Ansible reboot Linux machine or server with playbooks

Indeed to reboot the VM or bare metal Linux machine/server using Ansible and wait for it to come back but it does not work with the playbook as described here. How to reboot the Linux server with Ansible? How do I reboot and wait for the reboot to complete in the Ansible playbook for multiple Linux servers?

You can reboot a Linux or Unix-based machine, wait for it to go down (say for kernel update), come back up, and respond to commands. You can use either command or shell module to reboot the Linux server when the kernel is updated. However, now we have a reboot module to reboot a machine using Ansible.

I tested this module with:

1. Ubuntu Linux 16.04 / 18.04 / 20.04 LTS
2. CentOS Linux 7/8
3. Debian Linux 9.x/10.x/11.x
4. RHEL 7.x/8.x
5. OpenSUSE and SUSE 12.x/15.x
6. FreeBSD
7. OpenBSD

Prerequisite

Please note that you must have Ansible version 2.7 or above to work with the reboot module:

$ ansible –version

If not using Ansible version 2.7, try to update it using the dnf command/yum command/apt command/apt-get command as per your Linux distro version. Redhat and its variants use yum, and Debian and varients use apt/apt-get. Make sure this is thought of as you create playbooks.

$ sudo apt update ## Debian or Ubuntu box ##
$ sudo yum update ## RHEL/CentOS 7 ##

Ansible reboot Linux machine or server with playbooks

The syntax is pretty simple to do a reboot:

- name: Reboot the machine with all defaults using Ansible
  reboot:

Here is a sample hosts file displayed using the cat command:

[all:vars]
k_ver="linux-image-4.15.0-36-generic"
ansible_user='{{ my_c_user }}'
ansible_become=yes
ansible_become_method=sudo
ansible_become_pass='{{ my_c_sudo_pass }}'
 
[legacy]
do-de.public
 
[cluster]
ln.cbz01 
ln.cbz02 
ln.cbz04 
ln.forum 
 
[lxd]
ln.cbz01
ln.cbz02
ln.cbz04
do-de.public 
 
[vpn:vars]
ansible_python_interpreter='/usr/bin/env python3'
 
[vpn]
do-blr-vpn
 
[backup]
gcvm.backup
 
[nodes:children]
vpn
backup
cluster
legacy
 
[isrestart:children]
backup
cluster
vpn

Here is my reboot .yml file:

---
- hosts: isrestart
  become: true
  become_user: root
  tasks:
          - name: Rebooting the cloud server/bare metal box
            reboot:

How to use Ansible reboot module playbook to reboot the box

Now all you have to do is run the playbook

$ ansible-playbook -i hosts --ask-vault-pass --extra-vars '@cluster.data.yml' reboot.yml

How to reboot a machine and set a time out value

By default, the Ansible reboot module waits 600 seconds. You can increase value using the following syntax:

- name: Reboot a Linux machine 
  reboot:
    reboot_timeout: 1800

How to set command to run on the rebooted host and expect success from to determine the machine is ready for further tasks

By default whoami command is used by Ansible. You can change it as follows:

- name: Reboot a Linux machine 
  reboot:
    test_command: uptime

OR

- name: Reboot a Linux machine 
  reboot:
    test_command: ping -c 4 192.168.2.254

How to set pre and post-reboot delay

One can force Ansible to wait after the reboot was successful and the connection was re-established in seconds as follows:

- name: Unconditionally reboot the machine with all defaults
  reboot:
    post_reboot_delay: 180

The above is useful if you won’t want to wait for additional networking/storage or server VPN to kick in despite your connection already working. You can also set a time for the shutdown to wait before requesting a reboot:

- name: Unconditionally reboot the machine with all defaults
  reboot:
    pre_reboot_delay: 180

View reboot log history on the Linux server

I am doing a conditional reboot of my Ubuntu or Debian Linux box. For instance, my Ansible playbook:

      - name: Check if a reboot is needed on AWS EC2 Ubuntu/Debian based servers
        register: reboot_required_file
        stat: path=/var/run/reboot-required get_md5=no
 
      - name: Reboot the box if kernel updated/installed on EC2 
        reboot:
          msg: "Reboot initiated by Ansible for kernel updates"
          connect_timeout: 5
          reboot_timeout: 300
          pre_reboot_delay: 0
          post_reboot_delay: 30
          test_command: uptime
        when: reboot_required_file.stat.exists

We can search for “Reboot initiated by Ansible for kernel updates” on our server to see when my box was rebooted using the grep command/zgrep command:

$ ssh [email protected]
$ sudo grep 'reboot' /var/log/auth.log
$ sudo zgrep 'Reboot initiated by Ansible for kernel updates' /var/log/auth.log*
$ sudo zgrep 'reboot' /var/log/auth.log*

Sample outputs:

auth.log:Jun  9 11:06:57 ls-debian-10 systemd-logind[488]: System is rebooting (Reboot initiated by Ansible for kernel updates).
auth.log.2.gz:May 27 04:55:54 ip-172-26-14-129 sudo:    admin : TTY=pts/0 ; PWD=/home/admin ; USER=root ; COMMAND=/sbin/reboot

Another option is to run the last command:

$ sudo last -x "reboot"

Reboot history:

reboot   system boot  4.19.0-9-amd64   Wed Jun 10 03:51   still running
reboot   system boot  4.19.0-9-amd64   Tue Jun  9 11:07 - 03:51  (16:43)
reboot   system boot  4.19.0-9-amd64   Wed May 27 04:56 - 11:06 (13+06:10)
reboot   system boot  4.9.0-12-amd64   Wed May 27 04:15 - 04:55  (00:40)
reboot   system boot  4.9.0-8-amd64    Wed May 27 04:09 - 04:14  (00:05)
reboot   system boot  4.9.0-8-amd64    Wed May 27 04:08 - 04:09  (00:01)

wtmp begins Wed May 27 04:08:01 2020

See “How To Find Out Last System Linux Reboot Time and Date Command” for more information

Conclusion

You just learned how to reboot Linux/Unix box and wait for the reboot to complete in the Ansible playbook. For more info, see Ansible docs.