Thoughts about ipv6 white listing

[Leen Besselink]

>>        it is prudent to use the same transport the query used for your response. :)
>>
> Also, i think the original design that was proposed at the top of the
> thread can be better met by this BIND feature that only serves AAAA to
> IPv6 transport, likely of the recursive server which is not a
> definitive indication of the end-node's ipv6 ability
>
> http://www.ietf.org/proceedings/77/slides/dnsop-7.pdf
I think the way it is proposed in the pdf would be mostly OK, where the
access-provider deployed IPv6 on his own network for his users and the
recursor of the access-provider only answers AAAA records for those
users that ask over IPv6.

The original poster of this thread was talking about an authoritive
server. Where you really have to little information about the users
situation to make any decision.