Dealing with nightly high load

[mark.potter at academy.com]

Cell: 281-734-6965


nagios-users-bounces at lists.sourceforge.net wrote on 11/29/2007 02:30:20 
AM:

> All,
> 
> I'm trying to deal with a bunch of servers that generate some high load
> during the night. That is, between 00:00 and 06:00 a series of job is
> run that will generate load from 10 to 15. Outside this windows, load
> should be no more than 3.
> 
> I have read all the documentation, but I cant find a way to implement
> this in Nagios. I have found 2 options:
> - defining 2 services (load_daily and load_nightly) with different
> tresholds. This will make 2 services show up in nagios that are
> essential the same
> - using escalations to supress messages during the night. This will
> however still show errors in the webinterface and spoil you reports with
> fake errors
> 
> How can I tell Nagios that during the day the load should be max 3
> (warning) and max 5 (critical) but during the night, it's ok if they are
> 8 (warning) and 10 (critical)? I'm sure other people have ran into this
> same problem and found a solution for it? Please share!
> 
> I'm running Nagios 2.10
> 
> Thanx for the help.
> 
> Met vriendelijke groet/With kind regards, 
> 
> Jasper Filon
> 
> CAIWAY Business
> 
<SNIP>

Jasper,

I have sort of the same issue and you have already stated the most 
expedient, and indeed only, solution that I found. Two services with 
different time periods for checking. It's doesn't look as pretty in the 
web interface but it works like a charm. The other method would be to 
write a wrapper that performs the check and alerts on the value you set 
based on time. This wouldn't be hard to do to be honest but it seems like 
re-inventing the wheel to some degree. The other question I have is if the 
servers are capable of withstanding a load that high why not just have 
your high threshold set the same all time? My situation isn't load based 
but with the standard Nagios plugin you can alert on any of the three 
points that uptime returns (here I make the assumption you are monitoring 
*nix boxen). If the situation calls for being able to be alerted to spikes 
during normal operating hours then you will either need two checks and 
time period for each or a wrapper that checks the time and returns an 
alert based on the time and load.

Best Regards,

Mark L. Potter
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.monitoring-lists.org/archive/users/attachments/20071129/0ee53644/attachment.html>
-------------- next part --------------
-------------------------------------------------------------------------
SF.Net email is sponsored by: The Future of Linux Business White Paper
from Novell.  From the desktop to the data center, Linux is going
mainstream.  Let it simplify your IT future.
http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
-------------- next part --------------
_______________________________________________
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