On Thu, 2005-08-25 at 15:19 -0500, Matthew Boehm wrote: > I have these setup just like in the tutorial, but I am getting wierd > results: > > UCD-SNMP-MIB::ssCpuRawUser.0 = Counter32: 3597620 > UCD-SNMP-MIB::ssCpuRawNice.0 = Counter32: 839 > UCD-SNMP-MIB::ssCpuRawSystem.0 = Counter32: 19131260 > UCD-SNMP-MIB::ssCpuRawIdle.0 = Counter32: 211001445
Ummm.... What exactly is wrong with these figures? They're not immediately meaningful on their own, I grant you. But that's because they're counters. You need to monitor the *difference* between successive readings. > As such, my graphs are wrong: > > http://cytrex.cytelcom.com/cgi-bin/14all.cgi?dir=asterisk/ > > If you look at the lower left graph (which is Idle vs User), it doesn't > even show Idle CPU. It says the number above is NaN. Double ummm.... I've just checked that link, and the graph looks fine to me. > Any ideas on why I am getting bad data? I'm using 1 min data intervals. I presume you did wait a minute to get the second reading, (and hence the first difference) before jumping to the conclusion that things were broken? When you're monitoring deltas (like with a Counter-based object), then one reading on it's own is no use whatsoever. It's only when there are two successive readings that MRTG can plot the difference between them. Dave ------------------------------------------------------- SF.Net email is Sponsored by the Better Software Conference & EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf _______________________________________________ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
