Introduction To Ansible

Image for post
Image for post
“person using silver laptop computer on desk” by John Schnobrich on Unsplash

Up till now, we have looked in Terraform for infrastructure provisioning and initial setup using provisioners. Now let’s look at ansible which is an open source automation platform. Ansible does configuration management, application deployment, along with infrastructure orchestration. Ansible is procedural rather than declarative. In ansible, we define what we want to do and ansible go through each and every step for that. In terraform, we specify what state we want to achieve and it makes sure we are at that state by creating, modifying or destroying needed resources. Ansible doesn’t manage any state so we need to define how we want to keep track of created resources using tags or other properties while terraform keeps the state of infrastructure so we don’t need to worry about duplicate resource creation. Personally, I recommend terraform for provisioning the infrastructure, and Ansible for configuring the software as terraform is much more intuitive for infrastructure orchestration.

Once upon a time, managing servers reliably and efficiently was a challenge. System administrators managed server by hand, installing software manually, changing configuration and managing services on servers. As managed servers grew and managed services become more complex, scaling manual process was time-consuming and hard. Then came Ansible which is helpful in creating the group of machines, define how to configure them, what action to be taken on them. All these configurations and actions can be triggered from a central location which can be your local system (named controller machine). Ansible uses SSH to connect to remote hosts and do the setup, no software needed to be installed beforehand on a remote host. It’s simple, agentless, powerful and flexible. It uses YAML in form of ansible playbook. Playbook is a file where automation is defined through tasks. A task is a single step to be performed like installing a package.

Ansible works by connecting to remote hosts (using SSH) defined in inventory file, which contains information about servers to be managed. Ansible then executes defined modules or tasks inside a playbook. Execution of playbook which is called the play. We can use predefined organised playbook called roles, which are used for sharing and reusing a provisioning.

Let’s have a look at some of the terminology used in ansible:

  1. Controller Machine: Machine where Ansible is installed
  2. Inventory: Information regarding servers to be managed
  3. Playbook: Automation is defined using tasks defined in YAML format
  4. Task: Procedure to be executed
  5. Module: Predefined commands executed directly on remote hosts
  6. Play: Execution of a playbook
  7. Role: a Pre-defined way for organizing playbooks
  8. Handlers: Tasks with unique names that will only be executed if notified by another task

As I am using Mac OS, so will be installing pip first using easy_install and then ansible using pip. Please look here to install for other platforms.

Once above command executed, run command below to make sure that ansible is installed properly.

The output should be something like below.

Ansible reads the ssh keys form ~/.ssh/id_rsa. We need to make sure we have public key setup on all remote hosts as we already done using terraform while creation of a remote EC2 instance.

For running ansible command, we need inventory file which is expected to be at a specified path: “/etc/ansible/hosts”. We can change its path using ansible config file (ansible.cfg file) in ansible workspace and define inventory file path there. We need to define username which we are going to use during ssh in ansible config file.

Create an inventory file and add the IP address (dummy)of a remote host.

Once this is done, let’s execute below command to ping all given remote host.

Ansible executes ping command to a remote host and gives below output:

We can even create groups in the inventory file and execute ansible commands by replacing all with a group name. In below example, the server is our group name specified in the inventory file.

Let’s look at playbooks to execute a series of actions. We need to make sure we define playbooks as idempotent so that they can run more than once without having any side effects. Ansible executes playbook in a sequential manner from top to bottom.

Sample playbook is like:

We are going to create a directory on our remote node using playbook for all hosts. Below mentioned playbook will create test directory in /home/ec2-user path.

When we execute above playbook using command “ansible-playbook playbook.yml” we get below result. In this, the first result is gathering facts. This happens as ansible executes a special module named “setup” before executing any task. This module connects to a remote host and gathers all kinds of information like IP address, disk space, CPU etc. Once this is done, our create directory task is executed to create the test directory.

There are many modules and commands available to be executed on remote hosts. With ansible, we can do a server setup, software installation and lot more tasks.

The complete code can be found in this git repository:

PS: If you liked the article, please support it with claps. Cheers

Written by

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store