[mrtg] Re: ucd-snmp and CPU Load Average (RH6.1)

[Uffe R. B. Andersen]

At 15:51 19-11-99 -0600, you wrote:
>I had a similar problem when I was monitoring CMU and had to switch to UCD.
>If I remember correctly, the value that UCD returns requires that you add
>the gauge option to the MRTG config file. Something else to try is to do an

The gauge option is already there. Btw. - the ucd-snmp in question is a
version 4.0.1 - that might have some importance (though I know of none).

>SNMPGET on the oid for CPU load and see what the value returned is, make
>sure you are getting proper numbers.  Lastly, please post a snip of your

[root at xerox mrtg-2.8.8]# snmpget xerox public .1.3.6.1.2.1.25.3.3.1.2.1
enterprises.ucdavis.mibRegistryTable.mrEntry.mrIndex.0 = OID: .ccitt.nullOID
[root at xerox mrtg-2.8.8]#

Which puzzles me a bit, because it should be 
host.hrDevice.hrProcessorTable.hrProcessorEntry.hrProcessorLoad 
and not enterprises.ucdavis... that is derived from x-ucd-snmp.

>config file so we can poke at it as well...

[excerpt from mrtg.cfg]
Target[xerox.twe.net.load]:
.1.3.6.1.2.1.25.3.3.1.2.1&.1.3.6.1.2.1.25.1.6.0:public at xerox.twe.net
MaxBytes[xerox.twe.net.load]: 100
Directory[xerox.twe.net.load]: xerox.twe.net
#WithPeak[xerox.twe.net.load]: wmy
Title[xerox.twe.net.load]: xerox.twe.net load
PageTop[xerox.twe.net.load]: <H1>Average load and proces count</H1>
Options[xerox.twe.net.load]: gauge nopercent
Unscaled[xerox.twe.net.load]: dwmy
YLegend[xerox.twe.net.load]: Load
#Suppress[xerox.twe.net.load]: y
ShortLegend[xerox.twe.net.load]: &nbsp;
LegendI[xerox.twe.net.load]: &nbsp;CPU-load:
LegendO[xerox.twe.net.load]: &nbsp;Processes:
Legend1[xerox.twe.net.load]: System load (1 minute sample, 100 times loadavg)
Legend2[xerox.twe.net.load]: Number of processes
[end of excerpt]

Don't bother trying to reach the machine - though the domain is real
enough, the machine is not currently reachable from the outside.

>-----Original Message-----
>From: mrtg-bounce at list.ee.ethz.ch [mailto:mrtg-bounce at list.ee.ethz.ch]On
>Behalf Of Uffe R. B. Andersen
>Sent: Friday, November 19, 1999 3:44 PM
>To: mrtg at list.ee.ethz.ch
>Subject: [mrtg] ucd-snmp and CPU Load Average (RH6.1)
>
>
>Hi Everyone
>
>Recently I upgraded an older RH to the new RH 6.1, which comes with
>ucd-snmp, as opposed to cmu-snmp in the former. It gave me quite some
>troubles, due to changes in OID's for disk-monitoring, the fact that
>ucd-snmp checks /etc/hosts.allow and so on. Almost everything is back
>online though, expedt one thing - monitoring CPU Load Average.
>
>This should be possible to do directly from the OID - I use
>.1.3.6.1.2.1.25.3.3.1.2 and .1.3.6.1.2.1.25.1.6.0, shown in same graph (the
>later is number of processes and works fine) - but it the first one returns
>nothing.
>
>Anyone have an idea? I know this is not strictly an mrtg issue, but I've
>been following this list for a while and not the ucd-snmp list - so I hope
>I'm forgiven and that someone knows the answer.

-- 
Med venlig hilsen - Sincerely
Uffe R. B. Andersen - mailto:urb at image.dk
http://www.image.dk/~urb/

--
* To unsubscribe from the mrtg mailing list, send a message with the
  subject: unsubscribe to mrtg-request at list.ee.ethz.ch
* The mailing list archive is at http://www.ee.ethz.ch/~slist/mrtg