Upgrades and reboots - question

Archives Ansible Project
1

Hi all!

We’ve been working on a playbook that does a dist upgrade and then performs a reboot. This has been problematic and results in a failure return status that sometimes means it worked and sometimes not. What is the trick to reboots – being able to smoothly reconnect after a reboot signal?

2

Just a guess but you might want to have a few seconds of delay set in your wait_for so that there is time for network services and sshd to come back up before ansible goes looking for it.

3

This works on my machine,

  • name: wait for server to come back
    local_action: shell while true; do echo “Waiting …” ; ssh -o ConnectTimeout=5 -o BatchMode=yes {{ inventory_hostname }} pwd ; [ $? -eq 0 ] && break || sleep 5; done
    sudo: false
4

Your server may still be in the process of shutting down when your “wait_for” is executed. So the wait_for could make a connection and think the server is back up when it is actually on its way down…

Try setting the delay parameter in the wait_for (e.g to 15) to ensure that the server is down/rebooting before waif_for starts checking.

John