[mrtg] Re: Returned values being rolling over, clipped or something

[Tim Schuh]

It sounds like you're rolling the counters over.  You'll probably need
to decrease your polling interval to preserve that data.  If you do
the math for the data rate you're seeing, I bet you'll find that 2^32
is smaller than your ifInOctets or ifOutOctets over the interval
you're using now.

ts


Rod Oliver wrote:
> Hi All,
> 
> A few weeks ago I had a problem when the values to be graphed by MRTG seemed
> to be lower than should be expected. The interfaces being graphed having
> this problem all regularly see in the order of at least 100Mb going through
> them. When the traffic over the interface gets up to around 100Mb MRTG
> starts to graph it as being much lower, as if suddenly there were a lot of
> dropped packets. I checked the interface and the number of packets through
> it did not compare to that shown by MRTG. When I change the frequency of the
> crontab for running MRTG from every 5 minutes to every minute the graphing
> becomes normal. I don't much like the idea of having to send snmp gets to
> the interfaces that frequently if I can otherwise help it. Is this due to a
> limitation of SNMP, of MRTG or something that I have overlooked?
> Rod Oliver
> 
> chello broadband nv
> 
> Network Operations
> 

-- 

Tim Schuh
Network Systems
Verio Central

--
* 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