[CentOS] Debugging system load - How to start?

Wed Oct 28 10:07:59 UTC 2009
Hakan Koseoglu <hakan at koseoglu.org>

Hi,

On Tue, Oct 27, 2009 at 9:35 PM, Gšötz Reinicke - IT Koordinator
<goetz.reinicke at filmakademie.de> wrote:
> we run an "old" mailserver system which was set up a couple of years
> ago. The systme dose "everything" what we need(ed). Over the last days I
> noticed an unnormal increase of the system load up to 10 and lots of
> users told me that there mailclient connections (sending and receiving)
> are dropped from time to time.
Try running Nmon over a couple of days collecting stats every 10-20
seconds (a bit excessive) or a minute. Then run the nmon stats through
the Nmon Analyser spreadsheet. That should show how the I/O, RAM and
CPU usage and top CPU consumers during those periods pretty nicely.

Download Nmon for RHEL (perfectly usable on CentOS) from IBM or
sourceforge : http://www.ibm.com/developerworks/wikis/display/WikiPtype/nmon
, http://nmon.sourceforge.net/pmwiki.php

and the analyser is here:
http://www.ibm.com/developerworks/wikis/display/WikiPtype/nmonanalyser

Here's a sample on how to run:
#Intensive, every 10 second. Will use a lot of CPU
nmon -fT -s 10 -c 8640

#Regular, a sample every minute
nmon -fT -s 60 -c 1440
-- 
Hakan (m1fcj) - http://www.hititgunesi.org