[c-nsp] Sup720-3BXL load balancing

[Rohit Shetti]

Hi,
	By default, CEF load balances on a per-source-destination ( I call
that 'per-flow') basis, which might result in certain interfaces forwarding
more traffic than others, despite the paths being of equal cost. If we have
more number of  flows then only there is possibility of cef doing almost
equal path load balancing.
 I doubt that the "problem" disappeared by a BGP reset at remote peer since
u had already done resets at your end. I rather suspect that at remote peer,
the flow has split and cef has done load-balancing on these flows.

Regards,
Rohit

-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net
[mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Kristofer Sigurdsson
Sent: Friday, April 07, 2006 3:42 PM
To: bep at whack.org
Cc: cisco-nsp at puck.nether.net
Subject: Re: [c-nsp] Sup720-3BXL load balancing

Hi,

> BGP and the main routing table are used to establish the CEF entries 
> (FIB) that are actually used for forwarding decisions.  FIB will share
based on
> src/dst address.   What does CEF show for that route?

I'm afraid the "problem" disappeared after a BGP reset on the other end
(weird, huh?).  CEF says the paths are equal and my graphs now show equal
traffic on both links...

I'm guessing they screwed something up on the other end and had just fixed
it when I gathered the data needed for my mail...

Thanks for the help,
Kristo

_______________________________________________
cisco-nsp mailing list  cisco-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/