[c-nsp] Sup720 CoPP, limits on CPU performance

[Gert Doering]

Hi,

On Wed, Mar 24, 2010 at 10:49:19PM -0400, Rodney Dunn wrote:
> Explain the glean to me again?

Someone "up-thread" reported about a problem with early(?) 3Bs that would
lead to a surprising interaction of mls rate limiters for glean traffic
and output ACLs of the input interface.  We haven't seen this, but I find
this scary.

Otherwise, using mls rate-limiting for glean and CoPP for the rest should
work out fine.

> Sorry..I'm overloaded but trying to catch up on the thread.
> 
> We, I need to go back and check, implemented periodic punts at one point 
> that would match the glean and all subsequent packets were dropped in 
> hw/under interrupt..it wasn't a 1:1 punt.

The issue wasn't "glean traffic overloading the CPU" but "certain
combinations of CoPP and mls rate-limit eating the wrong packets".

gert

-- 
USENET is *not* the non-clickable part of WWW!
                                                           //www.muc.de/~gert/
Gert Doering - Munich, Germany                             gert at greenie.muc.de
fax: +49-89-35655025                        gert at net.informatik.tu-muenchen.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 305 bytes
Desc: not available
URL: <https://puck.nether.net/pipermail/cisco-nsp/attachments/20100325/9574f300/attachment.bin>