[prev in list] [next in list] [prev in thread] [next in thread]
List: ipng
Subject: Re: [v6ops] DNS AAAA for free
From: Tim Chown <Tim.Chown () jisc ! ac ! uk>
Date: 2016-04-13 11:18:24
Message-ID: 90CC41CF-9F3D-44DB-8A23-AFB34321748E () jisc ! ac ! uk
[Download RAW message or body]
[Attachment #2 (text/plain)]
On 9 Apr 2016, at 14:11, Mark Smith \
<markzzzsmith@gmail.com<mailto:markzzzsmith@gmail.com>> wrote:
On 9 Apr 2016 8:55 PM, "Sander Steffann" \
<sander@steffann.nl<mailto:sander@steffann.nl>> wrote:
>
> Hi,
>
> > On 08 Apr 2016, at 13:27, otroan@employees.org<mailto:otroan@employees.org> \
> > wrote:
> > Here is an interesting draft being presented in dnsop today.
> > https://datatracker.ietf.org/doc/draft-vavrusa-dnsop-aaaa-for-free/?include_text=1
> >
> > The proposal is to allow AAAA records in response to A queries.
>
> One question: if clients send requests for both A and AAAA records simultaneously, \
> how will including the AAAA answer in the A request help? The client will just get \
> the AAAA answer twice: once on the A request and once on the AAAA request. If \
> lookups were serialised and the A request/response was completed before starting \
> the AAAA request it would make the second request unnecessary, but that's not what \
> I see in practice...
> It looks like good intentions, I'm just trying to figure out how this would work in \
> practice :)
Having briefly looked at it, it seems there is an efficiency benefit. However, \
answering an A query with both an A and AAAA response might incentivise just sending \
A queries.
Perhaps the opposite would be better - sending an AAAA query would result in an AAAA \
and A answer. That might create a small and subtle incentive to use and prefer IPv6, \
with Happy Eyeballs covering over IPv6 failures.
I am aware of ISPs who have deployed IPv6 to their customers, knowing they needed to \
scale their DNS provision, but have still had to hold further deployment until this \
was achieved. It's not a huge issue, but any measures to help address scalability as \
you add IPv6 are likely to be welcomed.
But as others have pointed out, there are practicalities with the proposal.
Tim
[Attachment #3 (text/html)]
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: \
after-white-space;" class=""> <div>
<blockquote type="cite" class="">
<div class="">On 9 Apr 2016, at 14:11, Mark Smith <<a \
href="mailto:markzzzsmith@gmail.com" class="">markzzzsmith@gmail.com</a>> \
wrote:</div> <div class="">
<p dir="ltr" class="">On 9 Apr 2016 8:55 PM, "Sander Steffann" <<a \
href="mailto:sander@steffann.nl" class="">sander@steffann.nl</a>> wrote:<br \
class=""> ><br class="">
> Hi,<br class="">
><br class="">
> > On 08 Apr 2016, at 13:27, <a href="mailto:otroan@employees.org" \
class="">otroan@employees.org</a> wrote:<br class=""> > ><br class="">
> > Here is an interesting draft being presented in dnsop today.<br class="">
> > <a href="https://datatracker.ietf.org/doc/draft-vavrusa-dnsop-aaaa-for-free/?include_text=1" \
class=""> https://datatracker.ietf.org/doc/draft-vavrusa-dnsop-aaaa-for-free/?include_text=1</a><br \
class=""> > ><br class="">
> > The proposal is to allow AAAA records in response to A queries.<br \
class=""> ><br class="">
> One question: if clients send requests for both A and AAAA records \
simultaneously, how will including the AAAA answer in the A request help? The client \
will just get the AAAA answer twice: once on the A request and once on the AAAA \
request. If lookups were serialised and the A request/response was completed before \
starting the AAAA request it would make the second request unnecessary, but that's \
not what I see in practice...<br class=""> ><br class="">
> It looks like good intentions, I'm just trying to figure out how this would work \
in practice :)<br class=""> </p>
<p dir="ltr" class="">Having briefly looked at it, it seems there is an efficiency \
benefit. However, answering an A query with both an A and AAAA response might \
incentivise just sending A queries.</p> <p dir="ltr" class="">Perhaps the opposite \
would be better - sending an AAAA query would result in an AAAA and A answer. That \
might create a small and subtle incentive to use and prefer IPv6, with Happy Eyeballs \
covering over IPv6 failures.</p> </div>
</blockquote>
<div class="">I am aware of ISPs who have deployed IPv6 to their customers, knowing \
they needed to scale their DNS provision, but have still had to hold further \
deployment until this was achieved. It's not a huge issue, but any measures to \
help address scalability as you add IPv6 are likely to be welcomed.</div>
<div class=""><br class="">
</div>
<div class="">But as others have pointed out, there are practicalities with the \
proposal. </div> <div class=""><br class="">
</div>
<div class="">Tim</div>
</div>
</body>
</html>
--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
--===============0004776784075022511==--
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic