Add functionality to place agent in downtime

[william-sy]

It would be great to add a feature where a host with all its services or just one service gets placed into downtime for X amount minutes.
This would be great for if the server is not ready yet with all its services while its being setup causing alerts / pages.
Also great for patching or other events that ansible will do with certain services that are monitored like upgrading said service.

[mkayontour]

Hi do you mean to set downtimes via the API for hosts and services?

[william-sy]

Hi,
Yes normally you could do something like this:

curl -k  -u user:password -H 'Accept: application/json' \
 -X POST 'https://icinga.internal:5665/v1/actions/schedule-downtime' \
 -d '{ "type": "Service", "filter": "host.name==\"server.internal\" && service.name==\"ping4\"", "start_time": 1592856490, "end_time": 1592866490, "all_services": true, "author": "ansible", "comment": "Patching!", "pretty": true }'

Which works in ansible if you put all kinds of variables there but is not very elegant!

It would be great if there were:

- name: "Place agent into downtime"
  icinga2_downtime:
     host: {{ inventory_hostname }}
     start: {{ some form of date and time }}
     end: {{ some form of date and time }}
     all_service: True/False 
     service: ping4
     comment: "Patching"
     type: flexible/fixed
     child_hosts: None/triggered/non-triggered 
  when: XX
  with_items: XX

[mkayontour]

I've done something likes this in the past, maybe I can provide an additional module to handle downtimes. But I think it needs some rewriting. If can excerpt the module and provide it in an separate repository before adding it to the collection.

https://github.com/mkayontour/icinga2_ansible_collection/blob/master/plugins/modules/icinga2_downtimes.py

[william-sy]

Thanks for that link, Looks like you have done quite a bit!

I spot a few todo's with the same subject: Rewrite filter and make second request for single services or multiple
Anything else you had an idea of rewriting ?

I guess a branch in this repo will do for the dev. of this module ?

[mkayontour]

Thanks, I need to get myself updated on this module. But I remember that I had problems with removing downtimes, because the API call always removes all downtimes. I'll put this in a branch.

[william-sy]

Ah that is a good feature to be aware of, Looking forward to a branch to colab on!