Updated April 6, 2023
Introduction to Ansible group_vars
In Ansible, we know that variables are very important as they store host-to-host data and usable to deploy some tasks based on remote host’s state-based. There are various ways to define variables like in the inventory file, in the playbook file, in a variable file imported in the playbook. Similarly, there is a way to define variables at a location from where the variables will be realized for a group of hosts. This location falls under group_vars and this can be under a predefined location. In this article, we will learn about group_vars with some examples and ways to use it.
What is Ansible group_vars?
It uses hosts file and group_vars directory to set variables for host groups and deploying Ansible plays/tasks against each host/group. Files under group_var directory are named after the host group’s name or all, accordingly, the variables will be assigned to that host group or all the hosts.
Although you can store the variables in inventory files and playbooks, having stored and defined variables in separate files and that too on a host group basis, adds simplicity and makes it easy to recognize, then use.
Files under group_vars are generically used to store basic identifying variables and then leverage these with other desirable variables by using include_vars and var_files.
How Does ANSIBLE group_vars work?
In this, group_vars directory is under Ansible base directory, which is by default /etc/ansible/. The files under group_vars can have extensions including ‘.yaml’, ‘.yml’, ‘.json’ or no file extension.
It loads hosts and group variable files by searching path which is relative to inventory file or playbook file. So which means if you have your hosts file at location /etc/ansible/hosts and there are three groups names grp1, grp2, and grp3 in the hosts’ file. Then below three files will be loaded for a host, which belongs to all three groups, is targeted for playbook execution.
Similarly, other combinations of these files are used, based on the host’s mapping to groups in the host’s file.
- /etc/ansible/group_vars/grp1.yaml
- /etc/ansible/group_vars/grp2.yml
- /etc/ansible/group_vars/grp3
If there is file /etc/ansible/group_vars/all present then variables in this file will be pulled for every execution. The definition of variables in these files is simple key: value pair format. An example is like below:
Example
---
# file location is
/etc/ansible/group_vars/grp1 fruit: apple
vegetable: potato
Also, there are a few points which should be noted when using group_vars:
- We can create directories under the group named directory, it will read all the files under these directories in lexicographical
- If there is some value which is true for your environment and for every server, the variable containing that value should be defined under /etc/ansible/group_vars/all file
- In AWS case, when we have a dynamic inventory where host groups are created and removed automatically, we need to tag the EC2 instances like “class:webserver”. Then variables defined under the /etc/ansible/group_vars/ec2_tag_class_webserver will be located file.
- If in hosts file, groups are organized in such an order that one group is a child of another, then the variables defined for children will have higher precedence over the variable with the same name defined for the parent
- When the same host is defined for several groups on the same level of the parent-children hierarchy, the variable file precedence with being on the name of groups in alphabetic order. That means if a host is mapped under groups alpha, beta, and gamma. Then for this host variables under /etc/ansible/group_vars/gamma will be pulled.
- We can use Ansible Vault for these files under group_vars, to protect the confidential data.
Examples
Now by using examples, we will try to learn about Ansible group_vars, which you might have to use in day to day operations. We will take some examples, but before going there, we will first understand our lab, we will use for testing purpose.
Here it control server named ansible-controller and two remotes hosts named host- one and host-two. We will create playbooks and run commands on the ansible-controller node and see the results on remote hosts.
Also, group_vars directory is defined as /etc/ansible/group_vars. The inventory file is /etc/ansible/hosts.
We created three files named as below: –
- /etc/ansible/group_vars/alpha
- /etc/ansible/group_vars/beta
- /etc/ansible/group_vars/gamma
The hosts file /etc/ansible/hosts have same groups named and the hosts mapping is like below
Example #1
[alpha] host-one
[beta] host-two
[gamma] host-one host-two 127.0.0.1
In this example, we put contents like below under /etc/ansible/group_vars/alpha to define variables and other two files
- /etc/ansible/group_vars/beta and
- /etc/ansible/group_vars/gamma are empty.
Example #2
fruit: apple
vegetable:
tomato
Now running debug module in ad-hoc command like below:
Example #3
ansible beta -m debug -a "var=fruit"
ansible gamma -m debug -a "var=fruit"
We get the following output where we can see that only variable defined in the file /etc/ansible/group_vars/alpha for alpha group means the host-one host will be pulled.
In this example, we have a playbook with below
Example #4
hosts: host-one tasks:
name: Here we copy the variable value to remote
copy:
content: "fruit variable valus is {{ fruit }}\n" dest: /tmp/sample.ini
Also, group_vars directory is defined as /etc/ansible/group_vars and the files with content under this directory are listed as below:
- cat /etc/ansible/group_vars/alpha
- cat /etc/ansible/group_vars/beta
- cat /etc/ansible/group_vars/gamma
When running playbook like below
Example #5
ansible-playbook ansible_group_vars.yaml
We get the following output where the variable value is copied to a file on the target host.
On checking, we will see that the value of the variable is pulled from
- /etc/ansible/group_vars/gamma.
- ssh host-one “cat /tmp/sample.ini”
- In this example, we have a playbook with below content. Using this playbook, we try to print variables from various directories in a hierarchy under /etc/ansible/group_vars.
Example #6
hosts: host-one tasks:
name: Here we print the variables from different debug:
msg: "Username is {{ username }} and connection port is {{ port }}"
- Also, group_vars directory is defined as /etc/ansible/group_vars, which have files like below
ll /etc/ansible/group_vars/*/*
- The files under this directory have content like below
cat /etc/ansible/group_vars/alpha/db_settings
cat /etc/ansible/group_vars/Charlie/connection_setting
- When running playbook like below
ansible-playbook ansible_group_vars_dir.yaml
- We get the following output where
Conclusion
As we can see in this article, using group_vars I easy to define and can be very helpful in cases where we need to work on a group of hosts. This not only saves our efforts to sort the hosts but also enhance the flexibility in our code. Also, the reusability of these group variables makes it more relevant to use in a production environment.
Recommended Articles
This is a guide to Ansible group_vars. Here we discuss an introduction to group_vars with syntax and respective examples to implement. You can also go through our other related articles to learn more –
