From: Ricardo M. <ric...@op...> - 2009-02-26 14:39:04
|
Hello, We had a strange behaviour with a service schedule last night; The service was scheduled to be checked only on February/2010(Fri Feb 26 00:08:36 2010). This is very strange, looking to the code we found the following piece: ------- base/checks.c: /* the service could not be rescheduled properly - set the next check time for next year, but don't actually reschedule it */ if(time_is_valid==FALSE && next_valid_time==preferred_time){ -------- Looking to status.log we saw that the last service check was performed at 22:48 of yesterday, and the time period defined for this service finish at 22:58 thus the next check could be scheduled to 22:53(theoretically) because the normal check interval was defined to 5 minutes. Considering that we have a high latency problem(about 300 seconds) Nagios theoretically tried to check the service OUT of time period scope(22:53 + ~300 seconds of latency problem) thus rescheduling the check for the next valid time following time period definition; Is the above reasoning right? If yes, why it was scheduled for the next year and not for the next valid time in time period definition? Below follow the timeperiod definition: define timeperiod{ timeperiod_name mytimeperiod alias a simple timeperiod definition sunday 00:05-23:58 monday 00:05-05:44,06:21-23:58 tuesday 00:05-23:58 wednesday 00:05-23:58 thursday 00:05-23:58 friday 00:05-23:58 saturday 00:05-23:58 } Any help is welcome. Thanks in advance. -rm |