[dnsoverhttp] Thinking about inlining DNS in HTTP

Patrick McManus <pmcmanus@mozilla.com> Fri, 16 September 2016 16:29 UTC

Return-Path: <pmcmanus@mozilla.com>
X-Original-To: dnsoverhttp@ietfa.amsl.com
Delivered-To: dnsoverhttp@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F16C812B2CA for <dnsoverhttp@ietfa.amsl.com>; Fri, 16 Sep 2016 09:29:48 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.234
X-Spam-Level:
X-Spam-Status: No, score=-1.234 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_SORBS_SPAM=0.001, SPF_HELO_PASS=-0.001, SPF_SOFTFAIL=0.665] autolearn=no 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 vdTA7zkZlXoa for <dnsoverhttp@ietfa.amsl.com>; Fri, 16 Sep 2016 09:29:47 -0700 (PDT)
Received: from linode64.ducksong.com (www.ducksong.com [192.155.95.102]) by ietfa.amsl.com (Postfix) with ESMTP id E3B9D12B2BB for <dnsoverhttp@ietf.org>; Fri, 16 Sep 2016 09:29:46 -0700 (PDT)
Received: from mail-io0-f170.google.com (mail-io0-f170.google.com [209.85.223.170]) by linode64.ducksong.com (Postfix) with ESMTPSA id 676CD3A06D for <dnsoverhttp@ietf.org>; Fri, 16 Sep 2016 12:29:38 -0400 (EDT)
Received: by mail-io0-f170.google.com with SMTP id q92so31208327ioi.1 for <dnsoverhttp@ietf.org>; Fri, 16 Sep 2016 09:29:38 -0700 (PDT)
X-Gm-Message-State: AE9vXwPcJC6cpLRf7Pa+f2JEPkMpN8V+c2hGShYBSQPOdX9a0pQoiqTCFGyEFSbI5Ke0m0aFyCZ8TLkEa3xNRg==
X-Received: by 10.107.19.37 with SMTP id b37mr25339449ioj.162.1474043377862; Fri, 16 Sep 2016 09:29:37 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.64.148.50 with HTTP; Fri, 16 Sep 2016 09:29:37 -0700 (PDT)
From: Patrick McManus <pmcmanus@mozilla.com>
Date: Fri, 16 Sep 2016 12:29:37 -0400
X-Gmail-Original-Message-ID: <CAOdDvNqBUacK5oCs3V4sYi+F3Yhqf2e5NiNC1LuPz+jAeuQ4gw@mail.gmail.com>
Message-ID: <CAOdDvNqBUacK5oCs3V4sYi+F3Yhqf2e5NiNC1LuPz+jAeuQ4gw@mail.gmail.com>
To: dnsoverhttp@ietf.org
Content-Type: multipart/alternative; boundary=001a113dea0eebda2b053ca2773d
Archived-At: <https://mailarchive.ietf.org/arch/msg/dnsoverhttp/AQRnfaBwcaZjPiNMjXOooOv_51Q>
Subject: [dnsoverhttp] Thinking about inlining DNS in HTTP
X-BeenThere: dnsoverhttp@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Discussion of DNS over HTTP <dnsoverhttp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/dnsoverhttp>, <mailto:dnsoverhttp-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/dnsoverhttp/>
List-Post: <mailto:dnsoverhttp@ietf.org>
List-Help: <mailto:dnsoverhttp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/dnsoverhttp>, <mailto:dnsoverhttp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 16 Sep 2016 16:29:49 -0000

Anyone want to talk about the general model of pushing DNS records to
clients as part of HTTP? This was originally part of the SPDY documents,
but was never fully implemented there and never made it into HTTP/2 due to
its experimental and speculative nature (unlike the rest of what became
h2). But a number of folks, including myself, think the idea still holds
significant promise.

Briefly, if www.example.com were to include references to content on
www.cdn.com it would be able to staple, unsolicited, at the protocol level
the necessary DNS records for the cdn into the http response containing the
reference. .presumably with a complete dnssec chain inline. Since
example.com pushes this information to different folks all the time, it can
have the chain cached suitably to the lifetime of the records in the chain.
Because the client can verify the information via dnssec it can go ahead
and connect www.cdn.com without the round trips necessary for resolution.

Remembering that this is bar bof fodder, If the HTTP folks were interested
in going down this path with an extension to h2 what would the guidance
from the dns folk be on corner cases to consider?

-Patrick