Re: [nsp] Interface counter mismatch

From: tony kam (kamt@hkg.hp.com)
Date: Thu Sep 09 1999 - 14:16:10 EDT


Hello BR,

We are not using CEF and I open a TAC case with Cisco but no solution yet.
Hopefully 11.1(28)CC can help.

Thanks.
Tony

----- Original Message -----
From: Bradley Reynolds <brad@qual.net>
To: <c.spurgeon@mail.utexas.edu>
Cc: tony kam <kamt@hkg.hp.com>
Sent: Friday, September 10, 1999 12:38 AM
Subject: Re: [nsp] Interface counter mismatch

> umm, there is/was a double counting bug in cef bits where
> packets would be accounted for twice. it was around 22CC
> (definitely in 19CC1)
>
> BR
>
> On Thu, 9 Sep 1999 c.spurgeon@mail.utexas.edu wrote:
>
> > >If you have a serial link between 2 routers, do you aware that the
counters
> > >are not match? For example, 5min output rate of your upstream is not
equal
> > >to the 5min input rate of your downstream, Sometimes there is a big
> > >difference!!!
> > >
> > >The router desscriptions are:
> > >
> > >-7500 series
> > >-VIP2-40
> > >-11.1(22)CC
> > >-fast switching, optimum switching and distributed switching are
enabled.
> > >-CSA
> > >-WFQ enabled
> > >
> > >Appreciated if you could let me know your experience and comments.
> > >
> >
> > We've seen this recently on a p2p atm link using subints on 7513s with
> > vip2/40 and 11.1(24)CC. CEF and dist switching were enabled, but no
> > fancy queuing.
> >
> > Interface counters in/out in one direction were totally out of whack,
> > with the input side on one end only showing about 50% of what you'd
> > expect given the output side on the other end. No high rates of input
> > queue drops and no ignore packets. The link was working fine with 0%
> > ping loss.
> >
> > Mrtg graphs showed the same thing on traffic. A search of the bug
> > database and queries to local Cisco SEs didn't turn up anything.
> >
> > Since rebooting the router with the bogus input side counters on
> > 11.1(28)CC we haven't seen the problem again.
> >
> > -Charles
> >
> > Charles E. Spurgeon
> > ACITS/Networking Services
> > c.spurgeon@mail.utexas.edu
> > 512.475.9265
> >
> BR
>
>



This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:12:05 EDT