|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |
Date: Thu, 18 Mar 1999 18:34:58 -0500 From: Jared Mauch <jared@puck.Nether.net> To: Tony Bourke <tonyb@globalcenter.net> Cc: nanog@merit.edu Mail-Followup-To: Tony Bourke <tonyb@globalcenter.net>, nanog@merit.edu In-Reply-To: <Pine.GSO.4.05.9903182326170.8380-100000@pobox.iad1.gctr.net>; from Tony Bourke on Thu, Mar 18, 1999 at 11:26:46PM +0000 Errors-To: owner-nanog-outgoing@merit.edu Yes, this is what I understand is the case. Most people I know that are talking about pushing this much traffic can use the PA-2FE and load-share, or use the POS OC3 pa. - jared On Thu, Mar 18, 1999 at 11:26:46PM +0000, Tony Bourke wrote: > > This looks kinda like the right place for this issue, apologies if not: > > True or False? > > FEC (Fast EtherChannel) is not suibtable for router to router trunking > because of the way FEC load balances > > I've heard and read in various places (although nothing conclusive) that > since FEC uses the last portion of a MAC address and hashs it to determine > which 100 meg link a packet goes. This is to prevent packets from > arriving on the other end out of order. > > This would seem to suggest that FEC wouldn't work between routers or > similar layouts because there is only one MAC address on either end, > therefor packets would always go over the same link. > > Anybody have concrete details on this issue? Does FEC actually determine > load balancing this say? is ther another way to configure this? > > I'm not real fond of FEC myself, and would like substantiated evidence in > an argument against it. -- Jared Mauch | pgp key available via finger from jared@puck.nether.net clue++; | http://puck.nether.net/~jared/ My statements are only mine.
|
|
|
|
|
|
|
|
|
|
|
|
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |