Notifications not 'reissued' when the notification period is active again

Hi everyone, we are currently using Checkmk Raw Edition 2.2.0p17

The email notifications actually work, with one exception that I don’t understand. I hope that you can help me.

Situation:

  • No mails should be sent outside service hours, but the checks should still be executed so that the current status is still visible in the dashboard.
  • If a service enters the WARN/CRIT status outside of service hours and retains this status until the start of service hours, the notification should be sent.
  • According to the documentation Notifications - via Email, SMS, ticket system and more, the rules “Notification period for hosts” and “Notification period for services” can be used for this purpose. Section 6.3 states “Monitoring events for an object that is not currently in its notification period will not be notified. Such notifications will be ‘reissued’ when the notification period is active again - if the host/service is still in a problem state.”

Problem:

  • In my case, no notifications are generated if the host/service goes into WARN/CRIT status outside the service period and retains this status until the service period begins.
  • If the status change takes place within the service times, the notifications are generated correctly.

My configuration:

  • Time period for the maintenance window: 00:00 - 05:30
  • Time period for the service time: 24/7 (with exclude 00:00 - 05:30)
  • One rule each for “Notification period for hosts” and “Notification period for services”, in which the time period for the service time is used
  • A rule under “Notifications” of type “mail” (without the condition “Match only during time period”, as the notifications are not sent subsequently in this case according to the documentation)

Have I misunderstood the documentation on the notification settings, or have I misconfigured something? Do you have any tips on what else I can check? Thank you very much.


A screenshot of an example: The server did not start up properly after a WSUS update. A notification should have been generated at the start of the service time at 05:30. When the server was restarted at 08:40 and the status had changed again, notifications were sent.

Today another example. The notification was triggered at 08:37 instead of 05:30

Screenshot of the “Effective parameters” of the host:

The time period used:

During scheduled downtime notifications are suppressed, so perhaps check if that is the case.

Otherwise, in Notifications - Chapter 7.3: Error diagnosis in the Nagios monitoring core is explained how to enable ~/var/nagios/debug.log
Maybe this log will give some info on why with new occurences there is no notification after 5:30.

Hi Yggy,
thanks for your reply.

Scheduled donwtimes are not available in the RAW Edition. But i checked the “Historic downtimes of host” for manually set downtimes - there are no entrys.

Enabling the debug log could be helpful, thanks for the link.

Scheduled downtimes are available in RAW, but recurring ones are not possible in GUI.

You can however script recurring scheduled downtimes: :upside_down_face:

But, downtimes is not the case for you. Hopefully debug gives info you need.

You are right, I was thinking about the recurring downtime. My mistake :slight_smile: