Re: Last Call: <draft-ietf-httpbis-http2-16.txt> (Hypertext Transfer Protocol version 2) to Proposed Standard

John C Klensin <> Sat, 03 January 2015 14:12 UTC

Return-Path: <>
Received: from localhost ( []) by (Postfix) with ESMTP id 8D30B1A1A93 for <>; Sat, 3 Jan 2015 06:12:49 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: 3
X-Spam-Level: ***
X-Spam-Status: No, score=3 tagged_above=-999 required=5 tests=[BAYES_20=-0.001, FH_HOST_EQ_D_D_D_D=0.765, HELO_MISMATCH_COM=0.553, HOST_EQ_STATICB=1.372, HOST_MISMATCH_NET=0.311] autolearn=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id n5Ha2w5TqeMN for <>; Sat, 3 Jan 2015 06:12:48 -0800 (PST)
Received: from ( []) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 62C531A1A8F for <>; Sat, 3 Jan 2015 06:12:48 -0800 (PST)
Received: from ([] helo=P5) by with esmtp (Exim 4.82 (FreeBSD)) (envelope-from <>) id 1Y7PRb-0002jc-Ks; Sat, 03 Jan 2015 09:12:43 -0500
Date: Sat, 03 Jan 2015 09:12:38 -0500
From: John C Klensin <>
To: Eliot Lear <>, Delan Azabani <>,
Subject: Re: Last Call: <draft-ietf-httpbis-http2-16.txt> (Hypertext Transfer Protocol version 2) to Proposed Standard
Message-ID: <AEE3A9C73AC6A1C6D6BFAD95@P5>
In-Reply-To: <>
References: <> <>
X-Mailer: Mulberry/4.0.8 (Win32)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: IETF-Discussion <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Sat, 03 Jan 2015 14:12:49 -0000

One comment in addition to Eliot's, which while I generally

--On Saturday, 03 January, 2015 13:09 +0100 Eliot Lear
<> wrote:

> On 1/2/15 9:29 AM, Delan Azabani wrote:

>> HTTP is simultaneously important enough that one can't simply
>> run a single server for any popular application, but not so
>> important that it deserves to necessarily be the A/AAAA
>> record for every hostname.

If I understand the above part of your comments correctly, "one
A/AAAA record..." hasn't been necessary since HTTP 1.1
introduced the HTTP "host" verb.  That was a lesson we learned
from SMTP, FTP, and elsewhere -- virtual hosts should not
require separate (either physical or virtual) interfaces and

That decision was made well before SRV records were sufficiently
deployed to be meaningful, but Eliot's point stands: there were
(and are) a number of complicated tradeoffs in this situation (I
don't think his list is complete, but both of you probably know
that), and the community made a rational decision about them.
Whether that decision should be reviewed today might be another
question, but the transition and deployment issues would
themselves become another tradeoff.