Page 1 of 1

Plugin timed out and service check timed out

Posted: Wed Sep 22, 2021 10:12 am
by arv
Whats the exact difference between plugin timed out and service check timed out after 60 secs.
We want to disable timeout alerts. Have set up the service_check_timeout_state=u in /usr/local/nagios/etc/nagios.cfg and stpped notifications for unknowns but we are still receiving Critical notifications for plugin timed out

Re: Plugin timed out and service check timed out

Posted: Thu Sep 23, 2021 4:50 pm
by benjaminsmith
Hi,

There are different state types for hosts and services, have you disabled unknown notification types for services and unreachable notification types for hosts.

I would check your configurations, this can be adjustedd at both the host/service and contact level. You can 'grep' through the nagios logs to see where the notifications are coming through and check the configs accordingly.

--Benjamin

Re: Plugin timed out and service check timed out

Posted: Thu Sep 30, 2021 10:49 am
by arv
Hi Benjamin

Notifications are disabled for unknown state and it works fine.
We had checks reporting service check timed out after 60 secs ,which was a Critical state by default. Added the service_check_timeout_state=u in nagios.cfg and ,as expected ,it works fine
But, for the checks which shows Plugin timed out. The state is still Critical ...It does not change to Unknown state. What needs to be configured to change the Plugin Timed out checks to Unknown state. Please guide

Thanks
Arvind

Re: Plugin timed out and service check timed out

Posted: Tue Oct 05, 2021 5:46 am
by arv
arv wrote:Hi Benjamin

Notifications are disabled for unknown state and it works fine.
We had checks reporting service check timed out after 60 secs ,which was a Critical state by default. Added the service_check_timeout_state=u in nagios.cfg and ,as expected ,it works fine
But, for the checks which shows Plugin timed out. The state is still Critical ...It does not change to Unknown state. What needs to be configured to change the Plugin Timed out checks to Unknown state. Please guide

Thanks
Arvind
Seems like a Nagios Bug to be fixed

Re: Plugin timed out and service check timed out

Posted: Tue Oct 05, 2021 10:36 am
by gormank
Nagios has a global check timeout of 60 seconds by default.
The check_nrpe (and I assume check_ncpa) have timeouts as well that would need to be less than the global timeout. NRPE (and I assume check_ncpa) can have the timeout set to make the status unknown in the command on the nagios hosts.
It sounds like the latter will resolve the issue. Maybe give that a try.

Re: Plugin timed out and service check timed out

Posted: Tue Oct 05, 2021 12:48 pm
by mcapra
A given Nagios plugin can have its own timeout logic that happens independent of the Nagios Core monitoring engine. That plugin's timeout logic may be triggering before Nagios Core's timeout logic, and the plugin may hard-code the status to CRITICAL instead of UNKNOWN. I can't say for sure if this is what is happening because I don't know what plugin you're using.

Bugs in Nagios Core can be filed as GitHub issues:
https://github.com/NagiosEnterprises/nagioscore/issues

Re: Plugin timed out and service check timed out

Posted: Thu Oct 07, 2021 4:59 am
by arv
Thanks Gormank/Mcapra for your inputs

The third party plugin used, check_sap , has no timeout settings and it is a direct SAP RFC connection (no NRPE/NCPA)
Will put up the BUG in github

Thanks
ARV