[c-nsp] 7200

[James]

On Fri, Sep 17, 2004 at 02:52:09PM +0300, Raimonds Treimanis wrote:
> Hi!
> 
> We have 7206VXR with NPE-300, 1Ge/1e I/O
> Its connected to our upstreem provider via STM-1 pos adapter and via gig ethernet to our backbone, no other PAs installed.
> Has full bgp table (140k prefixes). CEF enabled, NetfLow switching enabled.
> Running 12.2(18)S4 IOS
> Peak traffic is about 100Mbps, approx 30k pps
> And for some reason CPU load is 75% average with peaks of 100% when it starts dropping some packets.
> Question is - is it normal for such device, or there is some misconfiguration which eats up all the CPU?
> Cisco www says that this engine should be able to handle 20k pps process switching and 200k CEF/Fast.
> And it is fast switching, at least it says so:

Fast switching may not work well for you if you have traffic destined all over the
place.. It's mostly process switching with exception of running fast-cached routes
inside interrupt routine.

Have you tried turning cef enabled and ensure that it is turned on for the
relevant interfaces? Make sure 'ip route-cache cef' is turned on each interface..
First you need 'ip cef' globally tho..

HTH
-J


-- 
James Jun                                            TowardEX Technologies, Inc.
Technical Lead                        Network Design, Consulting, IT Outsourcing
james at towardex.com                  Boston-based Colocation & Bandwidth Services
cell: 1(978)-394-2867           web: http://www.towardex.com , noc: www.twdx.net