RE: [Perfparse-users] nagios restarting

[Cook, G. <GW...@ma...>]

I didn't see this original message on the PerfParse User's mailing list.
Did this thread begin somewhere else, or am I missing some messages?

per...@li... wrote:
>> Hi all.
>>=20
>> I've observed my nagios restart every perfparse.sh run. I use
>> perfparse-0.99.07=20

You should upgrade to 0.99.09. It may or may not help with this issue,
but there are a couple of other bugfixes/enhancements available in the
latest version. And don't forget to run
'/<path_to>/nagios/bin/perfparse-db-tool --update' after each upgrade...


>> [26-08-2004 12:35:00] Nagios 1.2 starting... (PID=3D12443)
>> [26-08-2004 12:35:00] Caught SIGHUP, restarting...
>>=20
>> Is this ok? If so, why is needed nagios restarting?
>=20
> If you are using --delete* option, yes, perfparse sends some SIGHUP
> signal to nagios to ask for restart and use a new serviceperf.log
> file.=20
> Otherwise, nothing to see with perfparse.

I've seen some postings (on Nagios Users or Devel lists) about Nagios
restarting when sent a HUP, whereas it should only re-read it's config
files. I don't remember exactly, but I think the original posters may
have been using Nagios 2.0 and/or Fedora Linux Distro. Are you using
either of these? What versions of Nagios/Linux are you using?


>> I'm trying to figure out why my nagios stops working after few hours
>> and I'd like to know if this "nagios restarting" has some relation
>> to "nagios stopping".
>
> perfparse does not ask nagios to stop, but only to start
> writing to a new
> serviceperf.log file, which also mean restarting, but not stopping.
> In your log, strange that nagios starts and restarts at the
> same time. SIGHUP caught by
> change at that time ?

If Nagios fails to HUP properly, it could be leaving you with more than
one Nagios process running. This can sometimes cause odd behavior, like
checks not being run and/or notifications not being sent.
Is Nagios actually shutting down? Or is it running but not performing
actions as expected?


>> When nagios stops its checks, it always has a zombie process.
>=20
> How long do they live ? (just asking for if others can help :)
>=20
> Yves
>=20
>>=20
>> Regards,
>>=20
>> Wilson

It looks like you've already answered the question that I asked above.
So, Nagios is running but not performing checks, correct? This is most
likely related to the zombie/second running Nagios process. This may be
due to the HUP sent by PerfParse, but not exactly caused by it. A HUP
sent to Nagios by PP, command line, or some other process should be
interpreted correctly and not cause this problem. If it's not working
correctly, I believe that the problem is related to Nagios itself. If
you can provide the version information that I asked above
(Nagios/Linux), I can look back through the archives and see what
information I can find. Or, you could do this if you are on Nagios
and/or Nagios Devel lists.

Garry W. Cook, CCNA
Network Infrastructure Manager
MACTEC, Inc. - http://www.mactec.com/
303.308.6228 (Office) - 720.220.1862 (Mobile)