msamon sucking up cpu resources

[Donald Thaler]

The top command shows that msamon is using 98% of the cpu, what is msamon and is this normal behaviour ?? Does msamon create a log file ?

[Tim Nelson · Accepted solution]

If you cannot find it in a search here or google I always revert to the following to try and find out what package it belonged to.

swlist -l file|grep msamon

looks like a Online Diag
Sup-Tool-Mgr.STM-UUT-RUN: /usr/sbin/stm/uut/bin/tools/monitor/msamon

Those logs are typically under /var/stm

Try to stop and restart diag to give it a kick.

[F Verschuren]

msamon is a part of ems
to monitor:
HP Storage Works Modular SAN array 1000 / 30 Monitor

so depending on your hartware you neet it to monitor the above

please take a look at your EMS warnings

[Tim Nelson]

Even better. When in doubt you may get lucky by doing this.

man msamon

( I was suprised a man page came up )

Now it makes sense. DIAG MSA disk array monitor :)

[Donald Thaler]

when u say stop and start diag do you mean

/sbin/rc2.d/S742diagnostic stop/start


or is there another command for diag ?

[Shah Sahib]

You obviously need to worry about your disks first, there is a problem, fix that and your cpu useage will drop, or kill the process.

[Andrew Merritt_2]

Hi Donald,
As the others have figured out, msamon is one of the EMS HW monitors, which watches MSA 30 and MSA 1000 devices.

The first thing to check is which version of OnlineDiags you have installed, since the early release of the msamon monitor had some problems. Run 'cstm' and see what version that shows. Check against this table - http://www.docs.hp.com/en/diag/stm/stm_upd.htm#table

If you don't have a current version, please update.

If you do have a current version, next thing to check is whether you actually have any MSA devices; also look in /var/opt/resmon/log/api.log in case there's a clue there. If you don't have MSA devices, please report the problem to HP Support, since msamon should not be running in that case. If you do have MSA devices, look in /var/opt/resmon/log/event.log to see if any events are being reported. If so, take the appropriate action; if not, again I think a call to HP support would be indicated.

Andrew