created on either or both of the mirroring partners.
However, they appear as a unified tree view in the Restore Database dialog box
irrespective of the partner from which they had been backed up. In addition, the
recovery script that is created for the database is created from images from both
partners as appropriate. To determine which partner the backup was taken from,
look at the property page for the image.
To view backup images you may look on the hosts that contains either of the
mirroring partners, provided that backups were done for that partner.
For example, suppose that mirroring partners are as follows:
■ Principal
Host name: HostA
SQL Server instance: Solaria
Database: Accounting
■ Mirror
184 Using NetBackup for SQL Server with database mirroring
Browsing for mirrored database backup images
Host name: HostB
SQL Server instance: Moonbeam
Database: Accounting
Moreover, suppose that all of the backups were done on HostB, though the principal
is currently on HostA. To browse backup images for the Accounting database, take
the following steps:
To browse for backup images
1 On HostA, open the NetBackup SQL Client.
2 From the File menu, select Restore SQL Server Objects.
3 In the Backup History Options dialog box, select HostB in the SQL Host
dropdown.
4 Click OK.
Note: In this example, if backup images had been created exclusively on HostA
or on both HostA and HostB, then all of the images from both partners would
be viewable. They are viewable by selecting HostA in the SQL Host dropdown.
You can browse and restore the mirrored database and expand the Accounting
database as usual. Restore operations that are created within a recovery script
are created with the redirected script. The redirected script is used as
necessary to the extent that any such operation identifies an image backed
up from the mirroring partner.
When you are ready to restore a mirrored database, you must remove the
mirroring attribute and then restore the database as usual.
Chapter 13
Troubleshooting
This chapter includes the following topics:
■ About progress reports created for NetBackup for SQL Server on the client
■ About sample progress report for NetBackup for SQL Server backup
■ About debug logging for SQL Server troubleshooting
■ Creating all NetBackup debug logs for SQL Server troubleshooting
■ About backup operation debug logging for SQL Server
■ About restore operation debug logging for SQL Server
■ Setting the maximum trace level for NetBackup for SQL Server
■ About NetBackup reports for SQL Server troubleshooting
■ About minimizing timeout failures on large SQL Server database restores
■ Troubleshooting VMware SQL Server jobs
About progress reports created for NetBackup for
SQL Server on the client
NetBackup for SQL Server creates a progress report for each operation that has
been initiated. This report contains summary information about the overall status
of your job. The reports are contained in directory
install_path\NetBackup\logs\user_ops\MsSql\logs and can be viewed by
choosing File > View status in the NetBackup Database Client graphical user
interface .
The progress report contains the following types of information:
■ The batch keywords and values which define the operation
186 Troubleshooting
About sample progress report for NetBackup for SQL Server backup
Information about the batch file syntax is available.
■ Summary information about the operation
■ Information about the operation as it progresses
■ Any error conditions or warnings that cause the operation to fail
■ The final outcome of the operation, whether it succeeded or failed and how
long it took
About sample progress report for NetBackup for SQL
Server backup
Table 13-1 shows a typical progress report that is created for a database backup.
Table 13-1 Progress report for NetBackup for SQL Server backup
Line Text
1 OPERATION backup
2 Database "DatabaseA"
3 SQLHOST "JUY"
4 SQLINSTANCE "NEWINSTANCE"
5 NBSERVER "JUY"
6 MAXTRANSFERSIZE 0
7 BLOCKSIZE 0
8 ENDOPER TRUE
9 INF - BACKUP started using
10 Microsoft SQL Server Yukon - 9.00.852 (Intel X86)
11 Jul 19 2004 22:09:12
12 Copyright (c) 1988-2003 Microsoft Corporation
13 Beta Edition on Windows NT 5.1 (Build 2600: Service Pack 2)
14 Batch = C:\Program
Files\Veritas\NetBackup\dbext\mssql\temp\__09_42_24_076_00.bch,
Op# = 1
Troubleshooting 187
About sample progress report for NetBackup for SQL Server backup
Table 13-1 Progress report for NetBackup for SQL Server backup (continued)
Line Text
15 INF - Using backup image
juy.MSSQL7.JUY\NEWINSTANCE.db.DatabaseA.~.7.001of001.20060701094227..C
16 INF - backup database "DatabaseA" to
VIRTUAL_DEVICE='VNBU0-5652-1224-1120228947' with stats = 10,
blocksize = 65536, maxtransfersize = 65536, buffercount = 1
17 INF - Number of stripes: 1, Number of buffers per stripe 1.
18 12:41:07 Initiating backup
19 12:41:10 INF - Starting bpbrm
20 12:41:12 INF - Data socket = juy.domain.com.4146
21 12:41:12 INF - Name socket = juy.domain.com.1527
22 12:41:12 INF - Job ID = 143
23 12:41:12 INF - Backup ID = juy_1058982070
24 12:41:12 INF - Backup time = 1058982070
25 12:41:12 INF - Policy name = SQL
26 12:41:12 INF - Snapshot = 0
27 12:41:12 INF - Frozen image = 0
28 12:41:12 INF - Backup copy = 0
29 12:41:12 INF - Master server = juy
30 12:41:12 INF - Media server = juy
31 12:41:12 INF - Multiplexing = 0
32 12:41:12 INF - New data socket = juy.domain.com.1388
33 12:41:12 INF - Use shared memory = 1
34 12:41:12 INF - Compression = 0
35 12:41:12 INF - Encrypt = 0
36 12:41:12 INF - Client read timeout = 300
37 12:41:12 INF - Media mount timeout = 0
188 Troubleshooting
About debug logging for SQL Server troubleshooting
Table 13-1 Progress report for NetBackup for SQL Server backup (continued)
Line Text
38 12:41:16 INF - Data buffer size = 262144
39 12:41:18 INF - Beginning backup on server juy of client juy
40 INF - Thread has been closed for stripe #0
41 12:41:25 INF - Server status = 0
42 12:41:26 INF - Backup by hao on client juy using policy SQL: the
requested operation was successfully completed.
Observe the following:
■ Lines 1 to 8 contain the batch syntax which drove this operation. This operation
was started from a batch file that the GUI created.
■ Line 9 indicates that the operation was a backup ("Dump"). The term ‘Load’
would indicate a restore.
■ Lines 10-13 provide the version of SQL Server and the Windows operating
system.
■ Line 14 provides the name of the batch file. Since the backup was launched
immediately, it was placed in install_path\NetBackup\DbExt\MsSql\Temp
directory.
■ Line 16 provides the actual SQL syntax that was used to launch the backup
command.
■ Line 17 indicates that this backup was a single stream backup.
■ Lines 18 to 42 provide NetBackup-based client parameters and statuses. Note
that the server status on line 40 indicates that the operation was completed
with status 0, which indicates success.
About debug logging for SQL Server troubleshooting
The NetBackup master server and client software offers a comprehensive set of
debug logs for troubleshooting the problems that can occur during NetBackup
operations. Debug logging is also available for SQL backup and restore operations.
After the cause of the problem is determined, you can disable debug logging.
You can control the amount of information that is written to debug logs.
For details on the contents of these debug logs, see the NetBackup Troubleshooting
Guide.
Troubleshooting 189
Creating all NetBackup debug logs for SQL Server troubleshooting
For additional NetBackup client logs and NetBackup master server logs, see the
online Help for the Backup, Archive, and Restore interface and the NetBackup
Administrator’s Guide, Volume I.
Creating all NetBackup debug logs for SQL Server
troubleshooting
You can use the following procedure to create all NetBackup debug logs.
To create all debug logs
◆ Run the following batch file:
install_path\NetBackup\logs\mklogdir.bat
See “About backup operation debug logging for SQL Server” on page 189.
See “About restore operation debug logging for SQL Server” on page 189.
About backup operation debug logging for SQL Server
The following logs pertain to back up operations:
install_path\NetBackup\logs\bphdb (scheduled backups only)
install_path\NetBackup\logs\dbclient
install_path\NetBackup\logs\bpbkar (Snapshot Client)
install_path\NetBackup\logs\bpfis (Snapshot Client)
Log names are formatted as mmddyy.log, where mm is the month, dd is the day,
and yy is the year.
About restore operation debug logging for SQL Server
The following logs pertain to restore operations:
install_path\NetBackup\logs\dbclient
install_path\NetBackup\logs\bpbkar (Snapshot Client)
install_path\NetBackup\logs\bpfis (Snapshot Client)
install_path\NetBackup\logs\bppfi (instant recovery)
190 Troubleshooting
Setting the maximum trace level for NetBackup for SQL Server
Log names are formatted as mmddyy.log, where mm is the month, dd is the day,
and yy is the year.
Setting the maximum trace level for NetBackup for
SQL Server
You can set the maximum trace level in the NetBackup SQL Client GUI or in the
batch file.
To set the maximum trace level in the NetBackup SQL Client GUI
1 Open the NetBackup SQL Client.
2 Select File > Set NetBackup client properties.
3 In the Client Trace Level group, select Maximum.
To set the maximum trace level in the batch file
1 Open the NetBackup SQL Client.
2 Select File > Manage script files.
3 Select the batch file you want to change and click Open File.
4 Add the following line:
TRACELEVEL MAX
5 Save the file.
About NetBackup reports for SQL Server
troubleshooting
The administrator has access to operational progress reports through
administrator interfaces. Reports may be generated for following: Backup Status,
Client Backups, Problems, All Log Entries, Media Lists, Media Contents, Images
on Media, Media Logs, Media Summary, and Media Written. These reports may
be generated for a specific time frame, client, or master server. See the NetBackup
Administrator’s Guide, Volume I for details.
About minimizing timeout failures on large SQL Server
database restores
A large SQL Server restore may fail with a Client Read Timeout error before any
data has been read from the NetBackup media. This error occurs because the SQL
Troubleshooting 191
Troubleshooting VMware SQL Server jobs
Server may need to pre-write the database files before the restore operation begins.
The time that is required for this process is a function of certain factors: the size
of the database files and the speed at which your host machine can write to disk.
For example, consider that your system can perform disk writes at the rate of 60
megabytes per second and you have a 2.4 terabyte database. Then it takes at least
12 hours for SQL Server to prep the disk before the actual restore can begin. In
reality, the delay may be even longer than what you calculate by as much as 20%
to 40%.
The timeout problem can be resolved by increasing the NetBackup Client Read
Timeout setting. Use the NetBackup Administration Console on the server to
change the properties of each client that contains a database you may need to
restore. The default for the Client Read Timeout setting is 300 seconds (5 minutes).
If you have any clients which contain large SQL Server databases, you may need
to set this value much higher.
If you use SQL Server 2005 or later, you can eliminate file initialization during
SQL Server restores.
See “About instant data file initialization” on page 52.
Troubleshooting VMware SQL Server jobs
Note the following when you perform VMware jobs:
■ Failure of an Application State Capture (ASC) job does not prevent a VMware
backup from continuing.
■ The ASC job can result in status 1. (For example, if there is a dismounted
database.)
■ Failure results in the discovery job or parent job exiting with status 1.
■ ASC messages are filtered to the ASC job details.
■ A Status 0 is returned if no applications are installed in guest virtual machine.
■ If neither the Symantec nor the VMware VSS Provider is installed at the time
of backup, the SQL Server databases are unquiesced.
192 Troubleshooting
Troubleshooting VMware SQL Server jobs
Chapter 14
Sample batch files
This chapter includes the following topics:
■ About sample batch files
About sample batch files
You can use batch files to initiate backup and restore operations. The batch file
examples show you how to perform a variety of tasks with NetBackup for SQL
Server batch files.
The following samples of batch files are available:
■ See “About NetBackup for SQL Server–Simple script to back up a database
named BUSINESS” on page 194.
■ See “About NetBackup for SQL Server–Simple script to restore a database
named pubs” on page 194.
■ See “About NetBackup for SQL Server–Perform a striped database backup and
allow multiple internal buffers per stripe” on page 195.
■ See “About NetBackup for SQL Server–Restore a database from multiple
stripes” on page 195.
■ See “ About NetBackup for SQL Server–Restore a database transaction log up
to a point in time” on page 196.
■ See “About NetBackup for SQL Server–Perform an operation and specify the
user ID and password to use to SQL Server” on page 196.
■ See “About NetBackup for SQL Server–Perform multiple operations in
sequence” on page 197.
■ See “About NetBackup for SQL Server–Perform a set of operations in parallel”
on page 198.
194 Sample batch files
About sample batch files
■ See “About NetBackup for SQL Server–Specify the maximum transfer size and
block size for a backup” on page 200.
■ See “About NetBackup for SQL Server–Stage a database restore from a database
backup, a differential backup, and a series of transaction backups” on page 200.
■ See “About NetBackup for SQL Server–Stage a database restore from a filegroup
backup, several file backups, and transaction log backups” on page 202.
■ See “About NetBackup for SQL Server–Using environment variables to exclude
instances and databases from backup” on page 205.
About NetBackup for SQL Server–Simple script to back up a database
named BUSINESS
Certain default values define the parameters for this operation. For example, there
is one backup stripe, minimum trace level, and the object type is a database (as
opposed to a transaction log).
OPERATION BACKUP
DATABASE "BUSINESS"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
ENDOPER TRUE
About NetBackup for SQL Server–Simple script to restore a database
named pubs
This sample restores a database that is called pubs, based upon the following
backup:
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.db.pubs.~.7.001of001.20060628123631..C"
To find out which backups you can restore, look at the dbclient log file created
when you did the backup or by use bplist.
See “About using bplist to retrieve SQL Server backups” on page 122.
OPERATION RESTORE
OBJECTTYPE DATABASE
DATABASE "pubs"
# The following image is type: Full
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.db.pubs.~.7.001of001.20060628123631..C"
Sample batch files 195
About sample batch files
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE RECOVERED
ENDOPER TRUE
About NetBackup for SQL Server–Perform a striped database backup
and allow multiple internal buffers per stripe
This example backs up the BUSINESS database using four data streams. Each data
stream uses two buffers.
OPERATION BACKUP
DATABASE "BUSINESS"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
STRIPES 4
NUMBUFS 2
MAXTRANSFERSIZE 0
BLOCKSIZE 7
ENDOPER TRUE
About NetBackup for SQL Server–Restore a database from multiple
stripes
For a striped restore, you must specify the number of stripes and the name of the
first backup image name. Notice that the backup image in this example is embedded
with the string .001of004, which indicates that it is the first of four backups.
OPERATION RESTORE
OBJECTTYPE DATABASE
DATABASE "Northwind"
NBIMAGE cadoo.MSSQL7.CADOO.db.Northwind.~.0.001of004.20060216151937..C
STRIPES 004
MAXTRANSFERSIZE 0
BLOCKSIZE 0
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
196 Sample batch files
About sample batch files
NBSERVER "CHISEL"
BROWSECLIENT "CADOO"
RECOVEREDSTATE RECOVERED
ENDOPER TRUE
About NetBackup for SQL Server–Restore a database transaction log
up to a point in time
This script is executed after the database is restored. The database is restored to
the specified point in time (Feb 16, 2006 at 2:03:00 P.M.). This time precedes the
date of the backup log (Feb 16, 2006 at 2:03:21 P.M.).
Note the following:
■ If STOPAT is not specified, then the database is restored to the date of the backup
log.
■ You do not need to manually stage the restoration of the database backup and
the associated log files. Create the script in the Restore Microsoft SQL Server
Objects dialog box.
■ Since RECOVEREDSTATE was not specified, the database is restored to a recovered
state following successful execution of this script.
OPERATION RESTORE
OBJECTTYPE TRXLOG
STOPAT 20060216/14:03:00
DATABASE Northwind
NBIMAGE cadoo.MSSQL7.CADOO.trx.Northwind.~.0.001of001.20060216140321..C
MAXTRANSFERSIZE 0
BLOCKSIZE 0
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
BROWSECLIENT "CADOO"
ENDOPER TRUE
About NetBackup for SQL Server–Perform an operation and specify
the user ID and password to use to SQL Server
Only specify a user ID and password if you use standard SQL Server security.
See “About SQL Server privileges” on page 46.
OPERATION BACKUP
DATABASE "BUSINESS"
Sample batch files 197
About sample batch files
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
USERID JSMITH
PASSWORD my.Pwd
ENDOPER TRUE
About NetBackup for SQL Server–Perform multiple operations in
sequence
In this sample batch file, five separate backups are performed sequentially.
Remember that each operation is required to be completely specified.
OPERATION BACKUP
DATABASE "BUSINESS"
OBJECTTYPE DATABASE
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
STRIPES 5
ENDOPER TRUE
OPERATION BACKUP
DATABASE "RECREATION"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
OBJECTTYPE TRXLOG
ENDOPER TRUE
OPERATION BACKUP
DATABASE "EDUCATION"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
198 Sample batch files
About sample batch files
STRIPES 2
ENDOPER TRUE
OPERATION BACKUP
DATABASE "GOVERNANCE"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
OBJECTYPE TRXLOG
ENDOPER TRUE
OPERATION BACKUP
DATABASE "SURVIVAL"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
OBJECTYPE TRXLOG
ENDOPER TRUE
About NetBackup for SQL Server–Perform a set of operations in parallel
This sample is identical to the following sample except that the first operation
contains BATCHSIZE 3.
See “About NetBackup for SQL Server–Perform multiple operations in sequence”
on page 197.
This setting tells NetBackup to start the first three operations in parallel. After
these are completed, NetBackup then begins the next set of 3. In this case, since
there are five operations, the second batch set contains two operations.
BATCHSIZE 3
OPERATION BACKUP
DATABASE "BUSINESS"
OBJECTTYPE DATABASE
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
Sample batch files 199
About sample batch files
STRIPES 5
ENDOPER TRUE
OPERATION BACKUP
DATABASE "RECREATION"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
OBJECTTYPE TRXLOG
ENDOPER TRUE
OPERATION BACKUP
DATABASE "EDUCATION"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
STRIPES 2
ENDOPER TRUE
OPERATION BACKUP
DATABASE "GOVERNANCE"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
OBJECTYPE TRXLOG
ENDOPER TRUE
OPERATION BACKUP
DATABASE "SURVIVAL"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
OBJECTYPE TRXLOG
ENDOPER TRUE
200 Sample batch files
About sample batch files
About NetBackup for SQL Server–Specify the maximum transfer size
and block size for a backup
This sample batch file backs up database "business" with a maximum transfer
size of 64 kilobytes bytes * 24 (1M). The maximum block size is 512 bytes * 26 (32
kilobytes).
OPERATION BACKUP
DATABASE "BUSINESS"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "CHISEL"
MAXTRANSFERSIZE 4
BLOCKSIZE 6
ENDOPER TRUE
About NetBackup for SQL Server–Stage a database restore from a
database backup, a differential backup, and a series of transaction
backups
This example shows a script that you generate in the Restore Microsoft SQL
Server Objects dialog box.
OPERATION RESTORE
OBJECTTYPE DATABASE
DATABASE "DatabaseA"
# The following image is type: Full
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.db.DatabaseA.~.7.001of001.20060701094227..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
OBJECTTYPE DATABASE
DUMPOPTION INCREMENTAL
DATABASE "DatabaseA"
# The following image is type: Full database differential
Sample batch files 201
About sample batch files
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.inc.DatabaseA.~.7.001of001.20060701103323..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
OBJECTTYPE TRXLOG
DATABASE "DatabaseA"
# The following image is type: transaction log
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.trx.DatabaseA.~.7.001of001.20060701090005..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
OBJECTTYPE TRXLOG
DATABASE "DatabaseA"
# The following image is type: transaction log
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.trx.DatabaseA.~.7.001of001.20060701100030..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
OBJECTTYPE TRXLOG
202 Sample batch files
About sample batch files
DATABASE "DatabaseA"
# The following image is type: transaction log
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.trx.DatabaseA.~.7.001of001.20060701110015..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
About NetBackup for SQL Server–Stage a database restore from a
filegroup backup, several file backups, and transaction log backups
This example shows a script for a full database restore that you generate in the
Restore Microsoft SQL Server Objects dialog box.
OPERATION RESTORE
OBJECTTYPE FILEGROUP
DATABASE "DatabaseR"
OBJECTNAME "PRIMARY"
# The following image is type: Filegroup
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.fg.DatabaseR.PRIMARY.7.001of001.20060701095634..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
OBJECTTYPE FILEGROUP
DATABASE "DatabaseR"
OBJECTNAME "DBR_FG2"
# The following image is type: Filegroup
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.fg.DatabaseR.DBR_FG2.7.001of001.20060701095425..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
Sample batch files 203
About sample batch files
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
OBJECTTYPE FILE
DATABASE "DatabaseR"
OBJECTNAME "DBR_FG1_File1"
# The following image is type: File
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.fil.DatabaseR.DBR_FG1_File1.7.001of001.
20060701100824..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
OBJECTTYPE FILE
DATABASE "DatabaseR"
OBJECTNAME "DBR_FG1_File2"
# The following image is type: File
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.fil.DatabaseR.DBR_FG1_File2.7.001of001.
20060701100908..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
204 Sample batch files
About sample batch files
OBJECTTYPE FILE
DATABASE "DatabaseR"
OBJECTNAME "DBR_FG1_File3"
# The following image is type: File
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.fil.DatabaseR.DBR_FG1_File3.7.001of001.
20060701100953..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
OBJECTTYPE TRXLOG
DATABASE "DatabaseR"
# The following image is type: transaction log
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.trx.DatabaseR.~.7.001of001.20060701100030..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
RECOVEREDSTATE NOTRECOVERED
ENDOPER TRUE
OPERATION RESTORE
OBJECTTYPE TRXLOG
DATABASE "DatabaseR"
# The following image is type: transaction log
NBIMAGE "cadoo.MSSQL7.CADOO\SECOND.trx.DatabaseR.~.7.001of001.20060701110015..C"
SQLHOST "CADOO"
SQLINSTANCE "SECOND"
NBSERVER "BOW"
BROWSECLIENT "CADOO"
MAXTRANSFERSIZE 0
BLOCKSIZE 7
RESTOREOPTION REPLACE
Sample batch files 205
About sample batch files
RECOVEREDSTATE RECOVERED
ENDOPER TRUE
About NetBackup for SQL Server–Using environment variables to
exclude instances and databases from backup
You can use SQLINSTANCE $ALL in your batch file to designate that all SQL Server
instances on your host be backed up. For example, the following batch file backs
up the master, model, and msdb databases. These databases are backed up on all
instances of SQL Server on the host on which the batch file is run.
SQLINSTANCE $ALL
OPERATION BACKUP
DATABASE "master"
NBSERVER "BEARING"
MAXTRANSFERSIZE 6
BLOCKSIZE 7
NUMBUFS 2
ENDOPER TRUE
OPERATION BACKUP
DATABASE "msdb"
NBSERVER "BEARING"
MAXTRANSFERSIZE 6
BLOCKSIZE 7
NUMBUFS 2
ENDOPER TRUE
OPERATION BACKUP
DATABASE "model"
NBSERVER "BEARING"
MAXTRANSFERSIZE 6
BLOCKSIZE 7
NUMBUFS 2
ENDOPER TRUE
To exclude SQL Server instances on your host from backup, create the Windows
environmental variable NB_SQL_INSTANCE_EXCLUDE. Specify a list of instances
names that you want to exclude. The list should consist of one or more names
that are separated by semi-colons.
For example, use the following value to indicate that you want to exclude the
default SQL Server instance and the instance named ABC-PRODUCTS from backup:
206 Sample batch files
About sample batch files
#DEFAULT#;ABC-PRODUCTS;
Note that the default SQL Server instance for the local host is designated as
#default#.
You can also exclude individual databases from backup by creating a Windows
environmental variable NB_SQL_DATABASE_EXCLUDE. For the value of the
variable, specify a list of database names.
For example, consider the following batch file:
SQLINSTANCE $ALL
OPERATION BACKUP
DATABASE $ALL
NBSERVER "BEARING"
MAXTRANSFERSIZE 6
BLOCKSIZE 7
NUMBUFS 2
ENDOPER TRUE
You can exclude the databases "master," "accounting," and "pubs" with the
NB_SQL_DATABASE_EXCLUDE environmental variable. For the value of the
variable, indicate the databases you want to exclude. Separate the database names
with semi-colons.
MASTER;ACCOUNTING;PUBS
The NB_SQL_DATABASE_EXCLUDE variable is applicable only for a batch file
that has DATABASE $ALL. It performs the same function as the keyword and
value pair EXCLUDE <database>. If both variables are used, they augment each
other to determine which databases to exclude.
Index
A backups (continued)
alternate buffer method 51 offhost 158
alternate client 158 SQL Server login 47
application backup schedule SQL Server security restrictions 48
overview 39 transaction logs 76
authorization backups, automatic
for scheduled operations 48 running manually 42
automatic backup schedule batch files
overview 39 BUFFER keyword 109
automatic retry of unsuccessful backups 118 defined in backup selections list 40
for backup operations 116
samples 194, 206
B specifying object to back up 110
Backup Exec STRIPES keyword 115
importing images 167 when required 103
redirected restores 175 BLOCKSIZE 50
restores bpbkar log 189
client. See specifying bpfis log 189
databases 180 bphdb log 189
filegroups 179 bplist
multiple databases 168 client parameter 122
options 169 server parameter 122
policy type. See specifying stripenumber 124
server. See specifying total stripes 124
SQL Master Database 176 using to retrieve list of backups 122
tables 168 bppfi log 189
transaction logs 174–175 buffer space parameters
supported SQL versions 168 BLOCKSIZE 50
backup media required 29 MAXTRANSFERSIZE 50
Backup Selections list NUMBER OF BUFFERS PER STRIPE 50
adding selections 40 buffers
overview 40 alternate buffer method 51
backups and transfer of data to NetBackup server 50
automatic retry of unsuccessful 118
browsing for 81
database filegroups 57, 77 C
database files 79 clients list, for backup policies 39
databases 56, 75 compatibility information 28
image names copying databases. See move operations.
colon and backslash in 123
in a Microsoft Cluster Server 137
in a Veritas Cluster Server 137
208 Index
D login parameters
database login parameters database
setting 71 setting 71
dbbackex 104 logs
and client-based schedulers 70 debug 188
dbclient log 189
DBMS (Database Management System) privileges 46 M
debug logs 188 manual backups
differential backups in a SAP environment 161
described 58 maximum jobs per client 34
disaster recovery MAXTRANSFERSIZE 50
of databases 133 Media multiplexing 43
overview 132 Microsoft Cluster Server (MSCS)
rebuilding the master database 133 configuring the NetBackup Server 136
starting SQL Server in single-user mode 133 SQL Server backups in 137
SQL Server restores in 138
E using with NetBackup for SQL Server 30
Enable Multiplexing 43 VIRTUALSERVER setting 136
MOVE keyword 109
move operations 91
F multi-interface network connections
file initialization 52 and clusters 138, 144
filegroups configuration in cluster environments 44, 139
protecting 58 configuration in clustered environments 46
recovery of 59 configuration in non-clustered
files environments 44, 46
protecting 58 overview
recovery of 59 for non-clustered environments 44
overview, for clustered environments 138
I specifying the public name of the client 46
installation multiple copies feature 38
adding a license key 31 multiplexed backups 43
prerequisites in a cluster 30 multiplexing
requirements for NetBackup clients 29 and performance 51
requirements for NetBackup servers 29
integrated security 46–47 N
NetBackup Client Service 47–48
J NetBackup for Microsoft SQL Server
Java interface 34 starting 71–72
NetBackup Server Scheduler 104
L network interface 81, 100
NUMBER OF BUFFERS PER STRIPE 50
license keys, adding 31
log shipping 53
login O
SQL Server ODBC data source names 107
for scheduled operations 47 offhost backups 158
Index 209
P restores (continued)
page verification 52 databases 86, 128
page-level restores 96 filegroup differentials 89
partial backups 80 filegroups 88
partial database restores 94 in a Microsoft Cluster Server 138
policy configuration in a SAP environment 161
adding clients 39 in a Veritas Cluster Server 138
attributes 36 page-level 96
backup selections list 40 R/3 database 164
for databases 36, 127 read-only filegroups 90
overview 35 redirecting. See redirected restores.
schedules 37 SQL Server login 47
testing 42 transaction logs
preventing timeout failures 191 within a staged recovery 87
progress reports 190 without staging a full recovery 91
protecting files and filegroups 58
S
R SAP environment
R/3 database backups 161–162
restoring 164 policy configuration 162
read-only filegroups scripts used 162
and reducing backup size 60–61 manual backups 161
backups of 53, 77 restores 161
restoring 90 transaction log backups 163
viewing backup sets 78 schedulers
read-write filegroups client-based
backups 78 and dbbackex 70
using to restore databases 89 schedules
recovering read-write backups 89 adding 37
recovery frequency 38
considerations for files and filegroups 59 properties 37
factors 61, 65 types of schedules 37
of a full database 87 security
staging 64 integrated 46–47
transaction logs 62 SQL Server restrictions 48
redirected restores standard SQL Server security 47, 49
of databases shared memory
to different locations 99 and performance 51
to different hosts 98 Source Client 81
reducing backup size 60 specifying a network interface 81, 100
remote folder button 41 SQL Host
remote hosts, browsing 120 selecting server that performed a backup 81
reports, progress 190 to login to 71
restores SQL instance
and striping 120 to login to 71
browsing for backup images 81 SQL Server login
database differentials 86, 128 for scheduled operations 47
database files 90 standard security 47
database move. See move operations. standby server 53
210 Index
STRIPES keyword 115
striping
and automatic restore 120
and number of buffers 109
and performance 51
configuring backup policy for 43
maximum jobs per client 34
T
testing policy configuration 42
timeout failures
preventing 191
troubleshooting 188
U
Use alternate client 158
V
Veritas Cluster Server (VCS)
configuring the NetBackup Server 136
SQL Server backups in 137
SQL Server restores in 138
VIRTUALSERVER 136
VMware
SQL Server backups 125
W
Windows interface 34