sure
the whole point of doing anything is to try and avoid mutliple lookup
hacks, like the existing concurrent A and AAAA lookups.
What about allowing the AAAA responses in an A query? Even if only in
additional.
Adrien
------ Original Message ------
From: "Rob Austein" <s...@hactrn.net>
To: "dnsop@ietf.org" <dnsop@ietf.org>
Sent: 12/04/2016 12:08:10 p.m.
Subject: Re: [DNSOP] AAAA4Free
At Tue, 12 Apr 2016 00:01:20 +0000, Evan Hunt wrote:
So, unless I'm missing something (certainly more than possible), I
don't
see server workloads or client latency being significantly reduced by
the
deployment of a mechanism like this.
Bingo.
Also keep in mind that A and AAAA for the same QNAME are in the same
zone, so by the time you've fetched (and validated, I hope) the chain
all the way down to one of them, you've done almost all of the work
involved in fetching (...) the other one.
I remain skeptical of the claim that the round trip time to the
iterative resolver is the critical factor. Or, to put it another way,
if that's the critical factor, you almost certainly have other
problems that a multiple query hack isn't going to solve.
_______________________________________________
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop
_______________________________________________
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop
