[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: ocean: do not boil
- To: Margaret Wasserman <mrw@windriver.com>
- Subject: RE: ocean: do not boil
- From: Suresh K Satapati <satapati@cisco.com>
- Date: Thu, 26 Sep 2002 09:24:05 -0700 (PDT)
- Cc: v6ops@ops.ietf.org
- Delivery-date: Thu, 26 Sep 2002 09:25:19 -0700
- Envelope-to: v6ops-data@psg.com
> - Running special DNS resolver code on the hosts and/or
> configuring DNS servers to return AAAA IPv4 mapped
> addresses for any A records.
DNS-ALG does this. resolver code does not have to chang in NAT-PT case
> - The NAT-PT solution involves changes to DNS (either to
> resolver on each host, or to the server), and the IPv4
> NAT solution does not.
both NAT and NAT-PT do not involve any changes to DNS resolver/server
>
> Are there some other technical advantages to the NAT-PT solution that I
> am missing?
technically, network has IPv6-only traffic. router either runs dual-stack+v4NAT
or dual-stack+NAT-PT; both may not be equal in terms of complexity, as Erik
already mentioned the ALG factor which is more in v4NAT today than in
NAT-PT.
there are clearly operational advantages, as mentioned already by folks.
>
> How will hosts know that they are running in a NAT-PT environment,
> and that they shouldn't send IPv4 traffic?
the problem here is dual-stack behavior is not clear. i tend to think they
would try to resolve using a AAAA first and then A, there might be
implementations that are doing A first, then AAAA
there is a problem for dual-stacks that are behind a NAT-PT, which i will
come to later.
>
> Is the routing set-up necessary to support multiple NAT-PT boxes the
> same as the routing set-up needed to support multiple IPv4 NATs? What
> are the key differences, if any?
no differences.
>
> I am concerned about the DNS modifications needed to make NAT-PT work
> correctly. We know that applications that currently work behind an
> IPv4 NAT will work properly with the IPv4 NAT choice. Are we _sure_
> that all of those applications will work properly in the NAT-PT case?
as i said, no DNS modifications are needed. just like in v4NAT, one would
point host resolver to a NAT-PT box. apps will work as they were with
v4NAT, as long as ALG support is there with NAT-PT
> Who has explored this in detail, and what did you find?
while Alain did explore this, there are solutions to overcome most
problems that Alain pointed out except DNS-SEC.
Now the dual-stack problem: with NAT-PT, dual-stack will always be forced
to use a IPv6-mapped address (translation path) that the DNS-ALG
synthesized for a v4 destination. dual-stack could very well use IPv4 to
talk to the destination, but is forced to use IPv6+NAT-PT because of
DNS-ALG's current behavior. this is assuming dual-stack is sending AAAA
first, and then A; but not true for other cases.
-Suresh