SIGTERM message

[reincarne]

Hi, I have this kind of message from time to time on all of me services when I check the history.
What is the meaning of this error? Does it mean that during this error the alert is not shown on the screen?
I need to investigate an issue and to understand if Nagios did not show a critical alert due this message or it means the notifications were not sent.\
If you check the log below, you can see that every 10 minutes it generated an alert, bu then a cut of 12 hours and nothing - what happens during these 12 hours?



November 08, 2014 18:00
[11-08-2014 18:42:41] SERVICE ALERT: Some_Service;OK;HARD;5;Ok. Value updated. (370 seconds)

November 08, 2014 17:00
Program Start[11-08-2014 17:33:00] Nagios 3.0.6 starting... (PID=20974)
Program End[11-08-2014 17:32:42] Caught SIGTERM, shutting down...
Program Start[11-08-2014 17:31:17] Nagios 3.0.6 starting... (PID=13331)
Program End[11-08-2014 17:30:53] Caught SIGTERM, shutting down...
Program Start[11-08-2014 17:09:14] Nagios 3.0.6 starting... (PID=9514)
Program End[11-08-2014 17:09:02] Caught SIGTERM, shutting down...

November 08, 2014 06:00
Service Critical[11-08-2014 06:31:36] SERVICE ALERT: Some_Service;CRITICAL;HARD;5;Error. Some_Service table doesn't updated! (2883 seconds)
Service Critical[11-08-2014 06:21:50] SERVICE ALERT: Some_Service;CRITICAL;SOFT;4;Error. Some_Service table doesn't updated! (2308 seconds)
Service Critical[11-08-2014 06:11:26] SERVICE ALERT: Some_Service;CRITICAL;SOFT;3;Error. Some_Service table doesn't updated! (1689 seconds)
Service Critical[11-08-2014 06:01:12] SERVICE ALERT: Some_Service;CRITICAL;SOFT;2;Error. Some_Service table doesn't updated! (1062 seconds)

[tmcdonald]

Those "SIGTERM" messages simply mean the Nagios process was restarted. As you can see, the process is starting back up within about 30 seconds each time. This happens if you run "service nagios restart" or if you have the Core Config Manager installed and you apply the changes. During the time the nagios process is not running, alerts will not be sent and checks will not be run, but once it starts back up everything will be rescheduled.

[reincarne]

But look, how come I didn't have an alert from morning until after noon, I know definitely that there was a problem. Does it mean Nagios didn't show this alert on the dashboard during this period?

[Box293]

I would turn on debugging (-1) in nagios.cfg and restart Nagios.

This will give you a better idea of exactly what is happening, it will be logged in nagios.debug.

[reincarne]

Yes, but now its too late for me. It's important only regarding the details I posted here because then my NOC engineer missed this alert due to this problem, he just didn't see it on screen .

[tmcdonald]

I think it's much more likely that the notification had issues than that the critical service did not show on the screen. From your posted logs I can't see anything around 06:00 that would indicate a failure to send, and the SIGTERMS do not show up until many hours later. If you want to post the relevant configs for that service (service config, commands, contacts, host, etc) we can certainly take a look and see what might have gone wrong.