Improve documentation of community.general.nagios

[miniwark]

Summary

The documentation of nagios module docs did not explain anything about the actual Nagios server who need to receive the nagios commands. All examples assume than the managed host is also a nagios server (or as at last than there is a nagios command on the managed host).

This will not work with many managed hosts (for example the one who a nrpe demon).

The way to work is probably to add delegate_to to the role for example:

- name: Schedule downtime for all services on host
  community.general.nagios:
    action: downtime
    minutes: 60
    service: all
    host: "{{ inventory_hostname }}"
    comment: "Mise à jour des applications Cocktail"
  delegate_to: nagios.example.com

(with nagios.example.com available via ssh)

Issue Type

Documentation Report

Component Name

nagios

Ansible Version

Community.general Version

version 10.3.0

Configuration

OS / Environment

No response

Additional Information

No response

Code of Conduct

• I agree to follow the Ansible Code of Conduct

[ansibullbot]

Files identified in the description:

• plugins/modules/nagios.py

If these files are incorrect, please update the component name section of the description or use the !component bot command.

click here for bot help

[ansibullbot]

cc @tbielawa @tgoetheyn
click here for bot help

[tgoetheyn]

I think it is quit obvious that a module that manages software needs to be ran on the server hosting the software.

@felixfontein , does this has to be explicitly added to the docs?

[felixfontein]

That usually depends on the software, and how the module can communicate and does communicate with the software. I don't know whether Nagios even has an admin API that can be used from another host.

Regarding the docs: it doesn't have to be added, but it also isn't disallowed. Whether it makes sense probably depends on Nagios.

[miniwark]

Hello,

it's was not completely obvious at last for me. If i simply follow one of the examples from the docs:

- name: Set 30 minutes of apache downtime
  community.general.nagios:
    action: downtime
    minutes: 30
    service: httpd
    host: '{{ inventory_hostname }}'

For example, for managing Apache web server (with a nrpe demon, but without a nagios command line)
I will get the error:

“msg”: “unable to locate nagios.cfg”

Witch is perfectly normal, because the managed host is of course not a Nagios server.
(My fault was probably of course to expect some sort of magic)

That said, before this module, i never heard nor have the need for the delegate_to option.

It's thanks to this forum page than i was able to use the module:
https://forum.ansible.com/t/nagios-module-scheduling-downtime/30191

And i am not the fist one to encounter the issue:
https://forum.ansible.com/t/using-nagios-module-from-ansible-galaxy-in-awx/36048

At last, one example with the delegate_to option would be nice.

[russoz]

hi @miniwark thanks for reporting. Would you be willing to contribute with a Pull Request to improve the documentation? TIA