In an IPv6 future, how will you solve IPv4 connectivity?

[Martin Millnert]

On Sun, 2010-10-10 at 16:05 +0200, Roger Wiklund wrote:
> Also, as these new users are IPv6 only, how can IPv4 hosts communicate
> with them? 4to6 NAT? 

Absent some elsewhere located state information and translation
procedures (say, a DNS46/4to6 NAT solution), it is impossible for a
IPv4-only host to arbitrarily initiate communication with any given IPv6
host. 2^128 >>> 2^32.
  DNS64 only solves DNS64-compliant communication, as you know. No
IPv4/IPv6 translation will enable pure IP communication like we are used
to from the legacy Internet. Eg., there will always be problems with
these sorts of hacks.
  (Additionally, IPv6 proponents have long awaited the point where a
IPv4-only host actually "must" communicate with a IPv6-only host, as
this incentives the IPv4-only host to get with the times and migrate.
Unless it affects their own living, I don't see why these proponents
would care if it couldn't. )

In the general case, the easiest and best solution to get IPv4-only
hosts to talk to IPv6-only hosts is likely to remain for said IPv4 hosts
to get IPv6 connectivity.

When that is not possible, you're left with the hacks, some of which
you've mentioned. The hacks have different "solution spaces" and it is
vital to understand that none can really cure the illness of the
connection not being natively routed IPv6-to-IPv6.

Just enable IPv6. ;)

Regards,
-- 
Martin Millnert <martin at millnert.se>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part
Url : http://lists.cluenet.de/pipermail/ipv6-ops/attachments/20101010/7615c50e/attachment.bin