[89065] in North American Network Operators' Group
second v6 multihoming paper
daemon@ATHENA.MIT.EDU (Lucy E. Lynch)
Wed Mar 1 22:38:44 2006
Date: Wed, 1 Mar 2006 19:38:09 -0800 (PST)
From: "Lucy E. Lynch" <llynch@darkwing.uoregon.edu>
To: nanog@merit.edu
Errors-To: owner-nanog@merit.edu
http://www.ecs.soton.ac.uk/~ajf101/irp-ajf101-multihoming.pdf
the money quote:
"The lack of a standardised solution to multihoming remains a large issue
frustrating wider-scale deployment of IPv6, as many large sites rely on
multihoming for connection reliability and optimality. The proposed multihoming
solutions presented in this paper can be broadly grouped into two categories:
those that support connection survivability, and those that do not. Those that
do, however, are invariably more complex, and require significant protocol
changes, thus breaking the "simplicity" aim as given in RFC3582 [23].
Therefore, it can be said that no proposed solution yet meets these goals.
Having said that, it is difficult to see how such a solution, that is both
simple and comprehensive, can be built, working in the new environment that is
IPv6. After at least five years of proposals, nothing that solves all goals has
yet been suggested."
Some of you may already know Alan Ford as a v6ops co-author:
http://www.potaroo.net/ietf/all-ids/draft-chown-v6ops-renumber-thinkabout-03.txt
and he also worked on the re-numbering presentation at the paris IETF
http://www3.ietf.org/proceedings/05aug/slides/v6ops-7.pdf
- lel