Re: Requirements [was Re: Transport level multihoming]

[Greg Maxwell <gmaxwell@martin.fl.us>]

On Tue, 3 Apr 2001, Brian E Carpenter wrote:

[snip]
> That's fair comment, but remember that we have legacy IPv6 stacks already.
> We certainly can't assume that host IPv6 stacks will all get upgraded
> rapidly to cover such changes - i.e. hosts must not lose connectivity if 
> they are not upgraded for multihoming, even if they don't get full benefit
> from the multihoming.

My idea involves an overlay on the standard API that allows existing
applications the ability to take advantage of the multihoming transport
with little or no changes. Obviously greater benefits could be achieved
by an application which understands it's transport is self-multihoming.

Legacy hosts (those not supporting a multihoming transport) will not
achieve the benifit of multihoming. Which I think would be lend a
considerable advantage to upgrading without being so important to require
a significantly seperate method of achieveing globally visiable 
which degrades aggregation multihoming in IPv6.

It look like I'm going to be working on a draft for this; The questions
I've recieved here are of great value and will contribute consideribly to
the points I address.