PE Hardware monitoring best practices for Linux

[Trond Hasle Amundsen]

SA <sa at claborn.net> writes:

> Can anyone give me advise, hopefully based on experience, for best proactive 
> hardware monitoring practices for PowerEdge servers under Linux
>
> What tools have you found reliable?

The most important "best practice" that I think most of us on this list
will agree on, is to install OpenManage Server Administrator (OMSA) on
the servers you want to monitor.

Then it's just the issue of getting alerts from OMSA whenever something
is wrong. You can configure OMSA to send SNMP traps to you favourite
trap destination, or you can use something that polls OMSA to check the
status. This can be Nagios, MOM, ITA, etc. or some homemade solution.

As others have pointed out, there is a wealth of Nagios plugins to
choose from. Even if you don't have a monitoring application at all,
Nagios plugins can easily be used in a homemade monitoring script. It's
simply about checking the exit value of the plugin and forward its
output to the sysadmin if the exit value is other than zero.

Best of luck :)

Cheers,
-- 
Trond Hasle Amundsen <t.h.amundsen at usit.uio.no>
Center for Information Technology Services, University of Oslo