[40813] in North American Network Operators' Group
RE: Routescience?
daemon@ATHENA.MIT.EDU (E.B. Dreger)
Thu Aug 23 01:02:54 2001
Date: Thu, 23 Aug 2001 05:01:57 +0000 (GMT)
From: "E.B. Dreger" <eddy+public+spam@noc.everquick.net>
To: nanog@merit.edu
Message-ID: <Pine.LNX.4.20.0108230459370.7083-100000@www.everquick.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: owner-nanog-outgoing@merit.edu
Sorry to respond/add to my own post, but there is another possibility
that I forgot to mention:
Create subinterfaces on the border router(s). Let's say that I'm
connected to three upstreams, and communicate with the LAN via f0/0.
Assume single border router.
I create f0/0.1, f0/0.2, and f0/0.3, each in separate /30s. When the
RouteScience box (presumed to operate using WCCP) receives a packet
that it wishes to test, it forwards to the appropriate border router
subint. The border router sets next-hop to a specific value that
correlates to the ingress f0/0.x subint.
Just don't round-robin packets from the same stream over different
paths. (!) Unacceptable even for a quick test.
Back to lurkerland,
On to dreamland,
Eddy
---------------------------------------------------------------------------
Brotsman & Dreger, Inc. - EverQuick Internet Division
Phone: +1 (316) 794-8922 Wichita/(Inter)national
Phone: +1 (785) 865-5885 Lawrence
---------------------------------------------------------------------------
Date: Mon, 21 May 2001 11:23:58 +0000 (GMT)
From: A Trap <blacklist@brics.com>
To: blacklist@brics.com
Subject: Please ignore this portion of my mail signature.
These last few lines are a trap for address-harvesting spambots. Do NOT
send mail to <blacklist@brics.com>, or you are likely to be blocked.
