3.0 - File Separation

3.1 - Intro to File Separation

3.1.1 - Host_Vars and Group_Vars

• Up until now, variables such as ansible_ssh_pass and ansible_host are all included in the same inventory file.
• This is not good practice for security purposes.
• What should be done is storing the variables under a separate YAML file under a new directory called host_vars.
• The YAML file name should match up with the host defined in the inventory file.
• An example follows:

# Inventory

db_and_web_server

# Sample variable file: host_vars/db_and_web_server.yaml

ansible_ssh_pass: <Password>
ansible_host: <host IP Address>

• When the Ansible playbook is triggered, it, by default, will look in this directory for the required YAML file.
• Similarly, group_vars should be used for group variables

3.1.2 - Include

• If you ever have a set of tasks that can be re-used, it's often beneficial to split these sets of actions into tasks.
• In the case of the simple-webapp example, you could have:
• Tasks to install the Database
• Tasks to set up the web server
• Each of these should be stored under a tasks directory and named accordingly e.g. deploy_db.yaml
• These can then be referenced from the primary playbook in a similar manner to the following:

- name: <Playbook Name>
  hosts: <hosts>
  tasks:
   - include: tasks/db_deploy.yaml
   - include: tasks/webapp_deploy.yaml