Automating Infrastructure With Ansible

YAML
# Basic server setup playbook
---
- name: Configure web servers
  hosts: webservers
  become: yes
  tasks:
    - name: Install nginx
      package:
        name: nginx
        state: present
    
    - name: Start and enable nginx
      systemd:
        name: nginx
        state: started
        enabled: yes
    
    - name: Configure firewall
      ufw:
        rule: allow
        port: '80,443'
        proto: tcp

1. Inventory Setup - Define target hosts in inventory.yml:

   YAML
   all:
     children:
       webservers:
         hosts:
           web1.example.com:
           web2.example.com:
       databases:
         hosts:
           db1.example.com:

2. Playbook Structure - Organize tasks logically:

   site.yml           # Main playbook
   group_vars/        # Variable files
   host_vars/         # Host-specific vars
   roles/             # Reusable components
   inventory.yml      # Host definitions
   

3. Task Development

   • Start with basic tasks
   • Add error handling with failed_when and ignore_errors
   • Use handlers for service restarts
   • Implement idempotency checks

4. Testing & Deployment

   • Test with --check --diff flags
   • Run on staging environment first
   • Use --limit for targeted deployments

Progress checklist for complex deployments:

• Inventory configured
• Variables defined
• Playbook syntax validated
• Dry run completed
• Staged deployment tested
• Production deployment executed
• Post-deployment verification

Example 1: Database Server Setup Input: "Set up PostgreSQL with custom config and backup user" Output:

YAML
- name: Setup PostgreSQL
  hosts: database
  become: yes
  vars:
    postgres_version: "14"
    backup_user: "backup_user"
  tasks:
    - name: Install PostgreSQL
      package:
        name: "postgresql-{{ postgres_version }}"
        state: present
    
    - name: Configure PostgreSQL
      template:
        src: postgresql.conf.j2
        dest: /etc/postgresql/{{ postgres_version }}/main/postgresql.conf
      notify: restart postgresql
    
    - name: Create backup user
      postgresql_user:
        name: "{{ backup_user }}"
        password: "{{ backup_password }}"
        role_attr_flags: REPLICATION
      become_user: postgres

Example 2: API-based Cloud Resource Input: "Create AWS S3 bucket with versioning and lifecycle policy" Output:

YAML
- name: Create S3 bucket
  amazon.aws.s3_bucket:
    name: "{{ bucket_name }}"
    state: present
    versioning: yes
    lifecycle:
      - id: delete_old_versions
        status: enabled
        noncurrent_version_expiration_days: 30

Example 3: CLI Tool Automation Input: "Deploy Docker containers across multiple hosts" Output:

YAML
- name: Deploy application containers
  hosts: docker_hosts
  tasks:
    - name: Pull latest image
      docker_image:
        name: "{{ app_image }}"
        tag: "{{ app_version }}"
        source: pull
    
    - name: Run application container
      docker_container:
        name: "{{ app_name }}"
        image: "{{ app_image }}:{{ app_version }}"
        state: started
        restart_policy: unless-stopped
        ports:
          - "8080:8080"
        env:
          DATABASE_URL: "{{ database_url }}"

• Use roles for reusable components across projects
• Encrypt secrets with ansible-vault for passwords and API keys
• Tag tasks for selective execution: ansible-playbook site.yml --tags "config"
• Variable precedence: host_vars > group_vars > playbook vars > role defaults
• Idempotency: Always use modules that check current state before making changes
• Error handling: Use block/rescue/always for complex error scenarios
• Documentation: Include meta/main.yml with role descriptions and dependencies

• Don't use shell module when specific modules exist (use package not shell: apt install)
• Avoid hardcoded values - use variables and templates instead
• Don't ignore return codes - always check task results for critical operations
• Skip become: yes on tasks that don't need elevated privileges
• Don't run without --check on production systems initially
• Avoid command modules for file operations - use copy, template, file modules instead