Print custom message from until loop in ansible - ansible

Question

I am trying to run a command multiple times and check if the output contains some string in it("hi"). I am purposefully simulating failure and expecting the until loop to fail. Everything is good till this point.
Now, I need to have some custom message stating why the until loop or the task failed. For Example: "Your command failed to print hi"
So Question is, How can I print custom message from until loop if the loop failed to pass withing the retries.
Playbook:
-->cat until.yml
---
- hosts: localhost
gather_facts: no
tasks:
- name: "check command"
shell: echo hello
register: var
until: var.stdout.find('hi') != -1
retries: 5
delay: 1
playbook output:
-->ansible-playbook until.yml
PLAY [localhost] *************************************************************************************************************************************************************************************************************************
TASK [check command] ********************************************************************************************************************************************************************************************************
FAILED - RETRYING: who triggered the playbook (5 retries left).
FAILED - RETRYING: who triggered the playbook (4 retries left).
FAILED - RETRYING: who triggered the playbook (3 retries left).
FAILED - RETRYING: who triggered the playbook (2 retries left).
FAILED - RETRYING: who triggered the playbook (1 retries left).
fatal: [localhost]: FAILED! => {
"attempts": 5,
"changed": true,
"cmd": "echo hello",
"delta": "0:00:00.003004",
"end": "2019-12-03 10:04:14.731488",
"rc": 0,
"start": "2019-12-03 10:04:14.728484"
}
STDOUT:
hello
PLAY RECAP *******************************************************************************************************************************************************************************************************************************
localhost : ok=0 changed=0 unreachable=0 failed=1

Answer 1

You can divide your task into two tasks:
First task will poll for the desired output using until loop. But we have used ignore_errors: True , so that until loop will not fail the playbook. We will just capture the result.
In second task , use assert to print success_msg for success case and fail_msg for failure case.
Following is tweaked ,minimum working example:
---
- hosts: localhost
gather_facts: no
tasks:
- name: "check command"
shell: echo hello
register: var
until: var.stdout.find('hi') != -1
retries: 5
delay: 1
ignore_errors: true
- name: "Print result"
assert:
that: var.stdout|regex_search('hi')
fail_msg: "COuld not find HI in command output"
success_msg: "Hi is present in Command output"

Answer 2

Have a look at block error handling which can be used for this purpose.
Basic overview:
- block:
- name: A task that may fail.
debug:
msg: "I may fail"
failed_when: true
register: might_fail_exec
rescue:
- name: fail nicely with a msg
fail:
msg: "The task that might fail has failed. Here is some info from the task: {{ might_fail_exec }}"