First, you need a way to get this information from the host. Either install a GOOD SNMP agent on the server, or install an agent. For windows, the pnsclient agent lets you query any Perfmon counter, and for UNIX, the NRPE agent allows you to run any general script or Nagios plugin.
Then you need a way to query the agent. SNMP is natively supported by MRTG. If you use pnsclient or NRPE, there are query tools in the routers2 package in the extras directory (mrtg-nrpe.pl and mrtg-pnsclient.pl)
Finally, you need to set up the .cfg file for the host. If using SNMP, then routers2 has 'cfgmaker-host' in the extras directory which will make a .cfg file for you. If using pnsclient or nrpe, then examples are with the query script or else you can see the example .cfg files on the demonstration routers2 server on this site.
Finally - look on the http://www.mrtg.com website. They have huge sections on how to configure MRTG to monitor all things under the sun. Also, look in the EXAMPLES and HOWTO documentation for routers2, which shows you how to set up special user defined graphs for special targets (eg, ping time ranges, temperatures, stacked graphs, etc)
Steve
Statistics: Posted by stevesh — Tue Nov 29, 2005 9:46 am
]]>
I'm a newbie here and would really appreciate the help on getting as subject above monitored on my MRTG.
Curently what i'm getting is only the bits in and out.
Thanks
Statistics: Posted by nfs — Mon Nov 28, 2005 10:41 pm
]]>