Page 1 of 1

scheduling problem, last_check=next_check in status.dat

Posted: Fri Nov 18, 2011 5:09 am
by roehle
Hi,

we came across a problem that puzzled us and we have not found a solution yet (we are using Nagios 3.2.1).
When looking at the scheduling queue we observed that certain host-checks (using check-host-alive) stay on top of the queue when the associated machines are unreachable (hard state).
The checks are configured as follows:

Code: Select all

[...]
	check_interval	240
	max_check_attempts	3
	retry_interval	40
	check_period		xerox_times
[...]
define timeperiod{
        timeperiod_name xerox_times
        alias           morning and afternoon
        monday          09:00-12:00,14:00-17:00
        tuesday         09:00-12:00,14:00-17:00
        wednesday       09:00-12:00,14:00-17:00
        thursday        09:00-12:00,14:00-17:00
        friday          09:00-12:00,14:00-17:00
        }
But according to ndoutils.hostchecks the check is executed every 5 minutes which has nothing to do with the above settings (we want one check in the morning, one in the afternoon) because normally the check_interval should be used once the machine is in hard state.
Inspecting the status.dat file we found for all these machines that the values of last_check and next_check are equal, e.g.

Code: Select all

last_check=1321354502
next_check=1321354502
Obviously there is something wrong... Is this a bug or just a misconfiguration? Any help/ideas appreciated...

Stefan