Re: [DNSOP] My "toxic" remark at the mic today

"Patrik Fältström " <> Fri, 06 November 2015 06:11 UTC

Return-Path: <>
Received: from localhost ( []) by (Postfix) with ESMTP id C7F5C1A8954 for <>; Thu, 5 Nov 2015 22:11:38 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.261
X-Spam-Status: No, score=-1.261 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HELO_EQ_SE=0.35, MIME_8BIT_HEADER=0.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id VB-8oeuYpm9F for <>; Thu, 5 Nov 2015 22:11:38 -0800 (PST)
Received: from ( [IPv6:2a02:80:3ffe::176]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id BF9091A8949 for <>; Thu, 5 Nov 2015 22:11:37 -0800 (PST)
Received: from [] ( []) by (Postfix) with ESMTPSA id 5AC281FDE4; Fri, 6 Nov 2015 07:11:36 +0100 (CET)
From: Patrik Fältström <>
To: Andrew Sullivan <>
Date: Fri, 06 Nov 2015 07:11:35 +0100
Message-ID: <>
In-Reply-To: <>
References: <> <> <>
MIME-Version: 1.0
Content-Type: multipart/signed; boundary="=_MailMate_1EF66AA5-976C-4C53-8D17-F4640EB6D293_="; micalg="pgp-sha1"; protocol="application/pgp-signature"
X-Mailer: MailMate (1.9.3r5164)
Archived-At: <>
Subject: Re: [DNSOP] My "toxic" remark at the mic today
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: IETF DNSOP WG mailing list <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Fri, 06 Nov 2015 06:11:39 -0000

On 6 Nov 2015, at 4:54, Andrew Sullivan wrote:

> On Thu, Nov 05, 2015 at 03:32:43PM +0900, Paul Hoffman wrote:
>> No, but there is an RFC from the IAB about what labels should not be into
>> the root without further consideration: RFC 4690. That has been widely
>> interpreted as "do not put X into the root", so it is similar to "do not put
>> RRtype Y into the root".
> With respect, labels and RRTYPEs are quite different things.  I think
> perhaps conflating these two things is not super helpful.  But I don't
> have a strong feeling about this, so long as we're not saying anything
> that can be construed as positive advice about what should go into the
> zone.

Paul, you are sure you talk about RFC4690 and not RFC5507?

Being one of the editors of RFC5507 (and 4690!) let me emphasize what Andrew wrote here. Yes, maybe people have conflated and simplified what we wrote, but the RFC is actually trying to be very very careful with wording and it would be said if text that once was very specific once again end up being watered down.

The key message in RFC5507 is that one should minimize the RRSet size, and think carefully if the selection criteria between RR is in the RRType or owner.