==========================
Configuration Management
==========================
=> Installing required softwares in the machines
=> Copy required files from one machine to another machine
=> OS Patching/Updates
=> We can perform configuration management in 2 ways
1) Manual Configuration Management
2) Automated Configuration Management
============================================
Problems with Manual Configuration Mgmt
============================================
1) Time Taking process
2) Repeated Work
3) Human Mistakes
Note: To overcome these problem we are going to automate configuration management
in the project.
=> To automate configuration management we have several tools
1) puppet
2) chef
3) Ansible (Trending)
================
What is Ansible
================
-> It is an open source software developed by Michael DeHaan and its ownership is
under RedHat.
=> Ansible was written in Python language.
-> Ansible is an automation tool that provides a way to define configuration as
code.
=======================
What Ansible can do ?
=======================
1) Automate Configuration Management
2) App Deployments
=======================
Ansible Architecture
=======================
1) Control Node
2) Managed Nodes
3) Host Inventory File
4) Playbooks
=> The machine which contains ansible software is called as Controlling Node.
=> The machines which are managing by Controlling Node are called as Managed Nodes.
=> Host inventory file contains managed nodes information.
=> Playbook is a YML/YAML which contains set of taks.
===============
Ansible Setup
===============
URL : https://github.com/ashokitschool/DevOps-Documents/blob/main/11-Ansible-
Setup.md
===========================
Ansible Ad-Hoc Commands
===========================
=> To run ad-hoc commands we will follow below syntax
Syntax :
$ ansible [all/group-name/private-ip] -m <module> -a <args>
Ex: $ ansible all -m ping
$ ansible webservers -m ping
$ ansible dbservers -m ping
=> We have several modules in ansible to perform configuration management
1) ping
2) shell
3) yum / apt
4) service
5) copy
$ ansible all -m ping
$ ansible all -m shell -a date
$ ansible all -m yum -a "name=git"
$ ansible webservers -m yum -a "name=httpd"
===========
Assignment
===========
=> Ansible Modules and purpose
=> What is YML
===================
Ansible Playbooks
===================
=> Playbook is a YAML file
=> Playbook contains one or more tasks
=> Using playbook we can define what tasks to performed and where to be performed.
=> We will give playbook as input for ansible control node to perform tasks in
managed nodes.
Note: To write Ansible playbooks, we should learn YAML first.
================
YML or YAML
================
=> YML/YAML stands for Yet another markup language.
=> It is used to store the data in human & machine readable format.
=> YML/YAML files will have extension as .yml or .yaml
-> Official Website : https://yaml.org/
============================
01 - Sample YML file data
============================
Note: indent spacing is very important
---
id: 101
name: Ashok
gender: Male
hobbies:
- music
- chess
- cricket
...
===========================
02 - Sample YML file data
===========================
---
person:
id: 101
name: Ashok
address:
city: hyd
state: TG
country: India
hobbies:
- cricket
- chess
- music
...
=============================
Write YML file to represent employee data with company and job details.
emp -> id, name, company and job
company -> name
job -> exp, type (permanent | contract)
---
emp:
id: 101
name: Ashok
company:
name: Microsoft
job:
exp: 11 Years
type: permanent
...
#### Website To validate YML syntax : https://www.yamllint.com/
################### Use VS Code IDE to write YML Files ######################
==================
Writing Playbooks
==================
=> Playbook contains 3 sections
1) Host Section
2) Variable Section
3) Task Section
=> Host Section Represents target machines to execute tasks.
=> Variables Section is used to declare variables required for playbook execution.
=> Task section is used to define what operations we want to perform using Ansible.
Note: In single playbook we can specify multiple tasks also.
=> To execute playbook we will use below syntax
$ ansible-playbook <playbook-yml-file>
=================================
Playbook to ping managed nodes
=================================
---
- hosts: all
tasks:
- name: ping all managed nodes
ping:
...
# It will check the syntax of a playbook
$ ansible-playbook <playbook-yml-file> --syntax-check
# It will display which hosts would be effected by a playbook before run
$ ansible-playbook <playbook-yml-file> --list-hosts
# Run the playbook Using below command
$ ansible-playbook <playbook-yml-file>
# It execute one-step-at-a-time, confirm each task before running with
(N)o/(y)es/(c)ontinue
$ ansible-playbook <playbook-yml-file> --step
# Run the playbook in verbose mode
$ ansible-playbook <playbook-yml-file> -vvv
====================
---
- hosts: all
tasks:
- name: create a file
file:
path: /home/ansible/ashokit.txt
state: touch
...
========================
---
- hosts: all
tasks:
- name: copy content to file
copy: content="welcome to ashokit\n" dest="/home/ansible/ashokit.txt"
...
===========================
---
- hosts: webservers
become: true #use it if you need sudo priviliges
tasks:
- name: install httpd package
yum:
name: httpd
state: latest
- name: copy index.html file
copy:
src: index.html
dest: /var/www/html/index.html
- name: start httpd service
service:
name: httpd
state: started
...
=================
Handlers & Tags
=================
-> In playbook all tasks will be executed by default in sequential order.
=> Using Handlers we can execute tasks based on other tasks status.
Note: If 2nd task status is changed then only execute 3rd task.
-> Handlers are used to notify the tasks to execute.
=> 'notify' keyword we will use to inform handler to execute.
-> Using Tag we can map task to a tag-name
-> Using tag name we can execute particular task and we can skip particular task
available in our playbook.
# to display all tags available in playbook
$ ansible-playbook handlers_tags.yml --list-tags
# Execute a task whose tag name is install
$ ansible-playbook handlers_tags.yml --tags "install"
# Execute the tasks whose tags names are install and copy
$ ansible-playbook handlers_tags.yml --tags "install,copy"
# Execute all the tasks in playbook by skipping install task
$ ansible-playbook handlers_tags.yml --skip-tags "install,copy"
============
Variables
============
=> Variables are used to store the data in key-value format
Ex: id=100
name=ashok
age=20
gender=male
=> In Ansible, we can use variables in 4 ways
1) Runtime Variables
2) Playbook variables
3) Group Variables
4) Host Variables
==================
Runtime Variables
==================
=> We can pass variable value in runtime like below
---
- hosts: webservers
become: true
tasks:
- name: install package
yum:
name: "{{package_name}}"
state: latest
...
$ ansible-playbook <yml> --extra-vars package_name=httpd
===================
Playbook Variables
===================
=> We can declare variable value with in the playbook
---
- hosts: webservers
become: true
vars:
package_name: httpd
tasks:
- name: install package
yum:
name: "{{package_name}}"
state: latest
...
=========================================================================
Requirement : Write ansible playbook to install below softwares
In webservers group : install java
In dbservers group : install mysql
---
- hosts: all
become: true
tasks:
- name: install soft
yum:
name: "{{package_name}}"
state: latest
...
=> To achieve above requirement we need to use group vars concept
============
Group Vars
============
=> group vars concept is used to specify variable value for group of managed nodes
as per inventory file.
=> Managed nodes we are configuring host inventory file like below
[webservers]
webserver1 ansible_host=172.31.0.95
webserver2 ansible_host=172.31.0.96
[dbservers]
172.31.5.185
172.31.5.186
=> While executing above playbook for webservers group i want to pass one package
name and for dbservers group i want to pass another package name.
Note: We need to create variables based on group name
Ex :
webservers.yml
dbservers.yml
Note: group_vars related yml files we should create in host inventory file location
Host Inventory file location : /etc/ansible/hosts
webservers variable file : /etc/ansible/group_vars/webservers.yml
dbservers variable file : /etc/ansible/group_vars/dbservers.yml
===============
Host Variables
===============
=> host variables are used to specify variable value at host level (or) machine
level
=> host vars we will create in below location
Location : /etc/ansible/host_vars
webserver1.yml
package_name: java
webserver2.yml
package_name: python
Note-1: host variables will take precendence over group variables
Note-2: Variables defined in playbook override both host_vars and group_vars.
===============
Ansible Vault
===============
=> It is used to secure our playbooks
=> Using Ansible vault concept, we can encrypt & decrypt our playbooks
Encryption : Convert data from readable format to un-readable format
DeCryption : Convert data from un-readable format to readable format
# Encrypt our playbook
ansible-vault encrypt <yml-file-name>
Note: To encrypt a playbook we need to set one vault password
# see encrypted playbook
cat <yml-file-name>
# see orignal content of playbook
ansible-vault view <yml-file-name>
# to edit encrypted playbook
ansible-vault edit <<yml-file-name>>
# how to run encrypted playbook
ansible-playbook <yml-file-name> --ask-vault-pass
# decrypt playbook
ansible-vault decrypt <yml-file-name>
===============
Ansible Roles
===============
=> If we add more functionalities in a playbook then it will become very lengthy
and it will be difficult to manage and maintain that playbook.
=> Using Roles concept we can break down large playbooks into smaller chunks.
=> Roles will provide abstraction for ansible configuration in a modular and re-
usable format.
=> Below playbook we will divide into small chunks using Role concept
---
- hosts: webservers
become: true #use it if you need sudo priviliges
tasks:
- name: install httpd package
yum:
name: httpd
state: latest
- name: copy index.html file
copy:
src: index.html
dest: /var/www/html/index.html
- name: start httpd service
service:
name: httpd
state: started
...
# To create a role we can use below command
Syntax : $ ansible-galaxy init <role-name>
==========================
Working with Ansible Role
==========================
### Step-1: Connect with control node and switch to ansible user
$ sudo su ansible
$ cd ~
### Step-2 : Create a role using 'ansible-galaxy'
$ mkdir roles
$ cd roles
$ ansible-galaxy init apache
$ sudo yum install tree
$ tree apache
### Step-3 : Create tasks inside "tasks/main.yml" like below
---
# tasks file for apache
- name: install httpd
yum:
name: httpd
state: latest
- name: copy index.html
copy:
src=index.html
dest=/var/www/html/
notify:
- restart apache
...
### Step-4 : Copy required files into "files" directory
Note: keep index.html file in files directory
### Step-5 : configure handlers in "handler/main.yml"
---
# handlers file for apache
- name: restart apache
service:
name: httpd
state: restarted
...
Note: With above 5 steps our "apache" role is ready now we can execute that role
like below
### Step-6 : Create main playbook to invoke role using role name
$ cd ~
$ vi invoke-roles.yml
---
- hosts: all
become: true
roles:
- apache
...
=================================
What is gather facts in ansible
=================================
=> In Ansible, gathering facts refers to the process of collecting information
about the target hosts before executing tasks.
Ex: OS, memory, cpu architecture etc....
=> This information will be collected automatically using "setup" module.
---
- hosts: all
gather_facts: yes
tasks:
- name: ping all managed nodes
ping:
...
=================================
What is debug keyword in ansible
=================================
=> debug keyword is used to print a msg when playbook is getting executed.
---
- hosts: localhost
gather_facts: yes
tasks:
- name: print the name of OS family
debug:
msg: "The OS is {{ansible_os_family}}"
- name: print memory info
debug:
msg: "Total memory is {{ ansible_facts['memtotal_mb'] }} MB."
...
======================================
What is register keyword in ansible
======================================
=> register keyword in ansible allow you to capture the output of a task and store
it into a variable for later use.
Note: one task output we can register and we can use it in another task like below
---
- hosts: localhost
tasks:
- name: Run a command to get current date
command: date
register: date_output
- name: print date command output
debug:
msg: "The current Date is => {{ date_output.stdout }}"
- name: check command is success or not
debug:
msg: "command exeution successfull"
when: date_output.rc == 0
...
=================================================================
Write a playbook to install java in different OS family machines
=================================================================
MN-1 : Amazon linux ==> Red Hat family
MN-2 : Ubuntu ==> Debian family
---
- hosts: all
gather_facts: yes
tasks:
- name: install java in Red Hat family
yum:
name: java
state: latest
when: ansible_os_family == 'Red hat'
- name: install java in Debian family
apt:
name: java
state: latest
when: ansible_os_family == 'Debian'
...
==============================
Error Handling in Playbooks
==============================
=> If we get any error in task execution then playbook execution will be terminated
abnormally(in the middle).
=> To achieve graceful termination we need to handle that error.
---
- hosts: all
tasks:
- name: This is first task
command: dates
register: dates_output
ignore_errors: yes
- name: This is second task
debug:
msg: "Second task executed..."
when: dates_output.rc == 0
- name: this is third task
debug:
var: dates_output
...
===================================================================================
Requirement : Write a playbook to check maven is available or not.
If it is not available then install maven software in control node.
===================================================================================
---
- hosts: localhost
become: true
tasks:
- name: check maven version
command: mvn --version
register: output
ignore_errors: yes
- name: print ouput
debug:
var: output
- name: install maven
yum:
name: maven
state: latest
when: output.failed
...
============================================================
Assignment: How to create s3 bucket using ansible playbook
============================================================
## Step-1 : Install amazon.aws using ansible collection
$ ansible-galaxy collection install amazon.aws --force
## Step-2 : install boto3 using ansible ad-hoc command
$ ansible localhost -m pip -a "name=boto3 state=present" --become
## Step-3: Configure AWS Account Access Keys using export command
export AWS_ACCESS_KEY_ID='your-access-key-id'
export AWS_SECRET_ACCESS_KEY='your-secret-access-key'
export AWS_DEFAULT_REGION='ap-south-1'
## Step-4 : Create a playbook and execute that
---
- hosts: localhost
tasks:
- name: create s3 bucket
amazon.aws.s3_bucket:
name: ashokit7999 #unqiue name required
state: present
region: ap-south-1
register: s3_bucket_info
- name: print s3 bucket info
debug:
var: s3_bucket_info
...
## Step-5 : Check s3 service in aws conole, bucket is created or not.
==========================
Ansible Classes Summary
==========================
1) What is Configuration management
2) Ansible Introduction
3) Ansible Architecture
4) Ansible Setup
5) Ansible Ad-Hoc commands
6) Ansible Modules
7) YML file
8) Playbooks
9) Handlers & Tags
10) Variables
- Playbook variables
- Runtime variables
- host vars
- group vars
11) Ansible Vault
12) Ansible Roles
14) gather facts + register + debug + when
15) ignore errors in playbook execution
16) Ansible + AWS Cloud Integration (custom modules)
17) Ansible Tower (theory)
