Ansible Coder

⚠️ SIMPLICITY FIRST - Default to Flat Structure

ALWAYS start with the simplest approach. Only add complexity when explicitly requested.

Simple (DEFAULT) vs Overengineered

When to Use Simple Approach (90% of cases)

• Setting up 1-5 servers
• Standard stack (Docker, nginx, fail2ban, ufw)
• Single environment or identical servers
• No complex conditional logic per host

When Complexity is Justified (10% of cases)

• Large fleet with divergent configurations
• Multi-team requiring role isolation
• Complex orchestration with dependencies
• User explicitly requests modular structure

Rule: If you can fit everything in one 200-line playbook, DO IT.

When to Use Ansible vs Cloud-Init

Rule of thumb: Cloud-init for initial provisioning, Ansible for ongoing management.

Directory Structure

Simple Structure (DEFAULT)

infra/ansible/
├── playbook.yml          # Single playbook with all tasks inline
├── requirements.yml      # Galaxy dependencies (geerlingguy.*, etc.)
├── hosts.ini             # Inventory (git-ignored)
└── hosts.ini.example     # Inventory template

Complex Structure (only when justified)

infra/ansible/
├── playbook.yml          # Main playbook
├── requirements.yml      # Galaxy dependencies
├── hosts.ini             # Inventory (git-ignored)
├── hosts.ini.example     # Inventory template
├── group_vars/
│   └── all.yml           # Shared variables
└── roles/
    └── custom_role/
        ├── tasks/main.yml
        ├── handlers/main.yml
        └── templates/

Inventory

Static Inventory

# hosts.ini
[web]
192.168.1.1 ansible_user=root

[db]
192.168.1.2 ansible_user=root

[all:vars]
ansible_python_interpreter=/usr/bin/python3

Dynamic from Terraform

# Generate inventory from Terraform output
SERVER_IP=$(cd infra && tofu output -raw server_ip)
cat > infra/ansible/hosts.ini << EOF
[web]
$SERVER_IP ansible_user=root
EOF

Playbook Structure

Basic Playbook

---
- name: Configure web servers
  hosts: web
  become: true

  vars:
    timezone: "UTC"
    swap_size_mb: "2048"

  tasks:
    - name: Update apt cache
      ansible.builtin.apt:
        update_cache: true
        cache_valid_time: 3600

    - name: Install packages
      ansible.builtin.apt:
        name:
          - docker.io
          - fail2ban
          - ufw
        state: present

With Roles

---
- name: Configure web servers
  hosts: web
  become: true

  vars:
    security_autoupdate_reboot: true
    security_autoupdate_reboot_time: "03:00"

  roles:
    - role: geerlingguy.swap
      when: ansible_swaptotal_mb < 1
    - role: geerlingguy.docker
    - role: security

Common Tasks

Package Management

- name: Install required packages
  ansible.builtin.apt:
    name:
      - curl
      - ca-certificates
      - gnupg
      - fail2ban
      - ufw
      - ntp
    state: present
    update_cache: true

Docker Installation

- name: Check if Docker is installed
  ansible.builtin.command: docker --version
  register: docker_installed
  ignore_errors: true
  changed_when: false

- name: Install Docker via convenience script
  ansible.builtin.shell: curl -fsSL https://get.docker.com | sh
  when: docker_installed.rc != 0
  args:
    creates: /usr/bin/docker

- name: Ensure Docker is running
  ansible.builtin.systemd:
    name: docker
    state: started
    enabled: true

SSH Hardening

- name: Disable SSH password authentication
  ansible.builtin.lineinfile:
    path: /etc/ssh/sshd_config
    regexp: "^#?PasswordAuthentication"
    line: "PasswordAuthentication no"
  notify: Restart ssh

- name: Disable SSH root login with password
  ansible.builtin.lineinfile:
    path: /etc/ssh/sshd_config
    regexp: "^#?PermitRootLogin"
    line: "PermitRootLogin prohibit-password"
  notify: Restart ssh

handlers:
  - name: Restart ssh
    ansible.builtin.systemd:
      name: ssh  # Ubuntu uses 'ssh', not 'sshd'
      state: restarted

Fail2ban

- name: Configure fail2ban for SSH
  ansible.builtin.copy:
    dest: /etc/fail2ban/jail.local
    content: |
      [sshd]
      enabled = true
      port = ssh
      filter = sshd
      logpath = /var/log/auth.log
      maxretry = 5
      bantime = 3600
      findtime = 600
    mode: "0644"
  notify: Restart fail2ban

- name: Ensure fail2ban is running
  ansible.builtin.systemd:
    name: fail2ban
    state: started
    enabled: true

handlers:
  - name: Restart fail2ban
    ansible.builtin.systemd:
      name: fail2ban
      state: restarted

UFW Firewall

- name: Set UFW default policies
  community.general.ufw:
    direction: "{{ item.direction }}"
    policy: "{{ item.policy }}"
  loop:
    - { direction: incoming, policy: deny }
    - { direction: outgoing, policy: allow }

- name: Allow specified ports through UFW
  community.general.ufw:
    rule: allow
    port: "{{ item }}"
    proto: tcp
  loop:
    - 22   # SSH
    - 80   # HTTP
    - 443  # HTTPS

- name: Enable UFW
  community.general.ufw:
    state: enabled

Kernel Tuning

- name: Configure sysctl for performance
  ansible.posix.sysctl:
    name: "{{ item.name }}"
    value: "{{ item.value }}"
    state: present
    reload: true
  loop:
    - { name: vm.swappiness, value: "10" }
    - { name: net.core.somaxconn, value: "65535" }

Timezone

- name: Set timezone
  community.general.timezone:
    name: "{{ timezone }}"

Remove Snap (Ubuntu bloat)

- name: Remove snapd
  ansible.builtin.apt:
    name: snapd
    state: absent
    purge: true
  ignore_errors: true

- name: Remove snap directories
  ansible.builtin.file:
    path: "{{ item }}"
    state: absent
  loop:
    - /snap
    - /var/snap
    - /var/lib/snapd

Galaxy Dependencies

requirements.yml

---
roles:
  - name: geerlingguy.swap
    version: 2.0.0
  - name: geerlingguy.docker
    version: 7.4.1

collections:
  - name: community.general
  - name: ansible.posix

Installation

ansible-galaxy install -r requirements.yml --force

Running Playbooks

Basic Execution

ANSIBLE_HOST_KEY_CHECKING=False ansible-playbook -i hosts.ini playbook.yml

With Variables

ansible-playbook -i hosts.ini playbook.yml \
  -e "timezone=Europe/Berlin" \
  -e "swap_size_mb=4096"

Dry Run

ansible-playbook -i hosts.ini playbook.yml --check --diff

Limit to Specific Hosts

ansible-playbook -i hosts.ini playbook.yml --limit web

Kamal Server Preparation

Complete playbook for Kamal deployment servers (based on kamal-ansible-manager):

---
- name: Prepare server for Kamal deployment
  hosts: web
  become: true

  vars:
    swap_file_size_mb: "2048"
    timezone: "UTC"
    ufw_allowed_ports: [22, 80, 443]

  roles:
    - role: geerlingguy.swap
      when: ansible_swaptotal_mb < 1

  tasks:
    # System updates
    - name: Update and upgrade packages
      ansible.builtin.apt:
        update_cache: true
        upgrade: dist

    # Remove bloat
    - name: Remove snapd
      ansible.builtin.apt:
        name: snapd
        state: absent
        purge: true
      ignore_errors: true

    # Essential packages
    - name: Install required packages
      ansible.builtin.apt:
        name: [curl, ca-certificates, fail2ban, ufw, ntp]
        state: present

    # Docker
    - name: Install Docker
      ansible.builtin.shell: curl -fsSL https://get.docker.com | sh
      args:
        creates: /usr/bin/docker

    - name: Enable Docker
      ansible.builtin.systemd:
        name: docker
        state: started
        enabled: true

    # Security
    - name: Configure fail2ban
      ansible.builtin.copy:
        dest: /etc/fail2ban/jail.local
        content: |
          [sshd]
          enabled = true
          maxretry = 5
          bantime = 3600
        mode: "0644"
      notify: Restart fail2ban

    - name: Configure UFW
      community.general.ufw:
        rule: allow
        port: "{{ item }}"
        proto: tcp
      loop: "{{ ufw_allowed_ports }}"

    - name: Enable UFW
      community.general.ufw:
        state: enabled
        policy: deny
        direction: incoming

    # SSH hardening
    - name: Harden SSH
      ansible.builtin.lineinfile:
        path: /etc/ssh/sshd_config
        regexp: "{{ item.regexp }}"
        line: "{{ item.line }}"
      loop:
        - { regexp: "^#?PasswordAuthentication", line: "PasswordAuthentication no" }
        - { regexp: "^#?PermitRootLogin", line: "PermitRootLogin prohibit-password" }
      notify: Restart ssh

    # Performance
    - name: Tune kernel
      ansible.posix.sysctl:
        name: "{{ item.name }}"
        value: "{{ item.value }}"
        reload: true
      loop:
        - { name: vm.swappiness, value: "10" }
        - { name: net.core.somaxconn, value: "65535" }

    - name: Set timezone
      community.general.timezone:
        name: "{{ timezone }}"

  handlers:
    - name: Restart fail2ban
      ansible.builtin.systemd:
        name: fail2ban
        state: restarted

    - name: Restart ssh
      ansible.builtin.systemd:
        name: ssh
        state: restarted

Integration with Terraform

Provision Script Pattern

#!/usr/bin/env bash
# infra/bin/provision

# 1. Terraform creates server
cd infra && tofu apply
SERVER_IP=$(tofu output -raw server_ip)

# 2. Wait for SSH
until ssh -o ConnectTimeout=5 root@$SERVER_IP true 2>/dev/null; do
  sleep 5
done

# 3. Generate inventory
echo "[web]\n$SERVER_IP ansible_user=root" > ansible/hosts.ini

# 4. Run Ansible
cd ansible
ansible-galaxy install -r requirements.yml
ANSIBLE_HOST_KEY_CHECKING=False ansible-playbook -i hosts.ini playbook.yml

# 5. Kamal bootstrap
cd ../..
bundle exec kamal server bootstrap

Troubleshooting