Re: [Bundled-domain-names] A suggestion for 2017

"John Levine" <johnl@taugh.com> Wed, 28 December 2016 21:17 UTC

Return-Path: <johnl@taugh.com>
X-Original-To: bundled-domain-names@ietfa.amsl.com
Delivered-To: bundled-domain-names@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 405711296A4 for <bundled-domain-names@ietfa.amsl.com>; Wed, 28 Dec 2016 13:17:26 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level:
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9ZDUb7QfMLx6 for <bundled-domain-names@ietfa.amsl.com>; Wed, 28 Dec 2016 13:17:25 -0800 (PST)
Received: from miucha.iecc.com (abusenet-1-pt.tunnel.tserv4.nyc4.ipv6.he.net [IPv6:2001:470:1f06:1126::2]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id DA08712969D for <bundled-domain-names@ietf.org>; Wed, 28 Dec 2016 13:17:24 -0800 (PST)
Received: (qmail 99775 invoked from network); 28 Dec 2016 21:17:29 -0000
Received: from unknown (64.57.183.18) by mail1.iecc.com with QMQP; 28 Dec 2016 21:17:29 -0000
Date: Wed, 28 Dec 2016 21:17:01 -0000
Message-ID: <20161228211701.30364.qmail@ary.lan>
From: John Levine <johnl@taugh.com>
To: bundled-domain-names@ietf.org
In-Reply-To: <20161228195030.GA7469@laperouse.bortzmeyer.org>
Organization:
X-Headerized: yes
Mime-Version: 1.0
Content-type: text/plain; charset="utf-8"
Content-transfer-encoding: 8bit
Archived-At: <https://mailarchive.ietf.org/arch/msg/bundled-domain-names/MWnrYIBWaNXGt7_zkpq9fPqUlJk>
Cc: bortzmeyer@nic.fr
Subject: Re: [Bundled-domain-names] A suggestion for 2017
X-BeenThere: bundled-domain-names@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: "Discussion of \"bundled domain names\"" <bundled-domain-names.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/bundled-domain-names>, <mailto:bundled-domain-names-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/bundled-domain-names/>
List-Post: <mailto:bundled-domain-names@ietf.org>
List-Help: <mailto:bundled-domain-names-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/bundled-domain-names>, <mailto:bundled-domain-names-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 28 Dec 2016 21:17:26 -0000

In article <20161228195030.GA7469@laperouse.bortzmeyer.org> you write:
>There have been many attempts to do something with the
>bundle/alias/variant issue, often with proposals to improve/remplace
>CNAME/DNAME/xNAME.
>
>One thing is missing: a repository of previous attempts, with
>explanations of the problems they raise.

That should be helpful, since people do seem to keep reinventing the
same well known bad ideas (WKBI's).  The conclusion we came to at the
most recent meeting is that almost all, and perhaps completely all, of
the problems with provisioning variants can be solved with better DNS
provisioning software, rather than looking for a magic DNS bullet that
will solve the problem for us.

> Case in point:
>draft-sury-dnsext-cname-at-apex was published six years ago, saw
>almost no discussion and expired. Why?

This is an excellent case in point.  It attempts to solve the problem
that you can't put a CNAME at the zone apex, by proposing an
incompatible change to CNAME that lets it work at a zone apex.  An
obvious failing is that there is no way to tell whether or not a
server or cache supports new CNAMEs, so there's no way to deploy it.

More importantly, we've already solved this problem by improving our
provisioning software.  Lots of DNS systems let you say this in a zone
file or equivalent:

foo.example. ANAME bar.example.

That tells the provisioning software to create A records for
foo.example with the contents of the A records at bar.example.
It polls now and then to see if bar.example has changed.

On my system, I have a somewhat overimplemented scheme where
I can say things like this:

  IN A [RMTIP:bar.example]

That means to replace the [RMTIP...] with records for with
the contents of records at bar.example.  If there's more
than one record it repeats the record, and as a special case, if
there are AAAA records it creates AAAA records as well.

www  IN  A   [RMTIP:www.example.org:ns1.somehost.com:A]

That says to replace the [RMTIP...] with the A record(s) for
www.example.org fetched from the name server at ns1.somehost.com.
This solves an annoying problem when I have a web site hosted
somewhere, the mail somewhere else, but the web host insists I use
their nameservers, even though there's no way to put the required MX
records on their nameservers.  Well, OK.

This sort of provisioning stuff is not hard.  I added the RMTIP stuff
to my system in an afternoon.

R's,
John