In a previous tutorial, we talked about expect command and we saw how to automate scripts in Linux using its scripting language. Today, we will take a step further in our automation techniques and talk about a tool that automates tasks more professionally and for different platforms, this tool is Ansible.
We will also talk about some Ansible features such as playbook, inventory, vault, role, and container.
What is Ansible?
Ansible is an open-source IT tool provided by Red-Hat Enterprise Linux (RHEL) that helps in configuration management, orchestration service, task, and application deployment automation.
This tool is aimed to help system administrators who are seeking to minimize recurring tasks, seamless deployment, and easy automation.
Similar tools to Ansible are Puppet, SaltStack, and Chef, which are the main configuration management tools available on the market.
Each one of these tools has its advantages and disadvantages, so choosing the right one can be a bit challenging, depending on which features are needed or which programming language is preferred.
From the advantages of Ansible compared to other tools, Ansible is sort of a new tool that is built on Python and uses YAML templates for scripting its jobs.
YAML stands for “YAML Ain’t Markup Language,” which is a very easy human-readable language. This helps new users to understand it easily.
Another advantage is to use Ansible; there is no need to install an agent in the hosts, which enhances the communication speed as it is using both push and pull models to send commands to its Linux nodes and for Windows nodes, the WinRM protocol is used.
As we stated above, since it’s a new tool from its disadvantages is that it has a poor GUI, un-customized, and immature platform when compared to other tools.
Even though Ansible considered to be used more frequently than ever and there is an increase in downloading it.
Ansible setup on Ubuntu
As we previously mentioned that there is no need to install an agent in the hosts which is unlike other tools. For Ansible, it’s a master node installation only which lacks background process, database dependency, and always running service, and that makes it extremely light.
It is recommended to use the default package manager for Ubuntu while installing Ansible, which will help to install the latest stable version.
Before starting the installation process and for the Linux package installation, you have to make sure that Python 2 (version 2.6 or later) or Python 3 (version 3.5 or later) is installed.
Even though that most Linux OS package managers when asked to download Ansible, they will download the best Python version and its dependencies automatically.
And for the source installation, the development suite may be needed, like the build-essential package for Ubuntu.
We can install Ansible on Ubuntu using one of the following two methods:
The first method through the Ubuntu package manager
First, add Ansible PPA for Ubuntu using the following command:
sudo apt-add-repository ppa:ansible/ansible
Second, press Enter to confirm the key server setup.
Third, update the package manager using the following command:
sudo apt update
Fourth, Ansible is ready to be installed using the next command:
sudo apt install ansible
The second method of installing Ansible is from its source:
This method is sometimes helpful for users who need some particular requirements like for example you need to install the beta or development version of Ansible even if this may grant you early access to new features and future modules, but also you need to be careful it is an unstable version that is still under development and testing.
Also, this method is helpful if you don’t need to install Ansible through the package manager.
So, to get the Ansible source package, you can use one of the following techniques.
First, download the .tar file:
Download the .tar file
wget -c https://releases.ansible.com/ansible/ansible-2.6.0rc3.tar.gz
tar -xzvf ./ansible-2.6.0rc3.tar.gz
Second through the GitHub source:
But we will need to install first the git command:
sudo apt install -y git
Then get Ansible
git clone https://github.com/ansible/ansible.git --recursive
After downloading the Ansible source by using one of the previous techniques, we will start building Ansible but as we previously mentioned that we would need to install Python.
So, we can use the following commands to install Python to make sure that Ansible requirements are met:
Go to Ansible Source directory:
Install Python using easy_install
sudo easy_install pip
Install Python requirements
sudo pip install -r ./requirements.txt
Setup the environment to use Ansible
If you are using GitHub Source you can update the Ansible project and its submodules as following:
git pull --rebase git submodule update --init --recursive
For every time you execute the previous step, you will need to be sure that the environment is already set up properly through the next two commands:
echo "export ANSIBLE_HOSTS=/etc/ansible/hosts" >> ~/.bashrc echo "source ~/ansible/hacking/env-setup" >> ~/.bashrc
Finally, the Ansible inventory can be usually found in /etc/ansible/hosts, and its configuration file is usually found in /etc/ansible/ansible.cfg
Ansible master node configuration
Usually the Ansible configuration file (ansible.cfg) is located in /etc/ansible/ansible.cfg or in the home directory which belongs to the user who installed Ansible.
As soon as you have installed Ansible you can start using it with its default configuration. Next, we will be discussing the most important and useful Ansible configurations that will improve your Ansible work experience.
Starting from Ansible 2.4 and later “ansible-config” is a command that is used by Ansible users to list the enable Ansible options with their values.
Ansible configuration file is divided into several sections but in this article, we will only focus on [defaults] general section. So, Let’s have a look at the basic parameters.
Using your favorite text editor (Gedit, vi, nano…) you can open the Ansible.cfg configuration file:
sudo nano /etc/ansible/ansible.cfg
- inventory: points out to the location of the inventory that Ansible uses to know the available hosts
inventory = /etc/ansible/hosts
- roles_path: points out to the location where the Ansible playbook has to search for extra roles
roles_path = /etc/ansible/roles
- log_path: points out to the location where Ansible log file is stored. Permission to write in this file should be given to Ansible users.
log_path = /var/log/ansible.log
- retry_files_enabled: indicates the retry feature which allows Ansible to create a .retry file anytime a playbook fails. It’s recommended leaving this option disabled unless you want it because if it is enabled it will create multiple files that will take space
retry_files_enabled = False
- host_keychecking: this parameter is used in constantly changing environments where old hosts machines are deleted and new hosts take their place. This parameter is usually used in a cloud or a virtualized environment.
host_key_checking = False
- forks: Indicates the number of parallel tasks that can be executed to the client host. By default, its value is 5 and this to save system resources and network bandwidth but in case you have enough resources and a good bandwidth you can increase the number.
forks = 5
- remote_port: contains the port number used by SSH on the hosts
remote_port = 22
- nocolor: It gives you the ability to use different colors for Ansible playbook and tasks that show errors and success
nocolor = 0
Node Configuration for Linux client
OpenSSH-server is the only important and required a tool to be installed on the client node and by default, all new versions of Linux use SSH as the main remote access tool. So, you need to check the following points carefully:
- SSH service is always up and running.
- SSH port which is 22 by default should be allowed in the system’s firewall.
Node Configuration for Windows client
To make Ansible able to remotely manage Windows host the following applications should be installed on Windows nodes:
- PowerShell version 3.0 or higher
- .NET version 4.0
For missing requirements, there is an Ansible already made PowerShell script that can carry out this installation automatically you can find it in the following link https://github.com/jborean93/ansible-windows/blob/master/scripts/Upgrade-PowerShell.ps1
But before running the previous script you need to change the execution policy to be unrestricted by executing the following script and you need to run it with administrator privilege:
$link = "https://raw.githubusercontent.com/jborean93/ansible-windows/master/scripts/Upgrade-PowerShell.$script = "$env:temp\Upgrade-PowerShell.ps1" $username = "Admin" password = "secure_password" (New-Object -TypeName System.Net.WebClient).DownloadFile($link, $script) Set-ExecutionPolicy -ExecutionPolicy Unrestricted -Force script -Version 5.1 -Username $username -Password $password -Verbose Set-ExecutionPolicy -ExecutionPolicy Restricted -Force $reg_winlogon_path = "HKLM:\Software\Microsoft\Windows NT\CurrentVersion\Winlogon" Set-ItemProperty -Path $reg_winlogon_path -Name AutoAdminLogon -Value 0 Remove-ItemProperty -Path $reg_winlogon_path -Name DefaultUserName -ErrorAction SilentlyContinue Remove-ItemProperty -Path $reg_winlogon_path -Name DefaultPassword -ErrorAction SilentlyContinue
After that run, the Ansible already made PowerShell script then run the execution policy script again to return it to restricted.
Another important script that needs to be run to configure WinRM to make it up and running to listen to Ansible commands, this script is also Ansible already made and you can find it in the following link: https://github.com/ansible/ansible/blob/devel/examples/scripts/ConfigureRemotingForAnsible.ps1
Similarly, this script needed to be run under administrator privileges and execution policy to be unrestricted you can use the following piece of code:
$link = "https://raw.githubusercontent.com/ansible/ansible/devel/examples/scripts/ConfigureRemotingForAnsible.$script = "$env:temp\ConfigureRemotingForAnsible.ps1" (New-Object -TypeName System.Net.WebClient).DownloadFile($link, $script) powershell.exe -ExecutionPolicy ByPass -File $script
If no errors are thrown then this Ansible now should be able to manage this node.
As we have previously mentioned that YAML is a human-friendly language that can be used to manage data. Next, we will talk about YAML basics and will show you how to write a code using YAML.
Guidelines to create a YAML file:
- YAML uses spaces instead of tabs.
- YAML is case sensitive
- YAML file should be saved with the .yaml extension
- YAML file sometimes starts with “—” and ends with “…” but it is optional.
Since YAML can be used to write Ansible playbooks so next we will show you how YAML is easy to use. So in the following example where we need to copy user configuration. If you are using an Ansible command it will look like that:
- name: Copy user configuration copy: src=/home/admin/setup.conf dest=/usr/local/projects/ owner=setup group=dev mode=0677 backup=yes
But in case you are using YAML it will be like this:
- name: Copy user configuration copy: src: /home/admin/setup.conf dest: /usr/local/projects/ owner: setup group: dev mode: 0677 backup: yes
Another example is a .ini inventory file can be as following:
node0.lab.edu [lab1servers] node1.lab.edu node2.lab.edu [lab2servers] node3.lab.edu
But in case you are using YAML it will look like this:
all: hosts: node0.lab.edu children: lab1servers: hosts: node1.lab.edu node2.lab.edu lab2server: hosts: node3.lab.edu
So from the previous two examples, you will find that YAML is easy to use, human-friendly, neat, and good looking language.
It is a .ini file that consists of records of IP addresses, hostnames of the host clients. It may also contain some other variables about the hosts.
In general, these file contents are organized in groups and each group has a name, this name is written between two square brackets like for example [Group1].
The location of Ansible inventory file is by default can be found in /etc/ansible/hosts. But it is recommended to put all the Ansible configuration files in a folder in the home directory of the user and this to allow the user to add and modify their configuration according to their needs. So, next is an example for opening the Ansible configuration file and setting the inventory:
sudo nano /etc/ansible/ansible.cfg inventory = /home/user1/ansible/hosts
Also, you can choose an Ansible inventory file while executing a command by adding -i option to the command:
ansible -m ping -i ~/ansible/hosts
There are two Ansible inventory types static and dynamic. Static inventory can be used in small organizations that have small to medium infrastructure.
While Dynamic inventory can be used in large organizations where there is a huge number of hosts, complicated tasks to be done and enormous errors may start to appear.
If you are trying to add hosts with a similar style to Ansible inventory you use a counter block like the next example:
Inventory file with similar style hosts:
[servers] node0.lec node1.lec node2.lec node3.lec node4.lec
Inventory using a counter block:
Ansible playbook simply is a systematic group of scripts that is using Ansible commands in a more organized method that can install and configure systems. Ansible playbook can perform the following tasks and delegate them to other servers:
- Reorder multi-tier system roll-outs.
- Applying application and systems patches.
- Collecting data from client hosts and depending on the collected data it starts sending instant actions to servers, devices, and load balancers.
Ansible playbook is written in YAML which is a very simple human-readable language compared to other traditional coding languages. YAML also can allow users to share their code easily.
While converting Ansible playbooks into roles will give you the ability to change a set of configuration management tasks into reusable modules with multiple configurations which will be easily shared when needed.
Structure of Ansible role is very simple it consists of many folders each one of them consists of a lot of YAML files that by default have one main.yml file but they can have more than one file.
It is an Ansible encryption tool that allows users to encrypt various variables. This Ansible vault produces encrypted files to save variables; those files can be moved to another location when needed.
Ansible vault can encrypt any different forms of data that are found in Ansible roles and playbooks. Also, it can encrypt task files in case you need to hide a variable name.
It is an open-source tool that allows users to automate everything about their containers from building to deployment to management. Ansible container allows better code management and implementing containers on any cloud registries.
By default, Ansible container is not installed from the beginning as a part of Ansible installation, so you will need to set it up on a container host, and during the installation process, you will need to choose a container engine to work on.
Mokhtar is the founder of LikeGeeks.com. He works as a Linux system administrator since 2010. He is responsible for maintaining, securing, and troubleshooting Linux servers for multiple clients around the world. He loves writing shell and Python scripts to automate his work.