Red Hat Enterprise Linux Automation with Ansible

Red Hat Enterprise Linux provides several mechanisms that you can use to schedule commands to run at some point in the future. The at command schedules jobs that run once at a specified time. The Cron subsystem schedules jobs to run on a recurring schedule, either in a user's personal crontab file, in the system Cron configuration in /etc/crontab, or as a file in /etc/cron.d. The systemd subsystem also provides timer units that can start service units on a set schedule.

- name: Remove tempuser
  ansible.posix.at:
    command: userdel -r tempuser
    count: 20
    units: minutes
    unique: yes

- name: Schedule backups for my home directory
  ansible.builtin.cron:
    name: Backup my home directory
    user: testing
    job: /home/testing/bin/backup-home-dir
    minute: 0
    hour: 16
    weekday: 5

- name: Schedule job to flush the Bolt cache
  ansible.builtin.cron:
    name: Flush Bolt cache
    cron_file: flush_bolt
    user: "root"
    minute: 45
    hour: 11
    job: "php ./app/nut cache:clear"

- name: Disable dnf makecache
  ansible.builtin.systemd:
    name: dnf-makecache.timer
    state: stopped
    enabled: no

You can choose between two modules to manage services or reload daemons: ansible.builtin.systemd and ansible.builtin.service.

The ansible.builtin.service module is intended to work with a number of service-management systems, including systemd, Upstart, SysVinit, BSD init, and others. Because it provides a generic interface to the initialization system, it offers a basic set of options to start, stop, restart, and enable services and other daemons.

- name: Start and enable nginx
  ansible.builtin.service:
    name: nginx
    state: started
    enabled: yes

The ansible.builtin.systemd module is designed to work with systemd only, but it offers additional configuration options specific to that system and service manager.

The following example that uses ansible.builtin.systemd is equivalent to the preceding example that used ansible.builtin.service:

- name: Start nginx
  ansible.builtin.systemd:
    name: nginx
    state: started
    enabled: yes

The next example reloads the httpd daemon, but before it does that it runs systemctl daemon-reload to reload the entire systemd configuration.

- name: Reload web server
  ansible.builtin.systemd:
    name: httpd
    state: reloaded
    daemon_reload: yes

The ansible.builtin.systemd module cannot set the default boot target. You can use the ansible.builtin.command module to set the default boot target.

- name: Change default systemd target
  hosts: all
  gather_facts: false

  vars:
    systemd_target: "multi-user.target" 

  tasks:
    - name: Get current systemd target
      ansible.builtin.command:
        cmd: systemctl get-default 
      changed_when: false 
      register: target 

    - name: Set default systemd target
      ansible.builtin.command:
        cmd: systemctl set-default {{ systemd_target }} 
      when: systemd_target not in target['stdout'] 
      become: true 

You can use the dedicated ansible.builtin.reboot module to reboot managed hosts during playbook execution. This module reboots the managed host, and waits until the managed host comes back up before continuing with playbook execution. The module determines that a managed host is back up by waiting until Ansible can run a command on the managed host.

The following simple example immediately triggers a reboot:

- name: Reboot now
  ansible.builtin.reboot:

By default, the playbook waits up to 600 seconds before deciding that the reboot failed, and another 600 seconds before deciding that the test command failed. You can adjust this value so that the timeouts are each 180 seconds. For example:

- name: Reboot, shorten timeout
  ansible.builtin.reboot:
    reboot_timeout: 180

Some other useful options to the module include: