[c-nsp] Cisco 7301 running at 99% CPU doing iBGP update

Rodney Dunn rodunn at cisco.com
Fri Jun 30 08:17:45 EDT 2006


If you are switching packets at process level on a 65xx you've
got a bigger problem.

Note I said: 

"Traceroutes going to process level is one thing."

On Fri, Jun 30, 2006 at 09:25:18AM +0200, Marian Durkovic wrote:
> 	version=3.1.1
> X-Spam-Checker-Version: SpamAssassin 3.1.1 (2006-03-10) on us.svf.stuba.sk
> 
> On Thu, Jun 29, 2006 at 11:03:54PM -0400, Rodney Dunn wrote:
> > I've never yet seen anone put that in the lab and prove there is
> > transit packet loss when this happens.
> > 
> > Traceroutes going to process level is one thing. Packet switched
> > under interrupt is another.
> 
> Well, not sure about 7301, but on Cat6500 running the -SXF train
> during BGP scanner runs the box certainly drops some of the process-switched
> packets due to SPD queue overflows. This can be reproduced by e.g.
> configuring TTL threshold for IP multicast (which forces multicast packets
> to process-switching path) and sending e.g. 8 Mbps video stream.
> 
> 
> 	With kind regards, 
> 
> --------------------------------------------------------------------------
> ----                                                                  ----
> ----   Marian Durkovic                       network  manager         ----
> ----                                                                  ----
> ----   Slovak Technical University           Tel: +421 2 524 51 301   ----
> ----   Computer Centre, Nam. Slobody 17      Fax: +421 2 524 94 351   ----
> ----   812 43 Bratislava, Slovak Republic    E-mail/sip: md at bts.sk    ----
> ----                                                                  ----
> --------------------------------------------------------------------------


More information about the cisco-nsp mailing list