Re: 6to4 relays [Re: WG Review: IPv6 Operations (v6ops)]

[David Terrell <dbt@meat.net>]

On Fri, Sep 13, 2002 at 05:54:28PM -0400, Keith Moore wrote:
>>To realize a "generalized host-based 6to4", one would need to
>>incorporate the NAT traversal mechanisms first pioneered by TEREDO
>>and the two-stage (end-to-edge; edge-to-internet) tunneling mechanism
>>first pioneered by ISATAP.
>
>No, that's not "generalized host-based 6to4", that's "6to4 trying
>to accomodate NAT brain-damage".  A solution which works well (okay,
>as well as possible) in the presence of NATs will be very sub-optimal
>between hosts whose communications is not hampered by NAT.  Which
>is why something like 6to4 and something like Teredo are both needed.

to diverge slightly on this topic, it would be kind of nice if 6to4
could turn into a more general transition mechanism for non-NAT
servers; i.e. if many 6to4 aware machines could implicitly bind to
2002:ipv4::something[*], v6only servers could try to connect to
that as an alternative to something like the site-local faith --
that could of course be a fallback for things that don't understand
this mechanism ("implicit host-based 6to4").

I was hoping to zip out a draft on this behavior before the deadline
but obviously I didn't make it.

* - Someone mentioned that Win XP is listening on 2002:ipv4::ipv4?

-- 
David Terrell             | "War is peace, 
Prime Minister, Nebcorp   | freedom is slavery, 
dbt@meat.net              | ignorance is strength 
http://wwn.nebcorp.com/   | Dishes are clean." - Chris Fester