I am trying to print a custom message, when certain host is unreachable. My problem is, when the host is unreachable, it will be skipped on the next task, thus the fail module will never be triggered.
And the host that is reachable, will also skip the task due to the when condition are not met.
I have tried ignore_unreachable: false, still the same.
Any idea would be appreciated.
---
- hosts: prod
gather_facts: no
tasks:
- name: fail when not reachable
action: ping
register: ping_result
# any_errors_fatal: true
- fail:
msg: "please make sure all server up and run again"
when: ping_result.unreachable is defined
any_errors_fatal: true
You are looking for the keyword ignore_unreachable, that should be set to true, because you do want to ignore the unreachable host in order to go forward to your fail task.
Given the playbook:
- hosts: nodes
gather_facts: no
tasks:
- ping:
register: ping_result
ignore_unreachable: true
- fail:
msg: "please make sure all server up and run again"
when: ping_result.unreachable is defined
any_errors_fatal: true
This yields:
TASK [ping] *******************************************************
fatal: [node1]: UNREACHABLE! => changed=false
msg: 'Failed to connect to the host via ssh'
skip_reason: Host node1 is unreachable
unreachable: true
fatal: [node2]: UNREACHABLE! => changed=false
msg: 'Failed to connect to the host via ssh'
skip_reason: Host node2 is unreachable
unreachable: true
fatal: [node3]: UNREACHABLE! => changed=false
msg: 'Failed to connect to the host via ssh'
skip_reason: Host node3 is unreachable
unreachable: true
TASK [fail] *******************************************************
fatal: [node1]: FAILED! => changed=false
msg: please make sure all server up and run again
fatal: [node2]: FAILED! => changed=false
msg: please make sure all server up and run again
fatal: [node3]: FAILED! => changed=false
msg: please make sure all server up and run again
NO MORE HOSTS LEFT ************************************************