Re: Tunnel-to-NAT scenario

[Remi Denis-Courmont <rdenis@simphalempin.com>]

On Tue, 17 Jun 2008 11:09:11 +1200, Brian E Carpenter

<brian.e.carpenter@gmail.com> wrote:

> I believe that we also need to come up with a solution for an

> IPv4 initiator reaching a server with IPv6-only connectivity.

> My question is whether we can be satisfied with a solution

> that requires that server to be dual stack, so that it can

> tunnel IPv4 in IPv6 to a conventional IPv4-to-IPv4 NAT.

> (Crude diagram below.)



Operating systems will continue to ship with IPv4 built-in for as long as

it is useful (i.e. very long). On top of that, IPv6 and IPv4 support tend

to be intertwined if only for the sake of dual-stack functions (mapped

addresses, address selection, automatic transition mechanism setup, etc).



So I guess we can assume IPv4 support for as long as it is potentially

useful. This may also be a self-realizing prophecy: if dual-stack is needed

on IPv6-only nodes, IPv6-only nodes will continue to have dual-stack.



> That seems a lot simpler than developing complete MNAT-PT

> or SHANTI solutions, which in can case can never really

> offer more transparency that conventional NAT.



Indeed.



-- 

Rémi Denis-Courmont

http://www.remlab.net