big performance issue with Nagios 2.5

[Robert Hajime Lanning]

<quote who="Daniel Meyer">
> So, whats the issue? When i came back into the office this
> morning i found nagios having a service check latency of about
> 350 sec (maximum and also average). I've had this last week to,
> and its "fixed" by simply restarting the nagios daemon.
>
> I've set up some performance monitoring of nagios and the general
> linux system, and this is what i see:
>
> - The load goes up to 2-2.5 within 12h of nagios running (not the
> peaks, but the 15 minute average)
> - nagios runs fine for about 15-18 hours (not general numbers,
> but what happened this weekend)
> - no unusual notifications over the whole weekend
>
> Any ideas where to look?

That is interesting.  I have 2.5 running on an Sun E250 with Sol8.
I have a constant CPU utilization of 45% in kernel.  Constant load
average of 2 - 2.5.  (with two CPUs)

This is for 83 hosts and 693 services, with 3 minute intervals.

Service latency will jump to 300's after about 3 days.  I have a
cron job that restarts Nagios twice a week.  I have scheduling
not carried across restarts.

My theory is that Solaris is having issues forking all the checks.
Does anyone know if Solaris does "copy on write" for the process
memory when forking?  I use a lot of ePN checks.

-- 
And, did Galoka think the Ulus were too ugly to save?
                                         -Centauri


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null