
From nobody Fri Sep  1 02:02:48 2017
Return-Path: <stephen.farrell@cs.tcd.ie>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 43DD9132418 for <acme@ietfa.amsl.com>; Fri,  1 Sep 2017 02:02:46 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.301
X-Spam-Level: 
X-Spam-Status: No, score=-4.301 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cs.tcd.ie
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 nO21LweAClvh for <acme@ietfa.amsl.com>; Fri,  1 Sep 2017 02:02:43 -0700 (PDT)
Received: from mercury.scss.tcd.ie (mercury.scss.tcd.ie [134.226.56.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 79EBF132EFC for <acme@ietf.org>; Fri,  1 Sep 2017 02:02:43 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by mercury.scss.tcd.ie (Postfix) with ESMTP id 0CFD2BEAA; Fri,  1 Sep 2017 10:02:39 +0100 (IST)
X-Virus-Scanned: Debian amavisd-new at scss.tcd.ie
Received: from mercury.scss.tcd.ie ([127.0.0.1]) by localhost (mercury.scss.tcd.ie [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fUuJBhN1QKM1; Fri,  1 Sep 2017 10:02:30 +0100 (IST)
Received: from [10.244.2.100] (95-45-153-252-dynamic.agg2.phb.bdt-fng.eircom.net [95.45.153.252]) by mercury.scss.tcd.ie (Postfix) with ESMTPSA id 7EBB5BE80; Fri,  1 Sep 2017 10:02:30 +0100 (IST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cs.tcd.ie; s=mail; t=1504256550; bh=ZvTnvikjRWsgcvb9M1wF1dbeqLhIlQ512GEjDpi8w0k=; h=Subject:To:Cc:References:From:Date:In-Reply-To:From; b=ITDiHPNPCSskgNtf8JTuilvqUs4qLHuCtzGbNYBEI1HneHtz/mYkb1aCiFvhUwQDS p+QP+Fn6RQKuwm6wH/pEl3xrtKgmJFxcp2QqdtTXyhrBDywbmPpblIKnyz9cotCEGE AiFp45VV287U1KgY2Tmp9YNCHAZmH33hNWQW+5VQ=
To: Martin Thomson <martin.thomson@gmail.com>, Adam Roach <adam@nostrum.com>
Cc: Ben Schwartz <bemasc@google.com>, "acme@ietf.org" <acme@ietf.org>
References: <CAHbrMsB_8x4C2Y0OjO=a2H8LkFqeHgELAX4_imW2ip7Y7Vj21Q@mail.gmail.com> <d69735c5-3e46-753e-9a68-f27c68707ce5@cs.tcd.ie> <85c8aa7f-52a6-38fc-3700-d8607103889d@nostrum.com> <CABkgnnV0CWKQMo_nxP8rv71f=obtSOLwfHj4YU7hy9mxUgN0PA@mail.gmail.com>
From: Stephen Farrell <stephen.farrell@cs.tcd.ie>
Openpgp: id=D66EA7906F0B897FB2E97D582F3C8736805F8DA2; url=
Message-ID: <03ce0062-78c8-4c6f-2e1f-9d37e8120488@cs.tcd.ie>
Date: Fri, 1 Sep 2017 10:02:29 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.2.1
MIME-Version: 1.0
In-Reply-To: <CABkgnnV0CWKQMo_nxP8rv71f=obtSOLwfHj4YU7hy9mxUgN0PA@mail.gmail.com>
Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="sQaP4lVXg3vSOWfCHKKSlU1T9GEvobjcn"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/PecMZ44Lu1KthYUy5bhCyBFRLH8>
Subject: Re: [Acme] On draft-ietf-acme-ip
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 01 Sep 2017 09:02:46 -0000

This is an OpenPGP/MIME signed message (RFC 4880 and 3156)
--sQaP4lVXg3vSOWfCHKKSlU1T9GEvobjcn
Content-Type: multipart/mixed; boundary="TWesaVHWgV5VFwBl2fQxO0UuUtJSSd6FC";
 protected-headers="v1"
From: Stephen Farrell <stephen.farrell@cs.tcd.ie>
To: Martin Thomson <martin.thomson@gmail.com>, Adam Roach <adam@nostrum.com>
Cc: Ben Schwartz <bemasc@google.com>, "acme@ietf.org" <acme@ietf.org>
Message-ID: <03ce0062-78c8-4c6f-2e1f-9d37e8120488@cs.tcd.ie>
Subject: Re: [Acme] On draft-ietf-acme-ip
References: <CAHbrMsB_8x4C2Y0OjO=a2H8LkFqeHgELAX4_imW2ip7Y7Vj21Q@mail.gmail.com>
 <d69735c5-3e46-753e-9a68-f27c68707ce5@cs.tcd.ie>
 <85c8aa7f-52a6-38fc-3700-d8607103889d@nostrum.com>
 <CABkgnnV0CWKQMo_nxP8rv71f=obtSOLwfHj4YU7hy9mxUgN0PA@mail.gmail.com>
In-Reply-To: <CABkgnnV0CWKQMo_nxP8rv71f=obtSOLwfHj4YU7hy9mxUgN0PA@mail.gmail.com>

--TWesaVHWgV5VFwBl2fQxO0UuUtJSSd6FC
Content-Type: text/plain; charset=utf-8
Content-Language: en-GB
Content-Transfer-Encoding: quoted-printable



On 01/09/17 03:58, Martin Thomson wrote:
> On Fri, Sep 1, 2017 at 10:47 AM, Adam Roach <adam@nostrum.com> wrote:
>> On 8/31/17 19:25, Stephen Farrell wrote:
>>>
>>> I really like the idea that the acme WG aims to figure out a way to e=
nable
>>> people at home to use https with their home n/w routers.
> ...
>> There was some musing at the W3C TPAC in Lisbon last year on this topi=
c. The
>> tricky part is figuring out what kind of model makes sense for the cer=
ts at
>> all. I suspect we'd need to come to some agreement on that issue befor=
e
>> trying to work out how ACME can be used to issue them. There's some
>> background reading at
>> <https://www.w3.org/wiki/TPAC2016/session-https-local-summary>, mostly=
 in
>> the form of slide decks.
>=20
> I don't see acme-ip being the solution here.  Everyone has - or could
> have - a 10.0.0.1.  The same applies to .local (see below).  The
> movement needs to come from the relying party side.
>=20
> Thanks for sharing the link Adam, I was not aware of this.  For the
> benefit of folks in the galleries, the three talks discuss two
> options.
>=20
> The first two talk about providing *real* names for the devices
> (<device-id>.<manufacturer>.com for example).  The nice thing with
> that is that that solution already works today.  With ACME, if the
> manufacturer is willing to answer the challenges, the device only
> needs some way to talk to the manufacturer when it wants a
> certificate, not have an actual online presence.  (Insert usual
> concerns about the manufacturer going out of business, etc...)
>=20

There's also the concern of how to get that scheme to work for
openwrt devices and similar, which is related. It'd be a fail
if that couldn't work I reckon.

Separately, I took a VPN-approach [1] in my highly-specific
environment - I wonder if something like that (replacing the
names and IPv6 addresses I used with device-id based names
and addresses) might address some of the privacy issues with
the first approach described. [2]

S.

[1]
https://my-own.net/ab/forum/question/3/how-did-you-get-https-within-the-h=
ome/
[2] https://www.w3.org/wiki/images/4/43/Http-migration-in-local-network.p=
df



> I'm not sure that I fully grok the last one, but it talks about an
> ACME-like protocol that is mediated by a browser.  It also talks about
> creating certificates for non-unique names on .local, so I'm not sure
> that it's feasible.
>=20
> Not discussed here, but we've talked a bit about using key continuity
> for network-local devices and changing the "bad certificate" page we
> show on first connection (with a different page when a different key
> is presented by the device).
>=20
> _______________________________________________
> Acme mailing list
> Acme@ietf.org
> https://www.ietf.org/mailman/listinfo/acme
>=20


--TWesaVHWgV5VFwBl2fQxO0UuUtJSSd6FC--

--sQaP4lVXg3vSOWfCHKKSlU1T9GEvobjcn
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQEcBAEBCAAGBQJZqSIlAAoJEC88hzaAX42iDSoH/0jtEz9qLDzPRuFOCdzj/bsl
evLcpT/+RTe1CILmevuadJeSn/T229QZFZXaRyC9Wm5lqs2EiYygfi1yxFxHdlYq
WBQDZ4mIVsJhBr4jbsWltH7mQNDjbpWZPiX4Ed5M9HqBbVCDhupj2fQKgVeVAxUK
pB3ksFsL+kAsXdskAd8nmpPvyxnJo+DaWlsYhxwAkEpel8bR2PKvksrrWJmIJPn7
JMhH9gdPo4v3gMw9Z+4CC3bPvO5caQstRF9n4kwSpZ3yfyLVGRs8Dyr8/CgwRoVw
WoGUa9NvzuyJweUsHF9/U5IPtLzKBc2AMOyPec4QlhjCIYyk0whpI22ojPWFBN8=
=Kb7L
-----END PGP SIGNATURE-----

--sQaP4lVXg3vSOWfCHKKSlU1T9GEvobjcn--


From nobody Fri Sep  1 03:52:43 2017
Return-Path: <martin.thomson@gmail.com>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AD6021331B7 for <acme@ietfa.amsl.com>; Fri,  1 Sep 2017 03:52:42 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
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 lrYbcA9WIXNs for <acme@ietfa.amsl.com>; Fri,  1 Sep 2017 03:52:40 -0700 (PDT)
Received: from mail-oi0-x235.google.com (mail-oi0-x235.google.com [IPv6:2607:f8b0:4003:c06::235]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B53201331A2 for <acme@ietf.org>; Fri,  1 Sep 2017 03:52:40 -0700 (PDT)
Received: by mail-oi0-x235.google.com with SMTP id r203so17613473oih.0 for <acme@ietf.org>; Fri, 01 Sep 2017 03:52:40 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;  h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=6FwL9QpH23QjkPXhWS/lsxBTpKcuEp7x6Kfzh6HCHyU=; b=EqtbsLqE2UUkqKv7fUbLYXGijnw0LTXBTmc5nbfykXwnbk0MyFVXqrb4uruGqdT+jg fJVhzOGzkQm5B/59/ACqC7eUULT2IuojNB0HzFeTA28/ZgOh+hG8vgh7fRt9kLKm31MN TKEiRTOahAuWZXblzleV2ueZOipkd+GhJ8RczUH7kDbjNwi3BX6k6vkh2qDkTKsAxDgi 4A+PGqmNGDtmF2agFYLUNrAFrAiLnRrf1/2bthyD+oKeYXT41b6AUpY/q8A+2aknZfNe PS7s8qZHaRdwv9lumF/jnkNXze5XBSYpe6kKIIXLP4nqJ7BZgvKAhSsy44Ie2zhPhT8Y u9uQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=6FwL9QpH23QjkPXhWS/lsxBTpKcuEp7x6Kfzh6HCHyU=; b=rXxSzsbGojmCca2MwM/9MYddzYli1GNjvX7HeAN8gFS+55kfpDsBJCLyJVG3nlXIEt uJzoZ4j8Ng/KIUUS2uOJzkfyRHJAIt9+dc7hGjYch7IDkfoWBXkaqEwM6dMvisvJjzF6 Vmi3E8HrJU5SPRTkZ2CtrPQ0rqaxwt5RmA6n2qo36noMs4UmLk0nbt1GSB6S6uCsUB0P KYZUKMAZd7Ugumra2jdG4gwN2yMKOr+zgrqVIc89ADogC/EsapFmq9s0PJw+45yy9eKp F2kyIFWR25hDl4AZpq/CDNBn+78X0KAJvzcvkjTsLEnERrlLYsGEntGfgUUKQtvV7oC5 w8Xw==
X-Gm-Message-State: AHPjjUjvkZDNPS0wkXYucDLCK1T4KQ0v4h/czsQvnFbfMBuEiG1Qsm4/ v8S2nnWYpQ0dkdS2vCZS0IFtJYjuFZcn
X-Google-Smtp-Source: ADKCNb5b5n/gQO4XYjq7vhG+Xx4JlUg6rVBxG8fKZJBooSCoSMpua1MZF7RdRmV9vXLGWTaukYSYz2a1sE083oD1jhE=
X-Received: by 10.202.4.130 with SMTP id 124mr1686590oie.246.1504263160007; Fri, 01 Sep 2017 03:52:40 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.157.63.163 with HTTP; Fri, 1 Sep 2017 03:52:39 -0700 (PDT)
In-Reply-To: <03ce0062-78c8-4c6f-2e1f-9d37e8120488@cs.tcd.ie>
References: <CAHbrMsB_8x4C2Y0OjO=a2H8LkFqeHgELAX4_imW2ip7Y7Vj21Q@mail.gmail.com> <d69735c5-3e46-753e-9a68-f27c68707ce5@cs.tcd.ie> <85c8aa7f-52a6-38fc-3700-d8607103889d@nostrum.com> <CABkgnnV0CWKQMo_nxP8rv71f=obtSOLwfHj4YU7hy9mxUgN0PA@mail.gmail.com> <03ce0062-78c8-4c6f-2e1f-9d37e8120488@cs.tcd.ie>
From: Martin Thomson <martin.thomson@gmail.com>
Date: Fri, 1 Sep 2017 20:52:39 +1000
Message-ID: <CABkgnnWgQH6kyAch1p+KT9OmWJXAscRUjAGcy2UCOKdsX2E7Zg@mail.gmail.com>
To: Stephen Farrell <stephen.farrell@cs.tcd.ie>
Cc: Adam Roach <adam@nostrum.com>, Ben Schwartz <bemasc@google.com>, "acme@ietf.org" <acme@ietf.org>
Content-Type: text/plain; charset="UTF-8"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/zmZdlNAybHyXRVwe6suXJU99kdk>
Subject: Re: [Acme] On draft-ietf-acme-ip
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 01 Sep 2017 10:52:43 -0000

On Fri, Sep 1, 2017 at 7:02 PM, Stephen Farrell
<stephen.farrell@cs.tcd.ie> wrote:
> There's also the concern of how to get that scheme to work for
> openwrt devices and similar, which is related. It'd be a fail
> if that couldn't work I reckon.

Exactly the same scheme works for those devices.  All you need is
someone with a real name who is willing to give you that name.  You
don't even need to put the name in the public DNS (no DNSSEC
required), because usually that device runs a DNS server that can
produce the mapping on demand.

The same drawbacks apply of course, the biggest one being that users
need to learn the name somehow.


From nobody Mon Sep 18 12:15:11 2017
Return-Path: <internet-drafts@ietf.org>
X-Original-To: acme@ietf.org
Delivered-To: acme@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 58A06132320; Mon, 18 Sep 2017 12:15:03 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: <i-d-announce@ietf.org>
Cc: acme@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 6.62.0
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <150576210331.15691.3080913852885263681@ietfa.amsl.com>
Date: Mon, 18 Sep 2017 12:15:03 -0700
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/7tU3LVSwdm_-FylgvLrXaRUc5tM>
Subject: [Acme] I-D Action: draft-ietf-acme-ip-01.txt
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 18 Sep 2017 19:15:03 -0000

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Automated Certificate Management Environment WG of the IETF.

        Title           : ACME IP Identifier Validation Extension
        Author          : Roland Bracewell Shoemaker
	Filename        : draft-ietf-acme-ip-01.txt
	Pages           : 7
	Date            : 2017-09-18

Abstract:
   This document specifies identifiers and challenges required to enable
   the Automated Certificate Management Environment (ACME) to issue
   certificates for IP addresses.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-acme-ip/

There are also htmlized versions available at:
https://tools.ietf.org/html/draft-ietf-acme-ip-01
https://datatracker.ietf.org/doc/html/draft-ietf-acme-ip-01

A diff from the previous version is available at:
https://www.ietf.org/rfcdiff?url2=draft-ietf-acme-ip-01


Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

Internet-Drafts are also available by anonymous FTP at:
ftp://ftp.ietf.org/internet-drafts/


From nobody Mon Sep 18 12:30:36 2017
Return-Path: <roland@letsencrypt.org>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DC6CA1321D8 for <acme@ietfa.amsl.com>; Mon, 18 Sep 2017 12:30:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.001
X-Spam-Level: 
X-Spam-Status: No, score=-2.001 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=letsencrypt.org
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 MTDJ8S6daBMz for <acme@ietfa.amsl.com>; Mon, 18 Sep 2017 12:30:33 -0700 (PDT)
Received: from mail-pf0-x244.google.com (mail-pf0-x244.google.com [IPv6:2607:f8b0:400e:c00::244]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id ED550132320 for <acme@ietf.org>; Mon, 18 Sep 2017 12:30:32 -0700 (PDT)
Received: by mail-pf0-x244.google.com with SMTP id q76so565095pfq.5 for <acme@ietf.org>; Mon, 18 Sep 2017 12:30:32 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=letsencrypt.org; s=google; h=subject:to:references:from:message-id:date:user-agent:mime-version :in-reply-to:content-language:content-transfer-encoding; bh=OT7ZyiGWBZFP5CYx+rf5nBgQuHre1gOcPAboGSTaLdk=; b=WYcEq1PYQ5A11gsCuqoNT0BT/6FrfWI5D3mtmqofcOmYa+SeDFFSMR6qHQ+lpi9xTa io3lesrKfZAF+51n8+0hjGqRsA+fLFVXYTFdFoQDuzSSpe9Rrfhlk9e0n2TGwyvMHEoQ uN1IqwgdAFblpjrraJoX1swMhi2SQYeS5k80s=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:subject:to:references:from:message-id:date :user-agent:mime-version:in-reply-to:content-language :content-transfer-encoding; bh=OT7ZyiGWBZFP5CYx+rf5nBgQuHre1gOcPAboGSTaLdk=; b=MOFZkVsLDigi8Dc4Lf2c11Oa5dxRn1t2fAyBmxXnB3bpYnesarTB5w2qnLgqKRfEGD qUnD1c7hFZx/Ygear3Y4fZ7YWngTnQur6qq4t67kpbQjHfVOxx6BaQV+kLA/HyNZ43qq l9Y29sgqBjMHKUN42pvFY2EhHFtgg1SCTGLEBhAmQodtvx1bU93Es2c2s9I86iSSf93k 9UVJCVCaLQJ4a+0n01e8TlNpXuEYWghR6Zx8e1cAuRcSsVuY5CeUmE2K3h2p+101H9WS pjWbsPziFsNnRvO1pWofXuhLfLzBjtB+I56snU1jAA6ip1uEdd5on6hUVeCMsPpbdNVN 407g==
X-Gm-Message-State: AHPjjUh4zMENh72YG1LxWhvx9rBFYXAXcb//ipVGcRcD3lca1ADLiIQ9 fXp/pCAuAH2OmXpGEazBXKXu
X-Google-Smtp-Source: ADKCNb5PQ7DHiuXKCRyiRwV0+tcc9cppiIR7iy9XwbtVjOaxEAVQd/48exbSPZwjfBd6CO5GNLsCwA==
X-Received: by 10.84.235.6 with SMTP id o6mr24408625plk.295.1505763032114; Mon, 18 Sep 2017 12:30:32 -0700 (PDT)
Received: from [10.12.108.69] ([216.222.84.164]) by smtp.gmail.com with ESMTPSA id x9sm217622pfk.40.2017.09.18.12.30.29 for <acme@ietf.org> (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 18 Sep 2017 12:30:30 -0700 (PDT)
To: acme@ietf.org
References: <150576210331.15691.3080913852885263681@ietfa.amsl.com>
From: Roland Shoemaker <roland@letsencrypt.org>
Message-ID: <473268b2-378a-7aaa-b36e-32f6c751a778@letsencrypt.org>
Date: Mon, 18 Sep 2017 12:30:28 -0700
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.2.1
MIME-Version: 1.0
In-Reply-To: <150576210331.15691.3080913852885263681@ietfa.amsl.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/0wv40jpMKICo_IjNnHDwbxBD9Zc>
Subject: Re: [Acme] I-D Action: draft-ietf-acme-ip-01.txt
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 18 Sep 2017 19:30:35 -0000

Little difference from the last draft, mostly small cleanups.

There was some previous discussion about how to handle policy decisions
for issuing certificates for IP addresses. It was suggested that this
draft should contain some stronger language that would allow default
denial of certificate issuance for IP addresses. I think there should
definitely be some process for communicating these kinds of policy
decisions but I don't think this document is the right place for it, nor
do I think this document should attempt to dictate CA policy by
requiring something like this. I believe doing so would be a step back
for any CA implementing this document as they are all currently able to,
and many do, issue certificates for any IP address as long as a user is
able to prove control of it.

I believe we (or the IETF more generally) should instead focus on
developing standards for communicating a policy about issuance for IP
addresses to CAs such as a CAA lookup mechanism that can handle them
(i.e. something like
https://tools.ietf.org/html/draft-shoemaker-caa-ip-01, note this lacks
the tree climbing behavior which after bouncing it around a bit I've
come to the decision that it does actually require).

(I also totally forgot to incorporate the reference to 5952 for IPv6
textual representation, only saw my note about doing that after
submitting the docs, I'll make sure to resolve this in the next version!)

On 09/18/2017 12:15 PM, internet-drafts@ietf.org wrote:
> 
> A New Internet-Draft is available from the on-line Internet-Drafts directories.
> This draft is a work item of the Automated Certificate Management Environment WG of the IETF.
> 
>         Title           : ACME IP Identifier Validation Extension
>         Author          : Roland Bracewell Shoemaker
> 	Filename        : draft-ietf-acme-ip-01.txt
> 	Pages           : 7
> 	Date            : 2017-09-18
> 
> Abstract:
>    This document specifies identifiers and challenges required to enable
>    the Automated Certificate Management Environment (ACME) to issue
>    certificates for IP addresses.
> 
> 
> The IETF datatracker status page for this draft is:
> https://datatracker.ietf.org/doc/draft-ietf-acme-ip/
> 
> There are also htmlized versions available at:
> https://tools.ietf.org/html/draft-ietf-acme-ip-01
> https://datatracker.ietf.org/doc/html/draft-ietf-acme-ip-01
> 
> A diff from the previous version is available at:
> https://www.ietf.org/rfcdiff?url2=draft-ietf-acme-ip-01
> 
> 
> Please note that it may take a couple of minutes from the time of submission
> until the htmlized version and diff are available at tools.ietf.org.
> 
> Internet-Drafts are also available by anonymous FTP at:
> ftp://ftp.ietf.org/internet-drafts/
> 
> _______________________________________________
> Acme mailing list
> Acme@ietf.org
> https://www.ietf.org/mailman/listinfo/acme
> 

-- 
Roland Bracewell Shoemaker
Software Engineer
Linux Foundation / Internet Security Research Group


From nobody Tue Sep 19 12:26:07 2017
Return-Path: <dmccarney@letsencrypt.org>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 805CC134376 for <acme@ietfa.amsl.com>; Tue, 19 Sep 2017 12:26:05 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=letsencrypt.org
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 IEkb-xjqQW_p for <acme@ietfa.amsl.com>; Tue, 19 Sep 2017 12:26:02 -0700 (PDT)
Received: from mail-it0-x231.google.com (mail-it0-x231.google.com [IPv6:2607:f8b0:4001:c0b::231]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 59EE2134373 for <acme@ietf.org>; Tue, 19 Sep 2017 12:26:02 -0700 (PDT)
Received: by mail-it0-x231.google.com with SMTP id 85so726302ith.2 for <acme@ietf.org>; Tue, 19 Sep 2017 12:26:02 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=letsencrypt.org; s=google; h=mime-version:reply-to:from:date:message-id:subject:to; bh=Tvlctq+KuF+Q31Q1AlVCvXuJWZW8gKf5EbOaEa21BTs=; b=dejxEGyi7EkfoBYdXOXZVC9ccg3knB8H1EddAZ3HmsH5CG2F43yMe8ZzxKynDRl5uu LlUjdXTSQEXP6mJ0XukDw7xjzX1beQrHjflAOSckclckMOazuIdj9y6fw4ApCGOyVKzy LTkLkKmgo5XivO5EbiQP+Shr8gDSqsGiMaLTY=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:reply-to:from:date:message-id :subject:to; bh=Tvlctq+KuF+Q31Q1AlVCvXuJWZW8gKf5EbOaEa21BTs=; b=MnJbC3CZGrF9KpgHiaRjx/T3C0FYfhG4RqmZxUCOfDpNl8SCB6RGeVzHDbzBxmRtre hYVAnvdWfoMKlSWVYtUjQKZB9gLG6zZSWjHkQHMTX7viOSV/Ypp6reeRYRvXJ8aqSKNx k0rUlADj3gddh0dJrrIdJ9ns94BvGZi6ZwHykjy2Nmit7Enx57kOU+NXv1eAnukcPuQj WrTkGHeb8I9pw7xxpLa5jhE2usp2SgKHhkSr1pS0di6oMULFAo220OEg9KW1Bjrm2elg DiTCxLfeA3z29/vWZvJnx/fymGWiAfdFbcB7nyewSQcFeY7vH3/RuzCSzsIiJiQji8vN Ju3A==
X-Gm-Message-State: AHPjjUiQJvtj9kR6HRzR9LW1oVblhuOzrGCoKkZWpty89GXa1e4VpTjy xb1oeo9wBtqpKAysPUcQ+anCiyOhwIcxz6DsJPs8vNNQp40=
X-Google-Smtp-Source: AOwi7QDuxSFV2+eQh2pl18gLC7KTE/6biqDvp37Lr+YqwtHSM+Z1q2jZmF5Z5fQW3ZmmJYuaJhaG2v3ooek+hbTferI=
X-Received: by 10.36.39.142 with SMTP id g136mr3508272ita.73.1505849161041; Tue, 19 Sep 2017 12:26:01 -0700 (PDT)
MIME-Version: 1.0
Reply-To: cpu@letsencrypt.org
Received: by 10.107.17.229 with HTTP; Tue, 19 Sep 2017 12:26:00 -0700 (PDT)
From: Daniel McCarney <cpu@letsencrypt.org>
Date: Tue, 19 Sep 2017 15:26:00 -0400
Message-ID: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com>
To: IETF ACME <acme@ietf.org>
Content-Type: multipart/alternative; boundary="001a1147cc665461a805598fd4de"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/DIjJEB06J5cFyuOlGPVcY2I51vg>
Subject: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 19 Sep 2017 19:26:05 -0000

--001a1147cc665461a805598fd4de
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi folks,

Just over a year ago in a thread titled "Proactive vs On-Finalization
Certificate Issuance"[0] we reached the consensus on the question of
whether to
issue a certificate "on-finalization" or "proactively" with the conclusion
to
standardize on proactive issuance.

Implementation experience shows that proactive issuance is slow in
practice. An
order can be associated with many authorizations and an authorization can b=
e
associated with many orders. This interacts poorly with optimizations Let's
Encrypt has developed from in-the-field experience with ACME and it is
difficult
to rate-limit effectively. I propose that we return to the on-finalization
style
approach and that we provide the CSR during finalization..

## Issues with proactive issuance

### Authorization Reuse

Proactive issuance requires that when an authorization is switched to a
valid
state by the result of a challenge update that we find the associated
orders and
iterate each order's authorizations to see if all of the authorizations are
valid and issuance can occur.

Let's Encrypt reuses existing authorizations[1][2] whenever possible, as
encouraged by ACME. This has been of huge value in reducing resource
consumption
(both in terms of database storage and challenge/API requests). Many client=
s
create pending authorizations they do not finalize and similarly re-create
pending authorizations for identifiers that the account already has valid
authorizations for. Unchecked this leads to significant database growth and
resource consumption, even with aggressive rate-limits[3].

The combination of authorization reuse and proactive issuance means a given
authorization can be associated with many orders. This necessitates an
expensive
many-to-many query at each authorization update to see if any associated
orders are now complete.

One possible workaround is a rate-limit that would constrain the number of
orders an authorization can be associated with. However establishing the
value
of the limit to both reduce server-side load & also effectively support
Let's
Encrypt's large-volume integrators and their issuance patterns is
challenging.

### CSR-first new-order flow

To support proactive issuance the CA also needs to have the CSR for the
order
at-hand when checking each authorization in case issuance can proceed. The
`new-order` request that creates the order therefore carries the full CSR
that
was previously delivered by the "certificate finalize" message.

We see users create many more authorizations than they are able to finalize
successfully. With the legacy "new-cert" flow our authorization reuse
dramatically helped reduce the DB space used up by authorizations that will
never lead to issuance because no CSR is stored until all identifiers are
validated.

With the new-order flow we have to accept and store a full CSR before
handing
back authorizations to the client. Reuse of authorizations does not prevent
the
DB bloat that will occur from users submitting new order requests and not
fulfilling them. The CSR contains a full public key and is one of the singl=
e
largest contributors to DB disk space usage. Addressing this will require
another strict rate limit on outstanding new-orders and is again difficult
to
=E2=80=9Cdial-in=E2=80=9D in to both prevent excessive resource consumption=
 and also
maintain
the ability for big integrators to process large-scale issuance using our
API.

# Proposed Change

I believe we should change the `new-order` endpoint to not accept a full
CSR but
to instead accept a list of identifiers that the ACME client wishes to
authenticate & issue for. Authorizations can be created and returned for
those
domains as happens today.

The order resource should be changed to have a =E2=80=9CfinalizeURL=E2=80=
=9D field which
provides the client a URL to POST a CSR to indicate that the CA should
issue the
certificate. Proactive issuance should be removed outright.

This will address both the problem of scanning existing many-to-many orders
& their authorizations as well as the db bloat that will occur from
front-loading the delivery of the CSR.

It may be tempting to leave proactive issuance as an optional feature but
this
will require maintaining the CSR in the new-order and I also believe that
having
two very distinct methods of issuance will complicate the client ecosystem
and
returns us to the question of how to determine which will occur that
spawned the
original "Proactive vs On-finalization Certificate Issuance"
thread[0].

[0] https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY
[1] https://community.letsencrypt.org/t/upcoming-api-changes/17947
[2]
https://community.letsencrypt.org/t/automatic-recycling-of-pending-authoriz=
ations/41321
[3] https://letsencrypt.org/docs/rate-limits/

--001a1147cc665461a805598fd4de
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi folks,</div><div><br></div><div>Just over a year a=
go in a thread titled &quot;Proactive vs On-Finalization</div><div>Certific=
ate Issuance&quot;[0] we reached the consensus on the question of whether t=
o</div><div>issue a certificate &quot;on-finalization&quot; or &quot;proact=
ively&quot; with the conclusion to</div><div>standardize on proactive issua=
nce.</div><div><br></div><div>Implementation experience shows that proactiv=
e issuance is slow in practice. An</div><div>order can be associated with m=
any authorizations and an authorization can be</div><div>associated with ma=
ny orders. This interacts poorly with optimizations Let&#39;s</div><div>Enc=
rypt has developed from in-the-field experience with ACME and it is difficu=
lt</div><div>to rate-limit effectively. I propose that we return to the on-=
finalization style</div><div>approach and that we provide the CSR during fi=
nalization..</div><div><br></div><div>## Issues with proactive issuance</di=
v><div><br></div><div>### Authorization Reuse</div><div><br></div><div>Proa=
ctive issuance requires that when an authorization is switched to a valid</=
div><div>state by the result of a challenge update that we find the associa=
ted orders and</div><div>iterate each order&#39;s authorizations to see if =
all of the authorizations are</div><div>valid and issuance can occur.</div>=
<div><br></div><div>Let&#39;s Encrypt reuses existing authorizations[1][2] =
whenever possible, as</div><div>encouraged by ACME. This has been of huge v=
alue in reducing resource consumption</div><div>(both in terms of database =
storage and challenge/API requests). Many clients</div><div>create pending =
authorizations they do not finalize and similarly re-create</div><div>pendi=
ng authorizations for identifiers that the account already has valid</div><=
div>authorizations for. Unchecked this leads to significant database growth=
 and</div><div>resource consumption, even with aggressive rate-limits[3].</=
div><div><br></div><div>The combination of authorization reuse and proactiv=
e issuance means a given</div><div>authorization can be associated with man=
y orders. This necessitates an expensive</div><div>many-to-many query at ea=
ch authorization update to see if any associated</div><div>orders are now c=
omplete.</div><div><br></div><div>One possible workaround is a rate-limit t=
hat would constrain the number of</div><div>orders an authorization can be =
associated with. However establishing the value</div><div>of the limit to b=
oth reduce server-side load &amp; also effectively support Let&#39;s</div><=
div>Encrypt&#39;s large-volume integrators and their issuance patterns is c=
hallenging.</div><div><br></div><div>### CSR-first new-order flow</div><div=
><br></div><div>To support proactive issuance the CA also needs to have the=
 CSR for the order</div><div>at-hand when checking each authorization in ca=
se issuance can proceed. The</div><div>`new-order` request that creates the=
 order therefore carries the full CSR that</div><div>was previously deliver=
ed by the &quot;certificate finalize&quot; message.</div><div><br></div><di=
v>We see users create many more authorizations than they are able to finali=
ze</div><div>successfully. With the legacy &quot;new-cert&quot; flow our au=
thorization reuse</div><div>dramatically helped reduce the DB space used up=
 by authorizations that will</div><div>never lead to issuance because no CS=
R is stored until all identifiers are</div><div>validated.</div><div><br></=
div><div>With the new-order flow we have to accept and store a full CSR bef=
ore handing</div><div>back authorizations to the client. Reuse of authoriza=
tions does not prevent the</div><div>DB bloat that will occur from users su=
bmitting new order requests and not</div><div>fulfilling them. The CSR cont=
ains a full public key and is one of the single</div><div>largest contribut=
ors to DB disk space usage. Addressing this will require</div><div>another =
strict rate limit on outstanding new-orders and is again difficult to</div>=
<div>=E2=80=9Cdial-in=E2=80=9D in to both prevent excessive resource consum=
ption and also maintain</div><div>the ability for big integrators to proces=
s large-scale issuance using our API.</div><div><br></div><div># Proposed C=
hange</div><div><br></div><div>I believe we should change the `new-order` e=
ndpoint to not accept a full CSR but</div><div>to instead accept a list of =
identifiers that the ACME client wishes to</div><div>authenticate &amp; iss=
ue for. Authorizations can be created and returned for those</div><div>doma=
ins as happens today.</div><div><br></div><div>The order resource should be=
 changed to have a =E2=80=9CfinalizeURL=E2=80=9D field which</div><div>prov=
ides the client a URL to POST a CSR to indicate that the CA should issue th=
e</div><div>certificate. Proactive issuance should be removed outright.</di=
v><div><br></div><div>This will address both the problem of scanning existi=
ng many-to-many orders</div><div>&amp; their authorizations as well as the =
db bloat that will occur from</div><div>front-loading the delivery of the C=
SR.</div><div><br></div><div>It may be tempting to leave proactive issuance=
 as an optional feature but this</div><div>will require maintaining the CSR=
 in the new-order and I also believe that having</div><div>two very distinc=
t methods of issuance will complicate the client ecosystem and</div><div>re=
turns us to the question of how to determine which will occur that spawned =
the</div><div>original &quot;Proactive vs On-finalization Certificate Issua=
nce&quot;</div><div>thread[0].</div><div><br></div><div>[0] <a href=3D"http=
s://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY">https:/=
/mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY</a></div><d=
iv>[1] <a href=3D"https://community.letsencrypt.org/t/upcoming-api-changes/=
17947">https://community.letsencrypt.org/t/upcoming-api-changes/17947</a></=
div><div>[2] <a href=3D"https://community.letsencrypt.org/t/automatic-recyc=
ling-of-pending-authorizations/41321">https://community.letsencrypt.org/t/a=
utomatic-recycling-of-pending-authorizations/41321</a></div><div>[3] <a hre=
f=3D"https://letsencrypt.org/docs/rate-limits/">https://letsencrypt.org/doc=
s/rate-limits/</a></div><div><br></div></div>

--001a1147cc665461a805598fd4de--


From nobody Tue Sep 19 13:34:57 2017
Return-Path: <logan.widick@gmail.com>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 25E2F1329B5 for <acme@ietfa.amsl.com>; Tue, 19 Sep 2017 13:34:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.698
X-Spam-Level: 
X-Spam-Status: No, score=-2.698 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
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 o0ZS58AjwsAw for <acme@ietfa.amsl.com>; Tue, 19 Sep 2017 13:34:52 -0700 (PDT)
Received: from mail-io0-x22c.google.com (mail-io0-x22c.google.com [IPv6:2607:f8b0:4001:c06::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4D6B713306C for <acme@ietf.org>; Tue, 19 Sep 2017 13:34:52 -0700 (PDT)
Received: by mail-io0-x22c.google.com with SMTP id 21so2008678iof.6 for <acme@ietf.org>; Tue, 19 Sep 2017 13:34:52 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;  h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=0ispkZSNbFpQgKDS14bpImxKztfC85ZIc973Q1IJoiw=; b=S15cck4hFzPY7ZIZmJT79fRwCgUwvWSC9KKL+fu27FdZZfe6ka13jF2aP/aFECNkf0 Sxo4V+XO+I8ntIdbccUUfYUDi56Cv7nSavWs+qvYhdYctlSRAQB0AwC8onc2ZAvcsbSg cw38hVpi31g4zgz1FzfEge25YfkURTxBNl5etzWcaKrkX64DLI4h2stGRndSANeju+u5 SaaB1Kop4Lef5jH7JXRn+nvOSNFEoOtIbHYDXL8rB9rAppzK1dkS2o11A5g+MP4ZC99q GVgWK1oQybmnj8ANUX1Wd8eZU3z429+zxBM9tlp5thRalM5xBRYVCQvX6o/fFBm7bZ1q J8SA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=0ispkZSNbFpQgKDS14bpImxKztfC85ZIc973Q1IJoiw=; b=r3cyNR1D6u+oBfVyDnIJ5jDHsKZRH/i4cftUvnBosBWIncRasiyFLk3YO4sWKuKdHO USMRtk4ALGw3rgT6SvjIhYzI2XNvuetBn2jADRMFHy28ZE7yx+D+6dku3cwZ6X0+QJ90 93Eet3lxiQs7yPc7PcrXteli1R6qOXFlvaU3AXgwiu1n6/+D3UxV8XDermiV2jwrEHVG UXt6Td7aSliBs4rxHW0lDGM+qtFxzqFLsL3vheKcuJTVPmc6ntQC/Vs5HdPxJ8SXrVU1 dT9yUrXT5j4/M525L+YUQDfRdNgPGVCbw3oP7IuGBk5H+9fqRHJ2URj2EK6pm/9I3opK hEkQ==
X-Gm-Message-State: AHPjjUifdx9I4dka/bQ8onKSVt0pvXvn0xFHvATVvSRagIpY5F1NKmje c7P5oXKUiY4n35kdUXtEY1pHkKTGt+9N3JCyIDc=
X-Google-Smtp-Source: AOwi7QBgnD5zKEQD0MuM4qR+2rJgCvjhXkWKzNXkbAV74QDb+2ulz6G9g8MM2Mr2Gl+ckukbDo+Vmx/hgqgcqXpusi4=
X-Received: by 10.107.11.89 with SMTP id v86mr4148099ioi.293.1505853291465; Tue, 19 Sep 2017 13:34:51 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.2.82.1 with HTTP; Tue, 19 Sep 2017 13:34:50 -0700 (PDT)
Received: by 10.2.82.1 with HTTP; Tue, 19 Sep 2017 13:34:50 -0700 (PDT)
In-Reply-To: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com>
From: Logan Widick <logan.widick@gmail.com>
Date: Tue, 19 Sep 2017 15:34:50 -0500
Message-ID: <CAMmAzE+bXN=5VsY092mBx+XxFOg6o4FHx3YkOoa9Uh3LPi1wfg@mail.gmail.com>
To: cpu@letsencrypt.org
Cc: ACME WG <acme@ietf.org>
Content-Type: multipart/alternative; boundary="001a113ee3a8859614055990cab1"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/U1jhZs2wBWdOBHy68BhXdvQN5p8>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 19 Sep 2017 20:34:55 -0000

--001a113ee3a8859614055990cab1
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Would it be possible to extract the key and identifiers from the CSR, add
the key  to the database if it doesn't already exist, find or create the
authorizations for the identifiers, not store the CSR, and then assemble
the certificate from the (valid) authorizations and key later?

Alternatively, could one get rid of the CSRs altogether, and have the
new-order be a nested JWS? The inner JWS would be signed by the key that
goes on the certificate and would contain everything required to generate
the certificate (identifiers, requested dates if applicable, etc.). The
outer JWS would be signed by the ACME account key and link the inner JWS to
the ACME account.

What goes in a CSR or certificate that can't be reconstructed later?

Logan

On Sep 19, 2017 2:26 PM, "Daniel McCarney" <cpu@letsencrypt.org> wrote:

Hi folks,

Just over a year ago in a thread titled "Proactive vs On-Finalization
Certificate Issuance"[0] we reached the consensus on the question of
whether to
issue a certificate "on-finalization" or "proactively" with the conclusion
to
standardize on proactive issuance.

Implementation experience shows that proactive issuance is slow in
practice. An
order can be associated with many authorizations and an authorization can b=
e
associated with many orders. This interacts poorly with optimizations Let's
Encrypt has developed from in-the-field experience with ACME and it is
difficult
to rate-limit effectively. I propose that we return to the on-finalization
style
approach and that we provide the CSR during finalization..

## Issues with proactive issuance

### Authorization Reuse

Proactive issuance requires that when an authorization is switched to a
valid
state by the result of a challenge update that we find the associated
orders and
iterate each order's authorizations to see if all of the authorizations are
valid and issuance can occur.

Let's Encrypt reuses existing authorizations[1][2] whenever possible, as
encouraged by ACME. This has been of huge value in reducing resource
consumption
(both in terms of database storage and challenge/API requests). Many client=
s
create pending authorizations they do not finalize and similarly re-create
pending authorizations for identifiers that the account already has valid
authorizations for. Unchecked this leads to significant database growth and
resource consumption, even with aggressive rate-limits[3].

The combination of authorization reuse and proactive issuance means a given
authorization can be associated with many orders. This necessitates an
expensive
many-to-many query at each authorization update to see if any associated
orders are now complete.

One possible workaround is a rate-limit that would constrain the number of
orders an authorization can be associated with. However establishing the
value
of the limit to both reduce server-side load & also effectively support
Let's
Encrypt's large-volume integrators and their issuance patterns is
challenging.

### CSR-first new-order flow

To support proactive issuance the CA also needs to have the CSR for the
order
at-hand when checking each authorization in case issuance can proceed. The
`new-order` request that creates the order therefore carries the full CSR
that
was previously delivered by the "certificate finalize" message.

We see users create many more authorizations than they are able to finalize
successfully. With the legacy "new-cert" flow our authorization reuse
dramatically helped reduce the DB space used up by authorizations that will
never lead to issuance because no CSR is stored until all identifiers are
validated.

With the new-order flow we have to accept and store a full CSR before
handing
back authorizations to the client. Reuse of authorizations does not prevent
the
DB bloat that will occur from users submitting new order requests and not
fulfilling them. The CSR contains a full public key and is one of the singl=
e
largest contributors to DB disk space usage. Addressing this will require
another strict rate limit on outstanding new-orders and is again difficult
to
=E2=80=9Cdial-in=E2=80=9D in to both prevent excessive resource consumption=
 and also
maintain
the ability for big integrators to process large-scale issuance using our
API.

# Proposed Change

I believe we should change the `new-order` endpoint to not accept a full
CSR but
to instead accept a list of identifiers that the ACME client wishes to
authenticate & issue for. Authorizations can be created and returned for
those
domains as happens today.

The order resource should be changed to have a =E2=80=9CfinalizeURL=E2=80=
=9D field which
provides the client a URL to POST a CSR to indicate that the CA should
issue the
certificate. Proactive issuance should be removed outright.

This will address both the problem of scanning existing many-to-many orders
& their authorizations as well as the db bloat that will occur from
front-loading the delivery of the CSR.

It may be tempting to leave proactive issuance as an optional feature but
this
will require maintaining the CSR in the new-order and I also believe that
having
two very distinct methods of issuance will complicate the client ecosystem
and
returns us to the question of how to determine which will occur that
spawned the
original "Proactive vs On-finalization Certificate Issuance"
thread[0].

[0] https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY
[1] https://community.letsencrypt.org/t/upcoming-api-changes/17947
[2] https://community.letsencrypt.org/t/automatic-recycling-of-p
ending-authorizations/41321
[3] https://letsencrypt.org/docs/rate-limits/


_______________________________________________
Acme mailing list
Acme@ietf.org
https://www.ietf.org/mailman/listinfo/acme

--001a113ee3a8859614055990cab1
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"auto"><div>Would it be possible to extract the key and identifi=
ers from the CSR, add the key =C2=A0to the database if it doesn&#39;t alrea=
dy exist, find or create the authorizations for the identifiers, not store =
the CSR, and then assemble the certificate from the (valid) authorizations =
and key later?=C2=A0</div><div dir=3D"auto"><br></div><div dir=3D"auto">Alt=
ernatively, could one get rid of the CSRs altogether, and have the new-orde=
r be a nested JWS? The inner JWS would be signed by the key that goes on th=
e certificate and would contain everything required to generate the certifi=
cate (identifiers, requested dates if applicable, etc.). The outer JWS woul=
d be signed by the ACME account key and link the inner JWS to the ACME acco=
unt.=C2=A0</div><div dir=3D"auto"><br></div><div dir=3D"auto">What goes in =
a CSR or certificate that can&#39;t be reconstructed later?</div><div dir=
=3D"auto"><br></div><div dir=3D"auto">Logan</div><div dir=3D"auto"><div cla=
ss=3D"gmail_extra" dir=3D"auto"><br><div class=3D"gmail_quote">On Sep 19, 2=
017 2:26 PM, &quot;Daniel McCarney&quot; &lt;<a href=3D"mailto:cpu@letsencr=
ypt.org" target=3D"_blank">cpu@letsencrypt.org</a>&gt; wrote:<br type=3D"at=
tribution"><blockquote class=3D"m_-4788304438553844519quote" style=3D"margi=
n:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir=3D"ltr">=
<div>Hi folks,</div><div><br></div><div>Just over a year ago in a thread ti=
tled &quot;Proactive vs On-Finalization</div><div>Certificate Issuance&quot=
;[0] we reached the consensus on the question of whether to</div><div>issue=
 a certificate &quot;on-finalization&quot; or &quot;proactively&quot; with =
the conclusion to</div><div>standardize on proactive issuance.</div><div><b=
r></div><div>Implementation experience shows that proactive issuance is slo=
w in practice. An</div><div>order can be associated with many authorization=
s and an authorization can be</div><div>associated with many orders. This i=
nteracts poorly with optimizations Let&#39;s</div><div>Encrypt has develope=
d from in-the-field experience with ACME and it is difficult</div><div>to r=
ate-limit effectively. I propose that we return to the on-finalization styl=
e</div><div>approach and that we provide the CSR during finalization..</div=
><div><br></div><div>## Issues with proactive issuance</div><div><br></div>=
<div>### Authorization Reuse</div><div><br></div><div>Proactive issuance re=
quires that when an authorization is switched to a valid</div><div>state by=
 the result of a challenge update that we find the associated orders and</d=
iv><div>iterate each order&#39;s authorizations to see if all of the author=
izations are</div><div>valid and issuance can occur.</div><div><br></div><d=
iv>Let&#39;s Encrypt reuses existing authorizations[1][2] whenever possible=
, as</div><div>encouraged by ACME. This has been of huge value in reducing =
resource consumption</div><div>(both in terms of database storage and chall=
enge/API requests). Many clients</div><div>create pending authorizations th=
ey do not finalize and similarly re-create</div><div>pending authorizations=
 for identifiers that the account already has valid</div><div>authorization=
s for. Unchecked this leads to significant database growth and</div><div>re=
source consumption, even with aggressive rate-limits[3].</div><div><br></di=
v><div>The combination of authorization reuse and proactive issuance means =
a given</div><div>authorization can be associated with many orders. This ne=
cessitates an expensive</div><div>many-to-many query at each authorization =
update to see if any associated</div><div>orders are now complete.</div><di=
v><br></div><div>One possible workaround is a rate-limit that would constra=
in the number of</div><div>orders an authorization can be associated with. =
However establishing the value</div><div>of the limit to both reduce server=
-side load &amp; also effectively support Let&#39;s</div><div>Encrypt&#39;s=
 large-volume integrators and their issuance patterns is challenging.</div>=
<div><br></div><div>### CSR-first new-order flow</div><div><br></div><div>T=
o support proactive issuance the CA also needs to have the CSR for the orde=
r</div><div>at-hand when checking each authorization in case issuance can p=
roceed. The</div><div>`new-order` request that creates the order therefore =
carries the full CSR that</div><div>was previously delivered by the &quot;c=
ertificate finalize&quot; message.</div><div><br></div><div>We see users cr=
eate many more authorizations than they are able to finalize</div><div>succ=
essfully. With the legacy &quot;new-cert&quot; flow our authorization reuse=
</div><div>dramatically helped reduce the DB space used up by authorization=
s that will</div><div>never lead to issuance because no CSR is stored until=
 all identifiers are</div><div>validated.</div><div><br></div><div>With the=
 new-order flow we have to accept and store a full CSR before handing</div>=
<div>back authorizations to the client. Reuse of authorizations does not pr=
event the</div><div>DB bloat that will occur from users submitting new orde=
r requests and not</div><div>fulfilling them. The CSR contains a full publi=
c key and is one of the single</div><div>largest contributors to DB disk sp=
ace usage. Addressing this will require</div><div>another strict rate limit=
 on outstanding new-orders and is again difficult to</div><div>=E2=80=9Cdia=
l-in=E2=80=9D in to both prevent excessive resource consumption and also ma=
intain</div><div>the ability for big integrators to process large-scale iss=
uance using our API.</div><div><br></div><div># Proposed Change</div><div><=
br></div><div>I believe we should change the `new-order` endpoint to not ac=
cept a full CSR but</div><div>to instead accept a list of identifiers that =
the ACME client wishes to</div><div>authenticate &amp; issue for. Authoriza=
tions can be created and returned for those</div><div>domains as happens to=
day.</div><div><br></div><div>The order resource should be changed to have =
a =E2=80=9CfinalizeURL=E2=80=9D field which</div><div>provides the client a=
 URL to POST a CSR to indicate that the CA should issue the</div><div>certi=
ficate. Proactive issuance should be removed outright.</div><div><br></div>=
<div>This will address both the problem of scanning existing many-to-many o=
rders</div><div>&amp; their authorizations as well as the db bloat that wil=
l occur from</div><div>front-loading the delivery of the CSR.</div><div><br=
></div><div>It may be tempting to leave proactive issuance as an optional f=
eature but this</div><div>will require maintaining the CSR in the new-order=
 and I also believe that having</div><div>two very distinct methods of issu=
ance will complicate the client ecosystem and</div><div>returns us to the q=
uestion of how to determine which will occur that spawned the</div><div>ori=
ginal &quot;Proactive vs On-finalization Certificate Issuance&quot;</div><d=
iv>thread[0].</div><div><br></div><div>[0] <a href=3D"https://mailarchive.i=
etf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY" target=3D"_blank">https:=
//mailarchive.ietf.org/a<wbr>rch/msg/acme/0lVmGl8e-rmSH0x7y<wbr>cDW5dj6GAY<=
/a></div><div>[1] <a href=3D"https://community.letsencrypt.org/t/upcoming-a=
pi-changes/17947" target=3D"_blank">https://community.letsencrypt.<wbr>org/=
t/upcoming-api-changes/179<wbr>47</a></div><div>[2] <a href=3D"https://comm=
unity.letsencrypt.org/t/automatic-recycling-of-pending-authorizations/41321=
" target=3D"_blank">https://community.letsencrypt.<wbr>org/t/automatic-recy=
cling-of-p<wbr>ending-authorizations/41321</a></div><div>[3] <a href=3D"htt=
ps://letsencrypt.org/docs/rate-limits/" target=3D"_blank">https://letsencry=
pt.org/docs/r<wbr>ate-limits/</a></div><div><br></div></div>
<br>______________________________<wbr>_________________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org" target=3D"_blank">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/l<wbr>istinfo/acme</a><br>
<br></blockquote></div><br></div></div></div>

--001a113ee3a8859614055990cab1--


From nobody Tue Sep 19 13:51:18 2017
Return-Path: <ilariliusvaara@welho.com>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 97CBC1321CB for <acme@ietfa.amsl.com>; Tue, 19 Sep 2017 13:51:17 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] 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 p3JJds4-sW9i for <acme@ietfa.amsl.com>; Tue, 19 Sep 2017 13:51:15 -0700 (PDT)
Received: from welho-filter3.welho.com (welho-filter3.welho.com [83.102.41.25]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9BE061321A4 for <acme@ietf.org>; Tue, 19 Sep 2017 13:51:15 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by welho-filter3.welho.com (Postfix) with ESMTP id 250A15DE7E; Tue, 19 Sep 2017 23:51:13 +0300 (EEST)
X-Virus-Scanned: Debian amavisd-new at pp.htv.fi
Received: from welho-smtp1.welho.com ([IPv6:::ffff:83.102.41.84]) by localhost (welho-filter3.welho.com [::ffff:83.102.41.25]) (amavisd-new, port 10024) with ESMTP id 4dAxeEV61po2; Tue, 19 Sep 2017 23:51:12 +0300 (EEST)
Received: from LK-Perkele-VII (87-92-19-27.bb.dnainternet.fi [87.92.19.27]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by welho-smtp1.welho.com (Postfix) with ESMTPSA id 5CDF9C4; Tue, 19 Sep 2017 23:51:09 +0300 (EEST)
Date: Tue, 19 Sep 2017 23:51:08 +0300
From: Ilari Liusvaara <ilariliusvaara@welho.com>
To: Logan Widick <logan.widick@gmail.com>
Cc: cpu@letsencrypt.org, ACME WG <acme@ietf.org>
Message-ID: <20170919205108.occqzaqbzvnn257k@LK-Perkele-VII>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAMmAzE+bXN=5VsY092mBx+XxFOg6o4FHx3YkOoa9Uh3LPi1wfg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <CAMmAzE+bXN=5VsY092mBx+XxFOg6o4FHx3YkOoa9Uh3LPi1wfg@mail.gmail.com>
User-Agent: NeoMutt/20170609 (1.8.3)
Sender: ilariliusvaara@welho.com
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/Yzd_2FxULdPmMOGBLA6N9Pk99vo>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 19 Sep 2017 20:51:17 -0000

On Tue, Sep 19, 2017 at 03:34:50PM -0500, Logan Widick wrote:
> Would it be possible to extract the key and identifiers from the CSR, add
> the key  to the database if it doesn't already exist, find or create the
> authorizations for the identifiers, not store the CSR, and then assemble
> the certificate from the (valid) authorizations and key later?

Most CSRs have fresh keys that are never reused (thanks to forced key-
rotation[1] in most ACME clients).

> Alternatively, could one get rid of the CSRs altogether, and have the
> new-order be a nested JWS? The inner JWS would be signed by the key that
> goes on the certificate and would contain everything required to generate
> the certificate (identifiers, requested dates if applicable, etc.). The
> outer JWS would be signed by the ACME account key and link the inner JWS to
> the ACME account.

Using the private key to sign a JWS is cryptographically very much non-
kosher.

Also, some systems can emit CSRs for non-exportable key. With those,
you are SoL unless you can send a CSR to the CA.

> What goes in a CSR or certificate that can't be reconstructed later?

For Let's Encrypt, AFAIK the only things in CSR that matter are:

- Requested domains
- Public key
- Signature (to verify proof-of-possession).


[1] Even if the client has mode that does not rotate the keys, that
mode is in most of the clients somehow crippled, so it does not
actually work well.




-Ilari


From nobody Wed Sep 20 08:43:45 2017
Return-Path: <jsha@eff.org>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A175B13300C for <acme@ietfa.amsl.com>; Wed, 20 Sep 2017 08:43:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -7.001
X-Spam-Level: 
X-Spam-Status: No, score=-7.001 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=eff.org
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 YVbhSvQSy5v7 for <acme@ietfa.amsl.com>; Wed, 20 Sep 2017 08:43:42 -0700 (PDT)
Received: from mail2.eff.org (mail2.eff.org [173.239.79.204]) (using TLSv1.2 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id AC3A513202D for <acme@ietf.org>; Wed, 20 Sep 2017 08:43:42 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=eff.org; s=mail2;  h=Content-Transfer-Encoding:Content-Type:In-Reply-To:MIME-Version:Date:Message-ID:From:References:Cc:To:Subject; bh=Pr1ibdT8Fo4VMRaw3eOFM9HfeodOeaFM3I4GTHuJmDw=;  b=iB5V7l9Flx5GzkDbbpmp3LyZNhP4ryM2ZDuO7bm+78IdjWTfzVd4wzIaMJcMGeGB1YrY14Go/WDmBnrHbfraZgecETGrCGxxf8OwiSL1aSEsP1I7HIaaJ+BX8v+wYYSqBSaBnz0Bejko4id5AVAHVAlBgYcs6eoHMCDCAar/1Zc=;
Received: ; Wed, 20 Sep 2017 08:43:39 -0700
To: Logan Widick <logan.widick@gmail.com>, cpu@letsencrypt.org
Cc: ACME WG <acme@ietf.org>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAMmAzE+bXN=5VsY092mBx+XxFOg6o4FHx3YkOoa9Uh3LPi1wfg@mail.gmail.com>
From: Jacob Hoffman-Andrews <jsha@eff.org>
Message-ID: <aeaef833-e7cb-dce7-c1c3-c1cb32a92913@eff.org>
Date: Wed, 20 Sep 2017 08:43:41 -0700
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.2.1
MIME-Version: 1.0
In-Reply-To: <CAMmAzE+bXN=5VsY092mBx+XxFOg6o4FHx3YkOoa9Uh3LPi1wfg@mail.gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Content-Language: en-US
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/rf4CZRlsOumUNiLxOefJV6Oew-0>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 20 Sep 2017 15:43:44 -0000

On 09/19/2017 01:34 PM, Logan Widick wrote:
> Would it be possible to extract the key and identifiers from the CSR,
> add the key  to the database if it doesn't already exist, find or
> create the authorizations for the identifiers, not store the CSR, and
> then assemble the certificate from the (valid) authorizations and key
> later?
Note that the size of the CSR for an RSA key is about half from the
signature, and half from the key. So extracting the key would only save
the size of the signature, about half. Also, this wouldn't solve the
more important problem, the overhead of reviewing all authzs on all
related orders on finalizing each authz, when an order may have a
hundred or more authzs.

I support Daniel's proposal.


From nobody Wed Sep 20 13:22:33 2017
Return-Path: <rlb@ipv.sx>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9AC21134336 for <acme@ietfa.amsl.com>; Wed, 20 Sep 2017 13:22:31 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level: 
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=ipv-sx.20150623.gappssmtp.com
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 22pspqFIfTVk for <acme@ietfa.amsl.com>; Wed, 20 Sep 2017 13:22:28 -0700 (PDT)
Received: from mail-wr0-x232.google.com (mail-wr0-x232.google.com [IPv6:2a00:1450:400c:c0c::232]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BC967134335 for <acme@ietf.org>; Wed, 20 Sep 2017 13:22:27 -0700 (PDT)
Received: by mail-wr0-x232.google.com with SMTP id z39so3055627wrb.8 for <acme@ietf.org>; Wed, 20 Sep 2017 13:22:27 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ipv-sx.20150623.gappssmtp.com; s=20150623; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=U1bjyxGt0SdVD/CKKfSzPorZxYmqsivyCpt1v/5dQxM=; b=z6A//RF9WYLn2B5tAkBLZOU0XWFUk9ILvsbOTciEGOA5QpdLu3YrReWiwTNYQUjdL/ PVaLWpyqsCbQsyg6eoupsci2OYAxisnLlMg3onC3SdNwRkfU7UUowi0jv7RBdcKkntWq IF1Lw/GVPnHTopEehSjd51O84fu5b/T8JIgcwhuNiYOXZs3oF2GQyci2CRSJ7TMJuHqp cBwcA19vMmJ10a+16mqaTL6ta+uHnIPX7ht/pYWebYXTrSw0mV/nsVuTsnU0jOvnDO/g SoTfD7ZOdsL2dtJPDtAmqPMCsrp4BZ+6R9lJyHJ3nPMe8C4ilIaf6jUyuyA/J8LttTEe 19AA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=U1bjyxGt0SdVD/CKKfSzPorZxYmqsivyCpt1v/5dQxM=; b=nB6e+w286pI+r7e4+rHmqm+EugqLKinezhidAU1dPU1jT0/R16YbZUIG3DmmjJo5Yw US7OSYLOoIagCljJx7dbujrdxBdAN8Bdt8FlpwQyLd4sEg/fdyZSQsVOq+niLDJ9UGUP +mZKrJ7x7zu601v1JYgv7ath4ws91nS4EHVXatlvatbpGyvg+YqpXWmbcIIRgnTdhZZw Q7F1s5oDjNxjk/KiYncPBezoRaKies4jszcGpwSEr9aL/1kfmjL1BC+ZdVxGYDO+Kpxk yQbgwYO73fyFFXdIHzywq3AFG2tmUyRHRllFX6vhukKTyqQgpcGf8SZzrVTTmrYD85rM xM4g==
X-Gm-Message-State: AHPjjUgFvNFx0j6JbtzDQ3Y0zOMkTjm07y+ga5rSfbjj+VAnp7Jp4D/M N0gy2wCNuJP6cOXw1jk1iKlkHODt17f3PzIIEeDQYkbJ
X-Google-Smtp-Source: AOwi7QCRhxVGQpMNmVQ1SB1YQphj5pACPnda+FZLvd5BYYwQ52rjeSNNFVlLcUvJIfmWggCKykY79V+B2g4lUhz/1m4=
X-Received: by 10.223.142.83 with SMTP id n77mr1554725wrb.108.1505938945968; Wed, 20 Sep 2017 13:22:25 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.28.159.200 with HTTP; Wed, 20 Sep 2017 13:22:25 -0700 (PDT)
In-Reply-To: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com>
From: Richard Barnes <rlb@ipv.sx>
Date: Wed, 20 Sep 2017 13:22:25 -0700
Message-ID: <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com>
To: Daniel McCarney <cpu@letsencrypt.org>
Cc: IETF ACME <acme@ietf.org>
Content-Type: multipart/alternative; boundary="f403045d6084eda24c0559a4bbfa"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/nmB9GEjSAxD6jWbDzOD0eu3H8dw>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 20 Sep 2017 20:22:31 -0000

--f403045d6084eda24c0559a4bbfa
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hey Daniel,

Thanks for the input.  I can see why you might be feeling some pain here,
but I'm not sure the proposed solution is quite right.

Note that ACME already has the "new-authorization" flow, which does pretty
much exactly what you're proposing.  The only difference is that instead of
authorizations being created all at once in response to a "new-cert"
request, they're created individually by the client.  But it's ultimately
the same number of RTTs (authz fetch ~ new-authz) -- or rather, one fewer
in the new-authz case because there's no new-cert request.  To what degree
would your problem be solved simply by having the affected integrators move
over to that flow?

I can sympathize that the new-order flow might not be suitable for all use
cases.  Maybe it would be useful to write down some thoughts on which use
cases are best addressed by which flows.

--Richard



On Tue, Sep 19, 2017 at 12:26 PM, Daniel McCarney <cpu@letsencrypt.org>
wrote:

> Hi folks,
>
> Just over a year ago in a thread titled "Proactive vs On-Finalization
> Certificate Issuance"[0] we reached the consensus on the question of
> whether to
> issue a certificate "on-finalization" or "proactively" with the conclusio=
n
> to
> standardize on proactive issuance.
>
> Implementation experience shows that proactive issuance is slow in
> practice. An
> order can be associated with many authorizations and an authorization can
> be
> associated with many orders. This interacts poorly with optimizations Let=
's
> Encrypt has developed from in-the-field experience with ACME and it is
> difficult
> to rate-limit effectively. I propose that we return to the on-finalizatio=
n
> style
> approach and that we provide the CSR during finalization..
>
> ## Issues with proactive issuance
>
> ### Authorization Reuse
>
> Proactive issuance requires that when an authorization is switched to a
> valid
> state by the result of a challenge update that we find the associated
> orders and
> iterate each order's authorizations to see if all of the authorizations a=
re
> valid and issuance can occur.
>
> Let's Encrypt reuses existing authorizations[1][2] whenever possible, as
> encouraged by ACME. This has been of huge value in reducing resource
> consumption
> (both in terms of database storage and challenge/API requests). Many
> clients
> create pending authorizations they do not finalize and similarly re-creat=
e
> pending authorizations for identifiers that the account already has valid
> authorizations for. Unchecked this leads to significant database growth a=
nd
> resource consumption, even with aggressive rate-limits[3].
>
> The combination of authorization reuse and proactive issuance means a giv=
en
> authorization can be associated with many orders. This necessitates an
> expensive
> many-to-many query at each authorization update to see if any associated
> orders are now complete.
>
> One possible workaround is a rate-limit that would constrain the number o=
f
> orders an authorization can be associated with. However establishing the
> value
> of the limit to both reduce server-side load & also effectively support
> Let's
> Encrypt's large-volume integrators and their issuance patterns is
> challenging.
>
> ### CSR-first new-order flow
>
> To support proactive issuance the CA also needs to have the CSR for the
> order
> at-hand when checking each authorization in case issuance can proceed. Th=
e
> `new-order` request that creates the order therefore carries the full CSR
> that
> was previously delivered by the "certificate finalize" message.
>
> We see users create many more authorizations than they are able to finali=
ze
> successfully. With the legacy "new-cert" flow our authorization reuse
> dramatically helped reduce the DB space used up by authorizations that wi=
ll
> never lead to issuance because no CSR is stored until all identifiers are
> validated.
>
> With the new-order flow we have to accept and store a full CSR before
> handing
> back authorizations to the client. Reuse of authorizations does not
> prevent the
> DB bloat that will occur from users submitting new order requests and not
> fulfilling them. The CSR contains a full public key and is one of the
> single
> largest contributors to DB disk space usage. Addressing this will require
> another strict rate limit on outstanding new-orders and is again difficul=
t
> to
> =E2=80=9Cdial-in=E2=80=9D in to both prevent excessive resource consumpti=
on and also
> maintain
> the ability for big integrators to process large-scale issuance using our
> API.
>
> # Proposed Change
>
> I believe we should change the `new-order` endpoint to not accept a full
> CSR but
> to instead accept a list of identifiers that the ACME client wishes to
> authenticate & issue for. Authorizations can be created and returned for
> those
> domains as happens today.
>
> The order resource should be changed to have a =E2=80=9CfinalizeURL=E2=80=
=9D field which
> provides the client a URL to POST a CSR to indicate that the CA should
> issue the
> certificate. Proactive issuance should be removed outright.
>
> This will address both the problem of scanning existing many-to-many orde=
rs
> & their authorizations as well as the db bloat that will occur from
> front-loading the delivery of the CSR.
>
> It may be tempting to leave proactive issuance as an optional feature but
> this
> will require maintaining the CSR in the new-order and I also believe that
> having
> two very distinct methods of issuance will complicate the client ecosyste=
m
> and
> returns us to the question of how to determine which will occur that
> spawned the
> original "Proactive vs On-finalization Certificate Issuance"
> thread[0].
>
> [0] https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GA=
Y
> [1] https://community.letsencrypt.org/t/upcoming-api-changes/17947
> [2] https://community.letsencrypt.org/t/automatic-recycling-of-
> pending-authorizations/41321
> [3] https://letsencrypt.org/docs/rate-limits/
>
>
> _______________________________________________
> Acme mailing list
> Acme@ietf.org
> https://www.ietf.org/mailman/listinfo/acme
>
>

--f403045d6084eda24c0559a4bbfa
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hey Daniel,</div><div><br></div><div>Thanks for the i=
nput.=C2=A0 I can see why you might be feeling some pain here, but I&#39;m =
not sure the proposed solution is quite right.</div><div><br></div><div>Not=
e that ACME already has the &quot;new-authorization&quot; flow, which does =
pretty much exactly what you&#39;re proposing.=C2=A0 The only difference is=
 that instead of authorizations being created all at once in response to a =
&quot;new-cert&quot; request, they&#39;re created individually by the clien=
t.=C2=A0 But it&#39;s ultimately the same number of RTTs (authz fetch ~ new=
-authz) -- or rather, one fewer in the new-authz case because there&#39;s n=
o new-cert request.=C2=A0 To what degree would your problem be solved simpl=
y by having the affected integrators move over to that flow?<br></div><div>=
<br></div><div>I can sympathize that the new-order flow might not be suitab=
le for all use cases.=C2=A0 Maybe it would be useful to write down some tho=
ughts on which use cases are best addressed by which flows.=C2=A0 <br></div=
><div><br></div><div>--Richard<br></div><div><br></div><div><br></div><div =
class=3D"gmail_extra"><br><div class=3D"gmail_quote">On Tue, Sep 19, 2017 a=
t 12:26 PM, Daniel McCarney <span dir=3D"ltr">&lt;<a href=3D"mailto:cpu@let=
sencrypt.org" target=3D"_blank">cpu@letsencrypt.org</a>&gt;</span> wrote:<b=
r><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:=
1px #ccc solid;padding-left:1ex"><div dir=3D"ltr"><div>Hi folks,</div><div>=
<br></div><div>Just over a year ago in a thread titled &quot;Proactive vs O=
n-Finalization</div><div>Certificate Issuance&quot;[0] we reached the conse=
nsus on the question of whether to</div><div>issue a certificate &quot;on-f=
inalization&quot; or &quot;proactively&quot; with the conclusion to</div><d=
iv>standardize on proactive issuance.</div><div><br></div><div>Implementati=
on experience shows that proactive issuance is slow in practice. An</div><d=
iv>order can be associated with many authorizations and an authorization ca=
n be</div><div>associated with many orders. This interacts poorly with opti=
mizations Let&#39;s</div><div>Encrypt has developed from in-the-field exper=
ience with ACME and it is difficult</div><div>to rate-limit effectively. I =
propose that we return to the on-finalization style</div><div>approach and =
that we provide the CSR during finalization..</div><div><br></div><div>## I=
ssues with proactive issuance</div><div><br></div><div>### Authorization Re=
use</div><div><br></div><div>Proactive issuance requires that when an autho=
rization is switched to a valid</div><div>state by the result of a challeng=
e update that we find the associated orders and</div><div>iterate each orde=
r&#39;s authorizations to see if all of the authorizations are</div><div>va=
lid and issuance can occur.</div><div><br></div><div>Let&#39;s Encrypt reus=
es existing authorizations[1][2] whenever possible, as</div><div>encouraged=
 by ACME. This has been of huge value in reducing resource consumption</div=
><div>(both in terms of database storage and challenge/API requests). Many =
clients</div><div>create pending authorizations they do not finalize and si=
milarly re-create</div><div>pending authorizations for identifiers that the=
 account already has valid</div><div>authorizations for. Unchecked this lea=
ds to significant database growth and</div><div>resource consumption, even =
with aggressive rate-limits[3].</div><div><br></div><div>The combination of=
 authorization reuse and proactive issuance means a given</div><div>authori=
zation can be associated with many orders. This necessitates an expensive</=
div><div>many-to-many query at each authorization update to see if any asso=
ciated</div><div>orders are now complete.</div><div><br></div><div>One poss=
ible workaround is a rate-limit that would constrain the number of</div><di=
v>orders an authorization can be associated with. However establishing the =
value</div><div>of the limit to both reduce server-side load &amp; also eff=
ectively support Let&#39;s</div><div>Encrypt&#39;s large-volume integrators=
 and their issuance patterns is challenging.</div><div><br></div><div>### C=
SR-first new-order flow</div><div><br></div><div>To support proactive issua=
nce the CA also needs to have the CSR for the order</div><div>at-hand when =
checking each authorization in case issuance can proceed. The</div><div>`ne=
w-order` request that creates the order therefore carries the full CSR that=
</div><div>was previously delivered by the &quot;certificate finalize&quot;=
 message.</div><div><br></div><div>We see users create many more authorizat=
ions than they are able to finalize</div><div>successfully. With the legacy=
 &quot;new-cert&quot; flow our authorization reuse</div><div>dramatically h=
elped reduce the DB space used up by authorizations that will</div><div>nev=
er lead to issuance because no CSR is stored until all identifiers are</div=
><div>validated.</div><div><br></div><div>With the new-order flow we have t=
o accept and store a full CSR before handing</div><div>back authorizations =
to the client. Reuse of authorizations does not prevent the</div><div>DB bl=
oat that will occur from users submitting new order requests and not</div><=
div>fulfilling them. The CSR contains a full public key and is one of the s=
ingle</div><div>largest contributors to DB disk space usage. Addressing thi=
s will require</div><div>another strict rate limit on outstanding new-order=
s and is again difficult to</div><div>=E2=80=9Cdial-in=E2=80=9D in to both =
prevent excessive resource consumption and also maintain</div><div>the abil=
ity for big integrators to process large-scale issuance using our API.</div=
><div><br></div><div># Proposed Change</div><div><br></div><div>I believe w=
e should change the `new-order` endpoint to not accept a full CSR but</div>=
<div>to instead accept a list of identifiers that the ACME client wishes to=
</div><div>authenticate &amp; issue for. Authorizations can be created and =
returned for those</div><div>domains as happens today.</div><div><br></div>=
<div>The order resource should be changed to have a =E2=80=9CfinalizeURL=E2=
=80=9D field which</div><div>provides the client a URL to POST a CSR to ind=
icate that the CA should issue the</div><div>certificate. Proactive issuanc=
e should be removed outright.</div><div><br></div><div>This will address bo=
th the problem of scanning existing many-to-many orders</div><div>&amp; the=
ir authorizations as well as the db bloat that will occur from</div><div>fr=
ont-loading the delivery of the CSR.</div><div><br></div><div>It may be tem=
pting to leave proactive issuance as an optional feature but this</div><div=
>will require maintaining the CSR in the new-order and I also believe that =
having</div><div>two very distinct methods of issuance will complicate the =
client ecosystem and</div><div>returns us to the question of how to determi=
ne which will occur that spawned the</div><div>original &quot;Proactive vs =
On-finalization Certificate Issuance&quot;</div><div>thread[0].</div><div><=
br></div><div>[0] <a href=3D"https://mailarchive.ietf.org/arch/msg/acme/0lV=
mGl8e-rmSH0x7ycDW5dj6GAY" target=3D"_blank">https://mailarchive.ietf.org/<w=
br>arch/msg/acme/0lVmGl8e-<wbr>rmSH0x7ycDW5dj6GAY</a></div><div>[1] <a href=
=3D"https://community.letsencrypt.org/t/upcoming-api-changes/17947" target=
=3D"_blank">https://community.letsencrypt.<wbr>org/t/upcoming-api-changes/<=
wbr>17947</a></div><div>[2] <a href=3D"https://community.letsencrypt.org/t/=
automatic-recycling-of-pending-authorizations/41321" target=3D"_blank">http=
s://community.letsencrypt.<wbr>org/t/automatic-recycling-of-<wbr>pending-au=
thorizations/41321</a></div><div>[3] <a href=3D"https://letsencrypt.org/doc=
s/rate-limits/" target=3D"_blank">https://letsencrypt.org/docs/<wbr>rate-li=
mits/</a></div><div><br></div></div>
<br>______________________________<wbr>_________________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/<wbr>listinfo/acme</a><br>
<br></blockquote></div><br></div></div>

--f403045d6084eda24c0559a4bbfa--


From nobody Thu Sep 21 11:40:11 2017
Return-Path: <rlb@ipv.sx>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1D9DE132924 for <acme@ietfa.amsl.com>; Thu, 21 Sep 2017 11:40:09 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.598
X-Spam-Level: 
X-Spam-Status: No, score=-2.598 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=ipv-sx.20150623.gappssmtp.com
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 GHkAGhYWwT9x for <acme@ietfa.amsl.com>; Thu, 21 Sep 2017 11:40:05 -0700 (PDT)
Received: from mail-wm0-x22f.google.com (mail-wm0-x22f.google.com [IPv6:2a00:1450:400c:c09::22f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 59C68132198 for <acme@ietf.org>; Thu, 21 Sep 2017 11:40:05 -0700 (PDT)
Received: by mail-wm0-x22f.google.com with SMTP id m72so2009576wmc.0 for <acme@ietf.org>; Thu, 21 Sep 2017 11:40:05 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ipv-sx.20150623.gappssmtp.com; s=20150623; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=g6WAhf4Nom5hunFuQNCD0au+57PeA08t+I8YZ4jU6Z8=; b=cgzn85KiA435jIolKRZxmohNjB4k/MG0sKpz9rNVjarcqnNuDe1Dkf7K3IyvmbpIFs 9AgrPaAyMIGtcR45VNC1C2pfHdQeZ4IyPsp097cIcAJCAVgldKVXzhzIHZVT3VUiQwGy M7UcDyFv6x1Uheov6T0TFr79p0/Qmuvh/VkrnDrc5PGDVJ1RFMuOH8MGF/X4/kf7eBLl 1tuaxlo+LGv4Dee5F4U/UaW/L/AI5k9bUG8y7TB3BL5G9lh7A07iG370XiYvGwpzwRcP ESnNw/0vD0PHtYbkj43sA8ZOJDHfq6Jcx/GFxBnFcPHbf+MQ66Fb3RAZ9chjYJ14VckQ m/5A==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=g6WAhf4Nom5hunFuQNCD0au+57PeA08t+I8YZ4jU6Z8=; b=oq5KQnxkoAKeTn6rq2JoRlm2cDEr/u36Kq1G+iCImp5YIMYH+3RoyvO6WWXPEVj+m9 fhBv2BwSiySu84f/Skt39t8Ko4dVWaBzGFAVf4miRMA2VmS561ZmpAUmlo+DV2MeWKdV Y3YHzBnlOJxdeooLOVrUS9JMvTXjI0x9dIkHRDk1aYbEGgXrcZ1P+kqGjcJzeRRdlsjX ufuzlbxpPWrcGQbxDqAHafzbeuyA2qFJmutm7tsZ3MaCbZG4n9iV4r/Sun9t2W2q2YP9 ApktmpispfJTvBQ6Zjhndj55eSE/itTvmb8EDd6WAhs68+3hfmwsvMH+U3rAkV4V7aLB aruQ==
X-Gm-Message-State: AHPjjUhv1wLWhwihLvm28eUfqHeQofnhkYAmdka0fmMIe7vQVDdkwvbd 6Vbho38tNSSDM/ltIaoj5IlyIFC0zhlpzizstAA6fmmR
X-Google-Smtp-Source: AOwi7QBzXAtI2SFVVS+90elWWt55XRyVHRYxyeP1wuFwnVYMXaqAki2lvC2Eh3gzjnGh0mPQhClMJN0X4ZHgCUxeYw4=
X-Received: by 10.28.236.212 with SMTP id h81mr1750957wmi.42.1506019203645; Thu, 21 Sep 2017 11:40:03 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.28.159.200 with HTTP; Thu, 21 Sep 2017 11:40:03 -0700 (PDT)
In-Reply-To: <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com>
From: Richard Barnes <rlb@ipv.sx>
Date: Thu, 21 Sep 2017 11:40:03 -0700
Message-ID: <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com>
To: Daniel McCarney <cpu@letsencrypt.org>
Cc: IETF ACME <acme@ietf.org>
Content-Type: multipart/alternative; boundary="001a11477a5ca898a50559b76b02"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/k9s5pitb6N48bU4TN4HfpPf4BR4>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 21 Sep 2017 18:40:09 -0000

--001a11477a5ca898a50559b76b02
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

I talked a bit with Daniel about this off-list. Here's a summary of where
we got to:

1. In the current "new-order" flow, the server doesn't *really* have to be
proactive -- instead of updating an order object when an authz changes
state, it can wait to update/issue until the client polls for the order
object, indicating that it's actually interested in having a certificate.
That doesn't alleviate the pain of storing the CSR, but it at least avoids
the expense of finding affected orders and issuing in cases where the
client abandons the order.  If we think this approach is OK, I would
propose that we revise the document to make clear that it is OK.  What do
folks think?  Daniel noted that there might be some issues with GET
idempotency here, but I don't think this actually makes GET non-idempotent.

2. Some CAs might want to support only the "new-authz" flow, i.e., never
allow a "new-order" request if the relevant authorizations haven't already
been done.  It seems like this could be addressed in a pretty
straightforward manner just by defining a new error code (say
"preauthRequired") that the server can return in this case.

Personally, it seems to me that if we make the clarification (1) and add
the error code (2) described here, that would address the pain points
Daniel raises, without a whole lot of new mechanism.  I'm not really seeing
much additional value to be had from an explicit "ok issue now" request
(vs. the GET poll).

--Richard



On Wed, Sep 20, 2017 at 1:22 PM, Richard Barnes <rlb@ipv.sx> wrote:

> Hey Daniel,
>
> Thanks for the input.  I can see why you might be feeling some pain here,
> but I'm not sure the proposed solution is quite right.
>
> Note that ACME already has the "new-authorization" flow, which does prett=
y
> much exactly what you're proposing.  The only difference is that instead =
of
> authorizations being created all at once in response to a "new-cert"
> request, they're created individually by the client.  But it's ultimately
> the same number of RTTs (authz fetch ~ new-authz) -- or rather, one fewer
> in the new-authz case because there's no new-cert request.  To what degre=
e
> would your problem be solved simply by having the affected integrators mo=
ve
> over to that flow?
>
> I can sympathize that the new-order flow might not be suitable for all us=
e
> cases.  Maybe it would be useful to write down some thoughts on which use
> cases are best addressed by which flows.
>
> --Richard
>
>
>
> On Tue, Sep 19, 2017 at 12:26 PM, Daniel McCarney <cpu@letsencrypt.org>
> wrote:
>
>> Hi folks,
>>
>> Just over a year ago in a thread titled "Proactive vs On-Finalization
>> Certificate Issuance"[0] we reached the consensus on the question of
>> whether to
>> issue a certificate "on-finalization" or "proactively" with the
>> conclusion to
>> standardize on proactive issuance.
>>
>> Implementation experience shows that proactive issuance is slow in
>> practice. An
>> order can be associated with many authorizations and an authorization ca=
n
>> be
>> associated with many orders. This interacts poorly with optimizations
>> Let's
>> Encrypt has developed from in-the-field experience with ACME and it is
>> difficult
>> to rate-limit effectively. I propose that we return to the
>> on-finalization style
>> approach and that we provide the CSR during finalization..
>>
>> ## Issues with proactive issuance
>>
>> ### Authorization Reuse
>>
>> Proactive issuance requires that when an authorization is switched to a
>> valid
>> state by the result of a challenge update that we find the associated
>> orders and
>> iterate each order's authorizations to see if all of the authorizations
>> are
>> valid and issuance can occur.
>>
>> Let's Encrypt reuses existing authorizations[1][2] whenever possible, as
>> encouraged by ACME. This has been of huge value in reducing resource
>> consumption
>> (both in terms of database storage and challenge/API requests). Many
>> clients
>> create pending authorizations they do not finalize and similarly re-crea=
te
>> pending authorizations for identifiers that the account already has vali=
d
>> authorizations for. Unchecked this leads to significant database growth
>> and
>> resource consumption, even with aggressive rate-limits[3].
>>
>> The combination of authorization reuse and proactive issuance means a
>> given
>> authorization can be associated with many orders. This necessitates an
>> expensive
>> many-to-many query at each authorization update to see if any associated
>> orders are now complete.
>>
>> One possible workaround is a rate-limit that would constrain the number =
of
>> orders an authorization can be associated with. However establishing the
>> value
>> of the limit to both reduce server-side load & also effectively support
>> Let's
>> Encrypt's large-volume integrators and their issuance patterns is
>> challenging.
>>
>> ### CSR-first new-order flow
>>
>> To support proactive issuance the CA also needs to have the CSR for the
>> order
>> at-hand when checking each authorization in case issuance can proceed. T=
he
>> `new-order` request that creates the order therefore carries the full CS=
R
>> that
>> was previously delivered by the "certificate finalize" message.
>>
>> We see users create many more authorizations than they are able to
>> finalize
>> successfully. With the legacy "new-cert" flow our authorization reuse
>> dramatically helped reduce the DB space used up by authorizations that
>> will
>> never lead to issuance because no CSR is stored until all identifiers ar=
e
>> validated.
>>
>> With the new-order flow we have to accept and store a full CSR before
>> handing
>> back authorizations to the client. Reuse of authorizations does not
>> prevent the
>> DB bloat that will occur from users submitting new order requests and no=
t
>> fulfilling them. The CSR contains a full public key and is one of the
>> single
>> largest contributors to DB disk space usage. Addressing this will requir=
e
>> another strict rate limit on outstanding new-orders and is again
>> difficult to
>> =E2=80=9Cdial-in=E2=80=9D in to both prevent excessive resource consumpt=
ion and also
>> maintain
>> the ability for big integrators to process large-scale issuance using ou=
r
>> API.
>>
>> # Proposed Change
>>
>> I believe we should change the `new-order` endpoint to not accept a full
>> CSR but
>> to instead accept a list of identifiers that the ACME client wishes to
>> authenticate & issue for. Authorizations can be created and returned for
>> those
>> domains as happens today.
>>
>> The order resource should be changed to have a =E2=80=9CfinalizeURL=E2=
=80=9D field which
>> provides the client a URL to POST a CSR to indicate that the CA should
>> issue the
>> certificate. Proactive issuance should be removed outright.
>>
>> This will address both the problem of scanning existing many-to-many
>> orders
>> & their authorizations as well as the db bloat that will occur from
>> front-loading the delivery of the CSR.
>>
>> It may be tempting to leave proactive issuance as an optional feature bu=
t
>> this
>> will require maintaining the CSR in the new-order and I also believe tha=
t
>> having
>> two very distinct methods of issuance will complicate the client
>> ecosystem and
>> returns us to the question of how to determine which will occur that
>> spawned the
>> original "Proactive vs On-finalization Certificate Issuance"
>> thread[0].
>>
>> [0] https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7y
>> cDW5dj6GAY
>> [1] https://community.letsencrypt.org/t/upcoming-api-changes/17947
>> [2] https://community.letsencrypt.org/t/automatic-recycling-of-p
>> ending-authorizations/41321
>> [3] https://letsencrypt.org/docs/rate-limits/
>>
>>
>> _______________________________________________
>> Acme mailing list
>> Acme@ietf.org
>> https://www.ietf.org/mailman/listinfo/acme
>>
>>
>

--001a11477a5ca898a50559b76b02
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>I talked a bit with Daniel about this off-list. Here&=
#39;s a summary of where we got to:</div><div><br></div><div>1. In the curr=
ent &quot;new-order&quot; flow, the server doesn&#39;t *really* have to be =
proactive -- instead of updating an order object when an authz changes stat=
e, it can wait to update/issue until the client polls for the order object,=
 indicating that it&#39;s actually interested in having a certificate.=C2=
=A0 That doesn&#39;t alleviate the pain of storing the CSR, but it at least=
 avoids the expense of finding affected orders and issuing in cases where t=
he client abandons the order.=C2=A0 If we think this approach is OK, I woul=
d propose that we revise the document to make clear that it is OK.=C2=A0 Wh=
at do folks think?=C2=A0 Daniel noted that there might be some issues with =
GET idempotency here, but I don&#39;t think this actually makes GET non-ide=
mpotent.<br></div><div><br></div><div>2. Some CAs might want to support onl=
y the &quot;new-authz&quot; flow, i.e., never allow a &quot;new-order&quot;=
 request if the relevant authorizations haven&#39;t already been done.=C2=
=A0 It seems like this could be addressed in a pretty straightforward manne=
r just by defining a new error code (say &quot;preauthRequired&quot;) that =
the server can return in this case.<br></div><div><br></div><div>Personally=
, it seems to me that if we make the clarification (1) and add the error co=
de (2) described here, that would address the pain points Daniel raises, wi=
thout a whole lot of new mechanism.=C2=A0 I&#39;m not really seeing much ad=
ditional value to be had from an explicit &quot;ok issue now&quot; request =
(vs. the GET poll).</div><div><br></div><div>--Richard<br></div><div><br></=
div><div><br></div></div><div class=3D"gmail_extra"><br><div class=3D"gmail=
_quote">On Wed, Sep 20, 2017 at 1:22 PM, Richard Barnes <span dir=3D"ltr">&=
lt;<a href=3D"mailto:rlb@ipv.sx" target=3D"_blank">rlb@ipv.sx</a>&gt;</span=
> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;bo=
rder-left:1px #ccc solid;padding-left:1ex"><div dir=3D"ltr"><div>Hey Daniel=
,</div><div><br></div><div>Thanks for the input.=C2=A0 I can see why you mi=
ght be feeling some pain here, but I&#39;m not sure the proposed solution i=
s quite right.</div><div><br></div><div>Note that ACME already has the &quo=
t;new-authorization&quot; flow, which does pretty much exactly what you&#39=
;re proposing.=C2=A0 The only difference is that instead of authorizations =
being created all at once in response to a &quot;new-cert&quot; request, th=
ey&#39;re created individually by the client.=C2=A0 But it&#39;s ultimately=
 the same number of RTTs (authz fetch ~ new-authz) -- or rather, one fewer =
in the new-authz case because there&#39;s no new-cert request.=C2=A0 To wha=
t degree would your problem be solved simply by having the affected integra=
tors move over to that flow?<br></div><div><br></div><div>I can sympathize =
that the new-order flow might not be suitable for all use cases.=C2=A0 Mayb=
e it would be useful to write down some thoughts on which use cases are bes=
t addressed by which flows.=C2=A0 <br></div><span class=3D"HOEnZb"><font co=
lor=3D"#888888"><div><br></div><div>--Richard<br></div><div><br></div><div>=
<br></div></font></span><div class=3D"gmail_extra"><br><div class=3D"gmail_=
quote"><div><div class=3D"h5">On Tue, Sep 19, 2017 at 12:26 PM, Daniel McCa=
rney <span dir=3D"ltr">&lt;<a href=3D"mailto:cpu@letsencrypt.org" target=3D=
"_blank">cpu@letsencrypt.org</a>&gt;</span> wrote:<br></div></div><blockquo=
te class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc so=
lid;padding-left:1ex"><div><div class=3D"h5"><div dir=3D"ltr"><div>Hi folks=
,</div><div><br></div><div>Just over a year ago in a thread titled &quot;Pr=
oactive vs On-Finalization</div><div>Certificate Issuance&quot;[0] we reach=
ed the consensus on the question of whether to</div><div>issue a certificat=
e &quot;on-finalization&quot; or &quot;proactively&quot; with the conclusio=
n to</div><div>standardize on proactive issuance.</div><div><br></div><div>=
Implementation experience shows that proactive issuance is slow in practice=
. An</div><div>order can be associated with many authorizations and an auth=
orization can be</div><div>associated with many orders. This interacts poor=
ly with optimizations Let&#39;s</div><div>Encrypt has developed from in-the=
-field experience with ACME and it is difficult</div><div>to rate-limit eff=
ectively. I propose that we return to the on-finalization style</div><div>a=
pproach and that we provide the CSR during finalization..</div><div><br></d=
iv><div>## Issues with proactive issuance</div><div><br></div><div>### Auth=
orization Reuse</div><div><br></div><div>Proactive issuance requires that w=
hen an authorization is switched to a valid</div><div>state by the result o=
f a challenge update that we find the associated orders and</div><div>itera=
te each order&#39;s authorizations to see if all of the authorizations are<=
/div><div>valid and issuance can occur.</div><div><br></div><div>Let&#39;s =
Encrypt reuses existing authorizations[1][2] whenever possible, as</div><di=
v>encouraged by ACME. This has been of huge value in reducing resource cons=
umption</div><div>(both in terms of database storage and challenge/API requ=
ests). Many clients</div><div>create pending authorizations they do not fin=
alize and similarly re-create</div><div>pending authorizations for identifi=
ers that the account already has valid</div><div>authorizations for. Unchec=
ked this leads to significant database growth and</div><div>resource consum=
ption, even with aggressive rate-limits[3].</div><div><br></div><div>The co=
mbination of authorization reuse and proactive issuance means a given</div>=
<div>authorization can be associated with many orders. This necessitates an=
 expensive</div><div>many-to-many query at each authorization update to see=
 if any associated</div><div>orders are now complete.</div><div><br></div><=
div>One possible workaround is a rate-limit that would constrain the number=
 of</div><div>orders an authorization can be associated with. However estab=
lishing the value</div><div>of the limit to both reduce server-side load &a=
mp; also effectively support Let&#39;s</div><div>Encrypt&#39;s large-volume=
 integrators and their issuance patterns is challenging.</div><div><br></di=
v><div>### CSR-first new-order flow</div><div><br></div><div>To support pro=
active issuance the CA also needs to have the CSR for the order</div><div>a=
t-hand when checking each authorization in case issuance can proceed. The</=
div><div>`new-order` request that creates the order therefore carries the f=
ull CSR that</div><div>was previously delivered by the &quot;certificate fi=
nalize&quot; message.</div><div><br></div><div>We see users create many mor=
e authorizations than they are able to finalize</div><div>successfully. Wit=
h the legacy &quot;new-cert&quot; flow our authorization reuse</div><div>dr=
amatically helped reduce the DB space used up by authorizations that will</=
div><div>never lead to issuance because no CSR is stored until all identifi=
ers are</div><div>validated.</div><div><br></div><div>With the new-order fl=
ow we have to accept and store a full CSR before handing</div><div>back aut=
horizations to the client. Reuse of authorizations does not prevent the</di=
v><div>DB bloat that will occur from users submitting new order requests an=
d not</div><div>fulfilling them. The CSR contains a full public key and is =
one of the single</div><div>largest contributors to DB disk space usage. Ad=
dressing this will require</div><div>another strict rate limit on outstandi=
ng new-orders and is again difficult to</div><div>=E2=80=9Cdial-in=E2=80=9D=
 in to both prevent excessive resource consumption and also maintain</div><=
div>the ability for big integrators to process large-scale issuance using o=
ur API.</div><div><br></div><div># Proposed Change</div><div><br></div><div=
>I believe we should change the `new-order` endpoint to not accept a full C=
SR but</div><div>to instead accept a list of identifiers that the ACME clie=
nt wishes to</div><div>authenticate &amp; issue for. Authorizations can be =
created and returned for those</div><div>domains as happens today.</div><di=
v><br></div><div>The order resource should be changed to have a =E2=80=9Cfi=
nalizeURL=E2=80=9D field which</div><div>provides the client a URL to POST =
a CSR to indicate that the CA should issue the</div><div>certificate. Proac=
tive issuance should be removed outright.</div><div><br></div><div>This wil=
l address both the problem of scanning existing many-to-many orders</div><d=
iv>&amp; their authorizations as well as the db bloat that will occur from<=
/div><div>front-loading the delivery of the CSR.</div><div><br></div><div>I=
t may be tempting to leave proactive issuance as an optional feature but th=
is</div><div>will require maintaining the CSR in the new-order and I also b=
elieve that having</div><div>two very distinct methods of issuance will com=
plicate the client ecosystem and</div><div>returns us to the question of ho=
w to determine which will occur that spawned the</div><div>original &quot;P=
roactive vs On-finalization Certificate Issuance&quot;</div><div>thread[0].=
</div><div><br></div><div>[0] <a href=3D"https://mailarchive.ietf.org/arch/=
msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY" target=3D"_blank">https://mailarchive=
.ietf.org/a<wbr>rch/msg/acme/0lVmGl8e-rmSH0x7y<wbr>cDW5dj6GAY</a></div><div=
>[1] <a href=3D"https://community.letsencrypt.org/t/upcoming-api-changes/17=
947" target=3D"_blank">https://community.letsencrypt.<wbr>org/t/upcoming-ap=
i-changes/179<wbr>47</a></div><div>[2] <a href=3D"https://community.letsenc=
rypt.org/t/automatic-recycling-of-pending-authorizations/41321" target=3D"_=
blank">https://community.letsencrypt.<wbr>org/t/automatic-recycling-of-p<wb=
r>ending-authorizations/41321</a></div><div>[3] <a href=3D"https://letsencr=
ypt.org/docs/rate-limits/" target=3D"_blank">https://letsencrypt.org/docs/r=
<wbr>ate-limits/</a></div><div><br></div></div>
<br></div></div><span class=3D"">______________________________<wbr>_______=
__________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org" target=3D"_blank">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/l<wbr>istinfo/acme</a><br>
<br></span></blockquote></div><br></div></div>
</blockquote></div><br></div>

--001a11477a5ca898a50559b76b02--


From nobody Thu Sep 21 16:25:50 2017
Return-Path: <rsalz@akamai.com>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1DCD21320D8 for <acme@ietfa.amsl.com>; Thu, 21 Sep 2017 16:25:49 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=akamai.com
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 wXNeOa__Zh7b for <acme@ietfa.amsl.com>; Thu, 21 Sep 2017 16:25:46 -0700 (PDT)
Received: from mx0a-00190b01.pphosted.com (mx0a-00190b01.pphosted.com [67.231.149.131]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id DAB1D13306F for <acme@ietf.org>; Thu, 21 Sep 2017 16:25:46 -0700 (PDT)
Received: from pps.filterd (m0122333.ppops.net [127.0.0.1]) by mx0a-00190b01.pphosted.com (8.16.0.21/8.16.0.21) with SMTP id v8LNLrTw010308; Fri, 22 Sep 2017 00:25:45 +0100
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=akamai.com; h=from : to : cc : subject : date : message-id : references : in-reply-to : content-type : mime-version; s=jan2016.eng; bh=QHSTlHgrJiUePjseJD4mV6MhfoO9j+1JhhAHbRbxc24=; b=VAOHttUGmm2KDifa5KuZnf60s9TvM7uEKw6vJ8otfoGPb6JUTL1DcknjWIizMJLdQxOu dsQzDGIut9nJ4KeDCvqEwCpFBzJkPXF2ggloXO2xnNYI1ZDtIe6KOKqto+qJ8tjqZ/O4 QTOhjX4oFOW9Bcbin29pVcvq3J4KEcMJPlHASQv7lOSu3IYp1TgrUnfAx4YUjNzUpRs2 SUtVmg4qMY3zAA9k7vGxOwQqIRTI6JR+jmD+fim8S/xedUiHPqWi4Z5Q+tffjSoHeVIE FhTnopGKBQVc8DFU3FhYdDqOFPcQ6OOq3kJUT3hVwWBLj+huemXlLEav4zgkhDgKfBPB YQ== 
Received: from prod-mail-ppoint3 ([96.6.114.86]) by mx0a-00190b01.pphosted.com with ESMTP id 2d4nkx8wye-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Fri, 22 Sep 2017 00:25:45 +0100
Received: from pps.filterd (prod-mail-ppoint3.akamai.com [127.0.0.1]) by prod-mail-ppoint3.akamai.com (8.16.0.21/8.16.0.21) with SMTP id v8LNLJiX012599; Thu, 21 Sep 2017 19:25:44 -0400
Received: from email.msg.corp.akamai.com ([172.27.25.31]) by prod-mail-ppoint3.akamai.com with ESMTP id 2d0y9v8ajc-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-SHA384 bits=256 verify=NOT); Thu, 21 Sep 2017 19:25:43 -0400
Received: from USTX2EX-DAG1MB1.msg.corp.akamai.com (172.27.27.101) by ustx2ex-dag1mb1.msg.corp.akamai.com (172.27.27.101) with Microsoft SMTP Server (TLS) id 15.0.1263.5; Thu, 21 Sep 2017 18:25:41 -0500
Received: from USTX2EX-DAG1MB1.msg.corp.akamai.com ([172.27.6.131]) by ustx2ex-dag1mb1.msg.corp.akamai.com ([172.27.6.131]) with mapi id 15.00.1263.000; Thu, 21 Sep 2017 18:25:41 -0500
From: "Salz, Rich" <rsalz@akamai.com>
To: Richard Barnes <rlb@ipv.sx>, Daniel McCarney <cpu@letsencrypt.org>
CC: IETF ACME <acme@ietf.org>
Thread-Topic: [Acme] Revisiting Proactive Issuance & new-order CSR
Thread-Index: AQHTMX0n4OR5qCsIX0ijdiL5aoXCT6K+jS+AgAF1u4CAAAzBAA==
Date: Thu, 21 Sep 2017 23:25:41 +0000
Message-ID: <20F8F23C-F7CA-4713-85B4-2FD69E3E6E25@akamai.com>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com> <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com>
In-Reply-To: <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
user-agent: Microsoft-MacOutlook/f.21.0.170409
x-ms-exchange-messagesentrepresentingtype: 1
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [172.19.237.189]
Content-Type: multipart/alternative; boundary="_000_20F8F23CF7CA471385B42FD69E3E6E25akamaicom_"
MIME-Version: 1.0
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:, , definitions=2017-09-21_06:, , signatures=0
X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 suspectscore=0 malwarescore=0 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1707230000 definitions=main-1709210314
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:, , definitions=2017-09-21_06:, , signatures=0
X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 priorityscore=1501 malwarescore=0 suspectscore=0 phishscore=0 bulkscore=0 spamscore=0 clxscore=1011 lowpriorityscore=0 impostorscore=0 adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1707230000 definitions=main-1709210314
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/yv_smfvoNYjyV518xQUhgvkkc0M>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 21 Sep 2017 23:25:49 -0000

--_000_20F8F23CF7CA471385B42FD69E3E6E25akamaicom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64

SWYgdGhlIHN1Z2dlc3Rpb24gaXMgZmluZSB3aXRoIERhbmllbCwgSSB0aGluayBlZGl0b3JpYWwg
Y2xhcmlmaWNhdGlvbiBhbmQgYSBuZXcgZXJyb3IgY29kZSBjYW4gYmUgYWRkZWQgd2l0aG91dCBk
aXNydXB0aW5nIHRoZSAgY3ljbGUgYW5kIEkgYW0gc3Ryb25nbHkgaW4gZmF2b3Igb2YgdGhhdC4N
Cg0K

--_000_20F8F23CF7CA471385B42FD69E3E6E25akamaicom_
Content-Type: text/html; charset="utf-8"
Content-ID: <7D8E4568AA25F241B163FF22EC3B7208@akamai.com>
Content-Transfer-Encoding: base64
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==

--_000_20F8F23CF7CA471385B42FD69E3E6E25akamaicom_--


From nobody Fri Sep 22 05:55:43 2017
Return-Path: <martin.thomson@gmail.com>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EF1FC134330 for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 05:55:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
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 ghmx1krzhsce for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 05:55:40 -0700 (PDT)
Received: from mail-io0-x22c.google.com (mail-io0-x22c.google.com [IPv6:2607:f8b0:4001:c06::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 62CFD134326 for <acme@ietf.org>; Fri, 22 Sep 2017 05:55:40 -0700 (PDT)
Received: by mail-io0-x22c.google.com with SMTP id i197so2799040ioe.9 for <acme@ietf.org>; Fri, 22 Sep 2017 05:55:40 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;  h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=nDx218sWCFKST40a0I74wgd5UItTFasX2jfC2mknQPo=; b=OPM9t5XJJhOv2KE8FfXKx9AzuRYBKYXGtD1TRlzNH+tfS6E8gWD4a/oo04YBV7qOs5 +lmc/mCcJPZqQtsI5mjlheSJ8w1zY2g+WpmCGQx8R+QPZTPos+v1yuKx+1+XdQGgeamO tssoRlDa0LO8n7swWSa3PsSDfwvGE5WaqQlPivPQXtCDko9vV48JQfTTaXgFlUkvacBD s/G7iLaQCoisqnXEBKC04s7kCdL9GQda/u+IGM3ap44bI6AolSS9Ibp2DYvchjlrAlaz RBBLCNIoj6tUMlg3iXEEF6DrS93+Q8fzFPyTYA+vGtJ0wBVVN4aZMbTvW4iFJMMw9506 Pl+Q==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=nDx218sWCFKST40a0I74wgd5UItTFasX2jfC2mknQPo=; b=YUcAnz/kd3TNxPcPcPg3FnQtNwk6fcBby6/FDHOyBiX16b71zLNGVx7SxNkqfMYU5W iPiDgchgCYvogDdhnliGkuME9gMH0rFBGiTQy4eJG+fn7NKcPgMM8jEoFUVfjD+nMPiZ oMR73C3KIjATmpLZvw+E32rc72qlWc9OWoKy0FRgB6u61kaoTT0u+62OJyKo4/K4jzKt X1uj3BOwZa8MGDnVM1mwT5CarTftmC4w53V2QD+vmBBqzBtuULkt/u3kafNujAeonb3h /o9U4PlzDmf6fmKl/vrjpXjLwZVua5Y6Ecx/Epr3dWRai1/DPBGAlj/YR4uJHv3ngkzb kGHw==
X-Gm-Message-State: AHPjjUi75+hHXMYIhgKaXi5Lr6C8RmlTlGiExHGD1NZFlQf5QBuGtuHZ ba+88d44boSdPdcdzjeh3VFd/Wo6jKlLQH9iS6c=
X-Google-Smtp-Source: AOwi7QDNof/N7Ej3t8hm0I/Z8GDA3Rys5jf+33XXBRqSWul6YgqPA9AZ2I1+ALWsPFLgv6tDva1asJCMlH4FwfUi7iM=
X-Received: by 10.202.67.139 with SMTP id q133mr5938586oia.257.1506084939674;  Fri, 22 Sep 2017 05:55:39 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.157.0.38 with HTTP; Fri, 22 Sep 2017 05:55:39 -0700 (PDT)
In-Reply-To: <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com> <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com>
From: Martin Thomson <martin.thomson@gmail.com>
Date: Fri, 22 Sep 2017 22:55:39 +1000
Message-ID: <CABkgnnXr_eGcyRCiye9_ObNC54Uf64UDUGcDSpSXtp-z_aroRw@mail.gmail.com>
To: Richard Barnes <rlb@ipv.sx>
Cc: Daniel McCarney <cpu@letsencrypt.org>, IETF ACME <acme@ietf.org>
Content-Type: text/plain; charset="UTF-8"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/ODF0R4MRS2vbPszuaXvYtliG098>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 22 Sep 2017 12:55:42 -0000

On Fri, Sep 22, 2017 at 4:40 AM, Richard Barnes <rlb@ipv.sx> wrote:
> Daniel noted that there might be some issues with GET idempotency here, but
> I don't think this actually makes GET non-idempotent.

It's still idempotent, because doing the GET twice has the same effect
as doing it once.  You are concerned about the GET no longer being
safe.  That is, it is no longer side-effect free.

It is OK to do things in response to a GET as long as the server owns
the consequences.  Servers frequently log GET requests, which isn't
safe or even idempotent, but servers do it because there is an
advantage to them that exceeds the costs.  As long as clients aren't
held accountable for these side-effects, it's OK.

Or, you could consider the resource to have been updated at some point
in the past as though the authorization triggered proactive issuance,
but it's just that the GET has taken a little while to "find" the
updated state.  Think of it as a lazy load, or maybe an extreme case
of the server cache needing to be populated.

If you like, we could also dive into whether this specific type of
request is safe to send in 0-RTT, but I don't have the brain cells for
that right now.


From nobody Fri Sep 22 06:42:40 2017
Return-Path: <dmccarney@letsencrypt.org>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 71F67134468 for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 06:42:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=letsencrypt.org
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 kGrlOji4sRCM for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 06:42:26 -0700 (PDT)
Received: from mail-io0-x22e.google.com (mail-io0-x22e.google.com [IPv6:2607:f8b0:4001:c06::22e]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B0E9C134458 for <acme@ietf.org>; Fri, 22 Sep 2017 06:42:25 -0700 (PDT)
Received: by mail-io0-x22e.google.com with SMTP id w94so3105297ioi.7 for <acme@ietf.org>; Fri, 22 Sep 2017 06:42:25 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=letsencrypt.org; s=google; h=mime-version:reply-to:in-reply-to:references:from:date:message-id :subject:to:cc; bh=n5SBjtoP0DSLaGR7oSG0X5NlkpKlfkTjV6u/UWCJ0jE=; b=HhzN25lPq0j6SUCCOPJPQ2cLh9LLhoKdN5e+HMTyP69D37R1swinxi7fa4BcQS0rcs xvgWv3Lw8VGnAXawU8GxUV10YdSsmkqTi+jVELEbVui8b4GSGKPSVLueLJ6/D7Qaar4j xwByoLn0LYyXkIwYf/Zky+tvEAoR3gMrliNmw=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:reply-to:in-reply-to:references :from:date:message-id:subject:to:cc; bh=n5SBjtoP0DSLaGR7oSG0X5NlkpKlfkTjV6u/UWCJ0jE=; b=QjUo2JAkHrI8hmIIq8NlyjmFjeu+d+vsPG47jbysih2PTlt+1iWN9J3bflRW6+8A51 RnP1d+p8vY1SN4fC3wpIkUd1cL11rx0cLT6tmQiVLjLkgF+BhR5CCp07tJ0HtTxHAUQY dkUDUawDEx7RXWvzmQ/gvmt7lc1iUWqe3UKBmu7Uh+jT0Xt1267z/JgV8ss7A2PCBdrg 6EW6ePzEPRvVyDNMO76kD4wKmNp087/bCK9Df1VBA3p1UJrBFis0lNvD7Y27v2fwkjwO wCMGHsESbKskyUF/DpP6OBR7j7XteR39osZZC6BcoxA/lZeX1Yo5KZvPq+nnEt5pjBDz LD5w==
X-Gm-Message-State: AHPjjUgZC3XexVpicM1gUACMBTovcl8cq/P9jlfymo7KYgfxTE2JeURh KsQ2NpdK5ly/+JCq356OwyMQ3mqkIPp4cPptDSbBwA==
X-Google-Smtp-Source: AOwi7QDAePJ35OMi4FApD6aB4mYD3PNNGwVEAQ6YUAFVS68hTCk3WRUVwbPxvviuEeu1EGt+XXGDAzMQzDVDX9OQ6RE=
X-Received: by 10.107.68.20 with SMTP id r20mr7405803ioa.187.1506087744745; Fri, 22 Sep 2017 06:42:24 -0700 (PDT)
MIME-Version: 1.0
Reply-To: cpu@letsencrypt.org
Received: by 10.107.17.229 with HTTP; Fri, 22 Sep 2017 06:42:24 -0700 (PDT)
In-Reply-To: <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com> <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com>
From: Daniel McCarney <cpu@letsencrypt.org>
Date: Fri, 22 Sep 2017 09:42:24 -0400
Message-ID: <CAKnbcLgz80aOGGiLQsXowja4vmjgQjLeZC6vZ2q+yFyAOXxTXQ@mail.gmail.com>
To: Richard Barnes <rlb@ipv.sx>
Cc: IETF ACME <acme@ietf.org>
Content-Type: multipart/alternative; boundary="089e0826371807b8830559c7614a"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/PE7z7cD3RyLmGs6hSj3j3BGfVgg>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 22 Sep 2017 13:42:37 -0000

--089e0826371807b8830559c7614a
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

RE: #1 - this does help, but will require changing the spec to clarify. The
current wording does not support this behaviour and cleary indicates the
server MUST issue for any order that has been satisfied.

RE: #2 - a CA relying on the new-authz flow to avoid the cost of accepting
a CSR up front would never be able to use the new-order flow to give the
client authorizations they didn't ask for or they are back to accepting
CSRs for orders that may never complete. That's why we need a new error to
return to reject new-order requests that aren't fully pre-authorized. If
the largest ACME CAs will not implement new-order as described outside of
pre-authorization why have it at all? Who will implement it?

The primary failing of #2 is that it causes those CAs to lose one
substantial benefit to the new-order flow that I would like to keep:
allowing for policy-based decisions on the CA side to influence which
authorizations are required. Without this property we will further have to
update the specification for wildcard issuance. The current spec doesn't
permit asking for authorizations for wildcard identifiers with new-authz so
you'd need some way of asking the server for a wildcard identifier and have
it respond with an authorization for something completely different (An
authorization for some wildcard resource say with a  DNS-01 challenge for
the base domain).

I agree that this is a late change but we are nearing the last chance we
have to address what I see as a fairly significant point of complication in
the protocol.  I note that no one has proposed any reasons why proactive
issuance must remain since this thread was restarted except that it is the
status-quo and we want to minimize change.

I am not in favour of the suggestions from Richard so far and would like to
hear more about why my proposal is not suitable. Richard's changes are
smaller but leave protocol complication and prevent large CAs from enjoying
the benefits of the new-order flow. My proposal addresses the problems and
leaves a simpler protocol without features that will not be used by large
providers.

We have to change the specification no matter whether my proposals or
Richards are taken so why not make changes that will result in the clearest
protocol with the least amount of cruft few are likely to implement?

- Daniel





On Thu, Sep 21, 2017 at 2:40 PM, Richard Barnes <rlb@ipv.sx> wrote:

> I talked a bit with Daniel about this off-list. Here's a summary of where
> we got to:
>
> 1. In the current "new-order" flow, the server doesn't *really* have to b=
e
> proactive -- instead of updating an order object when an authz changes
> state, it can wait to update/issue until the client polls for the order
> object, indicating that it's actually interested in having a certificate.
> That doesn't alleviate the pain of storing the CSR, but it at least avoid=
s
> the expense of finding affected orders and issuing in cases where the
> client abandons the order.  If we think this approach is OK, I would
> propose that we revise the document to make clear that it is OK.  What do
> folks think?  Daniel noted that there might be some issues with GET
> idempotency here, but I don't think this actually makes GET non-idempoten=
t.
>
> 2. Some CAs might want to support only the "new-authz" flow, i.e., never
> allow a "new-order" request if the relevant authorizations haven't alread=
y
> been done.  It seems like this could be addressed in a pretty
> straightforward manner just by defining a new error code (say
> "preauthRequired") that the server can return in this case.
>
> Personally, it seems to me that if we make the clarification (1) and add
> the error code (2) described here, that would address the pain points
> Daniel raises, without a whole lot of new mechanism.  I'm not really seei=
ng
> much additional value to be had from an explicit "ok issue now" request
> (vs. the GET poll).
>
> --Richard
>
>
>
> On Wed, Sep 20, 2017 at 1:22 PM, Richard Barnes <rlb@ipv.sx> wrote:
>
>> Hey Daniel,
>>
>> Thanks for the input.  I can see why you might be feeling some pain here=
,
>> but I'm not sure the proposed solution is quite right.
>>
>> Note that ACME already has the "new-authorization" flow, which does
>> pretty much exactly what you're proposing.  The only difference is that
>> instead of authorizations being created all at once in response to a
>> "new-cert" request, they're created individually by the client.  But it'=
s
>> ultimately the same number of RTTs (authz fetch ~ new-authz) -- or rathe=
r,
>> one fewer in the new-authz case because there's no new-cert request.  To
>> what degree would your problem be solved simply by having the affected
>> integrators move over to that flow?
>>
>> I can sympathize that the new-order flow might not be suitable for all
>> use cases.  Maybe it would be useful to write down some thoughts on whic=
h
>> use cases are best addressed by which flows.
>>
>> --Richard
>>
>>
>>
>> On Tue, Sep 19, 2017 at 12:26 PM, Daniel McCarney <cpu@letsencrypt.org>
>> wrote:
>>
>>> Hi folks,
>>>
>>> Just over a year ago in a thread titled "Proactive vs On-Finalization
>>> Certificate Issuance"[0] we reached the consensus on the question of
>>> whether to
>>> issue a certificate "on-finalization" or "proactively" with the
>>> conclusion to
>>> standardize on proactive issuance.
>>>
>>> Implementation experience shows that proactive issuance is slow in
>>> practice. An
>>> order can be associated with many authorizations and an authorization
>>> can be
>>> associated with many orders. This interacts poorly with optimizations
>>> Let's
>>> Encrypt has developed from in-the-field experience with ACME and it is
>>> difficult
>>> to rate-limit effectively. I propose that we return to the
>>> on-finalization style
>>> approach and that we provide the CSR during finalization..
>>>
>>> ## Issues with proactive issuance
>>>
>>> ### Authorization Reuse
>>>
>>> Proactive issuance requires that when an authorization is switched to a
>>> valid
>>> state by the result of a challenge update that we find the associated
>>> orders and
>>> iterate each order's authorizations to see if all of the authorizations
>>> are
>>> valid and issuance can occur.
>>>
>>> Let's Encrypt reuses existing authorizations[1][2] whenever possible, a=
s
>>> encouraged by ACME. This has been of huge value in reducing resource
>>> consumption
>>> (both in terms of database storage and challenge/API requests). Many
>>> clients
>>> create pending authorizations they do not finalize and similarly
>>> re-create
>>> pending authorizations for identifiers that the account already has val=
id
>>> authorizations for. Unchecked this leads to significant database growth
>>> and
>>> resource consumption, even with aggressive rate-limits[3].
>>>
>>> The combination of authorization reuse and proactive issuance means a
>>> given
>>> authorization can be associated with many orders. This necessitates an
>>> expensive
>>> many-to-many query at each authorization update to see if any associate=
d
>>> orders are now complete.
>>>
>>> One possible workaround is a rate-limit that would constrain the number
>>> of
>>> orders an authorization can be associated with. However establishing th=
e
>>> value
>>> of the limit to both reduce server-side load & also effectively support
>>> Let's
>>> Encrypt's large-volume integrators and their issuance patterns is
>>> challenging.
>>>
>>> ### CSR-first new-order flow
>>>
>>> To support proactive issuance the CA also needs to have the CSR for the
>>> order
>>> at-hand when checking each authorization in case issuance can proceed.
>>> The
>>> `new-order` request that creates the order therefore carries the full
>>> CSR that
>>> was previously delivered by the "certificate finalize" message.
>>>
>>> We see users create many more authorizations than they are able to
>>> finalize
>>> successfully. With the legacy "new-cert" flow our authorization reuse
>>> dramatically helped reduce the DB space used up by authorizations that
>>> will
>>> never lead to issuance because no CSR is stored until all identifiers a=
re
>>> validated.
>>>
>>> With the new-order flow we have to accept and store a full CSR before
>>> handing
>>> back authorizations to the client. Reuse of authorizations does not
>>> prevent the
>>> DB bloat that will occur from users submitting new order requests and n=
ot
>>> fulfilling them. The CSR contains a full public key and is one of the
>>> single
>>> largest contributors to DB disk space usage. Addressing this will requi=
re
>>> another strict rate limit on outstanding new-orders and is again
>>> difficult to
>>> =E2=80=9Cdial-in=E2=80=9D in to both prevent excessive resource consump=
tion and also
>>> maintain
>>> the ability for big integrators to process large-scale issuance using
>>> our API.
>>>
>>> # Proposed Change
>>>
>>> I believe we should change the `new-order` endpoint to not accept a ful=
l
>>> CSR but
>>> to instead accept a list of identifiers that the ACME client wishes to
>>> authenticate & issue for. Authorizations can be created and returned fo=
r
>>> those
>>> domains as happens today.
>>>
>>> The order resource should be changed to have a =E2=80=9CfinalizeURL=E2=
=80=9D field which
>>> provides the client a URL to POST a CSR to indicate that the CA should
>>> issue the
>>> certificate. Proactive issuance should be removed outright.
>>>
>>> This will address both the problem of scanning existing many-to-many
>>> orders
>>> & their authorizations as well as the db bloat that will occur from
>>> front-loading the delivery of the CSR.
>>>
>>> It may be tempting to leave proactive issuance as an optional feature
>>> but this
>>> will require maintaining the CSR in the new-order and I also believe
>>> that having
>>> two very distinct methods of issuance will complicate the client
>>> ecosystem and
>>> returns us to the question of how to determine which will occur that
>>> spawned the
>>> original "Proactive vs On-finalization Certificate Issuance"
>>> thread[0].
>>>
>>> [0] https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7y
>>> cDW5dj6GAY
>>> [1] https://community.letsencrypt.org/t/upcoming-api-changes/17947
>>> [2] https://community.letsencrypt.org/t/automatic-recycling-of-p
>>> ending-authorizations/41321
>>> [3] https://letsencrypt.org/docs/rate-limits/
>>>
>>>
>>> _______________________________________________
>>> Acme mailing list
>>> Acme@ietf.org
>>> https://www.ietf.org/mailman/listinfo/acme
>>>
>>>
>>
>

--089e0826371807b8830559c7614a
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><br>RE: #1 - this does help, but will require changing the=
 spec to clarify. The current wording does not support this behaviour and c=
leary indicates the server MUST issue for any order that has been satisfied=
.<br><br><div>RE: #2 - a CA relying on the new-authz flow to avoid the cost=
 of accepting a CSR up front would never be able to use the new-order flow =
to give the client authorizations they didn&#39;t ask for or they are back =
to accepting CSRs for orders that may never complete. That&#39;s why we nee=
d a new error to return to reject new-order requests that aren&#39;t fully =
pre-authorized. If the largest ACME CAs will not implement new-order as des=
cribed outside of pre-authorization why have it at all? Who will implement =
it? =C2=A0</div><div><br>The primary failing of #2 is that it causes those =
CAs to lose one substantial benefit to the new-order flow that I would like=
 to keep: allowing for policy-based decisions on the CA side to influence w=
hich authorizations are required. Without this property we will further hav=
e to update the specification for wildcard issuance. The current spec doesn=
&#39;t permit asking for authorizations for wildcard identifiers with new-a=
uthz so you&#39;d need some way of asking the server for a wildcard identif=
ier and have it respond with an authorization for something completely diff=
erent (An authorization for some wildcard resource say with a =C2=A0DNS-01 =
challenge for the base domain).</div><div><br>I agree that this is a late c=
hange but we are nearing the last chance we have to address what I see as a=
 fairly significant point of complication in the protocol.=C2=A0 I note tha=
t no one has proposed any reasons why proactive issuance must remain since =
this thread was restarted except that it is the status-quo and we want to m=
inimize change.=C2=A0<br><br>I am not in favour of the suggestions from Ric=
hard so far and would like to hear more about why my proposal is not suitab=
le. Richard&#39;s changes are smaller but leave protocol complication and p=
revent large CAs from enjoying the benefits of the new-order flow. My propo=
sal addresses the problems and leaves a simpler protocol without features t=
hat will not be used by large providers.<br><br>We have to change the speci=
fication no matter whether my proposals or Richards are taken so why not ma=
ke changes that will result in the clearest protocol with the least amount =
of cruft few are likely to implement?<br><br>- Daniel<br><br><br><br><br></=
div></div><div class=3D"gmail_extra"><br><div class=3D"gmail_quote">On Thu,=
 Sep 21, 2017 at 2:40 PM, Richard Barnes <span dir=3D"ltr">&lt;<a href=3D"m=
ailto:rlb@ipv.sx" target=3D"_blank">rlb@ipv.sx</a>&gt;</span> wrote:<br><bl=
ockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #=
ccc solid;padding-left:1ex"><div dir=3D"ltr"><div>I talked a bit with Danie=
l about this off-list. Here&#39;s a summary of where we got to:</div><div><=
br></div><div>1. In the current &quot;new-order&quot; flow, the server does=
n&#39;t *really* have to be proactive -- instead of updating an order objec=
t when an authz changes state, it can wait to update/issue until the client=
 polls for the order object, indicating that it&#39;s actually interested i=
n having a certificate.=C2=A0 That doesn&#39;t alleviate the pain of storin=
g the CSR, but it at least avoids the expense of finding affected orders an=
d issuing in cases where the client abandons the order.=C2=A0 If we think t=
his approach is OK, I would propose that we revise the document to make cle=
ar that it is OK.=C2=A0 What do folks think?=C2=A0 Daniel noted that there =
might be some issues with GET idempotency here, but I don&#39;t think this =
actually makes GET non-idempotent.<br></div><div><br></div><div>2. Some CAs=
 might want to support only the &quot;new-authz&quot; flow, i.e., never all=
ow a &quot;new-order&quot; request if the relevant authorizations haven&#39=
;t already been done.=C2=A0 It seems like this could be addressed in a pret=
ty straightforward manner just by defining a new error code (say &quot;prea=
uthRequired&quot;) that the server can return in this case.<br></div><div><=
br></div><div>Personally, it seems to me that if we make the clarification =
(1) and add the error code (2) described here, that would address the pain =
points Daniel raises, without a whole lot of new mechanism.=C2=A0 I&#39;m n=
ot really seeing much additional value to be had from an explicit &quot;ok =
issue now&quot; request (vs. the GET poll).</div><span class=3D"HOEnZb"><fo=
nt color=3D"#888888"><div><br></div><div>--Richard<br></div><div><br></div>=
<div><br></div></font></span></div><div class=3D"HOEnZb"><div class=3D"h5">=
<div class=3D"gmail_extra"><br><div class=3D"gmail_quote">On Wed, Sep 20, 2=
017 at 1:22 PM, Richard Barnes <span dir=3D"ltr">&lt;<a href=3D"mailto:rlb@=
ipv.sx" target=3D"_blank">rlb@ipv.sx</a>&gt;</span> wrote:<br><blockquote c=
lass=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;=
padding-left:1ex"><div dir=3D"ltr"><div>Hey Daniel,</div><div><br></div><di=
v>Thanks for the input.=C2=A0 I can see why you might be feeling some pain =
here, but I&#39;m not sure the proposed solution is quite right.</div><div>=
<br></div><div>Note that ACME already has the &quot;new-authorization&quot;=
 flow, which does pretty much exactly what you&#39;re proposing.=C2=A0 The =
only difference is that instead of authorizations being created all at once=
 in response to a &quot;new-cert&quot; request, they&#39;re created individ=
ually by the client.=C2=A0 But it&#39;s ultimately the same number of RTTs =
(authz fetch ~ new-authz) -- or rather, one fewer in the new-authz case bec=
ause there&#39;s no new-cert request.=C2=A0 To what degree would your probl=
em be solved simply by having the affected integrators move over to that fl=
ow?<br></div><div><br></div><div>I can sympathize that the new-order flow m=
ight not be suitable for all use cases.=C2=A0 Maybe it would be useful to w=
rite down some thoughts on which use cases are best addressed by which flow=
s.=C2=A0 <br></div><span class=3D"m_-6041448135189202427HOEnZb"><font color=
=3D"#888888"><div><br></div><div>--Richard<br></div><div><br></div><div><br=
></div></font></span><div class=3D"gmail_extra"><br><div class=3D"gmail_quo=
te"><div><div class=3D"m_-6041448135189202427h5">On Tue, Sep 19, 2017 at 12=
:26 PM, Daniel McCarney <span dir=3D"ltr">&lt;<a href=3D"mailto:cpu@letsenc=
rypt.org" target=3D"_blank">cpu@letsencrypt.org</a>&gt;</span> wrote:<br></=
div></div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;bord=
er-left:1px #ccc solid;padding-left:1ex"><div><div class=3D"m_-604144813518=
9202427h5"><div dir=3D"ltr"><div>Hi folks,</div><div><br></div><div>Just ov=
er a year ago in a thread titled &quot;Proactive vs On-Finalization</div><d=
iv>Certificate Issuance&quot;[0] we reached the consensus on the question o=
f whether to</div><div>issue a certificate &quot;on-finalization&quot; or &=
quot;proactively&quot; with the conclusion to</div><div>standardize on proa=
ctive issuance.</div><div><br></div><div>Implementation experience shows th=
at proactive issuance is slow in practice. An</div><div>order can be associ=
ated with many authorizations and an authorization can be</div><div>associa=
ted with many orders. This interacts poorly with optimizations Let&#39;s</d=
iv><div>Encrypt has developed from in-the-field experience with ACME and it=
 is difficult</div><div>to rate-limit effectively. I propose that we return=
 to the on-finalization style</div><div>approach and that we provide the CS=
R during finalization..</div><div><br></div><div>## Issues with proactive i=
ssuance</div><div><br></div><div>### Authorization Reuse</div><div><br></di=
v><div>Proactive issuance requires that when an authorization is switched t=
o a valid</div><div>state by the result of a challenge update that we find =
the associated orders and</div><div>iterate each order&#39;s authorizations=
 to see if all of the authorizations are</div><div>valid and issuance can o=
ccur.</div><div><br></div><div>Let&#39;s Encrypt reuses existing authorizat=
ions[1][2] whenever possible, as</div><div>encouraged by ACME. This has bee=
n of huge value in reducing resource consumption</div><div>(both in terms o=
f database storage and challenge/API requests). Many clients</div><div>crea=
te pending authorizations they do not finalize and similarly re-create</div=
><div>pending authorizations for identifiers that the account already has v=
alid</div><div>authorizations for. Unchecked this leads to significant data=
base growth and</div><div>resource consumption, even with aggressive rate-l=
imits[3].</div><div><br></div><div>The combination of authorization reuse a=
nd proactive issuance means a given</div><div>authorization can be associat=
ed with many orders. This necessitates an expensive</div><div>many-to-many =
query at each authorization update to see if any associated</div><div>order=
s are now complete.</div><div><br></div><div>One possible workaround is a r=
ate-limit that would constrain the number of</div><div>orders an authorizat=
ion can be associated with. However establishing the value</div><div>of the=
 limit to both reduce server-side load &amp; also effectively support Let&#=
39;s</div><div>Encrypt&#39;s large-volume integrators and their issuance pa=
tterns is challenging.</div><div><br></div><div>### CSR-first new-order flo=
w</div><div><br></div><div>To support proactive issuance the CA also needs =
to have the CSR for the order</div><div>at-hand when checking each authoriz=
ation in case issuance can proceed. The</div><div>`new-order` request that =
creates the order therefore carries the full CSR that</div><div>was previou=
sly delivered by the &quot;certificate finalize&quot; message.</div><div><b=
r></div><div>We see users create many more authorizations than they are abl=
e to finalize</div><div>successfully. With the legacy &quot;new-cert&quot; =
flow our authorization reuse</div><div>dramatically helped reduce the DB sp=
ace used up by authorizations that will</div><div>never lead to issuance be=
cause no CSR is stored until all identifiers are</div><div>validated.</div>=
<div><br></div><div>With the new-order flow we have to accept and store a f=
ull CSR before handing</div><div>back authorizations to the client. Reuse o=
f authorizations does not prevent the</div><div>DB bloat that will occur fr=
om users submitting new order requests and not</div><div>fulfilling them. T=
he CSR contains a full public key and is one of the single</div><div>larges=
t contributors to DB disk space usage. Addressing this will require</div><d=
iv>another strict rate limit on outstanding new-orders and is again difficu=
lt to</div><div>=E2=80=9Cdial-in=E2=80=9D in to both prevent excessive reso=
urce consumption and also maintain</div><div>the ability for big integrator=
s to process large-scale issuance using our API.</div><div><br></div><div>#=
 Proposed Change</div><div><br></div><div>I believe we should change the `n=
ew-order` endpoint to not accept a full CSR but</div><div>to instead accept=
 a list of identifiers that the ACME client wishes to</div><div>authenticat=
e &amp; issue for. Authorizations can be created and returned for those</di=
v><div>domains as happens today.</div><div><br></div><div>The order resourc=
e should be changed to have a =E2=80=9CfinalizeURL=E2=80=9D field which</di=
v><div>provides the client a URL to POST a CSR to indicate that the CA shou=
ld issue the</div><div>certificate. Proactive issuance should be removed ou=
tright.</div><div><br></div><div>This will address both the problem of scan=
ning existing many-to-many orders</div><div>&amp; their authorizations as w=
ell as the db bloat that will occur from</div><div>front-loading the delive=
ry of the CSR.</div><div><br></div><div>It may be tempting to leave proacti=
ve issuance as an optional feature but this</div><div>will require maintain=
ing the CSR in the new-order and I also believe that having</div><div>two v=
ery distinct methods of issuance will complicate the client ecosystem and</=
div><div>returns us to the question of how to determine which will occur th=
at spawned the</div><div>original &quot;Proactive vs On-finalization Certif=
icate Issuance&quot;</div><div>thread[0].</div><div><br></div><div>[0] <a h=
ref=3D"https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6G=
AY" target=3D"_blank">https://mailarchive.ietf.org/a<wbr>rch/msg/acme/0lVmG=
l8e-rmSH0x7y<wbr>cDW5dj6GAY</a></div><div>[1] <a href=3D"https://community.=
letsencrypt.org/t/upcoming-api-changes/17947" target=3D"_blank">https://com=
munity.letsencrypt.<wbr>org/t/upcoming-api-changes/179<wbr>47</a></div><div=
>[2] <a href=3D"https://community.letsencrypt.org/t/automatic-recycling-of-=
pending-authorizations/41321" target=3D"_blank">https://community.letsencry=
pt.<wbr>org/t/automatic-recycling-of-p<wbr>ending-authorizations/41321</a><=
/div><div>[3] <a href=3D"https://letsencrypt.org/docs/rate-limits/" target=
=3D"_blank">https://letsencrypt.org/docs/r<wbr>ate-limits/</a></div><div><b=
r></div></div>
<br></div></div><span>______________________________<wbr>_________________<=
br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org" target=3D"_blank">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/l<wbr>istinfo/acme</a><br>
<br></span></blockquote></div><br></div></div>
</blockquote></div><br></div>
</div></div></blockquote></div><br></div>

--089e0826371807b8830559c7614a--


From nobody Fri Sep 22 13:15:35 2017
Return-Path: <kathleen.moriarty.ietf@gmail.com>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2E182133038 for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 13:15:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
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 macbGDYwUZ9j for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 13:15:32 -0700 (PDT)
Received: from mail-pg0-x233.google.com (mail-pg0-x233.google.com [IPv6:2607:f8b0:400e:c05::233]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id ABF971323B8 for <acme@ietf.org>; Fri, 22 Sep 2017 13:15:32 -0700 (PDT)
Received: by mail-pg0-x233.google.com with SMTP id p5so1141163pgn.7 for <acme@ietf.org>; Fri, 22 Sep 2017 13:15:32 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;  h=mime-version:from:date:message-id:subject:to :content-transfer-encoding; bh=25Yi8j5rKYpMOzhzaXd5B6y8PP1r4jsh+9Eu88G0/Cw=; b=dVOZFRF8XVyvDzOIzSAHigwneCURFcLDKI+HizfYHrlUQ63BV57hVYdejSLgCGRt9y HgkvYRnMWq9+ne0lrw82K341AATdTJ4A2D+6Ztjr4FGtzTLKMuezeuZmKCEgWteGIwGe 9HapiH33XEvsAJd1wumubIWG5rUE+oGONos2+TDVepZb541CWTIDnNNpqiAq2XqhQOg3 kbZ+MhHlzrCpPA1T6hsS2UVnrB5G1GK0UYB/aDWCP82xb7xde7RL8QqWhS1rwiGL0PMQ 14Qaqk45UHHjImEDKs462Trx2CCwGSD6wz1PTBX+9GN9FYCXt2lXlKvSqdIahE5M1LRW oO/A==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:from:date:message-id:subject:to :content-transfer-encoding; bh=25Yi8j5rKYpMOzhzaXd5B6y8PP1r4jsh+9Eu88G0/Cw=; b=TIvGvFm7rEm49pAesBBayawar0y8WIJLJyNiXEVBbFpibXLOj89fpMulUjB3fTrFNG yfw5CX71tpHGvpSUtMuG+u1HdF1dVgVMPwOqJDjyL0dvkFMnKywSB+xe5LuqbsQNroIc MHESTkXSgY5pfTMEvT5lB3SEa0lVnZZ5nArbhMQlmU8zY+X/qoy8fCi2uqD+XbsFCGZL NzdzfQ1KPHAznUzh15+WpJyi6WCdN0ktgg31hdBniCnUjVAzTv45Dwj9HdTp/0ZF/0BK mn07LVRGxSIYbqBfMcNSNUy6dj4bzh/fPTTu62MIsEYaw/HuPY4R/alt1spHb04f7/ap I0Zw==
X-Gm-Message-State: AHPjjUgu6afe2sdxtErkcQWX3uLHRegkorPUawNCaBWRVrYezZhpXqlc zktC9L8XoymoesuDMk/SoXrWwZ+SmD8lzWLNLvr9LA==
X-Google-Smtp-Source: AOwi7QC7XwXzdcGLTdUeJXGjvca4ApdC96z8ZhT96590oR/KsFZI8rvkIcyGW8G854o9cKU8QFdr/BpdzLSVaZjltm8=
X-Received: by 10.98.102.82 with SMTP id a79mr269104pfc.109.1506111331786; Fri, 22 Sep 2017 13:15:31 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.100.144.1 with HTTP; Fri, 22 Sep 2017 13:14:51 -0700 (PDT)
From: Kathleen Moriarty <kathleen.moriarty.ietf@gmail.com>
Date: Fri, 22 Sep 2017 16:14:51 -0400
Message-ID: <CAHbuEH4ogM3NQNEUFW44JFjh3s5mgvBHuhExvei0F1Voe2T+xg@mail.gmail.com>
To: IETF ACME <acme@ietf.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/dTrh5koTxfCweCpIoDNr0uUEv0o>
Subject: [Acme] AD review of draft-ietf-acme-acme
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 22 Sep 2017 20:15:34 -0000

Hello,

Thank you to the editors and WG for your efforts on
draft-ietf-acme-acme, it's a well written and easy to understand
draft.  I do have a few comments, that need to be address by the
editors and SHEPHERD.

Please review the idnits.  There are a few warnings that should be
correctable and can be addressed, but more importantly, it calls out
several downrefs that are not in the shepherd writeup.  These need to
be in the writeup and then also mentioned in the IETF last call
announcement (I'll make sure the latter happens).

Here's my mostly editorial comments:

Introduction:
It reads very well, but it would be helpful to those unfamiliar with
the work to explicitly state that ACME is about DV certificates.  The
first sentence of the last paragraph seems like the best place to add
this in.
Current text:
   This document describes an extensible framework for automating the
   issuance and domain validation procedure, thereby allowing servers
   and infrastructural software to obtain certificates without user
   interaction.
Proposed:
   This document describes an extensible framework for automating the
   issuance and domain validation procedure for DV certificates,
thereby allowing servers
   and infrastructural software to obtain certificates without user
   interaction.

I do like how the introduction is framed as it=E2=80=99s clear the level of
security provided by the certificates issued via ACME.  Thanks for
that.

The introduction should also make it clear that ACME can be used for
other services using DV certificates, not just HTTP.  This is
mentioned in the Terminology section, but I think a clear sentence
upfront would be helpful.

Section 2: Nit/suggestion: Change the tense to read as a published RFC
in the last paragraph, last sentence (take it or leave it).
Current:
   Such close integration of ACME with HTTPS
   servers would allow the immediate and automated deployment of
   certificates as they are issued, sparing the human administrator from
   much of the time-consuming work described in the previous section.
Proposed:
   Such close integration of ACME with HTTPS
   servers allows the immediate and automated deployment of
   certificates as they are issued, sparing the human administrator from
   much of the time-consuming work described in the previous section.

IANA Section:
Everything looks good except that RFC5226 has been obsoleted, so the
new reference is RFC8126 and =E2=80=9Cspecification required=E2=80=9D
https://tools.ietf.org/html/rfc8126#section-4.6 still means the same
thing, so using that is fine.

Security considerations:
I think it would be good to mention here that they are DV certs so the
reader understands from he introduction and this section the level of
cert issued through ACME and doesn=E2=80=99t assume a higher level of
assurance.

s/ACME is a protocol for managing certificates/ACME is a protocol for
managing DV certificates/

10.1 - I know this is obvious to the people in the WG, but a reference
to RFC7525 to properly configure TLS 1.2 should be included to protect
against the mentioned attacks.  If you want to also reference TLS 1.3
and specify no 0-RTT that would be good too.  It=E2=80=99d be great to see
this get widely deployed, so there may be a number of newcomers
reading this to make their lives easier with cert
issuance/maintenance.

10.2 - what if the server that the account verifier has an account on
(client for ACME) is compromised and is used to request new
certificates or perform other actions?  I think this is one of the
larger risks, so mentioning this is possible and hardening measures
should be taken to prevent compromise would be prudent.  Hardening
measures or appropriate security controls should be broadly understood
terms (I think) so that you wouldn=E2=80=99t need to list out things like
turing off unnecessary services.

10.5 - I would think you=E2=80=99d see requests for new phishing domains
rather than known ones through this process.  What would you look for
there as that=E2=80=99s a complex problem - it would be hard to know all le=
git
names to know if one was a play off of the original.  This happened
with equifax - they provided a site name with a typo and a =E2=80=98white h=
at=E2=80=99
attacker had set up a site that looked just like theirs at that site.
But it was announced by equifax.  Tough problem as automating a check
on the DNS registration may not detect something unusual, but maybe
recommending this check could help?  RDP lessens what needs to be
shared from whois though, although I don=E2=80=99t think that=E2=80=99s wid=
ely
deployed.

--=20

Best regards,
Kathleen


From nobody Fri Sep 22 13:51:36 2017
Return-Path: <mary.ietf.barnes@gmail.com>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 80D99132F3F for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 13:51:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
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 6YfsgZylsIZn for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 13:51:30 -0700 (PDT)
Received: from mail-qk0-x22f.google.com (mail-qk0-x22f.google.com [IPv6:2607:f8b0:400d:c09::22f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 186131320D8 for <acme@ietf.org>; Fri, 22 Sep 2017 13:51:30 -0700 (PDT)
Received: by mail-qk0-x22f.google.com with SMTP id c69so2194586qke.8 for <acme@ietf.org>; Fri, 22 Sep 2017 13:51:30 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;  h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=/Jnw5C4tCrAYEtlr51P8WqpfCCtI2na0Q7jujG8aOyk=; b=uuyd3JVinsjTFN9zE4oH96udjX8W+mH1J/qmukKEFtGKCJYtZBfvUzscgCmWaIyTSS bKSmh1tDhm2EIAfNc+XJSf5KMtcCsaiPu/uZyPNYQ+Cr7jEvB+wrw3BbocdnX9Er8vm5 UDjUouwdTnZETtjyaJBWW/VbYjA3HQWYuNASX4D9NsF0VIp3BXhYc94f4uAAap0p0svo 6aIJNKZxDuQcLxcM3gGfNvlp/jhDZ5xE0u8x5SbzzJ+agHE3DH8gM6jS5PX9fvgCW+Nz cxrN3WiSPBa6aX6wy3RjMx1GfYSExlzXZ/f6646z7r1orKHcdPA0eVR72414jJUVMPsW Dctw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=/Jnw5C4tCrAYEtlr51P8WqpfCCtI2na0Q7jujG8aOyk=; b=MePEYpolS+n2vGxYKD+ZLyROnhxgcq+sCHykSP/S2G1WFIvRbOOGkWfMz6fpmVhpH+ YX83U3Ibhfo+uupBohqrlTmfFy+w8r7+85/gJmI5I7JU+rHCurj1xRL15vZLOn8N/lCd eMETUOqppFaL+qZiJmT9fMOZTcO/7+JpVVeWXWPBE9J7NWg2vHWi0yNXKT+R1BXM/uWn 6Dn0TubbsrTEetN8I8jV33QH3hkpOmYkQmaU0VV7D/fjF85E7z+tSdCr9y4L2Qw6vdXm 5Oen8VaclmVFpeZ8539vAfPO76MkTzS5u9tZ0oueS75In07vOozYbCnNd2sz5gSR5Fc9 0adw==
X-Gm-Message-State: AHPjjUjEIPeI1WPO/rUt9pj3mkR73TiwAnbovWjUlQYtyWr1Ba5H3Ydi O7G54xKlQOaKJrh7Y8l+MXdi/4o8asKqE7AYtXc=
X-Google-Smtp-Source: AOwi7QAakOE4C+avJvo8nMtoaD6qXIJKmgJyBpEknjKhpgNF62f52Dio2vBG5fBH4qB34hioeXxb/kIv60Y1lFO9Vec=
X-Received: by 10.55.47.6 with SMTP id v6mr594690qkh.181.1506113489210; Fri, 22 Sep 2017 13:51:29 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.140.38.16 with HTTP; Fri, 22 Sep 2017 13:51:28 -0700 (PDT)
In-Reply-To: <CAHbuEH4ogM3NQNEUFW44JFjh3s5mgvBHuhExvei0F1Voe2T+xg@mail.gmail.com>
References: <CAHbuEH4ogM3NQNEUFW44JFjh3s5mgvBHuhExvei0F1Voe2T+xg@mail.gmail.com>
From: Mary Barnes <mary.ietf.barnes@gmail.com>
Date: Fri, 22 Sep 2017 15:51:28 -0500
Message-ID: <CAHBDyN61y8KWmFY17YzXCTWfL4s8LTrxWvhHhbRZ32f-_uG8uA@mail.gmail.com>
To: Kathleen Moriarty <kathleen.moriarty.ietf@gmail.com>
Cc: IETF ACME <acme@ietf.org>
Content-Type: multipart/alternative; boundary="001a114f431684128f0559cd5fe1"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/h3YmoMcO4EPAH1a1sY9xW3MF2zM>
Subject: Re: [Acme] AD review of draft-ietf-acme-acme
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 22 Sep 2017 20:51:34 -0000

--001a114f431684128f0559cd5fe1
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Just to throw in my 1/2 cent.  We are using ACME for non-DV certificates in
the ATIS/SIP Forum SHAKEN framework as detailed in ATIS-1000080:
https://www.sipforum.org/download/joint-atissip-forum-standard-signature-ba=
sed-handling-asserted-information-using-tokens-shaken-governance-model-cert=
ificate-management-atis-1000080/?wpdmdl=3D3360

Regards,
Mary.

On Fri, Sep 22, 2017 at 3:14 PM, Kathleen Moriarty <
kathleen.moriarty.ietf@gmail.com> wrote:

> Hello,
>
> Thank you to the editors and WG for your efforts on
> draft-ietf-acme-acme, it's a well written and easy to understand
> draft.  I do have a few comments, that need to be address by the
> editors and SHEPHERD.
>
> Please review the idnits.  There are a few warnings that should be
> correctable and can be addressed, but more importantly, it calls out
> several downrefs that are not in the shepherd writeup.  These need to
> be in the writeup and then also mentioned in the IETF last call
> announcement (I'll make sure the latter happens).
>
> Here's my mostly editorial comments:
>
> Introduction:
> It reads very well, but it would be helpful to those unfamiliar with
> the work to explicitly state that ACME is about DV certificates.  The
> first sentence of the last paragraph seems like the best place to add
> this in.
> Current text:
>    This document describes an extensible framework for automating the
>    issuance and domain validation procedure, thereby allowing servers
>    and infrastructural software to obtain certificates without user
>    interaction.
> Proposed:
>    This document describes an extensible framework for automating the
>    issuance and domain validation procedure for DV certificates,
> thereby allowing servers
>    and infrastructural software to obtain certificates without user
>    interaction.
>
> I do like how the introduction is framed as it=E2=80=99s clear the level =
of
> security provided by the certificates issued via ACME.  Thanks for
> that.
>
> The introduction should also make it clear that ACME can be used for
> other services using DV certificates, not just HTTP.  This is
> mentioned in the Terminology section, but I think a clear sentence
> upfront would be helpful.
>
> Section 2: Nit/suggestion: Change the tense to read as a published RFC
> in the last paragraph, last sentence (take it or leave it).
> Current:
>    Such close integration of ACME with HTTPS
>    servers would allow the immediate and automated deployment of
>    certificates as they are issued, sparing the human administrator from
>    much of the time-consuming work described in the previous section.
> Proposed:
>    Such close integration of ACME with HTTPS
>    servers allows the immediate and automated deployment of
>    certificates as they are issued, sparing the human administrator from
>    much of the time-consuming work described in the previous section.
>
> IANA Section:
> Everything looks good except that RFC5226 has been obsoleted, so the
> new reference is RFC8126 and =E2=80=9Cspecification required=E2=80=9D
> https://tools.ietf.org/html/rfc8126#section-4.6 still means the same
> thing, so using that is fine.
>
> Security considerations:
> I think it would be good to mention here that they are DV certs so the
> reader understands from he introduction and this section the level of
> cert issued through ACME and doesn=E2=80=99t assume a higher level of
> assurance.
>
> s/ACME is a protocol for managing certificates/ACME is a protocol for
> managing DV certificates/
>
> 10.1 - I know this is obvious to the people in the WG, but a reference
> to RFC7525 to properly configure TLS 1.2 should be included to protect
> against the mentioned attacks.  If you want to also reference TLS 1.3
> and specify no 0-RTT that would be good too.  It=E2=80=99d be great to se=
e
> this get widely deployed, so there may be a number of newcomers
> reading this to make their lives easier with cert
> issuance/maintenance.
>
> 10.2 - what if the server that the account verifier has an account on
> (client for ACME) is compromised and is used to request new
> certificates or perform other actions?  I think this is one of the
> larger risks, so mentioning this is possible and hardening measures
> should be taken to prevent compromise would be prudent.  Hardening
> measures or appropriate security controls should be broadly understood
> terms (I think) so that you wouldn=E2=80=99t need to list out things like
> turing off unnecessary services.
>
> 10.5 - I would think you=E2=80=99d see requests for new phishing domains
> rather than known ones through this process.  What would you look for
> there as that=E2=80=99s a complex problem - it would be hard to know all =
legit
> names to know if one was a play off of the original.  This happened
> with equifax - they provided a site name with a typo and a =E2=80=98white=
 hat=E2=80=99
> attacker had set up a site that looked just like theirs at that site.
> But it was announced by equifax.  Tough problem as automating a check
> on the DNS registration may not detect something unusual, but maybe
> recommending this check could help?  RDP lessens what needs to be
> shared from whois though, although I don=E2=80=99t think that=E2=80=99s w=
idely
> deployed.
>
> --
>
> Best regards,
> Kathleen
>
> _______________________________________________
> Acme mailing list
> Acme@ietf.org
> https://www.ietf.org/mailman/listinfo/acme
>

--001a114f431684128f0559cd5fe1
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Just to throw in my 1/2 cent.=C2=A0 We are using ACME for =
non-DV certificates in the ATIS/SIP Forum SHAKEN framework as detailed in A=
TIS-1000080: =C2=A0<a href=3D"https://www.sipforum.org/download/joint-atiss=
ip-forum-standard-signature-based-handling-asserted-information-using-token=
s-shaken-governance-model-certificate-management-atis-1000080/?wpdmdl=3D336=
0">https://www.sipforum.org/download/joint-atissip-forum-standard-signature=
-based-handling-asserted-information-using-tokens-shaken-governance-model-c=
ertificate-management-atis-1000080/?wpdmdl=3D3360</a><div><br><div><div><di=
v>Regards,<br></div><div>Mary.</div></div></div></div></div><div class=3D"g=
mail_extra"><br><div class=3D"gmail_quote">On Fri, Sep 22, 2017 at 3:14 PM,=
 Kathleen Moriarty <span dir=3D"ltr">&lt;<a href=3D"mailto:kathleen.moriart=
y.ietf@gmail.com" target=3D"_blank">kathleen.moriarty.ietf@gmail.com</a>&gt=
;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 =
.8ex;border-left:1px #ccc solid;padding-left:1ex">Hello,<br>
<br>
Thank you to the editors and WG for your efforts on<br>
draft-ietf-acme-acme, it&#39;s a well written and easy to understand<br>
draft.=C2=A0 I do have a few comments, that need to be address by the<br>
editors and SHEPHERD.<br>
<br>
Please review the idnits.=C2=A0 There are a few warnings that should be<br>
correctable and can be addressed, but more importantly, it calls out<br>
several downrefs that are not in the shepherd writeup.=C2=A0 These need to<=
br>
be in the writeup and then also mentioned in the IETF last call<br>
announcement (I&#39;ll make sure the latter happens).<br>
<br>
Here&#39;s my mostly editorial comments:<br>
<br>
Introduction:<br>
It reads very well, but it would be helpful to those unfamiliar with<br>
the work to explicitly state that ACME is about DV certificates.=C2=A0 The<=
br>
first sentence of the last paragraph seems like the best place to add<br>
this in.<br>
Current text:<br>
=C2=A0 =C2=A0This document describes an extensible framework for automating=
 the<br>
=C2=A0 =C2=A0issuance and domain validation procedure, thereby allowing ser=
vers<br>
=C2=A0 =C2=A0and infrastructural software to obtain certificates without us=
er<br>
=C2=A0 =C2=A0interaction.<br>
Proposed:<br>
=C2=A0 =C2=A0This document describes an extensible framework for automating=
 the<br>
=C2=A0 =C2=A0issuance and domain validation procedure for DV certificates,<=
br>
thereby allowing servers<br>
=C2=A0 =C2=A0and infrastructural software to obtain certificates without us=
er<br>
=C2=A0 =C2=A0interaction.<br>
<br>
I do like how the introduction is framed as it=E2=80=99s clear the level of=
<br>
security provided by the certificates issued via ACME.=C2=A0 Thanks for<br>
that.<br>
<br>
The introduction should also make it clear that ACME can be used for<br>
other services using DV certificates, not just HTTP.=C2=A0 This is<br>
mentioned in the Terminology section, but I think a clear sentence<br>
upfront would be helpful.<br>
<br>
Section 2: Nit/suggestion: Change the tense to read as a published RFC<br>
in the last paragraph, last sentence (take it or leave it).<br>
Current:<br>
=C2=A0 =C2=A0Such close integration of ACME with HTTPS<br>
=C2=A0 =C2=A0servers would allow the immediate and automated deployment of<=
br>
=C2=A0 =C2=A0certificates as they are issued, sparing the human administrat=
or from<br>
=C2=A0 =C2=A0much of the time-consuming work described in the previous sect=
ion.<br>
Proposed:<br>
=C2=A0 =C2=A0Such close integration of ACME with HTTPS<br>
=C2=A0 =C2=A0servers allows the immediate and automated deployment of<br>
=C2=A0 =C2=A0certificates as they are issued, sparing the human administrat=
or from<br>
=C2=A0 =C2=A0much of the time-consuming work described in the previous sect=
ion.<br>
<br>
IANA Section:<br>
Everything looks good except that RFC5226 has been obsoleted, so the<br>
new reference is RFC8126 and =E2=80=9Cspecification required=E2=80=9D<br>
<a href=3D"https://tools.ietf.org/html/rfc8126#section-4.6" rel=3D"noreferr=
er" target=3D"_blank">https://tools.ietf.org/html/<wbr>rfc8126#section-4.6<=
/a> still means the same<br>
thing, so using that is fine.<br>
<br>
Security considerations:<br>
I think it would be good to mention here that they are DV certs so the<br>
reader understands from he introduction and this section the level of<br>
cert issued through ACME and doesn=E2=80=99t assume a higher level of<br>
assurance.<br>
<br>
s/ACME is a protocol for managing certificates/ACME is a protocol for<br>
managing DV certificates/<br>
<br>
10.1 - I know this is obvious to the people in the WG, but a reference<br>
to RFC7525 to properly configure TLS 1.2 should be included to protect<br>
against the mentioned attacks.=C2=A0 If you want to also reference TLS 1.3<=
br>
and specify no 0-RTT that would be good too.=C2=A0 It=E2=80=99d be great to=
 see<br>
this get widely deployed, so there may be a number of newcomers<br>
reading this to make their lives easier with cert<br>
issuance/maintenance.<br>
<br>
10.2 - what if the server that the account verifier has an account on<br>
(client for ACME) is compromised and is used to request new<br>
certificates or perform other actions?=C2=A0 I think this is one of the<br>
larger risks, so mentioning this is possible and hardening measures<br>
should be taken to prevent compromise would be prudent.=C2=A0 Hardening<br>
measures or appropriate security controls should be broadly understood<br>
terms (I think) so that you wouldn=E2=80=99t need to list out things like<b=
r>
turing off unnecessary services.<br>
<br>
10.5 - I would think you=E2=80=99d see requests for new phishing domains<br=
>
rather than known ones through this process.=C2=A0 What would you look for<=
br>
there as that=E2=80=99s a complex problem - it would be hard to know all le=
git<br>
names to know if one was a play off of the original.=C2=A0 This happened<br=
>
with equifax - they provided a site name with a typo and a =E2=80=98white h=
at=E2=80=99<br>
attacker had set up a site that looked just like theirs at that site.<br>
But it was announced by equifax.=C2=A0 Tough problem as automating a check<=
br>
on the DNS registration may not detect something unusual, but maybe<br>
recommending this check could help?=C2=A0 RDP lessens what needs to be<br>
shared from whois though, although I don=E2=80=99t think that=E2=80=99s wid=
ely<br>
deployed.<br>
<span class=3D"HOEnZb"><font color=3D"#888888"><br>
--<br>
<br>
Best regards,<br>
Kathleen<br>
<br>
______________________________<wbr>_________________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/<wbr>listinfo/acme</a><br>
</font></span></blockquote></div><br></div>

--001a114f431684128f0559cd5fe1--


From nobody Fri Sep 22 14:33:58 2017
Return-Path: <clint.t.wilson@gmail.com>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 69D08133053 for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 14:33:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.998
X-Spam-Level: 
X-Spam-Status: No, score=-1.998 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, URIBL_RED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
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 v7wStxvd411X for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 14:33:52 -0700 (PDT)
Received: from mail-qt0-x235.google.com (mail-qt0-x235.google.com [IPv6:2607:f8b0:400d:c0d::235]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 42A88132143 for <acme@ietf.org>; Fri, 22 Sep 2017 14:33:52 -0700 (PDT)
Received: by mail-qt0-x235.google.com with SMTP id o3so2370858qte.6 for <acme@ietf.org>; Fri, 22 Sep 2017 14:33:52 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;  h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=+1Cy9HuU1DUwkqJmmNm+59G2lK9jK9g/+m8WF7d70uo=; b=uRCwb8RQm9lNTRnLTg50GJeukA97UKzJSLryRX2Q3tjHS5hSRIqKz84M+Ff3K6aD+3 BFhF/vUbCWDkqFti9Fc7dHeYevaKR2hcrZId0fp5/phkt0gpjEMqHo1zHpv9Gh3qanOB fDypgTB5PoaTXJrm2ZhtfIZ0jwpVk+4WK5mYp/ZdoIAU6EKdTbs/NvvBFrDVq11jF2Js PwuKka8ReyQW/8fNOeCPHeHz53Iqvripi81RE4xJCrRJ7fQ5/72IOK3qNTj9DWBc451h hpjbbVyiOA6UiHQZ4+9Lp3UW59UjZwwyyL/rEmMqscXTew03zxpTfsiD05ibSlegHkgE bJHw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=+1Cy9HuU1DUwkqJmmNm+59G2lK9jK9g/+m8WF7d70uo=; b=IJvRW3qkvyVVETfBBWcqBP7bOulVstEBBcPAfDnYkU/ievsEdr1FvgNL3EzMStRZFJ 3OwxggS5d8hildzK7h7GBeQJ2fErxJiNsP62Gdi8vq83c+3m1n5pjTnRHKekl7i2yeKT thlFCeG6E2KqTgMG3WJyoRFhWcgY08M83C+2OjKr5Yn0YaQCAqeam4xxPtJsNbX25hYR jkixgjJrbS1I55wxTqhX4Ib7Vfxp81tDwOJpxoqK+J4520aUGQ/VtsmNSLS65YPgx5Vn zSbm2n81OriHwYwuitu9PGPmdan+xfDohg17WzMUpQQ0TEEDXpT6CtCWFFasKvHEeje+ AOtw==
X-Gm-Message-State: AHPjjUioYMW6byhbhpN2pyS1gtFgrEdw0UrOrWuMkMNh46nB+7T5I0dG nYwvOPoqKYVPGPKK4IPursv9HHoUGqPL81h5yZs=
X-Google-Smtp-Source: AOwi7QBnoCjpmdA8E11wCVjgnEInho9IGjEDgg4RhaadL12t1uzZo+nFhUI1WSBpojPrvqzlKOKgDNofFQJSt68qI4U=
X-Received: by 10.200.17.131 with SMTP id d3mr766673qtj.329.1506116031161; Fri, 22 Sep 2017 14:33:51 -0700 (PDT)
MIME-Version: 1.0
References: <CAHbuEH4ogM3NQNEUFW44JFjh3s5mgvBHuhExvei0F1Voe2T+xg@mail.gmail.com> <CAHBDyN61y8KWmFY17YzXCTWfL4s8LTrxWvhHhbRZ32f-_uG8uA@mail.gmail.com>
In-Reply-To: <CAHBDyN61y8KWmFY17YzXCTWfL4s8LTrxWvhHhbRZ32f-_uG8uA@mail.gmail.com>
From: Clint Wilson <clint.t.wilson@gmail.com>
Date: Fri, 22 Sep 2017 21:33:39 +0000
Message-ID: <CAJ=cBg1eh-27KzX8=zS+07iQr2E4m2-FaN=hh0Bi9U8SMeL8pA@mail.gmail.com>
To: Mary Barnes <mary.ietf.barnes@gmail.com>,  Kathleen Moriarty <kathleen.moriarty.ietf@gmail.com>
Cc: IETF ACME <acme@ietf.org>
Content-Type: multipart/alternative; boundary="089e082654640729ef0559cdf7a5"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/3-0UeRw1gZqjhuwV6jifGPkyIzQ>
Subject: Re: [Acme] AD review of draft-ietf-acme-acme
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 22 Sep 2017 21:33:55 -0000

--089e082654640729ef0559cdf7a5
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

An additional 1/2 cent from me; we intend to use ACME primarily, if not
solely, with non-DV certs.

On Fri, Sep 22, 2017, 2:51 PM Mary Barnes <mary.ietf.barnes@gmail.com>
wrote:

> Just to throw in my 1/2 cent.  We are using ACME for non-DV certificates
> in the ATIS/SIP Forum SHAKEN framework as detailed in ATIS-1000080:
> https://www.sipforum.org/download/joint-atissip-forum-standard-signature-=
based-handling-asserted-information-using-tokens-shaken-governance-model-ce=
rtificate-management-atis-1000080/?wpdmdl=3D3360
>
> Regards,
> Mary.
>
> On Fri, Sep 22, 2017 at 3:14 PM, Kathleen Moriarty <
> kathleen.moriarty.ietf@gmail.com> wrote:
>
>> Hello,
>>
>> Thank you to the editors and WG for your efforts on
>> draft-ietf-acme-acme, it's a well written and easy to understand
>> draft.  I do have a few comments, that need to be address by the
>> editors and SHEPHERD.
>>
>> Please review the idnits.  There are a few warnings that should be
>> correctable and can be addressed, but more importantly, it calls out
>> several downrefs that are not in the shepherd writeup.  These need to
>> be in the writeup and then also mentioned in the IETF last call
>> announcement (I'll make sure the latter happens).
>>
>> Here's my mostly editorial comments:
>>
>> Introduction:
>> It reads very well, but it would be helpful to those unfamiliar with
>> the work to explicitly state that ACME is about DV certificates.  The
>> first sentence of the last paragraph seems like the best place to add
>> this in.
>> Current text:
>>    This document describes an extensible framework for automating the
>>    issuance and domain validation procedure, thereby allowing servers
>>    and infrastructural software to obtain certificates without user
>>    interaction.
>> Proposed:
>>    This document describes an extensible framework for automating the
>>    issuance and domain validation procedure for DV certificates,
>> thereby allowing servers
>>    and infrastructural software to obtain certificates without user
>>    interaction.
>>
>> I do like how the introduction is framed as it=E2=80=99s clear the level=
 of
>> security provided by the certificates issued via ACME.  Thanks for
>> that.
>>
>> The introduction should also make it clear that ACME can be used for
>> other services using DV certificates, not just HTTP.  This is
>> mentioned in the Terminology section, but I think a clear sentence
>> upfront would be helpful.
>>
>> Section 2: Nit/suggestion: Change the tense to read as a published RFC
>> in the last paragraph, last sentence (take it or leave it).
>> Current:
>>    Such close integration of ACME with HTTPS
>>    servers would allow the immediate and automated deployment of
>>    certificates as they are issued, sparing the human administrator from
>>    much of the time-consuming work described in the previous section.
>> Proposed:
>>    Such close integration of ACME with HTTPS
>>    servers allows the immediate and automated deployment of
>>    certificates as they are issued, sparing the human administrator from
>>    much of the time-consuming work described in the previous section.
>>
>> IANA Section:
>> Everything looks good except that RFC5226 has been obsoleted, so the
>> new reference is RFC8126 and =E2=80=9Cspecification required=E2=80=9D
>> https://tools.ietf.org/html/rfc8126#section-4.6 still means the same
>> thing, so using that is fine.
>>
>> Security considerations:
>> I think it would be good to mention here that they are DV certs so the
>> reader understands from he introduction and this section the level of
>> cert issued through ACME and doesn=E2=80=99t assume a higher level of
>> assurance.
>>
>> s/ACME is a protocol for managing certificates/ACME is a protocol for
>> managing DV certificates/
>>
>> 10.1 - I know this is obvious to the people in the WG, but a reference
>> to RFC7525 to properly configure TLS 1.2 should be included to protect
>> against the mentioned attacks.  If you want to also reference TLS 1.3
>> and specify no 0-RTT that would be good too.  It=E2=80=99d be great to s=
ee
>> this get widely deployed, so there may be a number of newcomers
>> reading this to make their lives easier with cert
>> issuance/maintenance.
>>
>> 10.2 - what if the server that the account verifier has an account on
>> (client for ACME) is compromised and is used to request new
>> certificates or perform other actions?  I think this is one of the
>> larger risks, so mentioning this is possible and hardening measures
>> should be taken to prevent compromise would be prudent.  Hardening
>> measures or appropriate security controls should be broadly understood
>> terms (I think) so that you wouldn=E2=80=99t need to list out things lik=
e
>> turing off unnecessary services.
>>
>> 10.5 - I would think you=E2=80=99d see requests for new phishing domains
>> rather than known ones through this process.  What would you look for
>> there as that=E2=80=99s a complex problem - it would be hard to know all=
 legit
>> names to know if one was a play off of the original.  This happened
>> with equifax - they provided a site name with a typo and a =E2=80=98whit=
e hat=E2=80=99
>> attacker had set up a site that looked just like theirs at that site.
>> But it was announced by equifax.  Tough problem as automating a check
>> on the DNS registration may not detect something unusual, but maybe
>> recommending this check could help?  RDP lessens what needs to be
>> shared from whois though, although I don=E2=80=99t think that=E2=80=99s =
widely
>> deployed.
>>
>> --
>>
>> Best regards,
>> Kathleen
>>
>> _______________________________________________
>> Acme mailing list
>> Acme@ietf.org
>> https://www.ietf.org/mailman/listinfo/acme
>>
>
> _______________________________________________
> Acme mailing list
> Acme@ietf.org
> https://www.ietf.org/mailman/listinfo/acme
>

--089e082654640729ef0559cdf7a5
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<p dir=3D"ltr">An additional 1/2 cent from me; we intend to use ACME primar=
ily, if not solely, with non-DV certs.</p>
<br><div class=3D"gmail_quote"><div dir=3D"ltr">On Fri, Sep 22, 2017, 2:51 =
PM Mary Barnes &lt;<a href=3D"mailto:mary.ietf.barnes@gmail.com">mary.ietf.=
barnes@gmail.com</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" =
style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><di=
v dir=3D"ltr">Just to throw in my 1/2 cent.=C2=A0 We are using ACME for non=
-DV certificates in the ATIS/SIP Forum SHAKEN framework as detailed in ATIS=
-1000080: =C2=A0<a href=3D"https://www.sipforum.org/download/joint-atissip-=
forum-standard-signature-based-handling-asserted-information-using-tokens-s=
haken-governance-model-certificate-management-atis-1000080/?wpdmdl=3D3360" =
target=3D"_blank">https://www.sipforum.org/download/joint-atissip-forum-sta=
ndard-signature-based-handling-asserted-information-using-tokens-shaken-gov=
ernance-model-certificate-management-atis-1000080/?wpdmdl=3D3360</a><div><b=
r><div><div><div>Regards,<br></div><div>Mary.</div></div></div></div></div>=
<div class=3D"gmail_extra"><br><div class=3D"gmail_quote">On Fri, Sep 22, 2=
017 at 3:14 PM, Kathleen Moriarty <span dir=3D"ltr">&lt;<a href=3D"mailto:k=
athleen.moriarty.ietf@gmail.com" target=3D"_blank">kathleen.moriarty.ietf@g=
mail.com</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=
=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hello,<b=
r>
<br>
Thank you to the editors and WG for your efforts on<br>
draft-ietf-acme-acme, it&#39;s a well written and easy to understand<br>
draft.=C2=A0 I do have a few comments, that need to be address by the<br>
editors and SHEPHERD.<br>
<br>
Please review the idnits.=C2=A0 There are a few warnings that should be<br>
correctable and can be addressed, but more importantly, it calls out<br>
several downrefs that are not in the shepherd writeup.=C2=A0 These need to<=
br>
be in the writeup and then also mentioned in the IETF last call<br>
announcement (I&#39;ll make sure the latter happens).<br>
<br>
Here&#39;s my mostly editorial comments:<br>
<br>
Introduction:<br>
It reads very well, but it would be helpful to those unfamiliar with<br>
the work to explicitly state that ACME is about DV certificates.=C2=A0 The<=
br>
first sentence of the last paragraph seems like the best place to add<br>
this in.<br>
Current text:<br>
=C2=A0 =C2=A0This document describes an extensible framework for automating=
 the<br>
=C2=A0 =C2=A0issuance and domain validation procedure, thereby allowing ser=
vers<br>
=C2=A0 =C2=A0and infrastructural software to obtain certificates without us=
er<br>
=C2=A0 =C2=A0interaction.<br>
Proposed:<br>
=C2=A0 =C2=A0This document describes an extensible framework for automating=
 the<br>
=C2=A0 =C2=A0issuance and domain validation procedure for DV certificates,<=
br>
thereby allowing servers<br>
=C2=A0 =C2=A0and infrastructural software to obtain certificates without us=
er<br>
=C2=A0 =C2=A0interaction.<br>
<br>
I do like how the introduction is framed as it=E2=80=99s clear the level of=
<br>
security provided by the certificates issued via ACME.=C2=A0 Thanks for<br>
that.<br>
<br>
The introduction should also make it clear that ACME can be used for<br>
other services using DV certificates, not just HTTP.=C2=A0 This is<br>
mentioned in the Terminology section, but I think a clear sentence<br>
upfront would be helpful.<br>
<br>
Section 2: Nit/suggestion: Change the tense to read as a published RFC<br>
in the last paragraph, last sentence (take it or leave it).<br>
Current:<br>
=C2=A0 =C2=A0Such close integration of ACME with HTTPS<br>
=C2=A0 =C2=A0servers would allow the immediate and automated deployment of<=
br>
=C2=A0 =C2=A0certificates as they are issued, sparing the human administrat=
or from<br>
=C2=A0 =C2=A0much of the time-consuming work described in the previous sect=
ion.<br>
Proposed:<br>
=C2=A0 =C2=A0Such close integration of ACME with HTTPS<br>
=C2=A0 =C2=A0servers allows the immediate and automated deployment of<br>
=C2=A0 =C2=A0certificates as they are issued, sparing the human administrat=
or from<br>
=C2=A0 =C2=A0much of the time-consuming work described in the previous sect=
ion.<br>
<br>
IANA Section:<br>
Everything looks good except that RFC5226 has been obsoleted, so the<br>
new reference is RFC8126 and =E2=80=9Cspecification required=E2=80=9D<br>
<a href=3D"https://tools.ietf.org/html/rfc8126#section-4.6" rel=3D"noreferr=
er" target=3D"_blank">https://tools.ietf.org/html/rfc8126#section-4.6</a> s=
till means the same<br>
thing, so using that is fine.<br>
<br>
Security considerations:<br>
I think it would be good to mention here that they are DV certs so the<br>
reader understands from he introduction and this section the level of<br>
cert issued through ACME and doesn=E2=80=99t assume a higher level of<br>
assurance.<br>
<br>
s/ACME is a protocol for managing certificates/ACME is a protocol for<br>
managing DV certificates/<br>
<br>
10.1 - I know this is obvious to the people in the WG, but a reference<br>
to RFC7525 to properly configure TLS 1.2 should be included to protect<br>
against the mentioned attacks.=C2=A0 If you want to also reference TLS 1.3<=
br>
and specify no 0-RTT that would be good too.=C2=A0 It=E2=80=99d be great to=
 see<br>
this get widely deployed, so there may be a number of newcomers<br>
reading this to make their lives easier with cert<br>
issuance/maintenance.<br>
<br>
10.2 - what if the server that the account verifier has an account on<br>
(client for ACME) is compromised and is used to request new<br>
certificates or perform other actions?=C2=A0 I think this is one of the<br>
larger risks, so mentioning this is possible and hardening measures<br>
should be taken to prevent compromise would be prudent.=C2=A0 Hardening<br>
measures or appropriate security controls should be broadly understood<br>
terms (I think) so that you wouldn=E2=80=99t need to list out things like<b=
r>
turing off unnecessary services.<br>
<br>
10.5 - I would think you=E2=80=99d see requests for new phishing domains<br=
>
rather than known ones through this process.=C2=A0 What would you look for<=
br>
there as that=E2=80=99s a complex problem - it would be hard to know all le=
git<br>
names to know if one was a play off of the original.=C2=A0 This happened<br=
>
with equifax - they provided a site name with a typo and a =E2=80=98white h=
at=E2=80=99<br>
attacker had set up a site that looked just like theirs at that site.<br>
But it was announced by equifax.=C2=A0 Tough problem as automating a check<=
br>
on the DNS registration may not detect something unusual, but maybe<br>
recommending this check could help?=C2=A0 RDP lessens what needs to be<br>
shared from whois though, although I don=E2=80=99t think that=E2=80=99s wid=
ely<br>
deployed.<br>
<span class=3D"m_-4806915521565577366HOEnZb"><font color=3D"#888888"><br>
--<br>
<br>
Best regards,<br>
Kathleen<br>
<br>
_______________________________________________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org" target=3D"_blank">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/listinfo/acme</a><br>
</font></span></blockquote></div><br></div>
_______________________________________________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org" target=3D"_blank">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/listinfo/acme</a><br>
</blockquote></div>

--089e082654640729ef0559cdf7a5--


From nobody Fri Sep 22 15:03:49 2017
Return-Path: <kathleen.moriarty.ietf@gmail.com>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CA344133221 for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 15:03:47 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.697
X-Spam-Level: 
X-Spam-Status: No, score=-2.697 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, MIME_QP_LONG_LINE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001, URIBL_RED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
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 mh8sW6QUuV1N for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 15:03:44 -0700 (PDT)
Received: from mail-qk0-x232.google.com (mail-qk0-x232.google.com [IPv6:2607:f8b0:400d:c09::232]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 014D2133090 for <acme@ietf.org>; Fri, 22 Sep 2017 15:03:42 -0700 (PDT)
Received: by mail-qk0-x232.google.com with SMTP id u67so2324678qkg.6 for <acme@ietf.org>; Fri, 22 Sep 2017 15:03:42 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;  h=mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=IJg7JhVN7lGSIjUgey0+4gA7XA5vb5s3rt4tVFVyrgY=; b=BX7G5EL+2zLHGWri8TOWBfVsn5z8Up739RRo7TPRnLVBYGVe4FSUswgVrdyYcLYapZ c2yy/7qxcRi5Lno6EiGyDR+oefGJPo54FFJ/iuU4b7b3FUuyhgqHg6h1jT5AEjCzJQTM JJeiEcRBSzxTNPVBlm6GOdLKb1htzZsWvCVgzESKgnc5Mek0x/O7E7xpQRYidf0VYdtR eooATYmjfHP1TiBJmdmzIAIRrETje9TMtJkFwRXIEQq/IMXRvLmrPE5IklACAqbR6bWi WnZsIaRlcNlSC9igDaiR50HM+oaGXiljSWHqC+3yeZ7sWPgCIX9L56KUFFEPaxiwTqbX 82Ug==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=IJg7JhVN7lGSIjUgey0+4gA7XA5vb5s3rt4tVFVyrgY=; b=P/57Khy2738Ls9bLVyedUNd4VFvRFJGi20rsSpSEoYe3XYnbTNdcKBsq9MpKgHTrDL qBcq1dPycQ/s/xwjOd6nMQnsmZWp+XowzuTtCtfyVrC/A8zcQvg2LtLSWsOZspgCKGaO H1m2EuAS3NVbgymeLIUpOeLGG4b1FN6Jh1Tebfe688jBzHaDn6cVl5+FRNZE0JUMb5H2 2E/K1V6mF1z8JphSft/uOR5VddqkBRAkPcVdbxV5d21/twxgjIp/YQ5eq5f1xXU/brcc qYnaG+NBJX6JImGX2jZ9lG2P+FrrQQw267h53hCL8/dPeSWYTy4r5MoKYY89M42TGbsw cj3g==
X-Gm-Message-State: AHPjjUiIuzIBVjjYWRFnNj8IDU/A1kGx0QubwtgS2TY08TbwX6B/OGsw y83Qy4uKsLVUwE/pUcvFHaqzlO5K
X-Google-Smtp-Source: AOwi7QBppkziAOrvhgdJC5W3k6W2lArSRE4ClVTrZyVx9q5tPSd4afiaF7Rvtd8jxZO7pjXkkxlgqA==
X-Received: by 10.55.203.210 with SMTP id u79mr762009qkl.189.1506117821903; Fri, 22 Sep 2017 15:03:41 -0700 (PDT)
Received: from [192.168.1.6] (209-6-124-204.s3530.c3-0.arl-ubr1.sbo-arl.ma.cable.rcncustomer.com. [209.6.124.204]) by smtp.gmail.com with ESMTPSA id f48sm668703qtc.49.2017.09.22.15.03.41 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Fri, 22 Sep 2017 15:03:41 -0700 (PDT)
Content-Type: multipart/alternative; boundary=Apple-Mail-4C7D1F7E-0214-4415-AECC-B27C8281F3B0
Mime-Version: 1.0 (1.0)
From: Kathleen Moriarty <kathleen.moriarty.ietf@gmail.com>
X-Mailer: iPhone Mail (14F89)
In-Reply-To: <CAJ=cBg1eh-27KzX8=zS+07iQr2E4m2-FaN=hh0Bi9U8SMeL8pA@mail.gmail.com>
Date: Fri, 22 Sep 2017 18:03:40 -0400
Cc: Mary Barnes <mary.ietf.barnes@gmail.com>, IETF ACME <acme@ietf.org>
Content-Transfer-Encoding: 7bit
Message-Id: <0619953C-C4D0-4846-82ED-C814073F6DFC@gmail.com>
References: <CAHbuEH4ogM3NQNEUFW44JFjh3s5mgvBHuhExvei0F1Voe2T+xg@mail.gmail.com> <CAHBDyN61y8KWmFY17YzXCTWfL4s8LTrxWvhHhbRZ32f-_uG8uA@mail.gmail.com> <CAJ=cBg1eh-27KzX8=zS+07iQr2E4m2-FaN=hh0Bi9U8SMeL8pA@mail.gmail.com>
To: Clint Wilson <clint.t.wilson@gmail.com>
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/Chh-sojm9CxyJO1RsEAsbwrMvA4>
Subject: Re: [Acme] AD review of draft-ietf-acme-acme
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 22 Sep 2017 22:03:48 -0000

--Apple-Mail-4C7D1F7E-0214-4415-AECC-B27C8281F3B0
Content-Type: text/plain;
	charset=utf-8
Content-Transfer-Encoding: quoted-printable

Thanks for the input and the offline message, Mary.  An offline process is n=
eeded from what I can tell to use ACME for anything considered a higher assu=
rance than a DV cert, is that right?  If so, I think it's important to make t=
hat clear with a simple sentence.

Thank you,
Kathleen=20

Sent from my iPhone

> On Sep 22, 2017, at 5:33 PM, Clint Wilson <clint.t.wilson@gmail.com> wrote=
:
>=20
> An additional 1/2 cent from me; we intend to use ACME primarily, if not so=
lely, with non-DV certs.
>=20
>=20
>> On Fri, Sep 22, 2017, 2:51 PM Mary Barnes <mary.ietf.barnes@gmail.com> wr=
ote:
>> Just to throw in my 1/2 cent.  We are using ACME for non-DV certificates i=
n the ATIS/SIP Forum SHAKEN framework as detailed in ATIS-1000080:  https://=
www.sipforum.org/download/joint-atissip-forum-standard-signature-based-handl=
ing-asserted-information-using-tokens-shaken-governance-model-certificate-ma=
nagement-atis-1000080/?wpdmdl=3D3360
>>=20
>> Regards,
>> Mary.
>>=20
>>> On Fri, Sep 22, 2017 at 3:14 PM, Kathleen Moriarty <kathleen.moriarty.ie=
tf@gmail.com> wrote:
>>> Hello,
>>>=20
>>> Thank you to the editors and WG for your efforts on
>>> draft-ietf-acme-acme, it's a well written and easy to understand
>>> draft.  I do have a few comments, that need to be address by the
>>> editors and SHEPHERD.
>>>=20
>>> Please review the idnits.  There are a few warnings that should be
>>> correctable and can be addressed, but more importantly, it calls out
>>> several downrefs that are not in the shepherd writeup.  These need to
>>> be in the writeup and then also mentioned in the IETF last call
>>> announcement (I'll make sure the latter happens).
>>>=20
>>> Here's my mostly editorial comments:
>>>=20
>>> Introduction:
>>> It reads very well, but it would be helpful to those unfamiliar with
>>> the work to explicitly state that ACME is about DV certificates.  The
>>> first sentence of the last paragraph seems like the best place to add
>>> this in.
>>> Current text:
>>>    This document describes an extensible framework for automating the
>>>    issuance and domain validation procedure, thereby allowing servers
>>>    and infrastructural software to obtain certificates without user
>>>    interaction.
>>> Proposed:
>>>    This document describes an extensible framework for automating the
>>>    issuance and domain validation procedure for DV certificates,
>>> thereby allowing servers
>>>    and infrastructural software to obtain certificates without user
>>>    interaction.
>>>=20
>>> I do like how the introduction is framed as it=E2=80=99s clear the level=
 of
>>> security provided by the certificates issued via ACME.  Thanks for
>>> that.
>>>=20
>>> The introduction should also make it clear that ACME can be used for
>>> other services using DV certificates, not just HTTP.  This is
>>> mentioned in the Terminology section, but I think a clear sentence
>>> upfront would be helpful.
>>>=20
>>> Section 2: Nit/suggestion: Change the tense to read as a published RFC
>>> in the last paragraph, last sentence (take it or leave it).
>>> Current:
>>>    Such close integration of ACME with HTTPS
>>>    servers would allow the immediate and automated deployment of
>>>    certificates as they are issued, sparing the human administrator from=

>>>    much of the time-consuming work described in the previous section.
>>> Proposed:
>>>    Such close integration of ACME with HTTPS
>>>    servers allows the immediate and automated deployment of
>>>    certificates as they are issued, sparing the human administrator from=

>>>    much of the time-consuming work described in the previous section.
>>>=20
>>> IANA Section:
>>> Everything looks good except that RFC5226 has been obsoleted, so the
>>> new reference is RFC8126 and =E2=80=9Cspecification required=E2=80=9D
>>> https://tools.ietf.org/html/rfc8126#section-4.6 still means the same
>>> thing, so using that is fine.
>>>=20
>>> Security considerations:
>>> I think it would be good to mention here that they are DV certs so the
>>> reader understands from he introduction and this section the level of
>>> cert issued through ACME and doesn=E2=80=99t assume a higher level of
>>> assurance.
>>>=20
>>> s/ACME is a protocol for managing certificates/ACME is a protocol for
>>> managing DV certificates/
>>>=20
>>> 10.1 - I know this is obvious to the people in the WG, but a reference
>>> to RFC7525 to properly configure TLS 1.2 should be included to protect
>>> against the mentioned attacks.  If you want to also reference TLS 1.3
>>> and specify no 0-RTT that would be good too.  It=E2=80=99d be great to s=
ee
>>> this get widely deployed, so there may be a number of newcomers
>>> reading this to make their lives easier with cert
>>> issuance/maintenance.
>>>=20
>>> 10.2 - what if the server that the account verifier has an account on
>>> (client for ACME) is compromised and is used to request new
>>> certificates or perform other actions?  I think this is one of the
>>> larger risks, so mentioning this is possible and hardening measures
>>> should be taken to prevent compromise would be prudent.  Hardening
>>> measures or appropriate security controls should be broadly understood
>>> terms (I think) so that you wouldn=E2=80=99t need to list out things lik=
e
>>> turing off unnecessary services.
>>>=20
>>> 10.5 - I would think you=E2=80=99d see requests for new phishing domains=

>>> rather than known ones through this process.  What would you look for
>>> there as that=E2=80=99s a complex problem - it would be hard to know all=
 legit
>>> names to know if one was a play off of the original.  This happened
>>> with equifax - they provided a site name with a typo and a =E2=80=98whit=
e hat=E2=80=99
>>> attacker had set up a site that looked just like theirs at that site.
>>> But it was announced by equifax.  Tough problem as automating a check
>>> on the DNS registration may not detect something unusual, but maybe
>>> recommending this check could help?  RDP lessens what needs to be
>>> shared from whois though, although I don=E2=80=99t think that=E2=80=99s w=
idely
>>> deployed.
>>>=20
>>> --
>>>=20
>>> Best regards,
>>> Kathleen
>>>=20
>>> _______________________________________________
>>> Acme mailing list
>>> Acme@ietf.org
>>> https://www.ietf.org/mailman/listinfo/acme
>>=20
>> _______________________________________________
>> Acme mailing list
>> Acme@ietf.org
>> https://www.ietf.org/mailman/listinfo/acme

--Apple-Mail-4C7D1F7E-0214-4415-AECC-B27C8281F3B0
Content-Type: text/html;
	charset=utf-8
Content-Transfer-Encoding: quoted-printable

<html><head><meta http-equiv=3D"content-type" content=3D"text/html; charset=3D=
utf-8"></head><body dir=3D"auto"><div>Thanks for the input and the offline m=
essage, Mary. &nbsp;An offline process is needed from what I can tell to use=
 ACME for anything considered a higher assurance than a DV cert, is that rig=
ht? &nbsp;If so, I think it's important to make that clear with a simple sen=
tence.</div><div id=3D"AppleMailSignature"><br></div><div id=3D"AppleMailSig=
nature">Thank you,</div><div id=3D"AppleMailSignature">Kathleen&nbsp;<br><br=
>Sent from my iPhone</div><div><br>On Sep 22, 2017, at 5:33 PM, Clint Wilson=
 &lt;<a href=3D"mailto:clint.t.wilson@gmail.com">clint.t.wilson@gmail.com</a=
>&gt; wrote:<br><br></div><blockquote type=3D"cite"><div><p dir=3D"ltr">An a=
dditional 1/2 cent from me; we intend to use ACME primarily, if not solely, w=
ith non-DV certs.</p>
<br><div class=3D"gmail_quote"><div dir=3D"ltr">On Fri, Sep 22, 2017, 2:51 P=
M Mary Barnes &lt;<a href=3D"mailto:mary.ietf.barnes@gmail.com">mary.ietf.ba=
rnes@gmail.com</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" sty=
le=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div di=
r=3D"ltr">Just to throw in my 1/2 cent.&nbsp; We are using ACME for non-DV c=
ertificates in the ATIS/SIP Forum SHAKEN framework as detailed in ATIS-10000=
80: &nbsp;<a href=3D"https://www.sipforum.org/download/joint-atissip-forum-s=
tandard-signature-based-handling-asserted-information-using-tokens-shaken-go=
vernance-model-certificate-management-atis-1000080/?wpdmdl=3D3360" target=3D=
"_blank">https://www.sipforum.org/download/joint-atissip-forum-standard-sign=
ature-based-handling-asserted-information-using-tokens-shaken-governance-mod=
el-certificate-management-atis-1000080/?wpdmdl=3D3360</a><div><br><div><div>=
<div>Regards,<br></div><div>Mary.</div></div></div></div></div><div class=3D=
"gmail_extra"><br><div class=3D"gmail_quote">On Fri, Sep 22, 2017 at 3:14 PM=
, Kathleen Moriarty <span dir=3D"ltr">&lt;<a href=3D"mailto:kathleen.moriart=
y.ietf@gmail.com" target=3D"_blank">kathleen.moriarty.ietf@gmail.com</a>&gt;=
</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8=
ex;border-left:1px #ccc solid;padding-left:1ex">Hello,<br>
<br>
Thank you to the editors and WG for your efforts on<br>
draft-ietf-acme-acme, it's a well written and easy to understand<br>
draft.&nbsp; I do have a few comments, that need to be address by the<br>
editors and SHEPHERD.<br>
<br>
Please review the idnits.&nbsp; There are a few warnings that should be<br>
correctable and can be addressed, but more importantly, it calls out<br>
several downrefs that are not in the shepherd writeup.&nbsp; These need to<b=
r>
be in the writeup and then also mentioned in the IETF last call<br>
announcement (I'll make sure the latter happens).<br>
<br>
Here's my mostly editorial comments:<br>
<br>
Introduction:<br>
It reads very well, but it would be helpful to those unfamiliar with<br>
the work to explicitly state that ACME is about DV certificates.&nbsp; The<b=
r>
first sentence of the last paragraph seems like the best place to add<br>
this in.<br>
Current text:<br>
&nbsp; &nbsp;This document describes an extensible framework for automating t=
he<br>
&nbsp; &nbsp;issuance and domain validation procedure, thereby allowing serv=
ers<br>
&nbsp; &nbsp;and infrastructural software to obtain certificates without use=
r<br>
&nbsp; &nbsp;interaction.<br>
Proposed:<br>
&nbsp; &nbsp;This document describes an extensible framework for automating t=
he<br>
&nbsp; &nbsp;issuance and domain validation procedure for DV certificates,<b=
r>
thereby allowing servers<br>
&nbsp; &nbsp;and infrastructural software to obtain certificates without use=
r<br>
&nbsp; &nbsp;interaction.<br>
<br>
I do like how the introduction is framed as it=E2=80=99s clear the level of<=
br>
security provided by the certificates issued via ACME.&nbsp; Thanks for<br>
that.<br>
<br>
The introduction should also make it clear that ACME can be used for<br>
other services using DV certificates, not just HTTP.&nbsp; This is<br>
mentioned in the Terminology section, but I think a clear sentence<br>
upfront would be helpful.<br>
<br>
Section 2: Nit/suggestion: Change the tense to read as a published RFC<br>
in the last paragraph, last sentence (take it or leave it).<br>
Current:<br>
&nbsp; &nbsp;Such close integration of ACME with HTTPS<br>
&nbsp; &nbsp;servers would allow the immediate and automated deployment of<b=
r>
&nbsp; &nbsp;certificates as they are issued, sparing the human administrato=
r from<br>
&nbsp; &nbsp;much of the time-consuming work described in the previous secti=
on.<br>
Proposed:<br>
&nbsp; &nbsp;Such close integration of ACME with HTTPS<br>
&nbsp; &nbsp;servers allows the immediate and automated deployment of<br>
&nbsp; &nbsp;certificates as they are issued, sparing the human administrato=
r from<br>
&nbsp; &nbsp;much of the time-consuming work described in the previous secti=
on.<br>
<br>
IANA Section:<br>
Everything looks good except that RFC5226 has been obsoleted, so the<br>
new reference is RFC8126 and =E2=80=9Cspecification required=E2=80=9D<br>
<a href=3D"https://tools.ietf.org/html/rfc8126#section-4.6" rel=3D"noreferre=
r" target=3D"_blank">https://tools.ietf.org/html/rfc8126#section-4.6</a> sti=
ll means the same<br>
thing, so using that is fine.<br>
<br>
Security considerations:<br>
I think it would be good to mention here that they are DV certs so the<br>
reader understands from he introduction and this section the level of<br>
cert issued through ACME and doesn=E2=80=99t assume a higher level of<br>
assurance.<br>
<br>
s/ACME is a protocol for managing certificates/ACME is a protocol for<br>
managing DV certificates/<br>
<br>
10.1 - I know this is obvious to the people in the WG, but a reference<br>
to RFC7525 to properly configure TLS 1.2 should be included to protect<br>
against the mentioned attacks.&nbsp; If you want to also reference TLS 1.3<b=
r>
and specify no 0-RTT that would be good too.&nbsp; It=E2=80=99d be great to s=
ee<br>
this get widely deployed, so there may be a number of newcomers<br>
reading this to make their lives easier with cert<br>
issuance/maintenance.<br>
<br>
10.2 - what if the server that the account verifier has an account on<br>
(client for ACME) is compromised and is used to request new<br>
certificates or perform other actions?&nbsp; I think this is one of the<br>
larger risks, so mentioning this is possible and hardening measures<br>
should be taken to prevent compromise would be prudent.&nbsp; Hardening<br>
measures or appropriate security controls should be broadly understood<br>
terms (I think) so that you wouldn=E2=80=99t need to list out things like<br=
>
turing off unnecessary services.<br>
<br>
10.5 - I would think you=E2=80=99d see requests for new phishing domains<br>=

rather than known ones through this process.&nbsp; What would you look for<b=
r>
there as that=E2=80=99s a complex problem - it would be hard to know all leg=
it<br>
names to know if one was a play off of the original.&nbsp; This happened<br>=

with equifax - they provided a site name with a typo and a =E2=80=98white ha=
t=E2=80=99<br>
attacker had set up a site that looked just like theirs at that site.<br>
But it was announced by equifax.&nbsp; Tough problem as automating a check<b=
r>
on the DNS registration may not detect something unusual, but maybe<br>
recommending this check could help?&nbsp; RDP lessens what needs to be<br>
shared from whois though, although I don=E2=80=99t think that=E2=80=99s wide=
ly<br>
deployed.<br>
<span class=3D"m_-4806915521565577366HOEnZb"><font color=3D"#888888"><br>
--<br>
<br>
Best regards,<br>
Kathleen<br>
<br>
_______________________________________________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org" target=3D"_blank">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" ta=
rget=3D"_blank">https://www.ietf.org/mailman/listinfo/acme</a><br>
</font></span></blockquote></div><br></div>
_______________________________________________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org" target=3D"_blank">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" ta=
rget=3D"_blank">https://www.ietf.org/mailman/listinfo/acme</a><br>
</blockquote></div>
</div></blockquote></body></html>=

--Apple-Mail-4C7D1F7E-0214-4415-AECC-B27C8281F3B0--


From nobody Fri Sep 22 16:19:39 2017
Return-Path: <roland@letsencrypt.org>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4C7E0133090 for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 16:19:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.001
X-Spam-Level: 
X-Spam-Status: No, score=-2.001 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=letsencrypt.org
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 vO7IfFte82e5 for <acme@ietfa.amsl.com>; Fri, 22 Sep 2017 16:19:35 -0700 (PDT)
Received: from mail-pg0-x22f.google.com (mail-pg0-x22f.google.com [IPv6:2607:f8b0:400e:c05::22f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 13A9F133085 for <acme@ietf.org>; Fri, 22 Sep 2017 16:19:34 -0700 (PDT)
Received: by mail-pg0-x22f.google.com with SMTP id d8so1326103pgt.4 for <acme@ietf.org>; Fri, 22 Sep 2017 16:19:34 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=letsencrypt.org; s=google; h=subject:to:references:from:message-id:date:user-agent:mime-version :in-reply-to:content-language:content-transfer-encoding; bh=2NyEbMd7ViIpBRjRGf3O5bQCUcdyIbJhKVF1g+PXBWA=; b=cRTXGzVJAZ0IJfAmQ9R0wWWIPIrE3H3+qW4wADseB2C/+Xwqr36mt4KLHBNk0L70R6 cjehmeSZ1wYS6yHc3QWquS+UIJOl6nasJBqRdBNhqwD6uMLG9tBQ5Shb1qOaVY0mvFZ1 tVx33mFbaalub/iO56lznr2caPDw5Kg6Y6mMc=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:subject:to:references:from:message-id:date :user-agent:mime-version:in-reply-to:content-language :content-transfer-encoding; bh=2NyEbMd7ViIpBRjRGf3O5bQCUcdyIbJhKVF1g+PXBWA=; b=pgJkfepa/71VlYYUNUhGQUp1X/8l6fouXJKa6CAujwQxnXfdQMXa215dc7t/HPIDEi 5MJ0UV0botsGH7YysnXpRROMnQy280Oi3ccveOy8RfBadp6LHZccZxFo1HOeUC29UyXs 1wlse5WTJkV5JMY5tcGupxMQVByy1Cvbhnic+/2wIpArbDxVw5YcZlkz/0yIDZXxs+Al MSmv3WDCQ9C1yYdx24/D62TPRnEoYWlFCNuI9347qLCQv1yX2kyOVyUpQJh6zGB8aXxG 3USqzL0g/iWxFsC4DVLh7VBkpVljdHJ/s48K6kKTTnsG5c1Z8sCHDw1SCRSxkjlQTC5P 4scA==
X-Gm-Message-State: AHPjjUhZeGF2HxcjhSSsg4xXVitTo9+vjy1qCdehB58zkZFcAAz8iFU8 KJEUBvtT3HAJ5cOU88Lm9ry00cUB8DA=
X-Google-Smtp-Source: AOwi7QAZYo5uaA08pfXyG4o88oDnjCwp87/tyF0CbkHRc+FyfFyHULh3y+tq9oHElQyA2vrZ/j1+9g==
X-Received: by 10.99.181.23 with SMTP id y23mr572769pge.191.1506122373691; Fri, 22 Sep 2017 16:19:33 -0700 (PDT)
Received: from [192.168.42.99] ([50.1.57.72]) by smtp.gmail.com with ESMTPSA id u66sm1045710pfd.33.2017.09.22.16.19.30 for <acme@ietf.org> (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Fri, 22 Sep 2017 16:19:31 -0700 (PDT)
To: acme@ietf.org
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com> <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com>
From: Roland Bracewell Shoemaker <roland@letsencrypt.org>
Message-ID: <c9a26f93-b134-1d69-b21e-080d34bdcc30@letsencrypt.org>
Date: Fri, 22 Sep 2017 16:19:29 -0700
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.3.0
MIME-Version: 1.0
In-Reply-To: <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/ut2T5Yv8ZFKv4_QdSu8zdk5tpsk>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 22 Sep 2017 23:19:38 -0000

Broad points:

I understand that this group has been moving towards standardization for
a while but I really think that attempting to fix these issues with
creative readings of the existing language and minimal editorial changes
is asking for trouble and return trips to the WG in the future when
these band-aids turn out to not be fully thought out or themselves
introduce new problems.

These issues have come up during, as far as I'm aware, the only full-fat
implementation of the current version of this specification and that
implementation itself is maybe only 40-50% done. Assuming that these
will be the only issues that are discovered seems to be, again, asking
for trouble.

Comments on the proposed fixes:

1. Causing issuance really seems like it should be required to be an
authorized action, allowing this to happen by a GET allows anyone who is
able to determine the order URL to cause issuance to happen.
Implementing this behavior, even if not explicitly defined in the
specification, will almost definitely lead to users relying on it
(Hyrum's Law) which again will be a problem if another user is able to
simply send of a handful of GETs and cause another users certificate to
be issued.

2. Unless another implementer comes along this would mean that the only
existing implementation of the current specification before
standardization would not be implementing a major section of this
document. At this point why not just go back to the -03/-04 draft
new-authz flow. As Daniel has already pointed out this also completely
breaks one of the major reasons to use the new-order flow: policy based
authorization creation. How would a CA indicate that a user _may not_
need an authorization for www.example.com when issuing for example.com
or that they require a specific set of authorizations when issuing for
*.example.com?

Like I said at the start these band-aid editorial fixes seem to just
create another set of problems that cannot themselves be solved with
further minor editorial  changes.

On 09/21/2017 11:40 AM, Richard Barnes wrote:
> I talked a bit with Daniel about this off-list. Here's a summary of
> where we got to:
> 
> 1. In the current "new-order" flow, the server doesn't *really* have to
> be proactive -- instead of updating an order object when an authz
> changes state, it can wait to update/issue until the client polls for
> the order object, indicating that it's actually interested in having a
> certificate.  That doesn't alleviate the pain of storing the CSR, but it
> at least avoids the expense of finding affected orders and issuing in
> cases where the client abandons the order.  If we think this approach is
> OK, I would propose that we revise the document to make clear that it is
> OK.  What do folks think?  Daniel noted that there might be some issues
> with GET idempotency here, but I don't think this actually makes GET
> non-idempotent.
> 
> 2. Some CAs might want to support only the "new-authz" flow, i.e., never
> allow a "new-order" request if the relevant authorizations haven't
> already been done.  It seems like this could be addressed in a pretty
> straightforward manner just by defining a new error code (say
> "preauthRequired") that the server can return in this case.
> 
> Personally, it seems to me that if we make the clarification (1) and add
> the error code (2) described here, that would address the pain points
> Daniel raises, without a whole lot of new mechanism.  I'm not really
> seeing much additional value to be had from an explicit "ok issue now"
> request (vs. the GET poll).
> 
> --Richard
> 
> 
> 
> On Wed, Sep 20, 2017 at 1:22 PM, Richard Barnes <rlb@ipv.sx
> <mailto:rlb@ipv.sx>> wrote:
> 
>     Hey Daniel,
> 
>     Thanks for the input.  I can see why you might be feeling some pain
>     here, but I'm not sure the proposed solution is quite right.
> 
>     Note that ACME already has the "new-authorization" flow, which does
>     pretty much exactly what you're proposing.  The only difference is
>     that instead of authorizations being created all at once in response
>     to a "new-cert" request, they're created individually by the
>     client.  But it's ultimately the same number of RTTs (authz fetch ~
>     new-authz) -- or rather, one fewer in the new-authz case because
>     there's no new-cert request.  To what degree would your problem be
>     solved simply by having the affected integrators move over to that flow?
> 
>     I can sympathize that the new-order flow might not be suitable for
>     all use cases.  Maybe it would be useful to write down some thoughts
>     on which use cases are best addressed by which flows. 
> 
>     --Richard
> 
> 
> 
>     On Tue, Sep 19, 2017 at 12:26 PM, Daniel McCarney
>     <cpu@letsencrypt.org <mailto:cpu@letsencrypt.org>> wrote:
> 
>         Hi folks,
> 
>         Just over a year ago in a thread titled "Proactive vs
>         On-Finalization
>         Certificate Issuance"[0] we reached the consensus on the
>         question of whether to
>         issue a certificate "on-finalization" or "proactively" with the
>         conclusion to
>         standardize on proactive issuance.
> 
>         Implementation experience shows that proactive issuance is slow
>         in practice. An
>         order can be associated with many authorizations and an
>         authorization can be
>         associated with many orders. This interacts poorly with
>         optimizations Let's
>         Encrypt has developed from in-the-field experience with ACME and
>         it is difficult
>         to rate-limit effectively. I propose that we return to the
>         on-finalization style
>         approach and that we provide the CSR during finalization..
> 
>         ## Issues with proactive issuance
> 
>         ### Authorization Reuse
> 
>         Proactive issuance requires that when an authorization is
>         switched to a valid
>         state by the result of a challenge update that we find the
>         associated orders and
>         iterate each order's authorizations to see if all of the
>         authorizations are
>         valid and issuance can occur.
> 
>         Let's Encrypt reuses existing authorizations[1][2] whenever
>         possible, as
>         encouraged by ACME. This has been of huge value in reducing
>         resource consumption
>         (both in terms of database storage and challenge/API requests).
>         Many clients
>         create pending authorizations they do not finalize and similarly
>         re-create
>         pending authorizations for identifiers that the account already
>         has valid
>         authorizations for. Unchecked this leads to significant database
>         growth and
>         resource consumption, even with aggressive rate-limits[3].
> 
>         The combination of authorization reuse and proactive issuance
>         means a given
>         authorization can be associated with many orders. This
>         necessitates an expensive
>         many-to-many query at each authorization update to see if any
>         associated
>         orders are now complete.
> 
>         One possible workaround is a rate-limit that would constrain the
>         number of
>         orders an authorization can be associated with. However
>         establishing the value
>         of the limit to both reduce server-side load & also effectively
>         support Let's
>         Encrypt's large-volume integrators and their issuance patterns
>         is challenging.
> 
>         ### CSR-first new-order flow
> 
>         To support proactive issuance the CA also needs to have the CSR
>         for the order
>         at-hand when checking each authorization in case issuance can
>         proceed. The
>         `new-order` request that creates the order therefore carries the
>         full CSR that
>         was previously delivered by the "certificate finalize" message.
> 
>         We see users create many more authorizations than they are able
>         to finalize
>         successfully. With the legacy "new-cert" flow our authorization
>         reuse
>         dramatically helped reduce the DB space used up by
>         authorizations that will
>         never lead to issuance because no CSR is stored until all
>         identifiers are
>         validated.
> 
>         With the new-order flow we have to accept and store a full CSR
>         before handing
>         back authorizations to the client. Reuse of authorizations does
>         not prevent the
>         DB bloat that will occur from users submitting new order
>         requests and not
>         fulfilling them. The CSR contains a full public key and is one
>         of the single
>         largest contributors to DB disk space usage. Addressing this
>         will require
>         another strict rate limit on outstanding new-orders and is again
>         difficult to
>         “dial-in” in to both prevent excessive resource consumption and
>         also maintain
>         the ability for big integrators to process large-scale issuance
>         using our API.
> 
>         # Proposed Change
> 
>         I believe we should change the `new-order` endpoint to not
>         accept a full CSR but
>         to instead accept a list of identifiers that the ACME client
>         wishes to
>         authenticate & issue for. Authorizations can be created and
>         returned for those
>         domains as happens today.
> 
>         The order resource should be changed to have a “finalizeURL”
>         field which
>         provides the client a URL to POST a CSR to indicate that the CA
>         should issue the
>         certificate. Proactive issuance should be removed outright.
> 
>         This will address both the problem of scanning existing
>         many-to-many orders
>         & their authorizations as well as the db bloat that will occur from
>         front-loading the delivery of the CSR.
> 
>         It may be tempting to leave proactive issuance as an optional
>         feature but this
>         will require maintaining the CSR in the new-order and I also
>         believe that having
>         two very distinct methods of issuance will complicate the client
>         ecosystem and
>         returns us to the question of how to determine which will occur
>         that spawned the
>         original "Proactive vs On-finalization Certificate Issuance"
>         thread[0].
> 
>         [0]
>         https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY
>         <https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY>
>         [1]
>         https://community.letsencrypt.org/t/upcoming-api-changes/17947
>         <https://community.letsencrypt.org/t/upcoming-api-changes/17947>
>         [2]
>         https://community.letsencrypt.org/t/automatic-recycling-of-pending-authorizations/41321
>         <https://community.letsencrypt.org/t/automatic-recycling-of-pending-authorizations/41321>
>         [3] https://letsencrypt.org/docs/rate-limits/
>         <https://letsencrypt.org/docs/rate-limits/>
> 
> 
>         _______________________________________________
>         Acme mailing list
>         Acme@ietf.org <mailto:Acme@ietf.org>
>         https://www.ietf.org/mailman/listinfo/acme
>         <https://www.ietf.org/mailman/listinfo/acme>
> 
> 
> 
> 
> 
> _______________________________________________
> Acme mailing list
> Acme@ietf.org
> https://www.ietf.org/mailman/listinfo/acme
> 


From nobody Tue Sep 26 07:14:07 2017
Return-Path: <dmccarney@letsencrypt.org>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C2BE5132A89 for <acme@ietfa.amsl.com>; Tue, 26 Sep 2017 07:14:05 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=letsencrypt.org
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 JPWC59KQogVq for <acme@ietfa.amsl.com>; Tue, 26 Sep 2017 07:14:02 -0700 (PDT)
Received: from mail-it0-x22c.google.com (mail-it0-x22c.google.com [IPv6:2607:f8b0:4001:c0b::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 157F51326F6 for <acme@ietf.org>; Tue, 26 Sep 2017 07:14:01 -0700 (PDT)
Received: by mail-it0-x22c.google.com with SMTP id y138so3290155itc.5 for <acme@ietf.org>; Tue, 26 Sep 2017 07:14:01 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=letsencrypt.org; s=google; h=mime-version:reply-to:in-reply-to:references:from:date:message-id :subject:to; bh=6dKmVDvXBWQ4HxGrBD+6JILKEJDbGTWpMGv4E3RZleE=; b=WHNpF2G6/4VQdq1nyYkRgfBsJQjHTWey7CtS/nZ8LLAQFi8l4NrEuToURwQ5x1vUhu FGMSnRSbHBqc8clYmDt3/K6V89zYNEgUpLmeRICsDDzjPlpXSCTajGuU/WjskgDut22H gQtBV/zTtwJlvdgM/cpxwec5cYdluM9eYJ9YM=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:reply-to:in-reply-to:references :from:date:message-id:subject:to; bh=6dKmVDvXBWQ4HxGrBD+6JILKEJDbGTWpMGv4E3RZleE=; b=rRdCBkc/HXAYdC2Dd9ISQAkVxiX1E5M8R91qowmRIWryCN5Nq8sKYVbm3hJfupHDPz 1oPxwhPAfaCBmHZj8UUKV3r8cu8tprGqQbWUV0Be77mCHrGY+sR2OanVMKbnP57fWFoV u8at2nb9zvfmEBou0sfel8yxwHpWPRqZItfAY3yLvdrw9frFJKFpWzRq1xArlNRalwQ/ fisjyz22wmPjHC3OLyMLY7xjREadOxv5d0j/04mGJrbTvtJs1Lt8TOHKlDXQeaJt/bg9 LvIKXsEtc8x9SR1zlzYopCVvyPhT0GpcOnmqL5KBX0Y1yIqUKRn/nrndXRQVEZb4xpVB UI8A==
X-Gm-Message-State: AHPjjUiocm7nHZKGIZe2r2QRKNKr73Rld5onXxXas8L4d8lDw0kE7I1H AKiJDGArSi2L2XZppj5na0pQgAk8PUDVkbTNCkRGtTkZ4gA=
X-Google-Smtp-Source: AOwi7QAYcwLueDQcmmTR25hbrQjQ9RXR/PAtOvsNWoRY6/LSIhX92i+iawzF0mJ6r/YePs67e6tQnQypWKC92XTas8U=
X-Received: by 10.36.39.142 with SMTP id g136mr6368595ita.73.1506435240818; Tue, 26 Sep 2017 07:14:00 -0700 (PDT)
MIME-Version: 1.0
Reply-To: cpu@letsencrypt.org
Received: by 10.107.17.229 with HTTP; Tue, 26 Sep 2017 07:14:00 -0700 (PDT)
In-Reply-To: <c9a26f93-b134-1d69-b21e-080d34bdcc30@letsencrypt.org>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com> <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com> <c9a26f93-b134-1d69-b21e-080d34bdcc30@letsencrypt.org>
From: Daniel McCarney <cpu@letsencrypt.org>
Date: Tue, 26 Sep 2017 10:14:00 -0400
Message-ID: <CAKnbcLjm0wa-3wQN2YqJ2y2DfeOgQQvG_gX+fN4APcmPwKXtag@mail.gmail.com>
To: IETF ACME <acme@ietf.org>
Content-Type: multipart/alternative; boundary="001a1147cc66680b69055a184933"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/Wjc2nMpAVqkE8zXYWFTuCgXOIyI>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 26 Sep 2017 14:14:06 -0000

--001a1147cc66680b69055a184933
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

I've proposed a first cut at implementing the changes I described in my
original post as an ACME PR https://github.com/ietf-wg-acme/acme/pull/342



On Fri, Sep 22, 2017 at 7:19 PM, Roland Bracewell Shoemaker <
roland@letsencrypt.org> wrote:

> Broad points:
>
> I understand that this group has been moving towards standardization for
> a while but I really think that attempting to fix these issues with
> creative readings of the existing language and minimal editorial changes
> is asking for trouble and return trips to the WG in the future when
> these band-aids turn out to not be fully thought out or themselves
> introduce new problems.
>
> These issues have come up during, as far as I'm aware, the only full-fat
> implementation of the current version of this specification and that
> implementation itself is maybe only 40-50% done. Assuming that these
> will be the only issues that are discovered seems to be, again, asking
> for trouble.
>
> Comments on the proposed fixes:
>
> 1. Causing issuance really seems like it should be required to be an
> authorized action, allowing this to happen by a GET allows anyone who is
> able to determine the order URL to cause issuance to happen.
> Implementing this behavior, even if not explicitly defined in the
> specification, will almost definitely lead to users relying on it
> (Hyrum's Law) which again will be a problem if another user is able to
> simply send of a handful of GETs and cause another users certificate to
> be issued.
>
> 2. Unless another implementer comes along this would mean that the only
> existing implementation of the current specification before
> standardization would not be implementing a major section of this
> document. At this point why not just go back to the -03/-04 draft
> new-authz flow. As Daniel has already pointed out this also completely
> breaks one of the major reasons to use the new-order flow: policy based
> authorization creation. How would a CA indicate that a user _may not_
> need an authorization for www.example.com when issuing for example.com
> or that they require a specific set of authorizations when issuing for
> *.example.com?
>
> Like I said at the start these band-aid editorial fixes seem to just
> create another set of problems that cannot themselves be solved with
> further minor editorial  changes.
>
> On 09/21/2017 11:40 AM, Richard Barnes wrote:
> > I talked a bit with Daniel about this off-list. Here's a summary of
> > where we got to:
> >
> > 1. In the current "new-order" flow, the server doesn't *really* have to
> > be proactive -- instead of updating an order object when an authz
> > changes state, it can wait to update/issue until the client polls for
> > the order object, indicating that it's actually interested in having a
> > certificate.  That doesn't alleviate the pain of storing the CSR, but i=
t
> > at least avoids the expense of finding affected orders and issuing in
> > cases where the client abandons the order.  If we think this approach i=
s
> > OK, I would propose that we revise the document to make clear that it i=
s
> > OK.  What do folks think?  Daniel noted that there might be some issues
> > with GET idempotency here, but I don't think this actually makes GET
> > non-idempotent.
> >
> > 2. Some CAs might want to support only the "new-authz" flow, i.e., neve=
r
> > allow a "new-order" request if the relevant authorizations haven't
> > already been done.  It seems like this could be addressed in a pretty
> > straightforward manner just by defining a new error code (say
> > "preauthRequired") that the server can return in this case.
> >
> > Personally, it seems to me that if we make the clarification (1) and ad=
d
> > the error code (2) described here, that would address the pain points
> > Daniel raises, without a whole lot of new mechanism.  I'm not really
> > seeing much additional value to be had from an explicit "ok issue now"
> > request (vs. the GET poll).
> >
> > --Richard
> >
> >
> >
> > On Wed, Sep 20, 2017 at 1:22 PM, Richard Barnes <rlb@ipv.sx
> > <mailto:rlb@ipv.sx>> wrote:
> >
> >     Hey Daniel,
> >
> >     Thanks for the input.  I can see why you might be feeling some pain
> >     here, but I'm not sure the proposed solution is quite right.
> >
> >     Note that ACME already has the "new-authorization" flow, which does
> >     pretty much exactly what you're proposing.  The only difference is
> >     that instead of authorizations being created all at once in respons=
e
> >     to a "new-cert" request, they're created individually by the
> >     client.  But it's ultimately the same number of RTTs (authz fetch ~
> >     new-authz) -- or rather, one fewer in the new-authz case because
> >     there's no new-cert request.  To what degree would your problem be
> >     solved simply by having the affected integrators move over to that
> flow?
> >
> >     I can sympathize that the new-order flow might not be suitable for
> >     all use cases.  Maybe it would be useful to write down some thought=
s
> >     on which use cases are best addressed by which flows.
> >
> >     --Richard
> >
> >
> >
> >     On Tue, Sep 19, 2017 at 12:26 PM, Daniel McCarney
> >     <cpu@letsencrypt.org <mailto:cpu@letsencrypt.org>> wrote:
> >
> >         Hi folks,
> >
> >         Just over a year ago in a thread titled "Proactive vs
> >         On-Finalization
> >         Certificate Issuance"[0] we reached the consensus on the
> >         question of whether to
> >         issue a certificate "on-finalization" or "proactively" with the
> >         conclusion to
> >         standardize on proactive issuance.
> >
> >         Implementation experience shows that proactive issuance is slow
> >         in practice. An
> >         order can be associated with many authorizations and an
> >         authorization can be
> >         associated with many orders. This interacts poorly with
> >         optimizations Let's
> >         Encrypt has developed from in-the-field experience with ACME an=
d
> >         it is difficult
> >         to rate-limit effectively. I propose that we return to the
> >         on-finalization style
> >         approach and that we provide the CSR during finalization..
> >
> >         ## Issues with proactive issuance
> >
> >         ### Authorization Reuse
> >
> >         Proactive issuance requires that when an authorization is
> >         switched to a valid
> >         state by the result of a challenge update that we find the
> >         associated orders and
> >         iterate each order's authorizations to see if all of the
> >         authorizations are
> >         valid and issuance can occur.
> >
> >         Let's Encrypt reuses existing authorizations[1][2] whenever
> >         possible, as
> >         encouraged by ACME. This has been of huge value in reducing
> >         resource consumption
> >         (both in terms of database storage and challenge/API requests).
> >         Many clients
> >         create pending authorizations they do not finalize and similarl=
y
> >         re-create
> >         pending authorizations for identifiers that the account already
> >         has valid
> >         authorizations for. Unchecked this leads to significant databas=
e
> >         growth and
> >         resource consumption, even with aggressive rate-limits[3].
> >
> >         The combination of authorization reuse and proactive issuance
> >         means a given
> >         authorization can be associated with many orders. This
> >         necessitates an expensive
> >         many-to-many query at each authorization update to see if any
> >         associated
> >         orders are now complete.
> >
> >         One possible workaround is a rate-limit that would constrain th=
e
> >         number of
> >         orders an authorization can be associated with. However
> >         establishing the value
> >         of the limit to both reduce server-side load & also effectively
> >         support Let's
> >         Encrypt's large-volume integrators and their issuance patterns
> >         is challenging.
> >
> >         ### CSR-first new-order flow
> >
> >         To support proactive issuance the CA also needs to have the CSR
> >         for the order
> >         at-hand when checking each authorization in case issuance can
> >         proceed. The
> >         `new-order` request that creates the order therefore carries th=
e
> >         full CSR that
> >         was previously delivered by the "certificate finalize" message.
> >
> >         We see users create many more authorizations than they are able
> >         to finalize
> >         successfully. With the legacy "new-cert" flow our authorization
> >         reuse
> >         dramatically helped reduce the DB space used up by
> >         authorizations that will
> >         never lead to issuance because no CSR is stored until all
> >         identifiers are
> >         validated.
> >
> >         With the new-order flow we have to accept and store a full CSR
> >         before handing
> >         back authorizations to the client. Reuse of authorizations does
> >         not prevent the
> >         DB bloat that will occur from users submitting new order
> >         requests and not
> >         fulfilling them. The CSR contains a full public key and is one
> >         of the single
> >         largest contributors to DB disk space usage. Addressing this
> >         will require
> >         another strict rate limit on outstanding new-orders and is agai=
n
> >         difficult to
> >         =E2=80=9Cdial-in=E2=80=9D in to both prevent excessive resource=
 consumption and
> >         also maintain
> >         the ability for big integrators to process large-scale issuance
> >         using our API.
> >
> >         # Proposed Change
> >
> >         I believe we should change the `new-order` endpoint to not
> >         accept a full CSR but
> >         to instead accept a list of identifiers that the ACME client
> >         wishes to
> >         authenticate & issue for. Authorizations can be created and
> >         returned for those
> >         domains as happens today.
> >
> >         The order resource should be changed to have a =E2=80=9Cfinaliz=
eURL=E2=80=9D
> >         field which
> >         provides the client a URL to POST a CSR to indicate that the CA
> >         should issue the
> >         certificate. Proactive issuance should be removed outright.
> >
> >         This will address both the problem of scanning existing
> >         many-to-many orders
> >         & their authorizations as well as the db bloat that will occur
> from
> >         front-loading the delivery of the CSR.
> >
> >         It may be tempting to leave proactive issuance as an optional
> >         feature but this
> >         will require maintaining the CSR in the new-order and I also
> >         believe that having
> >         two very distinct methods of issuance will complicate the clien=
t
> >         ecosystem and
> >         returns us to the question of how to determine which will occur
> >         that spawned the
> >         original "Proactive vs On-finalization Certificate Issuance"
> >         thread[0].
> >
> >         [0]
> >         https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-
> rmSH0x7ycDW5dj6GAY
> >         <https://mailarchive.ietf.org/arch/msg/acme/0lVmGl8e-
> rmSH0x7ycDW5dj6GAY>
> >         [1]
> >         https://community.letsencrypt.org/t/upcoming-api-changes/17947
> >         <https://community.letsencrypt.org/t/upcoming-api-changes/17947=
>
> >         [2]
> >         https://community.letsencrypt.org/t/automatic-recycling-of-
> pending-authorizations/41321
> >         <https://community.letsencrypt.org/t/automatic-
> recycling-of-pending-authorizations/41321>
> >         [3] https://letsencrypt.org/docs/rate-limits/
> >         <https://letsencrypt.org/docs/rate-limits/>
> >
> >
> >         _______________________________________________
> >         Acme mailing list
> >         Acme@ietf.org <mailto:Acme@ietf.org>
> >         https://www.ietf.org/mailman/listinfo/acme
> >         <https://www.ietf.org/mailman/listinfo/acme>
> >
> >
> >
> >
> >
> > _______________________________________________
> > Acme mailing list
> > Acme@ietf.org
> > https://www.ietf.org/mailman/listinfo/acme
> >
>
> _______________________________________________
> Acme mailing list
> Acme@ietf.org
> https://www.ietf.org/mailman/listinfo/acme
>

--001a1147cc66680b69055a184933
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">I&#39;ve proposed a first cut at implementing the changes =
I described in my original post as an ACME PR=C2=A0<a href=3D"https://githu=
b.com/ietf-wg-acme/acme/pull/342">https://github.com/ietf-wg-acme/acme/pull=
/342</a><br><br><br></div><div class=3D"gmail_extra"><br><div class=3D"gmai=
l_quote">On Fri, Sep 22, 2017 at 7:19 PM, Roland Bracewell Shoemaker <span =
dir=3D"ltr">&lt;<a href=3D"mailto:roland@letsencrypt.org" target=3D"_blank"=
>roland@letsencrypt.org</a>&gt;</span> wrote:<br><blockquote class=3D"gmail=
_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:=
1ex">Broad points:<br>
<br>
I understand that this group has been moving towards standardization for<br=
>
a while but I really think that attempting to fix these issues with<br>
creative readings of the existing language and minimal editorial changes<br=
>
is asking for trouble and return trips to the WG in the future when<br>
these band-aids turn out to not be fully thought out or themselves<br>
introduce new problems.<br>
<br>
These issues have come up during, as far as I&#39;m aware, the only full-fa=
t<br>
implementation of the current version of this specification and that<br>
implementation itself is maybe only 40-50% done. Assuming that these<br>
will be the only issues that are discovered seems to be, again, asking<br>
for trouble.<br>
<br>
Comments on the proposed fixes:<br>
<br>
1. Causing issuance really seems like it should be required to be an<br>
authorized action, allowing this to happen by a GET allows anyone who is<br=
>
able to determine the order URL to cause issuance to happen.<br>
Implementing this behavior, even if not explicitly defined in the<br>
specification, will almost definitely lead to users relying on it<br>
(Hyrum&#39;s Law) which again will be a problem if another user is able to<=
br>
simply send of a handful of GETs and cause another users certificate to<br>
be issued.<br>
<br>
2. Unless another implementer comes along this would mean that the only<br>
existing implementation of the current specification before<br>
standardization would not be implementing a major section of this<br>
document. At this point why not just go back to the -03/-04 draft<br>
new-authz flow. As Daniel has already pointed out this also completely<br>
breaks one of the major reasons to use the new-order flow: policy based<br>
authorization creation. How would a CA indicate that a user _may not_<br>
need an authorization for <a href=3D"http://www.example.com" rel=3D"norefer=
rer" target=3D"_blank">www.example.com</a> when issuing for <a href=3D"http=
://example.com" rel=3D"noreferrer" target=3D"_blank">example.com</a><br>
or that they require a specific set of authorizations when issuing for<br>
*.<a href=3D"http://example.com" rel=3D"noreferrer" target=3D"_blank">examp=
le.com</a>?<br>
<br>
Like I said at the start these band-aid editorial fixes seem to just<br>
create another set of problems that cannot themselves be solved with<br>
further minor editorial=C2=A0 changes.<br>
<div><div class=3D"h5"><br>
On 09/21/2017 11:40 AM, Richard Barnes wrote:<br>
&gt; I talked a bit with Daniel about this off-list. Here&#39;s a summary o=
f<br>
&gt; where we got to:<br>
&gt;<br>
&gt; 1. In the current &quot;new-order&quot; flow, the server doesn&#39;t *=
really* have to<br>
&gt; be proactive -- instead of updating an order object when an authz<br>
&gt; changes state, it can wait to update/issue until the client polls for<=
br>
&gt; the order object, indicating that it&#39;s actually interested in havi=
ng a<br>
&gt; certificate.=C2=A0 That doesn&#39;t alleviate the pain of storing the =
CSR, but it<br>
&gt; at least avoids the expense of finding affected orders and issuing in<=
br>
&gt; cases where the client abandons the order.=C2=A0 If we think this appr=
oach is<br>
&gt; OK, I would propose that we revise the document to make clear that it =
is<br>
&gt; OK.=C2=A0 What do folks think?=C2=A0 Daniel noted that there might be =
some issues<br>
&gt; with GET idempotency here, but I don&#39;t think this actually makes G=
ET<br>
&gt; non-idempotent.<br>
&gt;<br>
&gt; 2. Some CAs might want to support only the &quot;new-authz&quot; flow,=
 i.e., never<br>
&gt; allow a &quot;new-order&quot; request if the relevant authorizations h=
aven&#39;t<br>
&gt; already been done.=C2=A0 It seems like this could be addressed in a pr=
etty<br>
&gt; straightforward manner just by defining a new error code (say<br>
&gt; &quot;preauthRequired&quot;) that the server can return in this case.<=
br>
&gt;<br>
&gt; Personally, it seems to me that if we make the clarification (1) and a=
dd<br>
&gt; the error code (2) described here, that would address the pain points<=
br>
&gt; Daniel raises, without a whole lot of new mechanism.=C2=A0 I&#39;m not=
 really<br>
&gt; seeing much additional value to be had from an explicit &quot;ok issue=
 now&quot;<br>
&gt; request (vs. the GET poll).<br>
&gt;<br>
&gt; --Richard<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt; On Wed, Sep 20, 2017 at 1:22 PM, Richard Barnes &lt;rlb@ipv.sx<br>
</div></div><span class=3D"">&gt; &lt;mailto:<a href=3D"mailto:rlb@ipv.sx">=
rlb@ipv.sx</a>&gt;&gt; wrote:<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0Hey Daniel,<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0Thanks for the input.=C2=A0 I can see why you might=
 be feeling some pain<br>
&gt;=C2=A0 =C2=A0 =C2=A0here, but I&#39;m not sure the proposed solution is=
 quite right.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0Note that ACME already has the &quot;new-authorizat=
ion&quot; flow, which does<br>
&gt;=C2=A0 =C2=A0 =C2=A0pretty much exactly what you&#39;re proposing.=C2=
=A0 The only difference is<br>
&gt;=C2=A0 =C2=A0 =C2=A0that instead of authorizations being created all at=
 once in response<br>
&gt;=C2=A0 =C2=A0 =C2=A0to a &quot;new-cert&quot; request, they&#39;re crea=
ted individually by the<br>
&gt;=C2=A0 =C2=A0 =C2=A0client.=C2=A0 But it&#39;s ultimately the same numb=
er of RTTs (authz fetch ~<br>
&gt;=C2=A0 =C2=A0 =C2=A0new-authz) -- or rather, one fewer in the new-authz=
 case because<br>
&gt;=C2=A0 =C2=A0 =C2=A0there&#39;s no new-cert request.=C2=A0 To what degr=
ee would your problem be<br>
&gt;=C2=A0 =C2=A0 =C2=A0solved simply by having the affected integrators mo=
ve over to that flow?<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0I can sympathize that the new-order flow might not =
be suitable for<br>
&gt;=C2=A0 =C2=A0 =C2=A0all use cases.=C2=A0 Maybe it would be useful to wr=
ite down some thoughts<br>
&gt;=C2=A0 =C2=A0 =C2=A0on which use cases are best addressed by which flow=
s.=C2=A0<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0--Richard<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0On Tue, Sep 19, 2017 at 12:26 PM, Daniel McCarney<b=
r>
</span><div><div class=3D"h5">&gt;=C2=A0 =C2=A0 =C2=A0&lt;<a href=3D"mailto=
:cpu@letsencrypt.org">cpu@letsencrypt.org</a> &lt;mailto:<a href=3D"mailto:=
cpu@letsencrypt.org">cpu@letsencrypt.org</a>&gt;&gt; wrote:<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Hi folks,<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Just over a year ago in a thread titl=
ed &quot;Proactive vs<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0On-Finalization<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Certificate Issuance&quot;[0] we reac=
hed the consensus on the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0question of whether to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0issue a certificate &quot;on-finaliza=
tion&quot; or &quot;proactively&quot; with the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0conclusion to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0standardize on proactive issuance.<br=
>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Implementation experience shows that =
proactive issuance is slow<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0in practice. An<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0order can be associated with many aut=
horizations and an<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0authorization can be<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0associated with many orders. This int=
eracts poorly with<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0optimizations Let&#39;s<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Encrypt has developed from in-the-fie=
ld experience with ACME and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0it is difficult<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0to rate-limit effectively. I propose =
that we return to the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0on-finalization style<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0approach and that we provide the CSR =
during finalization..<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0## Issues with proactive issuance<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0### Authorization Reuse<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Proactive issuance requires that when=
 an authorization is<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0switched to a valid<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0state by the result of a challenge up=
date that we find the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0associated orders and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0iterate each order&#39;s authorizatio=
ns to see if all of the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0authorizations are<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0valid and issuance can occur.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Let&#39;s Encrypt reuses existing aut=
horizations[1][2] whenever<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0possible, as<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0encouraged by ACME. This has been of =
huge value in reducing<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0resource consumption<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0(both in terms of database storage an=
d challenge/API requests).<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Many clients<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0create pending authorizations they do=
 not finalize and similarly<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0re-create<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0pending authorizations for identifier=
s that the account already<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0has valid<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0authorizations for. Unchecked this le=
ads to significant database<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0growth and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0resource consumption, even with aggre=
ssive rate-limits[3].<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0The combination of authorization reus=
e and proactive issuance<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0means a given<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0authorization can be associated with =
many orders. This<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0necessitates an expensive<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0many-to-many query at each authorizat=
ion update to see if any<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0associated<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0orders are now complete.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0One possible workaround is a rate-lim=
it that would constrain the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0number of<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0orders an authorization can be associ=
ated with. However<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0establishing the value<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0of the limit to both reduce server-si=
de load &amp; also effectively<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0support Let&#39;s<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Encrypt&#39;s large-volume integrator=
s and their issuance patterns<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0is challenging.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0### CSR-first new-order flow<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0To support proactive issuance the CA =
also needs to have the CSR<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0for the order<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0at-hand when checking each authorizat=
ion in case issuance can<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0proceed. The<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0`new-order` request that creates the =
order therefore carries the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0full CSR that<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0was previously delivered by the &quot=
;certificate finalize&quot; message.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0We see users create many more authori=
zations than they are able<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0to finalize<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0successfully. With the legacy &quot;n=
ew-cert&quot; flow our authorization<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0reuse<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0dramatically helped reduce the DB spa=
ce used up by<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0authorizations that will<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0never lead to issuance because no CSR=
 is stored until all<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0identifiers are<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0validated.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0With the new-order flow we have to ac=
cept and store a full CSR<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0before handing<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0back authorizations to the client. Re=
use of authorizations does<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0not prevent the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0DB bloat that will occur from users s=
ubmitting new order<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0requests and not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0fulfilling them. The CSR contains a f=
ull public key and is one<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0of the single<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0largest contributors to DB disk space=
 usage. Addressing this<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0will require<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0another strict rate limit on outstand=
ing new-orders and is again<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0difficult to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0=E2=80=9Cdial-in=E2=80=9D in to both =
prevent excessive resource consumption and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0also maintain<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0the ability for big integrators to pr=
ocess large-scale issuance<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0using our API.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0# Proposed Change<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0I believe we should change the `new-o=
rder` endpoint to not<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0accept a full CSR but<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0to instead accept a list of identifie=
rs that the ACME client<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0wishes to<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0authenticate &amp; issue for. Authori=
zations can be created and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0returned for those<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0domains as happens today.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0The order resource should be changed =
to have a =E2=80=9CfinalizeURL=E2=80=9D<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0field which<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0provides the client a URL to POST a C=
SR to indicate that the CA<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0should issue the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0certificate. Proactive issuance shoul=
d be removed outright.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0This will address both the problem of=
 scanning existing<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0many-to-many orders<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0&amp; their authorizations as well as=
 the db bloat that will occur from<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0front-loading the delivery of the CSR=
.<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0It may be tempting to leave proactive=
 issuance as an optional<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0feature but this<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0will require maintaining the CSR in t=
he new-order and I also<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0believe that having<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0two very distinct methods of issuance=
 will complicate the client<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0ecosystem and<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0returns us to the question of how to =
determine which will occur<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0that spawned the<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0original &quot;Proactive vs On-finali=
zation Certificate Issuance&quot;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0thread[0].<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0[0]<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"https://mailarchive.ietf.o=
rg/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY" rel=3D"noreferrer" target=3D"=
_blank">https://mailarchive.ietf.org/<wbr>arch/msg/acme/0lVmGl8e-<wbr>rmSH0=
x7ycDW5dj6GAY</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0&lt;<a href=3D"https://mailarchive.ie=
tf.org/arch/msg/acme/0lVmGl8e-rmSH0x7ycDW5dj6GAY" rel=3D"noreferrer" target=
=3D"_blank">https://mailarchive.ietf.org/<wbr>arch/msg/acme/0lVmGl8e-<wbr>r=
mSH0x7ycDW5dj6GAY</a>&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0[1]<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"https://community.letsencr=
ypt.org/t/upcoming-api-changes/17947" rel=3D"noreferrer" target=3D"_blank">=
https://community.letsencrypt.<wbr>org/t/upcoming-api-changes/<wbr>17947</a=
><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0&lt;<a href=3D"https://community.lets=
encrypt.org/t/upcoming-api-changes/17947" rel=3D"noreferrer" target=3D"_bla=
nk">https://community.<wbr>letsencrypt.org/t/upcoming-<wbr>api-changes/1794=
7</a>&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0[2]<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"https://community.letsencr=
ypt.org/t/automatic-recycling-of-pending-authorizations/41321" rel=3D"noref=
errer" target=3D"_blank">https://community.letsencrypt.<wbr>org/t/automatic=
-recycling-of-<wbr>pending-authorizations/41321</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0&lt;<a href=3D"https://community.lets=
encrypt.org/t/automatic-recycling-of-pending-authorizations/41321" rel=3D"n=
oreferrer" target=3D"_blank">https://community.<wbr>letsencrypt.org/t/autom=
atic-<wbr>recycling-of-pending-<wbr>authorizations/41321</a>&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0[3] <a href=3D"https://letsencrypt.or=
g/docs/rate-limits/" rel=3D"noreferrer" target=3D"_blank">https://letsencry=
pt.org/docs/<wbr>rate-limits/</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0&lt;<a href=3D"https://letsencrypt.or=
g/docs/rate-limits/" rel=3D"noreferrer" target=3D"_blank">https://letsencry=
pt.org/docs/<wbr>rate-limits/</a>&gt;<br>
&gt;<br>
&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0______________________________<wbr>__=
_______________<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Acme mailing list<br>
</div></div>&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"mailto:Acme@ie=
tf.org">Acme@ietf.org</a> &lt;mailto:<a href=3D"mailto:Acme@ietf.org">Acme@=
ietf.org</a>&gt;<br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0<a href=3D"https://www.ietf.org/mailm=
an/listinfo/acme" rel=3D"noreferrer" target=3D"_blank">https://www.ietf.org=
/mailman/<wbr>listinfo/acme</a><br>
&gt;=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0&lt;<a href=3D"https://www.ietf.org/m=
ailman/listinfo/acme" rel=3D"noreferrer" target=3D"_blank">https://www.ietf=
.org/mailman/<wbr>listinfo/acme</a>&gt;<br>
<div class=3D"HOEnZb"><div class=3D"h5">&gt;<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt; ______________________________<wbr>_________________<br>
&gt; Acme mailing list<br>
&gt; <a href=3D"mailto:Acme@ietf.org">Acme@ietf.org</a><br>
&gt; <a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferr=
er" target=3D"_blank">https://www.ietf.org/mailman/<wbr>listinfo/acme</a><b=
r>
&gt;<br>
<br>
______________________________<wbr>_________________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/<wbr>listinfo/acme</a><br>
</div></div></blockquote></div><br></div>

--001a1147cc66680b69055a184933--


From nobody Tue Sep 26 10:34:41 2017
Return-Path: <agwa@andrewayer.name>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7965B1330AD for <acme@ietfa.amsl.com>; Tue, 26 Sep 2017 10:34:39 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.001
X-Spam-Level: 
X-Spam-Status: No, score=-2.001 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=andrewayer.name
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 fTAryTyCskHL for <acme@ietfa.amsl.com>; Tue, 26 Sep 2017 10:34:38 -0700 (PDT)
Received: from alcazar.beanwood.com (alcazar.beanwood.com [70.85.129.230]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 7C7E1132D51 for <acme@ietf.org>; Tue, 26 Sep 2017 10:34:38 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=andrewayer.name; s=beanwood20160511; t=1506447277; bh=Ypn1oWvoiyrZNxJG1CBKF6wD/4jjWCTFR0bhZcE6Vks=; h=Date:From:To:Subject:In-Reply-To:References; b=IZGJyCty+c6GH+o0CCoFk4p2f5hky8CPxPHYw5WA4MiO0jLlWTYzPKi2ymf7NDlvw 3Vi5Y7U/FhM3yDr7v1BzHdMWSEzfw3EMFokAG2kLx56V/mjWv43bgduwntPU78f0GI ruuVBTmr0eV7e3887R7Yq3y4WdpMUZjQzb02V6IQOc+RCDCclowxySca+44ciNPE9E e/JtD7jq3Ma74NQSaD036pkqyGlb9/SariwYlrx062lGXfJj2q70l7kO72Zb8iWCYT 1BoFZ7lDWOsr3AovQQdZpV7II1juNh8QpU2g2zRNKhR+N150G91GuuWzbQfxl+ysys KlQJRaasCMFJA==
Date: Tue, 26 Sep 2017 10:34:36 -0700
From: Andrew Ayer <agwa@andrewayer.name>
To: acme@ietf.org
Message-Id: <20170926103436.53c38e1990aa7d43a7e8f780@andrewayer.name>
In-Reply-To: <c9a26f93-b134-1d69-b21e-080d34bdcc30@letsencrypt.org>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com> <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com> <c9a26f93-b134-1d69-b21e-080d34bdcc30@letsencrypt.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/wOBxjQhRyIKQKKlf7sGnSdD7mtg>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 26 Sep 2017 17:34:39 -0000

FWIW, I previously expressed support for proactive issuance because 
I thought it would lay the groundwork for a better renewal process,
especially for short-lived certificates.  However, this idea never
gained traction and the other necessary bits weren't added.  Instead,
STAR seems to be handling this use case.

Therefore, I no longer see a compelling reason for proactive issuance.
If it's causing problems it should be removed.  I also agree with Roland
that issuing upon an unauthenticated GET request is a bad idea.

Regards,
Andrew


From nobody Wed Sep 27 06:25:32 2017
Return-Path: <dmccarney@letsencrypt.org>
X-Original-To: acme@ietfa.amsl.com
Delivered-To: acme@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9138D1349E4 for <acme@ietfa.amsl.com>; Wed, 27 Sep 2017 06:25:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=letsencrypt.org
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 I6iVVtw-44PA for <acme@ietfa.amsl.com>; Wed, 27 Sep 2017 06:25:28 -0700 (PDT)
Received: from mail-io0-x230.google.com (mail-io0-x230.google.com [IPv6:2607:f8b0:4001:c06::230]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 85DD6134B89 for <acme@ietf.org>; Wed, 27 Sep 2017 06:24:12 -0700 (PDT)
Received: by mail-io0-x230.google.com with SMTP id w94so15624092ioi.7 for <acme@ietf.org>; Wed, 27 Sep 2017 06:24:12 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=letsencrypt.org; s=google; h=mime-version:reply-to:in-reply-to:references:from:date:message-id :subject:to; bh=dWmVBPsF0GVn+p159P284UDZ/6fmBmChLUHEkPbhWAQ=; b=ILNAtF4+e+ERkt3qpw97dmObUR/pjyLenkn7S4zVpjHL1TWtA3Ktc2OTB44iCuFoFH Dbirzg/2mJfWrJLe4xwCQ+QfOXaUIb03vzbhhC6K7mJDznmj66vzRX0oX+jrFJK3ToFw vl8IIMRd6RnYrCIUDYkkt1oIKGH+SBf00dCcI=
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:reply-to:in-reply-to:references :from:date:message-id:subject:to; bh=dWmVBPsF0GVn+p159P284UDZ/6fmBmChLUHEkPbhWAQ=; b=dt+wz5ktUCJxm+ZbuC5ToO4X+jpN4oSHmzjqkp9GzEpMfv4CxMxnvlH3wqxhIM6lYX LB5fTw+i6mtX8yHWfnqrvO+7Zz7pUR1W4e3Vp7XpQPAp/+VOzxqPaCsUeuxs9qXs70Hj yUKC62pe7olHq4pJnGQD91IYz08sFV77ACF9kr4jaIa3UyFADc4XAOxXVovx8nIo0GT7 Qpp4iQ2IDyf8PlX25nN2FeXMs50RStld0YdkhgDAnp1nNBW/sF0R6wbjIqvJAIcJMV2h 3NckbNMMppvf8THx34q29f+CgQ4xZt5MLl7oXdZfDjtY11CRj605byxisk6g8459h0Av Hejw==
X-Gm-Message-State: AMCzsaUGJ0uZ1jy0AnUfZNv7z0jwZz/tAjzfea0xf5aM5xIJ5Kg24hFx XKgG3fa1wAQhOE5NEK3VP+uvc93lJmEe3VvoFmlpnU3Q
X-Google-Smtp-Source: AOwi7QAD1xL9LkBWPwpnr9NKvtHBbQrqNSAvVQ1SqNI+sg+urbYbtmMbsXjoV5l/fr2bqwfvcNvVc2jEbDEAyWlMwNs=
X-Received: by 10.107.88.18 with SMTP id m18mr1940575iob.192.1506518651435; Wed, 27 Sep 2017 06:24:11 -0700 (PDT)
MIME-Version: 1.0
Reply-To: cpu@letsencrypt.org
Received: by 10.107.17.229 with HTTP; Wed, 27 Sep 2017 06:24:10 -0700 (PDT)
In-Reply-To: <20170926103436.53c38e1990aa7d43a7e8f780@andrewayer.name>
References: <CAKnbcLi20Z1pQ7-V3AfJ-97ULJOy5r0BFCUMnzUawS__m7q2zw@mail.gmail.com> <CAL02cgQ9xfVNj049K8JGzAWXDENORpvFPNTjPpyh+Ee_sJG=_g@mail.gmail.com> <CAL02cgSxYa7R8CGDuDZg+OvqFQwuf=FahDrcVLE+kTViMXMwvQ@mail.gmail.com> <c9a26f93-b134-1d69-b21e-080d34bdcc30@letsencrypt.org> <20170926103436.53c38e1990aa7d43a7e8f780@andrewayer.name>
From: Daniel McCarney <cpu@letsencrypt.org>
Date: Wed, 27 Sep 2017 09:24:10 -0400
Message-ID: <CAKnbcLh4vbKjbGB1BnpymfNeKQoW0gA=n_nJGb+-fMv0GpGJ3Q@mail.gmail.com>
To: IETF ACME <acme@ietf.org>, Richard Barnes <rlb@ipv.sx>
Content-Type: multipart/alternative; boundary="f403043cbcb4111a9d055a2bb530"
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/cvHgNOnzXzPPqPt293mQn57kvfI>
Subject: Re: [Acme] Revisiting Proactive Issuance & new-order CSR
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
Precedence: list
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 27 Sep 2017 13:25:31 -0000

--f403043cbcb4111a9d055a2bb530
Content-Type: text/plain; charset="UTF-8"

Hi again,

Richard: Do you have any subsequent thoughts now that Roland and myself
have given feedback on your proposed changes and why we still prefer the
suggestions I made initially?

Andrew's original support of proactive issuance was the most constructive
vote for why that behaviour was preferred during the initial discussion
period. Given that his support is now withdrawn and Roland and Jacob have
voiced support for my proposal I feel like we are approaching consensus.
There has not been any arguments made for the benefit of proactive issuance.

Boulder is beginning implementation of my proposed specification update and
I feel the best way to minimize Rich Salz's concerns about cycle disruption
is to resolve this thread quickly.

Thanks all,

On Tue, Sep 26, 2017 at 1:34 PM, Andrew Ayer <agwa@andrewayer.name> wrote:

> FWIW, I previously expressed support for proactive issuance because
> I thought it would lay the groundwork for a better renewal process,
> especially for short-lived certificates.  However, this idea never
> gained traction and the other necessary bits weren't added.  Instead,
> STAR seems to be handling this use case.
>
> Therefore, I no longer see a compelling reason for proactive issuance.
> If it's causing problems it should be removed.  I also agree with Roland
> that issuing upon an unauthenticated GET request is a bad idea.
>
> Regards,
> Andrew
>
> _______________________________________________
> Acme mailing list
> Acme@ietf.org
> https://www.ietf.org/mailman/listinfo/acme
>

--f403043cbcb4111a9d055a2bb530
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hi again,<br><br>Richard: Do you have any subsequent thoug=
hts now that Roland and myself have given feedback on your proposed changes=
 and why we still prefer the suggestions I made initially?<br><br>Andrew&#3=
9;s original support of proactive issuance was the most constructive vote f=
or why that behaviour was preferred during the initial discussion period. G=
iven that his support is now withdrawn and Roland and Jacob have voiced sup=
port for my proposal I feel like we are approaching consensus. There has no=
t been any arguments made for the benefit of proactive issuance.<br><br>Bou=
lder is beginning implementation of my proposed specification update and I =
feel the best way to minimize Rich Salz&#39;s concerns about cycle disrupti=
on is to resolve this thread quickly.<br><br>Thanks all,</div><div class=3D=
"gmail_extra"><br><div class=3D"gmail_quote">On Tue, Sep 26, 2017 at 1:34 P=
M, Andrew Ayer <span dir=3D"ltr">&lt;<a href=3D"mailto:agwa@andrewayer.name=
" target=3D"_blank">agwa@andrewayer.name</a>&gt;</span> wrote:<br><blockquo=
te class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc so=
lid;padding-left:1ex">FWIW, I previously expressed support for proactive is=
suance because<br>
I thought it would lay the groundwork for a better renewal process,<br>
especially for short-lived certificates.=C2=A0 However, this idea never<br>
gained traction and the other necessary bits weren&#39;t added.=C2=A0 Inste=
ad,<br>
STAR seems to be handling this use case.<br>
<br>
Therefore, I no longer see a compelling reason for proactive issuance.<br>
If it&#39;s causing problems it should be removed.=C2=A0 I also agree with =
Roland<br>
that issuing upon an unauthenticated GET request is a bad idea.<br>
<br>
Regards,<br>
Andrew<br>
<div class=3D"HOEnZb"><div class=3D"h5"><br>
______________________________<wbr>_________________<br>
Acme mailing list<br>
<a href=3D"mailto:Acme@ietf.org">Acme@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/acme" rel=3D"noreferrer" t=
arget=3D"_blank">https://www.ietf.org/mailman/<wbr>listinfo/acme</a><br>
</div></div></blockquote></div><br></div>

--f403043cbcb4111a9d055a2bb530--


From nobody Thu Sep 28 15:03:43 2017
Return-Path: <session-request@ietf.org>
X-Original-To: acme@ietf.org
Delivered-To: acme@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 027CE134992; Thu, 28 Sep 2017 15:03:41 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: IETF Meeting Session Request Tool <session-request@ietf.org>
To: <session-request@ietf.org>
Cc: Kathleen.Moriarty.ietf@gmail.com, acme@ietf.org, ynir.ietf@gmail.com, acme-chairs@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 6.63.0
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <150663622096.14108.15562835927070476473.idtracker@ietfa.amsl.com>
Date: Thu, 28 Sep 2017 15:03:40 -0700
Archived-At: <https://mailarchive.ietf.org/arch/msg/acme/qy5Z176PCjI5ZRUFBDoIjxo1Ja8>
Subject: [Acme] acme - New Meeting Session Request for IETF 100
X-BeenThere: acme@ietf.org
X-Mailman-Version: 2.1.22
List-Id: Automated Certificate Management Environment <acme.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/acme>, <mailto:acme-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/acme/>
List-Post: <mailto:acme@ietf.org>
List-Help: <mailto:acme-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/acme>, <mailto:acme-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 28 Sep 2017 22:03:41 -0000

A new meeting session request has just been submitted by Yoav Nir, a Chair of the acme working group.


---------------------------------------------------------
Working Group Name: Automated Certificate Management Environment
Area Name: Security Area
Session Requester: Yoav Nir

Number of Sessions: 1
Length of Session(s):  1.5 Hours
Number of Attendees: 70
Conflicts to Avoid: 
 First Priority:  tls curdle cfrg ipsecme i2nsf




People who must be present:
  Rich Salz
  Kathleen Moriarty
  Yoav Nir

Resources Requested:

Special Requests:
  
---------------------------------------------------------

