[16441] in North American Network Operators' Group
Re: Network Operators and smurf
daemon@ATHENA.MIT.EDU (Brian Holt)
Sat Apr 25 17:03:56 1998
Date: Sat, 25 Apr 1998 15:56:40 -0500
To: Al Reuben <alex@nac.net>, Havard.Eidnes@runit.sintef.no
From: Brian Holt <holt@nap.net>
Cc: jra@scfn.thpl.lib.fl.us, nanog@merit.edu
In-Reply-To: <Pine.BSF.3.96.980425122657.13390W-100000@iago.nac.net>
At 12:30 PM 4/25/98 -0400, Al Reuben wrote:
>The big question is, what do you do if most of your traffic _is_
>asymetrical? I mean, a more basic check could be, "Does the network that
>this packet was sourced from exist *at all*?", or "Do I have a route back
>to the source network through *any* interface?"
If your network is configured for shortest-exit routing, the peer routers
that you would most likely use this on would have a very symmetrical view
on traffic. This would appear to make this option look very enticing..
My major concern in using this is exactly how large of a CPU hit running
this function costs. Does anyone have a feel on what this does to RSP2s &
4s on 75xx series routers with a decent amount of tables and traffic?
Bri
--
Brian Holt
Network Engineer
Nap.Net, L.L.C.
(800)801-3920
