[mrtg] Re: Question regarding Serial interface on 3725/DS3

[Michael Markstaller]

> Is it just me, or is this quote garbled for everybody?
everybody, myself included.. sorry just used a default-crap outlook-xp.

>> Garry, sorry but this wasn't a very helpful answer..
>> 1. 32bit counters *could* wrap starting at ~114Mbit in 300 seconds so
>> this is not the problem here with a DS3

> Not quite so.  Above ~114Mbps the counter wrapping isn't noticable
> anymore.  It will look as if the counters did _not_ wrap and traffic
> is low.
Ok, my answer was misleading regarding the "could warp starting at"
should have said "counter wraps will lead to wrong values in RRDs 
above 114Mbit"

> At 6 Mbps sustained, the counters will increase with 6 * 1000 * 1000
> * 300 / 8 bytes which is 225,000,000.   After 19 normal updates, the
> counters _will_ wrap.  However, this means that during 19 intervals
> no problem should be seen.
yes and the 20th one will be guessed correctly by mrtg so this is still 
no problem, the values stored in rrds stay correct..

> I think the original poster is confusing bits and bytes.
yep :)

Just wanted to correct that using or not using 64bit-counters solves or
causes any problem within mrtg when having a 45Mbit-if..
i.e. on a 3725 you actually see ifHC*octets only when ifSpeed is > 
10Mbit which can be confusing; having a FastEther connect at 10Mbit 
makes graphs fail (I know there some fix inside mrtg but this causes 
useless polls and a delay then at least) anyway. using 64bit counters 
doesn't solve every problem but might also cause some..


Michael

--
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