Re: [nsp] Forwarding latency corellated to CPU utilization

From: Przemyslaw Karwasiecki (karwas@ifxcorp.com)
Date: Fri Jul 12 2002 - 15:05:00 EDT


Yes, that is what i think, it smells like a bug.

As ridiculous as it sounds, I cannot open a TAC case
 -- no contract, and boxes are after guarantee...
Welcome to new economy :-(

12.0.21ST is what I use all along, but in this application
it does not support LLQ, hence 12.2

Thanks anyway,

Przemek

On Fri, 2002-07-12 at 14:39, Jared Mauch wrote:
> Sounds like a bug. I'd open a tac case and get them to
> track the issue down for you.
>
> Also, you might want to try 12.0(21)ST for your mpls voodoo
> instead as it supports your hardware, unless there is something
> that requires you to run 12.2T that isn't listed here.
>
> (or even 12.0(21)S3 as it has been very stable in my
> experience).
>
> - jared
>
> On Fri, Jul 12, 2002 at 02:30:25PM -0400, Przemyslaw Karwasiecki wrote:
> > All,
> >
> > I have recently encountered interesting phenomenon:
> >
> > On cisco 7206 working in a typical configuration in ISP environment
> > (BGP, ISIS, MPLS), I have noticed that traffic forwarded from
> > ATM interface (PA-A3-OC3) to T1 interface (PA-8CT1) is prone
> > to increased forwarding latency (up to 700ms!) while CPU utilization
> > is high (during typical spikes while BGP scenner or ISIS SPF runs).
> >
> > This is happening only when forwarded packets are arriving as MPLS,
> > and egressing as normal IP HDLC frames. In other words, router is
> > removing MPLS label (penultimate hop).
> >
> > As soon as I disabled tag switching on ATM, problem is gone.
> >
> > IOS version is 12.2(8)T4.
> >
> > Is it a feature or bug?
> >
> > I can replicate it easily.
> >
> > Thanks,
> >
> > Przemek
>
> --
> Jared Mauch | pgp key available via finger from jared@puck.nether.net
> clue++; | http://puck.nether.net/~jared/ My statements are only mine.



This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:13:49 EDT