[mrtg] Re: [rrd-users] Re: data Spikes?

[Rich Adamson]

Marc,
Sounds like you are hitting the same problem that I have for the last
several months. Seems as though somewhere along v2.9.17 and up through
v2.9.18pre8 (or so) mrtg would cause a spike in the graph whenever it
missed a poll response.

I did some rather extensive testing that proved without a doubt that
was the case although I could not determine the logic problem within
mrtg. At one point folks thought the issue might be related to counter
roll-overs, but that's not the case. The answer appears to be related
to which previous value mrtg uses to calculate a point when it misses
a poll. All other snmp applications that I know of handles the missed
poll as a retransmission (etc), but mrtg doesn't. It trys to calculate
the missing data point based on historic data instead.

Hopefully that has been corrected somewhere around v2.9.18pre10 but
I've not tested that as yet.
Rich

------------------------
> Rob,
> 
> I agree. However, if the spike never hits the max value then it will not
> be recorded as a NaN. Our average load on a port may be only 20 - 30K.
> However, out of no where it would spike to 10 or 15mb randomly at all
> hours of the day on a 100mb port complete hiding the existing data. We
> have a fairly large implementation and it happens on all types of
> devices too at random times with no consistency. We confirmed it was not
> our network because we were using other products monitoring the same
> MIBs. I'm stumped. The only thing I can think of is it may be a counter
> wrapping issue.
> 
> Marc

--
Unsubscribe mailto:mrtg-request at list.ee.ethz.ch?subject=unsubscribe
Archive     http://www.ee.ethz.ch/~slist/mrtg
FAQ         http://faq.mrtg.org    Homepage     http://www.mrtg.org
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi