Master and slave servers for Nagios

[Jason Qualkenbush]

Wheeler, JF (Jonathan) wrote:
> As I have reported in the past I have 2 slave servers and a master
> server; all checks should be run from the slave servers and passed back
> to the master server.  I have been recently trying the understand why
> the master server still has kernel "Out of memory" problems such that
> the kernel starts killing active processes and, in some cases, panics
> because there are no more processes to kill (this happens perhaps once
> or twice per week usually around 4:50 - 5:10 in the morning).  As part
> of my investigations I have noticed that for a typical host 40% of tests
> are reported from the slave and 60% are run by the master.  I can tell
> this because 40% of messages for this typical host in /var/log/nagios on
> the master server begin "EXTERNAL_COMMAND" and 60% of messages begin
> "Warning:".   My question is why this should be ?  Here is a copy of
> nagios.log from the master server for one test of one host for today (so
> far):
>   

Sounds like this has to do more with the freshness of the passive 
check.  If the master server thinks the check isn't fresh, it will then 
run an active check to see for itself.  I'd tune in the freshness, and 
keep in mind the scheduling of the checks.  If you configure your 
freshness to expire at five minutes, and the slave server schedules that 
check for once every six minutes, you are going to get behavior like you 
mentioned.

-Jason

-- 
-jq


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
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