[49643] in North American Network Operators' Group
Re: Internet vulnerabilities
daemon@ATHENA.MIT.EDU (Bill Woodcock)
Mon Jul 8 02:31:44 2002
Date: Sun, 7 Jul 2002 23:31:00 -0700 (PDT)
From: Bill Woodcock <woody@zocalo.net>
To: Richard A Steenbergen <ras@e-gerbil.net>
Cc: Stephen Griffin <stephen.griffin@rcn.com>, <nanog@merit.edu>
In-Reply-To: <20020708020227.GJ99199@overlord.e-gerbil.net>
Errors-To: owner-nanog-outgoing@merit.edu
On Sun, 7 Jul 2002, Richard A Steenbergen wrote:
> I think the problem they are refering to is what happens if your routing
> topology changes (or worse, flaps). A stateful connection (like TCP) which
> would have stayed up during a routing change could potentially be shifted
> to a different server which obviously wouldn't know the other one's state.
Yes. As I said in a previous message in this thread, that's a common
objection brought up by people who've never run an anycast network and are
trying to think of reasons why it might be problematic. But since in the
real world that appears to happen two orders of magnitude less frequently
than connection failures due to loss of connectivity, _when you take no
steps to prevent it_, and the prevention is both trivial and necessary
with HTTP, which is the protocol most commonly anycasted, it's not an
issue at all.
-Bill
