RE: on NAT-PT

["Karim El-Malki (EAB)" <Karim.El-Malki@era.ericsson.se>]

Hi Margaret

 > In the 3G case, it seems likely that the NA(P)T-PT 
 > functionality would
 > co-reside with the GGSN function.

I would disagree, there's nothing that forces you to have the GGSN
and NAPT-PT co-located. I don't think this assumption is in any
of the 3gpp-ietf docs. Having the two co-located seems like a
special case to me.

 > 
 > So, I think that we should issue a strong recommendation 
 > against using NAT-PT
 > in any environment and move it to historical?

I think the 3G scenario is helping IPv6 deployment by mandating
IPv6 only for new multimedia applications. This helps in forcing people
to go to v6, but it also means we need to support communication to
ipv4 hosts. The 3g link-layer (pdp context) cannot support ipv4 and ipv6
together, so a phone having an ipv6 pdp context needs application proxies
and translators to communicate with ipv4 hosts. Application proxies can
handle the typical app.s (e.g. http) and the translator could be used for
the remaining. That would reduce the use of translators, but still they
would be needed. So I would disagree on such a strong recommendation.
Rather I would prefer recommending application proxies for most used app.s
and leave napt-pt (with adequate changes) for the remaining.

Rgds
/Karim