Hello, I can't bear to watch this anymore.

On 08/15/2017 02:29 PM, Mark Elkins wrote:
> [...]
> I think the cleanest way would be a new pseudo record (ANYA) as the
> reply would have to be a single complete resource set of all the
> possible answers (A's and AAAA's), all with one covering signature if
> DNSSEC is involved. One would then programmatically know then what was
> available.

Such changes would *not* help, really.  You can ask both A and AAAA in
separate parallel queries easily.  If something like ANYA was asked
instead, you would run into the risk of the server not implementing it
and introducing yet another round trip (in the better not-implemented
cases).

The main question on client side is how long to wait if one answer comes
and the other doesn't, but that does seem better decided by the client
(with separate queries) than by intermediate resolvers (with a single
query).  On resolver side, one could recommend some prefetching
strategies, as mentioned, and I can imagine having some such
recommendations in rfc6555bis (non-mandatory).

--Vladimir

_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to