Re: REVIEW NEEDED: draft-ietf-v6ops-ent-analysis-00.txt

[Tim Chown <tjc@ecs.soton.ac.uk>]

On Mon, Oct 25, 2004 at 02:03:57PM +0200, Brian E Carpenter wrote:
> Thanks Tim, just one follow-up
> 
> 
> >Why use 6to4 or a broker for a large enterprise when a specific tunnel
> >can be set up?
> 
> Indeed, that is better if possible. But with a disobliging ISP, it
> might not be possible.

Well, I admit in the academic networking world we have it easy(er) in
that our NRENs offer v4 and v6, so we just get a tunnel if needed (as
it happens we're native, if 6PE counts as native).

However, I don't see it as a big problem for a site to get a tunnel
set up.   A very successful IPv6 deployment in Europe was in Germany
on the 6WiN where they had 300+ commercial and academic tunnels to
end sites, thanks to the JOIN project.  
 
> >Well, "apart from updating the prefix" menas a whole site renumbering
> >exercise, and that's uncessary pain when you can use 2001: space from an
> >ISP who is quite likely to be your future native IPv6 provider.
> 
> Again, that doesn't work with a disobliging ISP. And we just *have* to
> make prefix updates a feasible option for IPv6; otherwise we will slowly
> slide back to a pre-CIDR state as sites wave money at ISPs.

I agree, hence draft-chown-v6ops-renumber-thinkabout-00 to complement Fred's
work... but I would rather recommend tools that reduce the likelihood of
renumbering in the first place.

-- 
Tim