
From nobody Wed Mar 11 11:08:56 2015
Return-Path: <stephen.farrell@cs.tcd.ie>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 67D141A1B8F for <unbearable@ietfa.amsl.com>; Wed, 11 Mar 2015 11:08:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.21
X-Spam-Level: 
X-Spam-Status: No, score=-4.21 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
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 WcBINf3ObkzF for <unbearable@ietfa.amsl.com>; Wed, 11 Mar 2015 11:08:52 -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 DB71B1A1BAC for <unbearable@ietf.org>; Wed, 11 Mar 2015 11:08:47 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by mercury.scss.tcd.ie (Postfix) with ESMTP id 6AB94BF08; Wed, 11 Mar 2015 18:08:46 +0000 (GMT)
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 IeHGw-Wd5xK4; Wed, 11 Mar 2015 18:08:46 +0000 (GMT)
Received: from [134.226.36.180] (stephen-think.dsg.cs.tcd.ie [134.226.36.180]) by mercury.scss.tcd.ie (Postfix) with ESMTPSA id 37989BF07; Wed, 11 Mar 2015 18:08:46 +0000 (GMT)
Message-ID: <550084AE.5060601@cs.tcd.ie>
Date: Wed, 11 Mar 2015 18:08:46 +0000
From: Stephen Farrell <stephen.farrell@cs.tcd.ie>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: "unbearable@ietf.org" <unbearable@ietf.org>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/ukY9S0JsekV6VluhsQxaSkiWtbY>
Cc: John Bradley <ve7jtb@ve7jtb.com>, Leif Johansson <leifj@sunet.se>
Subject: [Unbearable] tokbind wg chairs
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Mar 2015 18:08:55 -0000

Hi All,

I'm delighted to say that Leif Johansson and John Bradley
(cc'd) have agreed to co-chair the tokbind WG. I think
they'll be a great team to help us all get the work done
here, so please thank them for taking it on when you see
them.

The IESG has already approved the charter so you should
see the formal announcement go out in a day or so, and
Leif and John will start working on agenda items for the
Dallas meeting session (Tue 0900-1130) shortly. Please
direct any queries about that to Leif and John.

Thanks again to you all for getting this going and to
Leif and John for stepping up to chair,
Cheers,
S.


From nobody Wed Mar 11 12:13:04 2015
Return-Path: <leifj@sunet.se>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 102571A0242 for <unbearable@ietfa.amsl.com>; Wed, 11 Mar 2015 12:13:03 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.661
X-Spam-Level: 
X-Spam-Status: No, score=-1.661 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HELO_EQ_SE=0.35, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 dKcRXR8twA7W for <unbearable@ietfa.amsl.com>; Wed, 11 Mar 2015 12:13:00 -0700 (PDT)
Received: from e-mailfilter01.sunet.se (e-mailfilter01.sunet.se [IPv6:2001:6b0:8:2::201]) (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 6AEBE1A007D for <unbearable@ietf.org>; Wed, 11 Mar 2015 12:13:00 -0700 (PDT)
Received: from smtp1.sunet.se (smtp1.sunet.se [192.36.171.214]) by e-mailfilter01.sunet.se (8.14.4/8.14.4/Debian-4) with ESMTP id t2BJCrUu024416 (version=TLSv1/SSLv3 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Wed, 11 Mar 2015 20:12:53 +0100
Received: from kerio.sunet.se (kerio.sunet.se [192.36.171.210]) by smtp1.sunet.se (8.14.9/8.14.7) with ESMTP id t2BJCnk9008344 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Wed, 11 Mar 2015 20:12:52 +0100 (CET)
VBR-Info: md=sunet.se; mc=all; mv=swamid.se
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=sunet.se; s=default; t=1426101172; bh=MBDuKarCdMz77g0UO9+mn5x1wHzrp3v5tres1qJMyQY=; h=Date:From:To:CC:Subject:References:In-Reply-To; b=F4DraWPx+4pUPegv2IrLo6+pLwps3Qjeoz40pugvA37TrycZJTGnEdM1Y3U3WfN5o 2TZjyJfeuGQ164jNgolQMzGcPxeLBuPVKGveVvP30th+LV68tMU3mflHrtWDzIPCXs jXuBw1CRQwF+KE6TizvDIq2S6rmnXoBKAvjQE1vA=
X-Footer: c3VuZXQuc2U=
Received: from [10.0.0.107] ([62.102.145.131]) (authenticated user leifj@sunet.se) by kerio.sunet.se (Kerio Connect 8.3.4 patch 1) (using TLSv1.2 with cipher DHE-RSA-AES256-SHA (256 bits)); Wed, 11 Mar 2015 20:12:47 +0100
Message-ID: <550093AF.1060600@sunet.se>
Date: Wed, 11 Mar 2015 20:12:47 +0100
From: Leif Johansson <leifj@sunet.se>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: Stephen Farrell <stephen.farrell@cs.tcd.ie>, "unbearable@ietf.org" <unbearable@ietf.org>
References: <550084AE.5060601@cs.tcd.ie>
In-Reply-To: <550084AE.5060601@cs.tcd.ie>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
X-Bayes-Prob: 0.0001 (Score 0, tokens from: outbound, outbound-sunet-se:default, sunet-se:default, base:default, @@RPTN)
X-CanIt-Geo: ip=192.36.171.210; country=SE; latitude=59.3294; longitude=18.0686; http://maps.google.com/maps?q=59.3294,18.0686&z=6
X-CanItPRO-Stream: outbound-sunet-se:outbound (inherits from outbound-sunet-se:default, sunet-se:default, base:default)
X-Canit-Stats-ID: 09O2jcR9o - 395b1e587070 - 20150311
X-CanIt-Archive-Cluster: PfMRe/vJWMiXwM2YIH5BVExnUnw
Received-SPF: neutral (e-mailfilter01.sunet.se: 192.36.171.210 is neither permitted nor denied by domain leifj@sunet.se) receiver=e-mailfilter01.sunet.se; client-ip=192.36.171.210; envelope-from=<leifj@sunet.se>; helo=smtp1.sunet.se; identity=mailfrom
X-Scanned-By: CanIt (www . roaringpenguin . com) on 192.36.171.201
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/lZxIh6xoC2BOJD0wmqsCZKSPqW0>
Cc: John Bradley <ve7jtb@ve7jtb.com>
Subject: Re: [Unbearable] tokbind wg chairs
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Mar 2015 19:13:03 -0000

On 03/11/2015 07:08 PM, Stephen Farrell wrote:
> 
> Hi All,
> 
> I'm delighted to say that Leif Johansson and John Bradley
> (cc'd) have agreed to co-chair the tokbind WG. I think
> they'll be a great team to help us all get the work done
> here, so please thank them for taking it on when you see
> them.
> 
> The IESG has already approved the charter so you should
> see the formal announcement go out in a day or so, and
> Leif and John will start working on agenda items for the
> Dallas meeting session (Tue 0900-1130) shortly. Please
> direct any queries about that to Leif and John.
> 
> Thanks again to you all for getting this going and to
> Leif and John for stepping up to chair,
> Cheers,
> S.
> 

Thanks for the vote of confidence Stephen!

	Cheers Leif


From nobody Wed Mar 11 12:27:52 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7C5211A016A for <unbearable@ietfa.amsl.com>; Wed, 11 Mar 2015 12:27:50 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 i9IfqSek4YaW for <unbearable@ietfa.amsl.com>; Wed, 11 Mar 2015 12:27:48 -0700 (PDT)
Received: from mail-qg0-f42.google.com (mail-qg0-f42.google.com [209.85.192.42]) (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 59C6A1A00C7 for <unbearable@ietf.org>; Wed, 11 Mar 2015 12:27:48 -0700 (PDT)
Received: by qgdq107 with SMTP id q107so12675223qgd.6 for <unbearable@ietf.org>; Wed, 11 Mar 2015 12:27:47 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:message-id:references:to; bh=+ZniTdWa8JztoTIdbLCqrU2iJHiq09TlDOS5MPsRN4Y=; b=NLyFYwb9apALdEdtjgW9miOpQ4GlM7UfeS/8N4L4IAZvM9EvE7DJDxwUVhn3+57xbO mnTdVqJSN3xYRQ3Y+CDl9ZF6EARvNrOSsCk08hlKpPNUbsKM3ZH2m8YU//LYmyXZJbhR IQIYOUDhyaY4ssTqTIpZqDZ4xZkaiu+Wq7EpwAe213v503DA5sVwrxLIAkUvSxRdfM5K JejyKPzbD71ipzw/YD9fB+iQhT7/vjokREEsBolTds5sF4H3wTuptIN0xr9Jop56k7Cr DkeVqiWLBkJ+i6Pfy/eVu7EG08+hdFs9/YdNVwOKYDrVQ8ln9N41eBSfPmZ7rJcs515P fSrg==
X-Gm-Message-State: ALoCoQnY2p2OYksQ9RJSC9W7fD4jEgVUFmcv3Mg3C4DdR0WQXokNn+1FOyvdwB99WXYQghTGHvY3
X-Received: by 10.229.226.71 with SMTP id iv7mr16084988qcb.13.1426102067444; Wed, 11 Mar 2015 12:27:47 -0700 (PDT)
Received: from [192.168.8.100] ([181.202.7.165]) by mx.google.com with ESMTPSA id a92sm3194471qgf.11.2015.03.11.12.27.45 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 11 Mar 2015 12:27:46 -0700 (PDT)
Content-Type: multipart/signed; boundary="Apple-Mail=_20D062C7-491C-41F8-B023-32DC7EEE5F76"; protocol="application/pkcs7-signature"; micalg=sha1
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: John Bradley <ve7jtb@ve7jtb.com>
In-Reply-To: <550093AF.1060600@sunet.se>
Date: Wed, 11 Mar 2015 16:27:42 -0300
Message-Id: <A22A159B-258E-44AF-81EC-CC63EB2D19EC@ve7jtb.com>
References: <550084AE.5060601@cs.tcd.ie> <550093AF.1060600@sunet.se>
To: Leif Johansson <leifj@sunet.se>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/rEN36i3hvq2itv4KoDaGAbzRcoQ>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Stephen Farrell <stephen.farrell@cs.tcd.ie>
Subject: Re: [Unbearable] tokbind wg chairs
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Mar 2015 19:27:50 -0000

--Apple-Mail=_20D062C7-491C-41F8-B023-32DC7EEE5F76
Content-Transfer-Encoding: 7bit
Content-Type: text/plain;
	charset=us-ascii

Yes, thanks Stephen.

I am hoping that we can make quick progress on the specs.

John B.

> On Mar 11, 2015, at 4:12 PM, Leif Johansson <leifj@sunet.se> wrote:
> 
> On 03/11/2015 07:08 PM, Stephen Farrell wrote:
>> 
>> Hi All,
>> 
>> I'm delighted to say that Leif Johansson and John Bradley
>> (cc'd) have agreed to co-chair the tokbind WG. I think
>> they'll be a great team to help us all get the work done
>> here, so please thank them for taking it on when you see
>> them.
>> 
>> The IESG has already approved the charter so you should
>> see the formal announcement go out in a day or so, and
>> Leif and John will start working on agenda items for the
>> Dallas meeting session (Tue 0900-1130) shortly. Please
>> direct any queries about that to Leif and John.
>> 
>> Thanks again to you all for getting this going and to
>> Leif and John for stepping up to chair,
>> Cheers,
>> S.
>> 
> 
> Thanks for the vote of confidence Stephen!
> 
> 	Cheers Leif
> 


--Apple-Mail=_20D062C7-491C-41F8-B023-32DC7EEE5F76
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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--Apple-Mail=_20D062C7-491C-41F8-B023-32DC7EEE5F76--


From nobody Wed Mar 11 20:44:54 2015
Return-Path: <iesg-secretary@ietf.org>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4570D1A88CA; Wed, 11 Mar 2015 15:13:08 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -101.9
X-Spam-Level: 
X-Spam-Status: No, score=-101.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, USER_IN_WHITELIST=-100] autolearn=ham
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 Zyqc1_4plgPn; Wed, 11 Mar 2015 15:13:06 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id CD00C1A88D5; Wed, 11 Mar 2015 15:13:05 -0700 (PDT)
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
From: The IESG <iesg-secretary@ietf.org>
To: "IETF-Announce" <ietf-announce@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.1
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150311221305.2584.75514.idtracker@ietfa.amsl.com>
Date: Wed, 11 Mar 2015 15:13:05 -0700
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/wbq-1OUMrooHExzKUudoLD2_aZ0>
X-Mailman-Approved-At: Wed, 11 Mar 2015 20:44:52 -0700
Cc: tokbind WG <unbearable@ietf.org>
Subject: [Unbearable] WG Action: Formed Token Binding (tokbind)
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 11 Mar 2015 22:13:08 -0000

A new IETF working group has been formed in the Security Area. For
additional information please contact the Area Directors or the WG
Chairs.

Token Binding (tokbind)
------------------------------------------------
Current Status: Proposed WG

Chairs:
  John Bradley <ve7jtb@ve7jtb.com>
  Leif Johansson <leifj@sunet.se>

Assigned Area Director:
  Stephen Farrell <stephen.farrell@cs.tcd.ie>

Mailing list
  Address: unbearable@ietf.org
  To Subscribe: https://www.ietf.org/mailman/listinfo/unbearable
  Archive: http://www.ietf.org/mail-archive/web/unbearable/

Charter:


Web services generate various security tokens (e.g. HTTP cookies, OAuth
tokens, etc.) for web applications to access protected resources.
Currently these are bearer tokens, i.e. any party in possession of such
token gains access to the protected resource. Attackers export bearer
tokens from client machines or from compromised network connections,
present these bearer tokens to Web services, and impersonate
authenticated users. Token Binding enables defense against such attacks
by  cryptographically binding security tokens to a secret held by the
client.

The tasks of this working group are as follows:

1. Specify the Token Binding protocol v1.0.
2. Specify the use of the Token Binding protocol in combination with
HTTPS.

It is a goal of this working group to enable defense against attacks that
involve unauthorized replay of security tokens. Other issues associated
with the use of security tokens are out of scope. Another goal of this
working group is to design the Token Binding protocol such that it would
be also useable with application protocols other than HTTPS. Specifying
alternative application protocols is not a primary goal. 

The main design objectives for the Token Binding protocol, in no
particular order:

1. Allow applications and services to prevent unauthorized replay of
security tokens.
2. Allow strong key protection, e.g. using hardware-bound keys.
3. Support both first-party (server generates a token for later use with
this server) and federation (server generates a token for use with
another server) scenarios.
4. Preserve user privacy.
5. Make the Token Binding protocol useable in combination with a variety
of application protocols.
6. Allow the negotiation of the Token Binding protocol without additional
round-trips.
7. Allow the use of multiple cryptographic algorithms, so that a variety
of secure    hardware modules with different cryptographic capabilities
could be used with Token Binding.
8. Propose Token Binding specification that can be implemented in Web
browsers (but is not limited to them). E.g. Web browsers require that the
same bound security token must be presentable over multiple TLS sessions
and connections.

The working group will use the following documents as a starting point
for its work:

- draft-popov-token-binding-00;
- draft-balfanz-https-token-binding-00.

This WG will collaborate with other IETF WGs, in particular with the TLS,
HTTPbis and Oauth WGs and with the W3C webappsec WG.


Milestones:
  Jan 2016 - HTTPS Token Binding to IESG.
  Jan 2016 - WG document for the Token Binding Protocol v1.0.
  Jan 2016 - WG document for HTTPS Token Binding.
  Jan 2016 - Token Binding Protocol v1.0 to IESG.



From nobody Fri Mar 13 09:57:41 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8633E1A8715 for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 09:57:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 rAYLLqE4m9Up for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 09:57:36 -0700 (PDT)
Received: from mail-qc0-f178.google.com (mail-qc0-f178.google.com [209.85.216.178]) (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 C41DC1A8779 for <unbearable@ietf.org>; Fri, 13 Mar 2015 09:57:36 -0700 (PDT)
Received: by qcxm20 with SMTP id m20so28002770qcx.3 for <unbearable@ietf.org>; Fri, 13 Mar 2015 09:57:36 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:content-type:subject:message-id:date:to :mime-version; bh=CukZDOCZuS/cdCD9rJuNaBM40qHsgYVisFQUogyQWns=; b=AIqSFJlTecZ6JqxnpjZOcxjJNPJNV5Mrf1UEsNZ4/jPjlycyHjjq5C1yrXa4H8BSUy GMIhDGT0ExF/EyUNLviUsOr0g8IPEfAkQ2OTPS/HD85IkXbgHwobdTBaESxL6M1ptSkj 0jI6WOL5wy8H7nLasT7SyyeyncH9iJWdjCJTuy6PI3nJ1yvEV38jPNTyigiEb41zzVW2 +K5QD6UPVbMQml0pvcP3TFLslUE99dRYobDSmIfCUf9UeG7S84HSTqicygZQZOnjDVUZ TdcYvJrTqWJRSt6Y+Mog2R1ZpXRZL/+N8ina2lf+Dd6i3pewn3t4Z3+yBfzTfZRxkDa8 ZCBg==
X-Gm-Message-State: ALoCoQl1iHCcf30s91PDNjkXknkPVIo79Gd1EshfHCaOnU/i/kY/S1fq3PNH78GlCGYETTX/ioKR
X-Received: by 10.140.145.9 with SMTP id 9mr60800295qhr.104.1426265855964; Fri, 13 Mar 2015 09:57:35 -0700 (PDT)
Received: from [192.168.1.36] (186-79-112-76.baf.movistar.cl. [186.79.112.76]) by mx.google.com with ESMTPSA id n34sm1748221qge.27.2015.03.13.09.57.22 for <unbearable@ietf.org> (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Fri, 13 Mar 2015 09:57:35 -0700 (PDT)
From: John Bradley <ve7jtb@ve7jtb.com>
Content-Type: multipart/signed; boundary="Apple-Mail=_B7DF3D3D-C85E-4B04-B837-495722C94DF6"; protocol="application/pkcs7-signature"; micalg=sha1
Message-Id: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com>
Date: Fri, 13 Mar 2015 13:56:46 -0300
To: unbearable@ietf.org
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/-zvN8B8j8BfwaTlPlBrG7-tk6xo>
Subject: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Mar 2015 16:57:38 -0000

--Apple-Mail=_B7DF3D3D-C85E-4B04-B837-495722C94DF6
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

Now that we are officially a WG I have a question (no hat)

One thing that I identified as a potential issue talking to people who =
are working on early deployments, is the expectations around how keys =
are shared between applications.

I found some people that believe all apps sharing a common TLS library =
would be able to share the same ID/key for a origin.

Other folks were under the impression that each application would have =
its own key for an origin.

The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.

There was a middle view by the MDM people that the ID/key would be =
scoped to a profile/container so that all the enterprise apps would =
share ID/key but that social apps would not.

I don't know that there is necessarily one right answer, as this may be =
implementation dependent. =20

I don't think there is anything in the specs on this.   One reason I ask =
is that some of those differences impact on Native Application Single =
Signon and OAuth.

In some of the Windows prototypes I understand there is a Key Storage =
API that is used to share keys between the browser and apps. =20

Android, Windows, iOS , OSX may all have implementation specific =
details, but people building applications that leverage token binding =
will need to understand the differences as they may have security and =
privacy implications.

Figuring out where some of these issues get documented might be useful.

John B.=

--Apple-Mail=_B7DF3D3D-C85E-4B04-B837-495722C94DF6
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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--Apple-Mail=_B7DF3D3D-C85E-4B04-B837-495722C94DF6--


From nobody Fri Mar 13 10:19:00 2015
Return-Path: <Andrei.Popov@microsoft.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B6D301A8AA5 for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 10:18:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
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 JaMpiRwg6xLg for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 10:18:55 -0700 (PDT)
Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0123.outbound.protection.outlook.com [65.55.169.123]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2547E1A8AB2 for <unbearable@ietf.org>; Fri, 13 Mar 2015 10:18:55 -0700 (PDT)
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com (25.161.207.26) by BN3PR0301MB1249.namprd03.prod.outlook.com (25.161.207.25) with Microsoft SMTP Server (TLS) id 15.1.106.15; Fri, 13 Mar 2015 17:18:53 +0000
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) by BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) with mapi id 15.01.0106.007; Fri, 13 Mar 2015 17:18:52 +0000
From: Andrei Popov <Andrei.Popov@microsoft.com>
To: John Bradley <ve7jtb@ve7jtb.com>, "unbearable@ietf.org" <unbearable@ietf.org>
Thread-Topic: [Unbearable] Scoping of keys on the client.
Thread-Index: AQHQXa7TliKvYzf3qUabRBSpT6xKBZ0ao/aQ
Date: Fri, 13 Mar 2015 17:18:52 +0000
Message-ID: <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com>
In-Reply-To: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [2001:4898:80e8:ed31::2]
authentication-results: ve7jtb.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0301MB1249;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(377454003)(13464003)(52604005)(87936001)(2656002)(122556002)(40100003)(33656002)(92566002)(102836002)(2900100001)(46102003)(2501003)(2950100001)(99286002)(76576001)(62966003)(77156002)(106116001)(86362001)(50986999)(86612001)(76176999)(54356999)(19580405001)(19580395003)(107886001)(74316001)(3826002); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0301MB1249; H:BN3PR0301MB1250.namprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BN3PR0301MB124927B47247EE316D99F10B8C070@BN3PR0301MB1249.namprd03.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002009)(5005006); SRVR:BN3PR0301MB1249; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0301MB1249; 
x-forefront-prvs: 05143A8241
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: microsoft.onmicrosoft.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 13 Mar 2015 17:18:52.8448 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0301MB1249
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/7fJay3Jn9uRSxBNOgK91PA0rdYg>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Mar 2015 17:18:58 -0000

Thanks for raising this topic John.

> I found some people that believe all apps sharing a common TLS library wo=
uld be able to share the same ID/key for a origin.
In our prototypes so far (in Win10), the Token Binding Protocol implementat=
ion, the Token Binding key storage APIs, and the TLS library are three sepa=
rate components. Among other things, this allows 3-rd-party apps share Toke=
n Binding keys and achieve single sign-on regardless of the TLS library the=
y might be using (if they choose to do so).

> The different views impact security considerations in native environments=
.   As an example can you pass a token from one app to another and have it =
still work to the same origin, or is it a security feature that it won't wo=
rk and the token cant be stolen from a app and re-used by a different app o=
n the same device.
I think that single sign-on between apps and browsers is at least a useful =
option.

> I don't think there is anything in the specs on this.
> Figuring out where some of these issues get documented might be useful.
Documenting key scoping is indeed useful; on the other hand standardizing t=
he specifics might require standardizing user and application isolation mod=
els across platforms, which I don't think would be in scope for this forum.=
 Perhaps we could agree on some general principles and include implementati=
on guidance on Token Binding key scoping?

Cheers,

Andrei

-----Original Message-----
From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John Bra=
dley
Sent: Friday, March 13, 2015 9:57 AM
To: unbearable@ietf.org
Subject: [Unbearable] Scoping of keys on the client.

Now that we are officially a WG I have a question (no hat)

One thing that I identified as a potential issue talking to people who are =
working on early deployments, is the expectations around how keys are share=
d between applications.

I found some people that believe all apps sharing a common TLS library woul=
d be able to share the same ID/key for a origin.

Other folks were under the impression that each application would have its =
own key for an origin.

The different views impact security considerations in native environments. =
  As an example can you pass a token from one app to another and have it st=
ill work to the same origin, or is it a security feature that it won't work=
 and the token cant be stolen from a app and re-used by a different app on =
the same device.

There was a middle view by the MDM people that the ID/key would be scoped t=
o a profile/container so that all the enterprise apps would share ID/key bu=
t that social apps would not.

I don't know that there is necessarily one right answer, as this may be imp=
lementation dependent. =20

I don't think there is anything in the specs on this.   One reason I ask is=
 that some of those differences impact on Native Application Single Signon =
and OAuth.

In some of the Windows prototypes I understand there is a Key Storage API t=
hat is used to share keys between the browser and apps. =20

Android, Windows, iOS , OSX may all have implementation specific details, b=
ut people building applications that leverage token binding will need to un=
derstand the differences as they may have security and privacy implications=
.

Figuring out where some of these issues get documented might be useful.

John B.


From nobody Fri Mar 13 10:43:41 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5F90E1A8758 for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 10:43:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 kB-8TvOq640R for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 10:43:27 -0700 (PDT)
Received: from mail-qc0-f179.google.com (mail-qc0-f179.google.com [209.85.216.179]) (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 1E7CF1A1AB8 for <unbearable@ietf.org>; Fri, 13 Mar 2015 10:43:27 -0700 (PDT)
Received: by qcvx3 with SMTP id x3so28422339qcv.0 for <unbearable@ietf.org>; Fri, 13 Mar 2015 10:43:26 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:message-id:references:to; bh=2UVo1DrNvURGqDk33/VyS5SJY8OeDtnk0ZfKJy3aOBk=; b=nEBAPfWffxulLaDQ5zu9P15aVsJznXq/E9Ryqtd5y2pvjxp0J/B6U3Qp5X+ixj4gAa BoA3jykLNNy9SwnyvQ0RLPmRstqE2fDAQW2qkvyPn3GAAw+ZqAYGiIZ50kiqunG6Af3t 0DtFK+6H76m/oE29l+YcB+y20jWqgdHJ5mW8TxIxbZCdFLPwEynau+zLzV1ziWbpnLjf C6q0Z7EtSflH9VZsuAh/7WG8jkeRiEGuqiDgUVbiUoVssaLz2KyNSeDxoT2q2ox/OuPj JYdw8PXh8+LzmXPkExKIE5i/seM2b5TzXTRyR/RJSj/iIRb8RFWZQQ5MB8hgZ/bvueMS nOyg==
X-Gm-Message-State: ALoCoQm2RY5azEjnVT2DtPNu4awBFZKZP8m+Ng3Rm5vqN9dyCrpc93SguGUIjgtTTK9yQN9aRKk+
X-Received: by 10.140.134.198 with SMTP id 189mr61972138qhg.7.1426268606376; Fri, 13 Mar 2015 10:43:26 -0700 (PDT)
Received: from [192.168.8.101] ([181.202.25.120]) by mx.google.com with ESMTPSA id 74sm1843292qgp.9.2015.03.13.10.43.24 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Fri, 13 Mar 2015 10:43:25 -0700 (PDT)
Content-Type: multipart/signed; boundary="Apple-Mail=_B947BC32-ECBF-4169-9617-915530DCFF20"; protocol="application/pkcs7-signature"; micalg=sha1
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: John Bradley <ve7jtb@ve7jtb.com>
In-Reply-To: <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com>
Date: Fri, 13 Mar 2015 14:43:21 -0300
Message-Id: <B1944375-14CE-4870-9191-D7CD6E2FD434@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com>
To: Andrei Popov <Andrei.Popov@microsoft.com>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/7JVWwGxxnIjk2r2vmk01UZD3t6A>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Mar 2015 17:43:35 -0000

--Apple-Mail=_B947BC32-ECBF-4169-9617-915530DCFF20
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

Certainly in some potential OAuth use-cases where we want o encourage =
people to use the system browser vs a embedded web view, we may want to =
use the same key to bind a code in the browser and a native app.   =20

I think understanding how applications outside the browser or between =
browsers are going to share keys is worth putting some time into so that =
at a minimum developers will understand the issues.

In the Win10 prototype do you have some way to have a browser or app =
restrict the sharing of the key? =20
Sort of a non exportable flag, though in principal the keys themselves =
are never exportable but the use of them is shared.

John B.


> On Mar 13, 2015, at 2:18 PM, Andrei Popov <Andrei.Popov@microsoft.com> =
wrote:
>=20
> Thanks for raising this topic John.
>=20
>> I found some people that believe all apps sharing a common TLS =
library would be able to share the same ID/key for a origin.
> In our prototypes so far (in Win10), the Token Binding Protocol =
implementation, the Token Binding key storage APIs, and the TLS library =
are three separate components. Among other things, this allows =
3-rd-party apps share Token Binding keys and achieve single sign-on =
regardless of the TLS library they might be using (if they choose to do =
so).
>=20
>> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
> I think that single sign-on between apps and browsers is at least a =
useful option.
>=20
>> I don't think there is anything in the specs on this.
>> Figuring out where some of these issues get documented might be =
useful.
> Documenting key scoping is indeed useful; on the other hand =
standardizing the specifics might require standardizing user and =
application isolation models across platforms, which I don't think would =
be in scope for this forum. Perhaps we could agree on some general =
principles and include implementation guidance on Token Binding key =
scoping?
>=20
> Cheers,
>=20
> Andrei
>=20
> -----Original Message-----
> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of =
John Bradley
> Sent: Friday, March 13, 2015 9:57 AM
> To: unbearable@ietf.org
> Subject: [Unbearable] Scoping of keys on the client.
>=20
> Now that we are officially a WG I have a question (no hat)
>=20
> One thing that I identified as a potential issue talking to people who =
are working on early deployments, is the expectations around how keys =
are shared between applications.
>=20
> I found some people that believe all apps sharing a common TLS library =
would be able to share the same ID/key for a origin.
>=20
> Other folks were under the impression that each application would have =
its own key for an origin.
>=20
> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
>=20
> There was a middle view by the MDM people that the ID/key would be =
scoped to a profile/container so that all the enterprise apps would =
share ID/key but that social apps would not.
>=20
> I don't know that there is necessarily one right answer, as this may =
be implementation dependent. =20
>=20
> I don't think there is anything in the specs on this.   One reason I =
ask is that some of those differences impact on Native Application =
Single Signon and OAuth.
>=20
> In some of the Windows prototypes I understand there is a Key Storage =
API that is used to share keys between the browser and apps. =20
>=20
> Android, Windows, iOS , OSX may all have implementation specific =
details, but people building applications that leverage token binding =
will need to understand the differences as they may have security and =
privacy implications.
>=20
> Figuring out where some of these issues get documented might be =
useful.
>=20
> John B.


--Apple-Mail=_B947BC32-ECBF-4169-9617-915530DCFF20
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64

MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIINPDCCBjQw
ggQcoAMCAQICASAwDQYJKoZIhvcNAQEFBQAwfTELMAkGA1UEBhMCSUwxFjAUBgNVBAoTDVN0YXJ0
Q29tIEx0ZC4xKzApBgNVBAsTIlNlY3VyZSBEaWdpdGFsIENlcnRpZmljYXRlIFNpZ25pbmcxKTAn
BgNVBAMTIFN0YXJ0Q29tIENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA3MTAyNDIxMDI1NVoX
DTE3MTAyNDIxMDI1NVowgYwxCzAJBgNVBAYTAklMMRYwFAYDVQQKEw1TdGFydENvbSBMdGQuMSsw
KQYDVQQLEyJTZWN1cmUgRGlnaXRhbCBDZXJ0aWZpY2F0ZSBTaWduaW5nMTgwNgYDVQQDEy9TdGFy
dENvbSBDbGFzcyAyIFByaW1hcnkgSW50ZXJtZWRpYXRlIENsaWVudCBDQTCCASIwDQYJKoZIhvcN
AQEBBQADggEPADCCAQoCggEBAMsohUWcASz7GfKrpTOMKqANy9BV7V0igWdGxA8IU77L3aTxErQ+
fcxtDYZ36Z6GH0YFn7fq5RADteP0AYzrCA+EQTfi8q1+kA3m0nwtwXG94M5sIqsvs7lRP1aycBke
/s5g9hJHryZ2acScnzczjBCAo7X1v5G3yw8MDP2m2RCye0KfgZ4nODerZJVzhAlOD9YejvAXZqHk
sw56HzElVIoYSZ3q4+RJuPXXfIoyby+Y2m1E+YzX5iCZXBx05gk6MKAW1vaw4/v2OOLy6FZH3XHH
tOkzUreG//CsFnB9+uaYSlR65cdGzTsmoIK8WH1ygoXhRBm98SD7Hf/r3FELNvUCAwEAAaOCAa0w
ggGpMA8GA1UdEwEB/wQFMAMBAf8wDgYDVR0PAQH/BAQDAgEGMB0GA1UdDgQWBBSuVYNv7DHKufcd
+q9rMfPIHeOsuzAfBgNVHSMEGDAWgBROC+8apEBbpRdphzDKNGhD0EGu8jBmBggrBgEFBQcBAQRa
MFgwJwYIKwYBBQUHMAGGG2h0dHA6Ly9vY3NwLnN0YXJ0c3NsLmNvbS9jYTAtBggrBgEFBQcwAoYh
aHR0cDovL3d3dy5zdGFydHNzbC5jb20vc2ZzY2EuY3J0MFsGA1UdHwRUMFIwJ6AloCOGIWh0dHA6
Ly93d3cuc3RhcnRzc2wuY29tL3Nmc2NhLmNybDAnoCWgI4YhaHR0cDovL2NybC5zdGFydHNzbC5j
b20vc2ZzY2EuY3JsMIGABgNVHSAEeTB3MHUGCysGAQQBgbU3AQIBMGYwLgYIKwYBBQUHAgEWImh0
dHA6Ly93d3cuc3RhcnRzc2wuY29tL3BvbGljeS5wZGYwNAYIKwYBBQUHAgEWKGh0dHA6Ly93d3cu
c3RhcnRzc2wuY29tL2ludGVybWVkaWF0ZS5wZGYwDQYJKoZIhvcNAQEFBQADggIBADqpJw3I07QW
ke9plNBpxUxcffc7nUrIQpJHDci91DFG7fVhHRkMZ1J+BKg5UNUxIFJ2Z9B90Micc/NXcs7kPBRd
n6XGO/vPc87Y6R+cWS9Nc9+fp3Enmsm94OxOwI9wn8qnr/6o3mD4noP9JphwUPTXwHovjavRnhUQ
HLfo/i2NG0XXgTHXS2Xm0kVUozXqpYpAdumMiB/vezj1QHQJDmUdPYMcp+reg9901zkyT3fDW/iv
JVv6pWtkh6Pw2ytZT7mvg7YhX3V50Nv860cV11mocUVcqBLv0gcT+HBDYtbuvexNftwNQKD5193A
7zN4vG7CTYkXxytSjKuXrpEatEiFPxWgb84nVj25SU5q/r1Xhwby6mLhkbaXslkVtwEWT3Van49r
KjlK4XrUKYYWtnfzq6aSak5u0Vpxd1rY79tWhD3EdCvOhNz/QplNa+VkIsrcp7+8ZhP1l1b2U6Ma
xIVteuVMD3X0vziIwr7jxYae9FZjbxlpUemqXjcC0QaFfN7qI0JsQMALL7iGRBg7K0CoOBzECdD3
fuZil5kU/LP9cr1BK31U0Uy651bFnAMMMkqhAChIbn0ei72VnbpSsrrSdF0BAGYQ8vyHae5aCg+H
75dVCV33K6FuxZrf09yTz+Vx/PkdRUYkXmZz/OTfyJXsUOUXrym6KvI2rYpccSk5MIIHADCCBeig
AwIBAgICSAcwDQYJKoZIhvcNAQEFBQAwgYwxCzAJBgNVBAYTAklMMRYwFAYDVQQKEw1TdGFydENv
bSBMdGQuMSswKQYDVQQLEyJTZWN1cmUgRGlnaXRhbCBDZXJ0aWZpY2F0ZSBTaWduaW5nMTgwNgYD
VQQDEy9TdGFydENvbSBDbGFzcyAyIFByaW1hcnkgSW50ZXJtZWRpYXRlIENsaWVudCBDQTAeFw0x
NDAzMjQyMzU2MjNaFw0xNjAzMjUwOTM5MzFaMIGfMRkwFwYDVQQNExBxekYwMVhZQ1pNTDM4N2hE
MQswCQYDVQQGEwJDTDEiMCAGA1UECBMZTWV0cm9wb2xpdGFuYSBkZSBTYW50aWFnbzEWMBQGA1UE
BxMNSXNsYSBkZSBNYWlwbzEVMBMGA1UEAxMMSm9obiBCcmFkbGV5MSIwIAYJKoZIhvcNAQkBFhNq
YnJhZGxleUBpY2xvdWQuY29tMIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAtTL0o4QG
WC+jnmYa7xEjcBTAeIOt7ILy40qsnJHNedVaTH0EU5yHzoaEOGHuOuwJUz/C7r2TvXpJ/Ud4w6VO
HdOUGnnKUiH5MV/kIysZ7DpN5D1f+yEast00oKsEbf/D6flzfex2JFV9rT7AQ+FQaTdf3S9K7gM2
F5kODFg805BMYTGT+haw9VOMXju5s93VEjUQcnGrLy0RtoN76GM6ItxqNnEt/Ln+2GNq8JvPyUKe
JsAxfIlTyqIbw32VlusKXL4+jmgFi+LY6bsfg3VHLvy58QsQnCwHg15uARvy5X6owyGcG7xHwNml
fNWtBZ3DHNPh37HC9lmAy4iqw4PvNwIDAQABo4IDVTCCA1EwCQYDVR0TBAIwADALBgNVHQ8EBAMC
BLAwHQYDVR0lBBYwFAYIKwYBBQUHAwIGCCsGAQUFBwMEMB0GA1UdDgQWBBSUDb6BlJD7FIYgWj1w
4z+GsOXs7zAfBgNVHSMEGDAWgBSuVYNv7DHKufcd+q9rMfPIHeOsuzCBmQYDVR0RBIGRMIGOgRNq
YnJhZGxleUBpY2xvdWQuY29tgRNqYnJhZGxleUBpY2xvdWQuY29tgRdqb2huLmJyYWRsZXlAd2lu
Z2FhLmNvbYERdmU3anRiQHZlN2p0Yi5jb22BD2picmFkbGV5QG1lLmNvbYEQamJyYWRsZXlAbWFj
LmNvbYETamJyYWRsZXlAd2luZ2FhLmNvbTCCAUwGA1UdIASCAUMwggE/MIIBOwYLKwYBBAGBtTcB
AgMwggEqMC4GCCsGAQUFBwIBFiJodHRwOi8vd3d3LnN0YXJ0c3NsLmNvbS9wb2xpY3kucGRmMIH3
BggrBgEFBQcCAjCB6jAnFiBTdGFydENvbSBDZXJ0aWZpY2F0aW9uIEF1dGhvcml0eTADAgEBGoG+
VGhpcyBjZXJ0aWZpY2F0ZSB3YXMgaXNzdWVkIGFjY29yZGluZyB0byB0aGUgQ2xhc3MgMiBWYWxp
ZGF0aW9uIHJlcXVpcmVtZW50cyBvZiB0aGUgU3RhcnRDb20gQ0EgcG9saWN5LCByZWxpYW5jZSBv
bmx5IGZvciB0aGUgaW50ZW5kZWQgcHVycG9zZSBpbiBjb21wbGlhbmNlIG9mIHRoZSByZWx5aW5n
IHBhcnR5IG9ibGlnYXRpb25zLjA2BgNVHR8ELzAtMCugKaAnhiVodHRwOi8vY3JsLnN0YXJ0c3Ns
LmNvbS9jcnR1Mi1jcmwuY3JsMIGOBggrBgEFBQcBAQSBgTB/MDkGCCsGAQUFBzABhi1odHRwOi8v
b2NzcC5zdGFydHNzbC5jb20vc3ViL2NsYXNzMi9jbGllbnQvY2EwQgYIKwYBBQUHMAKGNmh0dHA6
Ly9haWEuc3RhcnRzc2wuY29tL2NlcnRzL3N1Yi5jbGFzczIuY2xpZW50LmNhLmNydDAjBgNVHRIE
HDAahhhodHRwOi8vd3d3LnN0YXJ0c3NsLmNvbS8wDQYJKoZIhvcNAQEFBQADggEBALscEldbrgeF
B1WC/hMdYxFT4Lc8ALtErgJryRozTdeMlzpsncIKyy8M54HhxQAMOqFe2HR+R9H7WeIzmkV95yJn
JY3bd4bxnnemhLrDyi1VlNjEjkK5kgegI8JavahFXl4FwJHHv8TOh71Wf3fiy0Do7d7TQmVDRrzt
1k/2w4CXKweQ2mdFw7fskiYoPGEK7pFiicGMFBzLiKRm61CqojS4IYShiP0nCZZWPwNJYs5lstxD
SSMaD+KccZVxkL7X2Qj9PJ+PCAQ6dMhvwTXrdcnrE7fI8PhFvHWrERjg7yIu1WI4Fgviy0u7437v
WzufSnfqMwbfz20fucO0chYq+tkxggNsMIIDaAIBATCBkzCBjDELMAkGA1UEBhMCSUwxFjAUBgNV
BAoTDVN0YXJ0Q29tIEx0ZC4xKzApBgNVBAsTIlNlY3VyZSBEaWdpdGFsIENlcnRpZmljYXRlIFNp
Z25pbmcxODA2BgNVBAMTL1N0YXJ0Q29tIENsYXNzIDIgUHJpbWFyeSBJbnRlcm1lZGlhdGUgQ2xp
ZW50IENBAgJIBzAJBgUrDgMCGgUAoIIBrTAYBgkqhkiG9w0BCQMxCwYJKoZIhvcNAQcBMBwGCSqG
SIb3DQEJBTEPFw0xNTAzMTMxNzQzMjFaMCMGCSqGSIb3DQEJBDEWBBRSzX9e2ey0Xwuqk98sf+pt
T1de7jCBpAYJKwYBBAGCNxAEMYGWMIGTMIGMMQswCQYDVQQGEwJJTDEWMBQGA1UEChMNU3RhcnRD
b20gTHRkLjErMCkGA1UECxMiU2VjdXJlIERpZ2l0YWwgQ2VydGlmaWNhdGUgU2lnbmluZzE4MDYG
A1UEAxMvU3RhcnRDb20gQ2xhc3MgMiBQcmltYXJ5IEludGVybWVkaWF0ZSBDbGllbnQgQ0ECAkgH
MIGmBgsqhkiG9w0BCRACCzGBlqCBkzCBjDELMAkGA1UEBhMCSUwxFjAUBgNVBAoTDVN0YXJ0Q29t
IEx0ZC4xKzApBgNVBAsTIlNlY3VyZSBEaWdpdGFsIENlcnRpZmljYXRlIFNpZ25pbmcxODA2BgNV
BAMTL1N0YXJ0Q29tIENsYXNzIDIgUHJpbWFyeSBJbnRlcm1lZGlhdGUgQ2xpZW50IENBAgJIBzAN
BgkqhkiG9w0BAQEFAASCAQCVIzKlP/kZQAw+tpsHLOR6q35Q6uYLrjsmPcdYdujdougf977/DldF
WOvJjodldNDPOyBIb0ec7ML1E7Ku8tTnwE4yDIW8/DUK8Ejjjv2XIJHOqxLdUWK+aJqCpYrPKXPy
8Jvpc75u5gbJjEunyz76RqPO+nNib6bVe2+8U8R+utqDXx/CQpGuYTWxPzlVhdjUrbvW8i8jw2/m
cZlPq224VdbCcmKl0t8N1x+YZNf5DxBYHR/1yxKKqsMldoSRXdCvKMwetbRFUI0+ml7t8KoFNEac
8zIiNqvo74DMV/rTLbW092HIxjs6wp8GJ06anPg6Q01eE6mSR3umBKa3mdi7AAAAAAAA
--Apple-Mail=_B947BC32-ECBF-4169-9617-915530DCFF20--


From nobody Fri Mar 13 10:47:34 2015
Return-Path: <phil.hunt@oracle.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 31DEC1A00CA for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 10:47:33 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.211
X-Spam-Level: 
X-Spam-Status: No, score=-4.211 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
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 r1JXNstgrjN2 for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 10:47:31 -0700 (PDT)
Received: from userp1040.oracle.com (userp1040.oracle.com [156.151.31.81]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6F2391A904A for <unbearable@ietf.org>; Fri, 13 Mar 2015 10:47:01 -0700 (PDT)
Received: from aserv0021.oracle.com (aserv0021.oracle.com [141.146.126.233]) by userp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2) with ESMTP id t2DHl0Vv027408 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Fri, 13 Mar 2015 17:47:00 GMT
Received: from userz7022.oracle.com (userz7022.oracle.com [156.151.31.86]) by aserv0021.oracle.com (8.13.8/8.13.8) with ESMTP id t2DHkxwL012622 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Fri, 13 Mar 2015 17:46:59 GMT
Received: from abhmp0009.oracle.com (abhmp0009.oracle.com [141.146.116.15]) by userz7022.oracle.com (8.14.5+Sun/8.14.4) with ESMTP id t2DHkwNq004313; Fri, 13 Mar 2015 17:46:58 GMT
Received: from [192.168.1.27] (/174.7.250.104) by default (Oracle Beehive Gateway v4.0) with ESMTP ; Fri, 13 Mar 2015 10:46:57 -0700
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (1.0)
From: Phil Hunt <phil.hunt@oracle.com>
X-Mailer: iPhone Mail (12B466)
In-Reply-To: <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com>
Date: Fri, 13 Mar 2015 10:46:55 -0700
Content-Transfer-Encoding: quoted-printable
Message-Id: <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com>
To: Andrei Popov <Andrei.Popov@microsoft.com>
X-Source-IP: aserv0021.oracle.com [141.146.126.233]
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/E99MJPS3XhpZ97LxCl2l04rH7_M>
Cc: John Bradley <ve7jtb@ve7jtb.com>, "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Mar 2015 17:47:33 -0000

The problem being that this just becomes device binding and app to app token=
 binding still isn't possible.=20

Do we care?

Phil

> On Mar 13, 2015, at 10:18, Andrei Popov <Andrei.Popov@microsoft.com> wrote=
:
>=20
> Thanks for raising this topic John.
>=20
>> I found some people that believe all apps sharing a common TLS library wo=
uld be able to share the same ID/key for a origin.
> In our prototypes so far (in Win10), the Token Binding Protocol implementa=
tion, the Token Binding key storage APIs, and the TLS library are three sepa=
rate components. Among other things, this allows 3-rd-party apps share Token=
 Binding keys and achieve single sign-on regardless of the TLS library they m=
ight be using (if they choose to do so).
>=20
>> The different views impact security considerations in native environments=
.   As an example can you pass a token from one app to another and have it s=
till work to the same origin, or is it a security feature that it won't work=
 and the token cant be stolen from a app and re-used by a different app on t=
he same device.
> I think that single sign-on between apps and browsers is at least a useful=
 option.
>=20
>> I don't think there is anything in the specs on this.
>> Figuring out where some of these issues get documented might be useful.
> Documenting key scoping is indeed useful; on the other hand standardizing t=
he specifics might require standardizing user and application isolation mode=
ls across platforms, which I don't think would be in scope for this forum. P=
erhaps we could agree on some general principles and include implementation g=
uidance on Token Binding key scoping?
>=20
> Cheers,
>=20
> Andrei
>=20
> -----Original Message-----
> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John Br=
adley
> Sent: Friday, March 13, 2015 9:57 AM
> To: unbearable@ietf.org
> Subject: [Unbearable] Scoping of keys on the client.
>=20
> Now that we are officially a WG I have a question (no hat)
>=20
> One thing that I identified as a potential issue talking to people who are=
 working on early deployments, is the expectations around how keys are share=
d between applications.
>=20
> I found some people that believe all apps sharing a common TLS library wou=
ld be able to share the same ID/key for a origin.
>=20
> Other folks were under the impression that each application would have its=
 own key for an origin.
>=20
> The different views impact security considerations in native environments.=
   As an example can you pass a token from one app to another and have it st=
ill work to the same origin, or is it a security feature that it won't work a=
nd the token cant be stolen from a app and re-used by a different app on the=
 same device.
>=20
> There was a middle view by the MDM people that the ID/key would be scoped t=
o a profile/container so that all the enterprise apps would share ID/key but=
 that social apps would not.
>=20
> I don't know that there is necessarily one right answer, as this may be im=
plementation dependent. =20
>=20
> I don't think there is anything in the specs on this.   One reason I ask i=
s that some of those differences impact on Native Application Single Signon a=
nd OAuth.
>=20
> In some of the Windows prototypes I understand there is a Key Storage API t=
hat is used to share keys between the browser and apps. =20
>=20
> Android, Windows, iOS , OSX may all have implementation specific details, b=
ut people building applications that leverage token binding will need to und=
erstand the differences as they may have security and privacy implications.
>=20
> Figuring out where some of these issues get documented might be useful.
>=20
> John B.
>=20
> _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org
> https://www.ietf.org/mailman/listinfo/unbearable


From nobody Fri Mar 13 12:58:32 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CA0A81A1BA2 for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 12:58:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 5FNfkD1WiiWa for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 12:58:28 -0700 (PDT)
Received: from mail-qc0-f174.google.com (mail-qc0-f174.google.com [209.85.216.174]) (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 86AD51A1B98 for <unbearable@ietf.org>; Fri, 13 Mar 2015 12:58:28 -0700 (PDT)
Received: by qcyl6 with SMTP id l6so29351782qcy.13 for <unbearable@ietf.org>; Fri, 13 Mar 2015 12:58:27 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:content-transfer-encoding:message-id:references :to; bh=QuzESwPsi+bm32YcbTWpne3e1DtH0RHi56zUVmjFt3M=; b=WnYnplcoR9D1zudDqkwxuRxYm5x+CGQQxS1pooV+vm6pWHFka43HK8wJxkR0EQsFeY 0BQ4qSpJAVYKdgEJ7U/POrRQs8Rq2xp8WeF4+rdRrLyhAgTLyvC0v1FMHMF3U+jGYc4S Uvs+yevQuyaJlBll4/N6AWMwIwMQ3Dqu30amLlTyJCtJJSmNdB6XArAi49n3IjxSaDzv xPMdpHVHoaXUV3b3/XV0IVrKFqt1BRU0dkt5JihuwwzY+SsdpzFrahWG9SEEAH3ycwIY K+5mp5fo7rahmD2OWUeJkyOwFSXP8i3W7GqyoB4jmcu4Bn8HYLWTvGCj10tghW3M9jfh ozEQ==
X-Gm-Message-State: ALoCoQlsfzss0hWOZQi6GrikE06GiUbx5KzYgtWdgM3jKMWxxdXQ6fWVc+iT9feF4upU/ZK2pE4A
X-Received: by 10.140.238.87 with SMTP id j84mr64158873qhc.78.1426276707676; Fri, 13 Mar 2015 12:58:27 -0700 (PDT)
Received: from [192.168.1.35] (186-79-112-76.baf.movistar.cl. [186.79.112.76]) by mx.google.com with ESMTPSA id b74sm2063063qga.40.2015.03.13.12.58.24 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Fri, 13 Mar 2015 12:58:26 -0700 (PDT)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (1.0)
From: John Bradley <ve7jtb@ve7jtb.com>
X-Mailer: iPhone Mail (12D508)
In-Reply-To: <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com>
Date: Fri, 13 Mar 2015 16:58:21 -0300
Content-Transfer-Encoding: quoted-printable
Message-Id: <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com>
To: Phil Hunt <phil.hunt@oracle.com>
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/bymseOCfn4aJyucdoagopxYeO3M>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Mar 2015 19:58:30 -0000

I think it is important to think through how the choices may impact how prot=
ocols can use token binding.=20

The NAAPS profile of connect is a possible candidate for using token binding=
 but that is highly dependent on how keys are shared on the client.=20

A simple OAuth use case is using token binding to secure a refresh token. =20=


How secure that is would depend on what other apps on the device may be able=
 to use that token if it leaked.=20

Understanding how keys are shared between apps is I think important.=20

John B


Sent from my iPhone

> On Mar 13, 2015, at 2:46 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>=20
> The problem being that this just becomes device binding and app to app tok=
en binding still isn't possible.=20
>=20
> Do we care?
>=20
> Phil
>=20
>> On Mar 13, 2015, at 10:18, Andrei Popov <Andrei.Popov@microsoft.com> wrot=
e:
>>=20
>> Thanks for raising this topic John.
>>=20
>>> I found some people that believe all apps sharing a common TLS library w=
ould be able to share the same ID/key for a origin.
>> In our prototypes so far (in Win10), the Token Binding Protocol implement=
ation, the Token Binding key storage APIs, and the TLS library are three sep=
arate components. Among other things, this allows 3-rd-party apps share Toke=
n Binding keys and achieve single sign-on regardless of the TLS library they=
 might be using (if they choose to do so).
>>=20
>>> The different views impact security considerations in native environment=
s.   As an example can you pass a token from one app to another and have it s=
till work to the same origin, or is it a security feature that it won't work=
 and the token cant be stolen from a app and re-used by a different app on t=
he same device.
>> I think that single sign-on between apps and browsers is at least a usefu=
l option.
>>=20
>>> I don't think there is anything in the specs on this.
>>> Figuring out where some of these issues get documented might be useful.
>> Documenting key scoping is indeed useful; on the other hand standardizing=
 the specifics might require standardizing user and application isolation mo=
dels across platforms, which I don't think would be in scope for this forum.=
 Perhaps we could agree on some general principles and include implementatio=
n guidance on Token Binding key scoping?
>>=20
>> Cheers,
>>=20
>> Andrei
>>=20
>> -----Original Message-----
>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John B=
radley
>> Sent: Friday, March 13, 2015 9:57 AM
>> To: unbearable@ietf.org
>> Subject: [Unbearable] Scoping of keys on the client.
>>=20
>> Now that we are officially a WG I have a question (no hat)
>>=20
>> One thing that I identified as a potential issue talking to people who ar=
e working on early deployments, is the expectations around how keys are shar=
ed between applications.
>>=20
>> I found some people that believe all apps sharing a common TLS library wo=
uld be able to share the same ID/key for a origin.
>>=20
>> Other folks were under the impression that each application would have it=
s own key for an origin.
>>=20
>> The different views impact security considerations in native environments=
.   As an example can you pass a token from one app to another and have it s=
till work to the same origin, or is it a security feature that it won't work=
 and the token cant be stolen from a app and re-used by a different app on t=
he same device.
>>=20
>> There was a middle view by the MDM people that the ID/key would be scoped=
 to a profile/container so that all the enterprise apps would share ID/key b=
ut that social apps would not.
>>=20
>> I don't know that there is necessarily one right answer, as this may be i=
mplementation dependent. =20
>>=20
>> I don't think there is anything in the specs on this.   One reason I ask i=
s that some of those differences impact on Native Application Single Signon a=
nd OAuth.
>>=20
>> In some of the Windows prototypes I understand there is a Key Storage API=
 that is used to share keys between the browser and apps. =20
>>=20
>> Android, Windows, iOS , OSX may all have implementation specific details,=
 but people building applications that leverage token binding will need to u=
nderstand the differences as they may have security and privacy implications=
.
>>=20
>> Figuring out where some of these issues get documented might be useful.
>>=20
>> John B.
>>=20
>> _______________________________________________
>> Unbearable mailing list
>> Unbearable@ietf.org
>> https://www.ietf.org/mailman/listinfo/unbearable


From nobody Fri Mar 13 13:15:17 2015
Return-Path: <leifj@sunet.se>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2C5491A90C2 for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 13:15:15 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.661
X-Spam-Level: 
X-Spam-Status: No, score=-1.661 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HELO_EQ_SE=0.35, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 ukoTyJvRH4-W for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 13:15:06 -0700 (PDT)
Received: from e-mailfilter01.sunet.se (e-mailfilter01.sunet.se [IPv6:2001:6b0:8:2::201]) (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 7A6E61A1F16 for <unbearable@ietf.org>; Fri, 13 Mar 2015 13:15:05 -0700 (PDT)
Received: from smtp1.sunet.se (smtp1.sunet.se [192.36.171.214]) by e-mailfilter01.sunet.se (8.14.4/8.14.4/Debian-4) with ESMTP id t2DKF2Ie031426 (version=TLSv1/SSLv3 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK) for <unbearable@ietf.org>; Fri, 13 Mar 2015 21:15:03 +0100
Received: from kerio.sunet.se (kerio.sunet.se [192.36.171.210]) by smtp1.sunet.se (8.14.9/8.14.7) with ESMTP id t2DKExK4009695 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for <unbearable@ietf.org>; Fri, 13 Mar 2015 21:15:02 +0100 (CET)
VBR-Info: md=sunet.se; mc=all; mv=swamid.se
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=sunet.se; s=default; t=1426277702; bh=ViYYricfxZNhpuyVXnqNoXtDXUPuq6cDybNw5ZU3orU=; h=Date:From:To:Subject; b=pktQvSjvco/c+gU3APDVGWIG/b0QbO0+h40lVVWbO5VU2UY7fugbzvLCAdJA4XL3B Rp6z6OPT92PlCrrvvf/i8rxZPgOQmhiqB8wYDLgAg5yM1vg5ExOpVryXpLJ7rwVL07 EpJ2N4IWoIcBN4q+Hb9s0atmw/OPBLhZTSuqUKeo=
X-Footer: c3VuZXQuc2U=
Received: from [10.0.0.107] ([62.102.145.131]) (authenticated user leifj@sunet.se) by kerio.sunet.se (Kerio Connect 8.3.4 patch 1) (using TLSv1.2 with cipher DHE-RSA-AES256-SHA (256 bits)) for unbearable@ietf.org; Fri, 13 Mar 2015 21:14:57 +0100
Message-ID: <55034540.6070800@sunet.se>
Date: Fri, 13 Mar 2015 21:14:56 +0100
From: Leif Johansson <leifj@sunet.se>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: "unbearable@ietf.org" <unbearable@ietf.org>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
X-Bayes-Prob: 0.0001 (Score 0, tokens from: outbound, outbound-sunet-se:default, sunet-se:default, base:default, @@RPTN)
X-CanIt-Geo: ip=192.36.171.210; country=SE; latitude=59.3294; longitude=18.0686; http://maps.google.com/maps?q=59.3294,18.0686&z=6
X-CanItPRO-Stream: outbound-sunet-se:outbound (inherits from outbound-sunet-se:default, sunet-se:default, base:default)
X-Canit-Stats-ID: 09O38f3GF - 008132074e08 - 20150313
X-CanIt-Archive-Cluster: PfMRe/vJWMiXwM2YIH5BVExnUnw
Received-SPF: neutral (e-mailfilter01.sunet.se: 192.36.171.210 is neither permitted nor denied by domain leifj@sunet.se) receiver=e-mailfilter01.sunet.se; client-ip=192.36.171.210; envelope-from=<leifj@sunet.se>; helo=smtp1.sunet.se; identity=mailfrom
X-Scanned-By: CanIt (www . roaringpenguin . com) on 192.36.171.201
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/P2BbFe1JooIRgd-hYOvFuHuxtVs>
Subject: [Unbearable] agenda for Dallas
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Mar 2015 20:15:15 -0000

Folks,

I've just posted a proposed agenda for Dallas.

http://www.ietf.org/proceedings/92/agenda/agenda-92-tokbind

The times are not to be taken literally.

The chairs hope for active and vigorous discussions to get
the WG off to a good start.

	Cheers Leif


From nobody Fri Mar 13 13:22:45 2015
Return-Path: <phil.hunt@oracle.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E213B1A017A for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 13:22:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.211
X-Spam-Level: 
X-Spam-Status: No, score=-4.211 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham
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 77NXBztiSGOT for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 13:22:39 -0700 (PDT)
Received: from userp1040.oracle.com (userp1040.oracle.com [156.151.31.81]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D74411A1A2C for <unbearable@ietf.org>; Fri, 13 Mar 2015 13:22:25 -0700 (PDT)
Received: from aserv0021.oracle.com (aserv0021.oracle.com [141.146.126.233]) by userp1040.oracle.com (Sentrion-MTA-4.3.2/Sentrion-MTA-4.3.2) with ESMTP id t2DKMOlK014882 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Fri, 13 Mar 2015 20:22:25 GMT
Received: from aserv0121.oracle.com (aserv0121.oracle.com [141.146.126.235]) by aserv0021.oracle.com (8.13.8/8.13.8) with ESMTP id t2DKMOCn012218 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=FAIL); Fri, 13 Mar 2015 20:22:24 GMT
Received: from abhmp0019.oracle.com (abhmp0019.oracle.com [141.146.116.25]) by aserv0121.oracle.com (8.13.8/8.13.8) with ESMTP id t2DKMOTZ030376; Fri, 13 Mar 2015 20:22:24 GMT
Received: from [192.168.1.9] (/174.7.250.104) by default (Oracle Beehive Gateway v4.0) with ESMTP ; Fri, 13 Mar 2015 13:22:24 -0700
Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: Phil Hunt <phil.hunt@oracle.com>
In-Reply-To: <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com>
Date: Fri, 13 Mar 2015 13:22:22 -0700
Content-Transfer-Encoding: quoted-printable
Message-Id: <2E82C711-8A32-48D0-913B-EFB09865AEDF@oracle.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com>
To: John Bradley <ve7jtb@ve7jtb.com>
X-Mailer: Apple Mail (2.2070.6)
X-Source-IP: aserv0021.oracle.com [141.146.126.233]
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/uWghpbH4Cg_DDxywYdPekDXoilA>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Mar 2015 20:22:44 -0000

Maybe I am recalling wrongly, but I don=E2=80=99t think OAuth WG =
discussed securing refresh tokens using binding (yet).  The OAuth POP =
use cases were focused on app-to-app access token security.

Phil

@independentid
www.independentid.com
phil.hunt@oracle.com

> On Mar 13, 2015, at 12:58 PM, John Bradley <ve7jtb@ve7jtb.com> wrote:
>=20
> I think it is important to think through how the choices may impact =
how protocols can use token binding.=20
>=20
> The NAAPS profile of connect is a possible candidate for using token =
binding but that is highly dependent on how keys are shared on the =
client.=20
>=20
> A simple OAuth use case is using token binding to secure a refresh =
token. =20
>=20
> How secure that is would depend on what other apps on the device may =
be able to use that token if it leaked.=20
>=20
> Understanding how keys are shared between apps is I think important.=20=

>=20
> John B
>=20
>=20
> Sent from my iPhone
>=20
>> On Mar 13, 2015, at 2:46 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>>=20
>> The problem being that this just becomes device binding and app to =
app token binding still isn't possible.=20
>>=20
>> Do we care?
>>=20
>> Phil
>>=20
>>> On Mar 13, 2015, at 10:18, Andrei Popov <Andrei.Popov@microsoft.com> =
wrote:
>>>=20
>>> Thanks for raising this topic John.
>>>=20
>>>> I found some people that believe all apps sharing a common TLS =
library would be able to share the same ID/key for a origin.
>>> In our prototypes so far (in Win10), the Token Binding Protocol =
implementation, the Token Binding key storage APIs, and the TLS library =
are three separate components. Among other things, this allows =
3-rd-party apps share Token Binding keys and achieve single sign-on =
regardless of the TLS library they might be using (if they choose to do =
so).
>>>=20
>>>> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
>>> I think that single sign-on between apps and browsers is at least a =
useful option.
>>>=20
>>>> I don't think there is anything in the specs on this.
>>>> Figuring out where some of these issues get documented might be =
useful.
>>> Documenting key scoping is indeed useful; on the other hand =
standardizing the specifics might require standardizing user and =
application isolation models across platforms, which I don't think would =
be in scope for this forum. Perhaps we could agree on some general =
principles and include implementation guidance on Token Binding key =
scoping?
>>>=20
>>> Cheers,
>>>=20
>>> Andrei
>>>=20
>>> -----Original Message-----
>>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of =
John Bradley
>>> Sent: Friday, March 13, 2015 9:57 AM
>>> To: unbearable@ietf.org
>>> Subject: [Unbearable] Scoping of keys on the client.
>>>=20
>>> Now that we are officially a WG I have a question (no hat)
>>>=20
>>> One thing that I identified as a potential issue talking to people =
who are working on early deployments, is the expectations around how =
keys are shared between applications.
>>>=20
>>> I found some people that believe all apps sharing a common TLS =
library would be able to share the same ID/key for a origin.
>>>=20
>>> Other folks were under the impression that each application would =
have its own key for an origin.
>>>=20
>>> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
>>>=20
>>> There was a middle view by the MDM people that the ID/key would be =
scoped to a profile/container so that all the enterprise apps would =
share ID/key but that social apps would not.
>>>=20
>>> I don't know that there is necessarily one right answer, as this may =
be implementation dependent. =20
>>>=20
>>> I don't think there is anything in the specs on this.   One reason I =
ask is that some of those differences impact on Native Application =
Single Signon and OAuth.
>>>=20
>>> In some of the Windows prototypes I understand there is a Key =
Storage API that is used to share keys between the browser and apps. =20
>>>=20
>>> Android, Windows, iOS , OSX may all have implementation specific =
details, but people building applications that leverage token binding =
will need to understand the differences as they may have security and =
privacy implications.
>>>=20
>>> Figuring out where some of these issues get documented might be =
useful.
>>>=20
>>> John B.
>>>=20
>>> _______________________________________________
>>> Unbearable mailing list
>>> Unbearable@ietf.org
>>> https://www.ietf.org/mailman/listinfo/unbearable


From nobody Fri Mar 13 13:42:29 2015
Return-Path: <wmills_92105@yahoo.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C74001A6FEB for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 13:42:12 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.509
X-Spam-Level: 
X-Spam-Status: No, score=-1.509 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_ENVFROM_END_DIGIT=0.25, FREEMAIL_FROM=0.001, FREEMAIL_REPLYTO_END_DIGIT=0.25, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 TF1kyR-ATgAJ for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 13:42:10 -0700 (PDT)
Received: from nm42-vm10.bullet.mail.bf1.yahoo.com (nm42-vm10.bullet.mail.bf1.yahoo.com [216.109.114.155]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D3DE11A6FF9 for <unbearable@ietf.org>; Fri, 13 Mar 2015 13:42:06 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s2048; t=1426279325; bh=flfudZ1PmN6PemC3W5f0VJUD28DmgNyu+DgViZyC4sg=; h=Date:From:Reply-To:To:Cc:In-Reply-To:References:Subject:From:Subject; b=WMKIu8VVszbmDB3fcbRnf9o2qRVKex6Q/Kko8M64NQQMxssdRi+aqxZKdUBXuFONVfEsRfd3h09fCHRzH30JUMVN4Xmr5ClsBZBFboduO5f6HSgJMELaxpUETEgew3nps0IVBhiQOjjiSPYdhDzZk/FDa1ESKyY8V/CvRDEgncSWvSSfjNPHv+aqZ4kizuGUh8pdSuvDfX+KbdafJbcpO+QfwM8c46GUBXZ5pFFpXkPPnLTDZLhfZKP7Oer2GD6IWEuJvSLrRghf1ZW57FqGweB3iobunzamOMWCtsm48OsSU8i5SARgz48y7tv1C7ZYux/ZyMnh7mpyrV4qorCa+g==
Received: from [98.139.215.142] by nm42.bullet.mail.bf1.yahoo.com with NNFMP;  13 Mar 2015 20:42:05 -0000
Received: from [98.139.212.197] by tm13.bullet.mail.bf1.yahoo.com with NNFMP;  13 Mar 2015 20:42:05 -0000
Received: from [127.0.0.1] by omp1006.mail.bf1.yahoo.com with NNFMP; 13 Mar 2015 20:42:05 -0000
X-Yahoo-Newman-Property: ymail-3
X-Yahoo-Newman-Id: 713087.70811.bm@omp1006.mail.bf1.yahoo.com
X-YMail-OSG: 0pflGEUVM1kaqQChZEi5BqIh6acDBZMRtwHVFPElmBCuwKS0yoxswbdeut6OKxR pHE1q8pdL9G.ARglAJ6HDEIWphwPdSzFp0vd19EzkGTcGMhPle5v4IRFBB7uzSCkVKl5l8Mpy0Ml JxTN037PhqanVI8emKRhHS_e2uE8kCb5Fin1_ZQc2Ev_NMjSGnCSdg5yzEZKHAPOsRkgZC1RkSpZ AXhe3JEKR_QiNNhzPFLaE_RAbnLFBW4GR01cU1nrBTzRnA1nYiEl5s0mjRSunKQUzujfGy_tqyA0 g8lcZ2E3T352QRbfuCDkHy8ieEW7Y3beYG4_RyGkNU2h_.aIyJtTPNh7XLkHcZn4u3UJ2YFx.XmH 00Jcfua68FB6iU3U1wvPplEwKmbHqwguR442lwvzTEGWgyvnmUpOIG_MMSj4DJLTVk4VaqtIA2ii ztz89aBWHgXWvXoQ9uN6MIHtBdGvMITfG338im8Xao_ZTaGlKrEozHAYJuuMddIfEQxis7szNpvj 2eobrlT0p.AZXGgR3poNxSeb.I7aZV_1TZSw6VLDWceFhs7YUVK3WkHYjQ85tCmRr0jTBPEY7jrA iLTmgqJbpiesFzu.8jA--
Received: by 66.196.80.114; Fri, 13 Mar 2015 20:42:05 +0000 
Date: Fri, 13 Mar 2015 20:42:03 +0000 (UTC)
From: Bill Mills <wmills_92105@yahoo.com>
To: Phil Hunt <phil.hunt@oracle.com>, John Bradley <ve7jtb@ve7jtb.com>
Message-ID: <2012209310.813417.1426279324059.JavaMail.yahoo@mail.yahoo.com>
In-Reply-To: <2E82C711-8A32-48D0-913B-EFB09865AEDF@oracle.com>
References: <2E82C711-8A32-48D0-913B-EFB09865AEDF@oracle.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;  boundary="----=_Part_813416_1237982517.1426279324053"
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/LztSDDEv8b-TJQjhGQi3wsogs2g>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: Bill Mills <wmills_92105@yahoo.com>
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Mar 2015 20:42:14 -0000

------=_Part_813416_1237982517.1426279324053
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

It would fit into client auth pretty cleanly.=20


     On Friday, March 13, 2015 1:22 PM, Phil Hunt <phil.hunt@oracle.com> wr=
ote:
  =20

 Maybe I am recalling wrongly, but I don=E2=80=99t think OAuth WG discussed=
 securing refresh tokens using binding (yet).=C2=A0 The OAuth POP use cases=
 were focused on app-to-app access token security.

Phil

@independentid
www.independentid.com
phil.hunt@oracle.com

> On Mar 13, 2015, at 12:58 PM, John Bradley <ve7jtb@ve7jtb.com> wrote:
>=20
> I think it is important to think through how the choices may impact how p=
rotocols can use token binding.=20
>=20
> The NAAPS profile of connect is a possible candidate for using token bind=
ing but that is highly dependent on how keys are shared on the client.=20
>=20
> A simple OAuth use case is using token binding to secure a refresh token.=
=C2=A0=20
>=20
> How secure that is would depend on what other apps on the device may be a=
ble to use that token if it leaked.=20
>=20
> Understanding how keys are shared between apps is I think important.=20
>=20
> John B
>=20
>=20
> Sent from my iPhone
>=20
>> On Mar 13, 2015, at 2:46 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>>=20
>> The problem being that this just becomes device binding and app to app t=
oken binding still isn't possible.=20
>>=20
>> Do we care?
>>=20
>> Phil
>>=20
>>> On Mar 13, 2015, at 10:18, Andrei Popov <Andrei.Popov@microsoft.com> wr=
ote:
>>>=20
>>> Thanks for raising this topic John.
>>>=20
>>>> I found some people that believe all apps sharing a common TLS library=
 would be able to share the same ID/key for a origin.
>>> In our prototypes so far (in Win10), the Token Binding Protocol impleme=
ntation, the Token Binding key storage APIs, and the TLS library are three =
separate components. Among other things, this allows 3-rd-party apps share =
Token Binding keys and achieve single sign-on regardless of the TLS library=
 they might be using (if they choose to do so).
>>>=20
>>>> The different views impact security considerations in native environme=
nts.=C2=A0 As an example can you pass a token from one app to another and h=
ave it still work to the same origin, or is it a security feature that it w=
on't work and the token cant be stolen from a app and re-used by a differen=
t app on the same device.
>>> I think that single sign-on between apps and browsers is at least a use=
ful option.
>>>=20
>>>> I don't think there is anything in the specs on this.
>>>> Figuring out where some of these issues get documented might be useful=
.
>>> Documenting key scoping is indeed useful; on the other hand standardizi=
ng the specifics might require standardizing user and application isolation=
 models across platforms, which I don't think would be in scope for this fo=
rum. Perhaps we could agree on some general principles and include implemen=
tation guidance on Token Binding key scoping?
>>>=20
>>> Cheers,
>>>=20
>>> Andrei
>>>=20
>>> -----Original Message-----
>>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John=
 Bradley
>>> Sent: Friday, March 13, 2015 9:57 AM
>>> To: unbearable@ietf.org
>>> Subject: [Unbearable] Scoping of keys on the client.
>>>=20
>>> Now that we are officially a WG I have a question (no hat)
>>>=20
>>> One thing that I identified as a potential issue talking to people who =
are working on early deployments, is the expectations around how keys are s=
hared between applications.
>>>=20
>>> I found some people that believe all apps sharing a common TLS library =
would be able to share the same ID/key for a origin.
>>>=20
>>> Other folks were under the impression that each application would have =
its own key for an origin.
>>>=20
>>> The different views impact security considerations in native environmen=
ts.=C2=A0 As an example can you pass a token from one app to another and ha=
ve it still work to the same origin, or is it a security feature that it wo=
n't work and the token cant be stolen from a app and re-used by a different=
 app on the same device.
>>>=20
>>> There was a middle view by the MDM people that the ID/key would be scop=
ed to a profile/container so that all the enterprise apps would share ID/ke=
y but that social apps would not.
>>>=20
>>> I don't know that there is necessarily one right answer, as this may be=
 implementation dependent.=C2=A0=20
>>>=20
>>> I don't think there is anything in the specs on this.=C2=A0 One reason =
I ask is that some of those differences impact on Native Application Single=
 Signon and OAuth.
>>>=20
>>> In some of the Windows prototypes I understand there is a Key Storage A=
PI that is used to share keys between the browser and apps.=C2=A0=20
>>>=20
>>> Android, Windows, iOS , OSX may all have implementation specific detail=
s, but people building applications that leverage token binding will need t=
o understand the differences as they may have security and privacy implicat=
ions.
>>>=20
>>> Figuring out where some of these issues get documented might be useful.
>>>=20
>>> John B.
>>>=20
>>> _______________________________________________
>>> Unbearable mailing list
>>> Unbearable@ietf.org
>>> https://www.ietf.org/mailman/listinfo/unbearable

_______________________________________________
Unbearable mailing list
Unbearable@ietf.org
https://www.ietf.org/mailman/listinfo/unbearable


  
------=_Part_813416_1237982517.1426279324053
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div style=3D"color:#000; background-color:#fff; font-family:HelveticaNeue,=
 Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif;font-size:12px=
"><div dir=3D"ltr" id=3D"yui_3_16_0_1_1426272734568_60912">It would fit int=
o client auth pretty cleanly.</div>  <br><div class=3D"qtdSeparateBR"><br><=
br></div><div class=3D"yahoo_quoted" style=3D"display: block;"> <div style=
=3D"font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Gr=
ande, sans-serif; font-size: 12px;"> <div style=3D"font-family: HelveticaNe=
ue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size:=
 16px;"> <div dir=3D"ltr"> <font size=3D"2" face=3D"Arial"> On Friday, Marc=
h 13, 2015 1:22 PM, Phil Hunt &lt;phil.hunt@oracle.com&gt; wrote:<br> </fon=
t> </div>  <br><br> <div class=3D"y_msg_container">Maybe I am recalling wro=
ngly, but I don=E2=80=99t think OAuth WG discussed securing refresh tokens =
using binding (yet).&nbsp; The OAuth POP use cases were focused on app-to-a=
pp access token security.<br clear=3D"none"><br clear=3D"none">Phil<br clea=
r=3D"none"><br clear=3D"none">@independentid<br clear=3D"none">www.independ=
entid.com<br clear=3D"none"><a shape=3D"rect" ymailto=3D"mailto:phil.hunt@o=
racle.com" href=3D"mailto:phil.hunt@oracle.com">phil.hunt@oracle.com</a><br=
 clear=3D"none"><div class=3D"yqt1361160970" id=3D"yqtfd19761"><br clear=3D=
"none">&gt; On Mar 13, 2015, at 12:58 PM, John Bradley &lt;<a shape=3D"rect=
" ymailto=3D"mailto:ve7jtb@ve7jtb.com" href=3D"mailto:ve7jtb@ve7jtb.com">ve=
7jtb@ve7jtb.com</a>&gt; wrote:<br clear=3D"none">&gt; <br clear=3D"none">&g=
t; I think it is important to think through how the choices may impact how =
protocols can use token binding. <br clear=3D"none">&gt; <br clear=3D"none"=
>&gt; The NAAPS profile of connect is a possible candidate for using token =
binding but that is highly dependent on how keys are shared on the client. =
<br clear=3D"none">&gt; <br clear=3D"none">&gt; A simple OAuth use case is =
using token binding to secure a refresh token.&nbsp; <br clear=3D"none">&gt=
; <br clear=3D"none">&gt; How secure that is would depend on what other app=
s on the device may be able to use that token if it leaked. <br clear=3D"no=
ne">&gt; <br clear=3D"none">&gt; Understanding how keys are shared between =
apps is I think important. <br clear=3D"none">&gt; <br clear=3D"none">&gt; =
John B<br clear=3D"none">&gt; <br clear=3D"none">&gt; <br clear=3D"none">&g=
t; Sent from my iPhone<br clear=3D"none">&gt; <br clear=3D"none">&gt;&gt; O=
n Mar 13, 2015, at 2:46 PM, Phil Hunt &lt;<a shape=3D"rect" ymailto=3D"mail=
to:phil.hunt@oracle.com" href=3D"mailto:phil.hunt@oracle.com">phil.hunt@ora=
cle.com</a>&gt; wrote:<br clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;&g=
t; The problem being that this just becomes device binding and app to app t=
oken binding still isn't possible. <br clear=3D"none">&gt;&gt; <br clear=3D=
"none">&gt;&gt; Do we care?<br clear=3D"none">&gt;&gt; <br clear=3D"none">&=
gt;&gt; Phil<br clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; On =
Mar 13, 2015, at 10:18, Andrei Popov &lt;<a shape=3D"rect" ymailto=3D"mailt=
o:Andrei.Popov@microsoft.com" href=3D"mailto:Andrei.Popov@microsoft.com">An=
drei.Popov@microsoft.com</a>&gt; wrote:<br clear=3D"none">&gt;&gt;&gt; <br =
clear=3D"none">&gt;&gt;&gt; Thanks for raising this topic John.<br clear=3D=
"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt;&gt; I found some people=
 that believe all apps sharing a common TLS library would be able to share =
the same ID/key for a origin.<br clear=3D"none">&gt;&gt;&gt; In our prototy=
pes so far (in Win10), the Token Binding Protocol implementation, the Token=
 Binding key storage APIs, and the TLS library are three separate component=
s. Among other things, this allows 3-rd-party apps share Token Binding keys=
 and achieve single sign-on regardless of the TLS library they might be usi=
ng (if they choose to do so).<br clear=3D"none">&gt;&gt;&gt; <br clear=3D"n=
one">&gt;&gt;&gt;&gt; The different views impact security considerations in=
 native environments.&nbsp;  As an example can you pass a token from one ap=
p to another and have it still work to the same origin, or is it a security=
 feature that it won't work and the token cant be stolen from a app and re-=
used by a different app on the same device.<br clear=3D"none">&gt;&gt;&gt; =
I think that single sign-on between apps and browsers is at least a useful =
option.<br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt;&gt; =
I don't think there is anything in the specs on this.<br clear=3D"none">&gt=
;&gt;&gt;&gt; Figuring out where some of these issues get documented might =
be useful.<br clear=3D"none">&gt;&gt;&gt; Documenting key scoping is indeed=
 useful; on the other hand standardizing the specifics might require standa=
rdizing user and application isolation models across platforms, which I don=
't think would be in scope for this forum. Perhaps we could agree on some g=
eneral principles and include implementation guidance on Token Binding key =
scoping?<br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; Che=
ers,<br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; Andrei<=
br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; -----Origina=
l Message-----<br clear=3D"none">&gt;&gt;&gt; From: Unbearable [mailto:<a s=
hape=3D"rect" ymailto=3D"mailto:unbearable-bounces@ietf.org" href=3D"mailto=
:unbearable-bounces@ietf.org">unbearable-bounces@ietf.org</a>] On Behalf Of=
 John Bradley<br clear=3D"none">&gt;&gt;&gt; Sent: Friday, March 13, 2015 9=
:57 AM<br clear=3D"none">&gt;&gt;&gt; To: <a shape=3D"rect" ymailto=3D"mail=
to:unbearable@ietf.org" href=3D"mailto:unbearable@ietf.org">unbearable@ietf=
.org</a><br clear=3D"none">&gt;&gt;&gt; Subject: [Unbearable] Scoping of ke=
ys on the client.<br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt=
;&gt; Now that we are officially a WG I have a question (no hat)<br clear=
=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; One thing that I ide=
ntified as a potential issue talking to people who are working on early dep=
loyments, is the expectations around how keys are shared between applicatio=
ns.<br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; I found =
some people that believe all apps sharing a common TLS library would be abl=
e to share the same ID/key for a origin.<br clear=3D"none">&gt;&gt;&gt; <br=
 clear=3D"none">&gt;&gt;&gt; Other folks were under the impression that eac=
h application would have its own key for an origin.<br clear=3D"none">&gt;&=
gt;&gt; <br clear=3D"none">&gt;&gt;&gt; The different views impact security=
 considerations in native environments.&nbsp;  As an example can you pass a=
 token from one app to another and have it still work to the same origin, o=
r is it a security feature that it won't work and the token cant be stolen =
from a app and re-used by a different app on the same device.<br clear=3D"n=
one">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; There was a middle view b=
y the MDM people that the ID/key would be scoped to a profile/container so =
that all the enterprise apps would share ID/key but that social apps would =
not.<br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; I don't=
 know that there is necessarily one right answer, as this may be implementa=
tion dependent.&nbsp; <br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none">&g=
t;&gt;&gt; I don't think there is anything in the specs on this.&nbsp;  One=
 reason I ask is that some of those differences impact on Native Applicatio=
n Single Signon and OAuth.<br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none=
">&gt;&gt;&gt; In some of the Windows prototypes I understand there is a Ke=
y Storage API that is used to share keys between the browser and apps.&nbsp=
; <br clear=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; Android, =
Windows, iOS , OSX may all have implementation specific details, but people=
 building applications that leverage token binding will need to understand =
the differences as they may have security and privacy implications.<br clea=
r=3D"none">&gt;&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; Figuring out where =
some of these issues get documented might be useful.<br clear=3D"none">&gt;=
&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; John B.<br clear=3D"none">&gt;&gt;=
&gt; <br clear=3D"none">&gt;&gt;&gt; ______________________________________=
_________<br clear=3D"none">&gt;&gt;&gt; Unbearable mailing list<br clear=
=3D"none">&gt;&gt;&gt; <a shape=3D"rect" ymailto=3D"mailto:Unbearable@ietf.=
org" href=3D"mailto:Unbearable@ietf.org">Unbearable@ietf.org</a><br clear=
=3D"none">&gt;&gt;&gt; <a shape=3D"rect" href=3D"https://www.ietf.org/mailm=
an/listinfo/unbearable" target=3D"_blank">https://www.ietf.org/mailman/list=
info/unbearable</a><br clear=3D"none"><br clear=3D"none">__________________=
_____________________________<br clear=3D"none">Unbearable mailing list<br =
clear=3D"none"><a shape=3D"rect" ymailto=3D"mailto:Unbearable@ietf.org" hre=
f=3D"mailto:Unbearable@ietf.org">Unbearable@ietf.org</a><br clear=3D"none">=
<a shape=3D"rect" href=3D"https://www.ietf.org/mailman/listinfo/unbearable"=
 target=3D"_blank">https://www.ietf.org/mailman/listinfo/unbearable</a><br =
clear=3D"none"></div><br><br></div>  </div> </div>  </div></div>
------=_Part_813416_1237982517.1426279324053--


From nobody Fri Mar 13 14:21:15 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4DEA11A8756 for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 14:21:13 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 dNQdT7QEHClT for <unbearable@ietfa.amsl.com>; Fri, 13 Mar 2015 14:21:09 -0700 (PDT)
Received: from mail-qg0-f44.google.com (mail-qg0-f44.google.com [209.85.192.44]) (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 279D31A886C for <unbearable@ietf.org>; Fri, 13 Mar 2015 14:21:06 -0700 (PDT)
Received: by qgfl89 with SMTP id l89so29121850qgf.11 for <unbearable@ietf.org>; Fri, 13 Mar 2015 14:21:05 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:message-id:references:to; bh=DncSVLDpanxDdcLy0F8O/4bOZ32JoyGOiyEYEYpaNSM=; b=ZtVkJU9JI/CjRPRmq0N5ox6JLSKhXVKnmPVbXN5EIVp6YwR+jkfPDPXzsV0aCMMLkG wUwOiQEycDuVhModrADg07KhOT+KbD8D86h5dc8nyy3I9nqH5y1l4QR8zhUmwMdRjC61 JkBa1uQeNFdwyDQF+KKjmzlholVKFiM+0LgocUcG++RvM5U1svTNFlnbkFKvVaIfY8qo bUnMMP9/Q3BtBfqRB2aCQ1kzj41cfzesjAu4GrsLRKetyn7Jtk/6rvvVZgiUHwXx/yfn TAuDLUQ5RkyD+GZ1yYEGiHUktwCwFkuUT1fdn+71qbvsqpPjy19hnVncDeiJm2KP5NY2 4EvA==
X-Gm-Message-State: ALoCoQmDnoNwd55ydk/orufAlmjj2nJ6mgLZr2EKTz3kmTT8a8cp8kDF3AP6BoM7XWxYFbGqCOBE
X-Received: by 10.229.68.136 with SMTP id v8mr61466487qci.16.1426281665342; Fri, 13 Mar 2015 14:21:05 -0700 (PDT)
Received: from [192.168.8.101] ([181.202.25.120]) by mx.google.com with ESMTPSA id u194sm2197692qhd.36.2015.03.13.14.21.03 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Fri, 13 Mar 2015 14:21:04 -0700 (PDT)
Content-Type: multipart/signed; boundary="Apple-Mail=_A15129A1-D546-4305-AD1F-38508FBFB473"; protocol="application/pkcs7-signature"; micalg=sha1
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: John Bradley <ve7jtb@ve7jtb.com>
In-Reply-To: <2E82C711-8A32-48D0-913B-EFB09865AEDF@oracle.com>
Date: Fri, 13 Mar 2015 18:20:59 -0300
Message-Id: <28F9966C-0101-4EDB-8445-BC35C549BFD5@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <2E82C711-8A32-48D0-913B-EFB09865AEDF@oracle.com>
To: Phil Hunt <phil.hunt@oracle.com>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/2YRyF9X-LqPhZHkxwOFH-EzJ7tc>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, "<oauth@ietf.org>" <oauth@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 13 Mar 2015 21:21:13 -0000

--Apple-Mail=_A15129A1-D546-4305-AD1F-38508FBFB473
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

I used that as an example.  I know one AS that is doing that for refresh =
tokens as it doesn't require any client or protocol change. =20

No the WG haven't discussed it.

The POP use case for AT is focused on app/client to RS.   Given that the =
AS and RS may not be in the same domain using token binding to secure =
the AT to the RS gets trickier.

In principal if the client knows the the URI of the resource server it =
could make a request to the RS and then take the public key and put it =
in a JWK to push to the AS as part of the existing POP key distribution =
spec.=20

The client would then get back a AT bound to that key.

That limits the client to only using a single RS as tin the current spec =
the key is only exchanged for code.

The key could be sent with refresh tokens if we wanted to allow multiple =
RS with different keys, however to do that securely you really want to =
secure the refresh token, to do that in a public client the AS could use =
token binding.

Those are issues the OAuth WG will need to look at.   To do that =
properly knowing things like if the RT leaks to another app on the =
device can it still be used?  Those questions need to be answered to =
formulate security considerations, and determine if token binding is the =
correct tool to use for a given use case.


John B.


> On Mar 13, 2015, at 5:22 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>=20
> Maybe I am recalling wrongly, but I don=E2=80=99t think OAuth WG =
discussed securing refresh tokens using binding (yet).  The OAuth POP =
use cases were focused on app-to-app access token security.
>=20
> Phil
>=20
> @independentid
> www.independentid.com
> phil.hunt@oracle.com
>=20
>> On Mar 13, 2015, at 12:58 PM, John Bradley <ve7jtb@ve7jtb.com> wrote:
>>=20
>> I think it is important to think through how the choices may impact =
how protocols can use token binding.=20
>>=20
>> The NAAPS profile of connect is a possible candidate for using token =
binding but that is highly dependent on how keys are shared on the =
client.=20
>>=20
>> A simple OAuth use case is using token binding to secure a refresh =
token. =20
>>=20
>> How secure that is would depend on what other apps on the device may =
be able to use that token if it leaked.=20
>>=20
>> Understanding how keys are shared between apps is I think important.=20=

>>=20
>> John B
>>=20
>>=20
>> Sent from my iPhone
>>=20
>>> On Mar 13, 2015, at 2:46 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>>>=20
>>> The problem being that this just becomes device binding and app to =
app token binding still isn't possible.=20
>>>=20
>>> Do we care?
>>>=20
>>> Phil
>>>=20
>>>> On Mar 13, 2015, at 10:18, Andrei Popov =
<Andrei.Popov@microsoft.com> wrote:
>>>>=20
>>>> Thanks for raising this topic John.
>>>>=20
>>>>> I found some people that believe all apps sharing a common TLS =
library would be able to share the same ID/key for a origin.
>>>> In our prototypes so far (in Win10), the Token Binding Protocol =
implementation, the Token Binding key storage APIs, and the TLS library =
are three separate components. Among other things, this allows =
3-rd-party apps share Token Binding keys and achieve single sign-on =
regardless of the TLS library they might be using (if they choose to do =
so).
>>>>=20
>>>>> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
>>>> I think that single sign-on between apps and browsers is at least a =
useful option.
>>>>=20
>>>>> I don't think there is anything in the specs on this.
>>>>> Figuring out where some of these issues get documented might be =
useful.
>>>> Documenting key scoping is indeed useful; on the other hand =
standardizing the specifics might require standardizing user and =
application isolation models across platforms, which I don't think would =
be in scope for this forum. Perhaps we could agree on some general =
principles and include implementation guidance on Token Binding key =
scoping?
>>>>=20
>>>> Cheers,
>>>>=20
>>>> Andrei
>>>>=20
>>>> -----Original Message-----
>>>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of =
John Bradley
>>>> Sent: Friday, March 13, 2015 9:57 AM
>>>> To: unbearable@ietf.org
>>>> Subject: [Unbearable] Scoping of keys on the client.
>>>>=20
>>>> Now that we are officially a WG I have a question (no hat)
>>>>=20
>>>> One thing that I identified as a potential issue talking to people =
who are working on early deployments, is the expectations around how =
keys are shared between applications.
>>>>=20
>>>> I found some people that believe all apps sharing a common TLS =
library would be able to share the same ID/key for a origin.
>>>>=20
>>>> Other folks were under the impression that each application would =
have its own key for an origin.
>>>>=20
>>>> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
>>>>=20
>>>> There was a middle view by the MDM people that the ID/key would be =
scoped to a profile/container so that all the enterprise apps would =
share ID/key but that social apps would not.
>>>>=20
>>>> I don't know that there is necessarily one right answer, as this =
may be implementation dependent. =20
>>>>=20
>>>> I don't think there is anything in the specs on this.   One reason =
I ask is that some of those differences impact on Native Application =
Single Signon and OAuth.
>>>>=20
>>>> In some of the Windows prototypes I understand there is a Key =
Storage API that is used to share keys between the browser and apps. =20
>>>>=20
>>>> Android, Windows, iOS , OSX may all have implementation specific =
details, but people building applications that leverage token binding =
will need to understand the differences as they may have security and =
privacy implications.
>>>>=20
>>>> Figuring out where some of these issues get documented might be =
useful.
>>>>=20
>>>> John B.
>>>>=20
>>>> _______________________________________________
>>>> Unbearable mailing list
>>>> Unbearable@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/unbearable
>=20


--Apple-Mail=_A15129A1-D546-4305-AD1F-38508FBFB473
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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--Apple-Mail=_A15129A1-D546-4305-AD1F-38508FBFB473--


From nobody Sat Mar 14 17:37:11 2015
Return-Path: <tonynad@microsoft.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3C9091A0404 for <unbearable@ietfa.amsl.com>; Sat, 14 Mar 2015 17:37:10 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
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 DPs4MeSEkP9e for <unbearable@ietfa.amsl.com>; Sat, 14 Mar 2015 17:37:08 -0700 (PDT)
Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0138.outbound.protection.outlook.com [65.55.169.138]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E3B441A03FF for <unbearable@ietf.org>; Sat, 14 Mar 2015 17:37:07 -0700 (PDT)
Received: from BN3PR0301MB1234.namprd03.prod.outlook.com (25.161.207.22) by BN3PR0301MB1252.namprd03.prod.outlook.com (25.161.207.28) with Microsoft SMTP Server (TLS) id 15.1.106.15; Sun, 15 Mar 2015 00:37:05 +0000
Received: from BN3PR0301MB1234.namprd03.prod.outlook.com ([25.161.207.22]) by BN3PR0301MB1234.namprd03.prod.outlook.com ([25.161.207.22]) with mapi id 15.01.0112.000; Sun, 15 Mar 2015 00:37:05 +0000
From: Anthony Nadalin <tonynad@microsoft.com>
To: John Bradley <ve7jtb@ve7jtb.com>, Phil Hunt <phil.hunt@oracle.com>
Thread-Topic: [Unbearable] Scoping of keys on the client.
Thread-Index: AQHQXa7T8BKibP80kEWurpiwHcJtZZ0aqC8AgAAH14CAACS5gIAB325g
Date: Sun, 15 Mar 2015 00:37:04 +0000
Message-ID: <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com>
In-Reply-To: <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [12.168.40.226]
authentication-results: ve7jtb.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0301MB1252;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(377454003)(51444003)(51704005)(13464003)(24454002)(52604005)(19580405001)(76576001)(46102003)(19580395003)(33656002)(86612001)(86362001)(2656002)(54356999)(87936001)(50986999)(40100003)(106116001)(99286002)(102836002)(15975445007)(77156002)(76176999)(122556002)(74316001)(2900100001)(66066001)(2950100001)(92566002)(62966003)(42262002); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0301MB1252; H:BN3PR0301MB1234.namprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BN3PR0301MB1252C053F48D91D9926AB326A6050@BN3PR0301MB1252.namprd03.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BN3PR0301MB1252; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0301MB1252; 
x-forefront-prvs: 05168A3970
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: microsoft.onmicrosoft.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 15 Mar 2015 00:37:04.5087 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0301MB1252
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/pdBBQoI-ncQRUMPwStFs0yzb8YA>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 00:37:10 -0000

Understanding how keys are shared between apps, is an app thing as there ca=
n be countless ways, the app that puts the key(s) there needs to be able to=
 document how it expects the sharing to occur.

-----Original Message-----
From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John Bra=
dley
Sent: Friday, March 13, 2015 12:58 PM
To: Phil Hunt
Cc: Andrei Popov; unbearable@ietf.org
Subject: Re: [Unbearable] Scoping of keys on the client.

I think it is important to think through how the choices may impact how pro=
tocols can use token binding.=20

The NAAPS profile of connect is a possible candidate for using token bindin=
g but that is highly dependent on how keys are shared on the client.=20

A simple OAuth use case is using token binding to secure a refresh token. =
=20

How secure that is would depend on what other apps on the device may be abl=
e to use that token if it leaked.=20

Understanding how keys are shared between apps is I think important.=20

John B


Sent from my iPhone

> On Mar 13, 2015, at 2:46 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>=20
> The problem being that this just becomes device binding and app to app to=
ken binding still isn't possible.=20
>=20
> Do we care?
>=20
> Phil
>=20
>> On Mar 13, 2015, at 10:18, Andrei Popov <Andrei.Popov@microsoft.com> wro=
te:
>>=20
>> Thanks for raising this topic John.
>>=20
>>> I found some people that believe all apps sharing a common TLS library =
would be able to share the same ID/key for a origin.
>> In our prototypes so far (in Win10), the Token Binding Protocol implemen=
tation, the Token Binding key storage APIs, and the TLS library are three s=
eparate components. Among other things, this allows 3-rd-party apps share T=
oken Binding keys and achieve single sign-on regardless of the TLS library =
they might be using (if they choose to do so).
>>=20
>>> The different views impact security considerations in native environmen=
ts.   As an example can you pass a token from one app to another and have i=
t still work to the same origin, or is it a security feature that it won't =
work and the token cant be stolen from a app and re-used by a different app=
 on the same device.
>> I think that single sign-on between apps and browsers is at least a usef=
ul option.
>>=20
>>> I don't think there is anything in the specs on this.
>>> Figuring out where some of these issues get documented might be useful.
>> Documenting key scoping is indeed useful; on the other hand standardizin=
g the specifics might require standardizing user and application isolation =
models across platforms, which I don't think would be in scope for this for=
um. Perhaps we could agree on some general principles and include implement=
ation guidance on Token Binding key scoping?
>>=20
>> Cheers,
>>=20
>> Andrei
>>=20
>> -----Original Message-----
>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John =
Bradley
>> Sent: Friday, March 13, 2015 9:57 AM
>> To: unbearable@ietf.org
>> Subject: [Unbearable] Scoping of keys on the client.
>>=20
>> Now that we are officially a WG I have a question (no hat)
>>=20
>> One thing that I identified as a potential issue talking to people who a=
re working on early deployments, is the expectations around how keys are sh=
ared between applications.
>>=20
>> I found some people that believe all apps sharing a common TLS library w=
ould be able to share the same ID/key for a origin.
>>=20
>> Other folks were under the impression that each application would have i=
ts own key for an origin.
>>=20
>> The different views impact security considerations in native environment=
s.   As an example can you pass a token from one app to another and have it=
 still work to the same origin, or is it a security feature that it won't w=
ork and the token cant be stolen from a app and re-used by a different app =
on the same device.
>>=20
>> There was a middle view by the MDM people that the ID/key would be scope=
d to a profile/container so that all the enterprise apps would share ID/key=
 but that social apps would not.
>>=20
>> I don't know that there is necessarily one right answer, as this may be =
implementation dependent. =20
>>=20
>> I don't think there is anything in the specs on this.   One reason I ask=
 is that some of those differences impact on Native Application Single Sign=
on and OAuth.
>>=20
>> In some of the Windows prototypes I understand there is a Key Storage AP=
I that is used to share keys between the browser and apps. =20
>>=20
>> Android, Windows, iOS , OSX may all have implementation specific details=
, but people building applications that leverage token binding will need to=
 understand the differences as they may have security and privacy implicati=
ons.
>>=20
>> Figuring out where some of these issues get documented might be useful.
>>=20
>> John B.
>>=20
>> _______________________________________________
>> Unbearable mailing list
>> Unbearable@ietf.org
>> https://www.ietf.org/mailman/listinfo/unbearable

_______________________________________________
Unbearable mailing list
Unbearable@ietf.org
https://www.ietf.org/mailman/listinfo/unbearable


From nobody Sat Mar 14 18:24:25 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AC2271A0687 for <unbearable@ietfa.amsl.com>; Sat, 14 Mar 2015 18:24:24 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 rIvdp3q0Zkit for <unbearable@ietfa.amsl.com>; Sat, 14 Mar 2015 18:24:22 -0700 (PDT)
Received: from mail-qc0-f181.google.com (mail-qc0-f181.google.com [209.85.216.181]) (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 1A7951A046D for <unbearable@ietf.org>; Sat, 14 Mar 2015 18:24:21 -0700 (PDT)
Received: by qcto4 with SMTP id o4so17984177qct.3 for <unbearable@ietf.org>; Sat, 14 Mar 2015 18:24:21 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:content-transfer-encoding:message-id:references :to; bh=67d5qi+RYTvC6mAmlylnyG3Un0ZgxTYlWIX2OALTMQ8=; b=PXr+YSnxhAk0uy994N2uiHtUu2mI5xy92Cq14Yooltuf+UwZ5ZKNyOvYzMOZw3j6Il hYXbv9wSyOp8cdMKMV7WDEIlVpTRlpJsO6tWkH0ORRhYXWr9QBpiEQEWbbUQHoMeyW30 n7565No/cvgo11WGV1+DW5ttY43xnZbgFXqucl0U95DkNLgQ05PF2KMNIKUa4ZytWliq UT6S0w24IxiYPrGhFZNAGZm2p5gpX4huBz6PI+7jgFlkWipzAdz/C3QCe3VOVRtJZ1nU pJ1OqweRGBhtrmhkLtI/L8NRK26vKC+vKxh45CdEgCEQUKeubZ72wnZv44rq+Ecom/dv dHzg==
X-Gm-Message-State: ALoCoQnKFpzCSMokPCSWWHxyM9TJ5D/zeua0EpCx3ZEsgi/6kHm9DNPat4gH/rXlLUWrzIAZa1sh
X-Received: by 10.140.25.208 with SMTP id 74mr65604903qgt.58.1426382661197; Sat, 14 Mar 2015 18:24:21 -0700 (PDT)
Received: from [10.234.3.249] ([201.220.243.110]) by mx.google.com with ESMTPSA id 10sm4533743qha.38.2015.03.14.18.24.19 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Sat, 14 Mar 2015 18:24:20 -0700 (PDT)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (1.0)
From: John Bradley <ve7jtb@ve7jtb.com>
X-Mailer: iPhone Mail (12D508)
In-Reply-To: <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com>
Date: Sat, 14 Mar 2015 22:24:18 -0300
Content-Transfer-Encoding: quoted-printable
Message-Id: <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com>
To: Anthony Nadalin <tonynad@microsoft.com>
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/W_OyhXX5eiOa4RJSdKzCDVJoxLU>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, Phil Hunt <phil.hunt@oracle.com>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 01:24:24 -0000

I was and until told otherwise under the impression that the private keys ar=
e explicitly not exposed to the application or JS in the browser.  That only=
 the public keys are.  =20

If the private key is exposed then it can be stolen along with the token, so=
rt of defeating the purpose of the specs.=20

I don't know if there is an intent to have a way for applications to control=
 what other apps have access (not or extract) to crate tls sessions with tho=
se private keys.=20

It seems like some environments are making the tls session keys available gl=
obally to any app on the device,  and some like Microsoft in Win10 prototype=
s may have a API for controlling access.  Nothing is documented on this that=
 has been submitted to the WG to this point.=20

This may not be something that we want to standardize,  but some understandi=
ng needs to happen around this so that people can understand how to use toke=
n binding safely.=20

John B.=20

Sent from my iPhone

> On Mar 14, 2015, at 9:37 PM, Anthony Nadalin <tonynad@microsoft.com> wrote=
:
>=20
> Understanding how keys are shared between apps, is an app thing as there c=
an be countless ways, the app that puts the key(s) there needs to be able to=
 document how it expects the sharing to occur.
>=20
> -----Original Message-----
> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John Br=
adley
> Sent: Friday, March 13, 2015 12:58 PM
> To: Phil Hunt
> Cc: Andrei Popov; unbearable@ietf.org
> Subject: Re: [Unbearable] Scoping of keys on the client.
>=20
> I think it is important to think through how the choices may impact how pr=
otocols can use token binding.=20
>=20
> The NAAPS profile of connect is a possible candidate for using token bindi=
ng but that is highly dependent on how keys are shared on the client.=20
>=20
> A simple OAuth use case is using token binding to secure a refresh token. =
=20
>=20
> How secure that is would depend on what other apps on the device may be ab=
le to use that token if it leaked.=20
>=20
> Understanding how keys are shared between apps is I think important.=20
>=20
> John B
>=20
>=20
> Sent from my iPhone
>=20
>> On Mar 13, 2015, at 2:46 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>>=20
>> The problem being that this just becomes device binding and app to app to=
ken binding still isn't possible.=20
>>=20
>> Do we care?
>>=20
>> Phil
>>=20
>>> On Mar 13, 2015, at 10:18, Andrei Popov <Andrei.Popov@microsoft.com> wro=
te:
>>>=20
>>> Thanks for raising this topic John.
>>>=20
>>>> I found some people that believe all apps sharing a common TLS library w=
ould be able to share the same ID/key for a origin.
>>> In our prototypes so far (in Win10), the Token Binding Protocol implemen=
tation, the Token Binding key storage APIs, and the TLS library are three se=
parate components. Among other things, this allows 3-rd-party apps share Tok=
en Binding keys and achieve single sign-on regardless of the TLS library the=
y might be using (if they choose to do so).
>>>=20
>>>> The different views impact security considerations in native environmen=
ts.   As an example can you pass a token from one app to another and have it=
 still work to the same origin, or is it a security feature that it won't wo=
rk and the token cant be stolen from a app and re-used by a different app on=
 the same device.
>>> I think that single sign-on between apps and browsers is at least a usef=
ul option.
>>>=20
>>>> I don't think there is anything in the specs on this.
>>>> Figuring out where some of these issues get documented might be useful.=

>>> Documenting key scoping is indeed useful; on the other hand standardizin=
g the specifics might require standardizing user and application isolation m=
odels across platforms, which I don't think would be in scope for this forum=
. Perhaps we could agree on some general principles and include implementati=
on guidance on Token Binding key scoping?
>>>=20
>>> Cheers,
>>>=20
>>> Andrei
>>>=20
>>> -----Original Message-----
>>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John B=
radley
>>> Sent: Friday, March 13, 2015 9:57 AM
>>> To: unbearable@ietf.org
>>> Subject: [Unbearable] Scoping of keys on the client.
>>>=20
>>> Now that we are officially a WG I have a question (no hat)
>>>=20
>>> One thing that I identified as a potential issue talking to people who a=
re working on early deployments, is the expectations around how keys are sha=
red between applications.
>>>=20
>>> I found some people that believe all apps sharing a common TLS library w=
ould be able to share the same ID/key for a origin.
>>>=20
>>> Other folks were under the impression that each application would have i=
ts own key for an origin.
>>>=20
>>> The different views impact security considerations in native environment=
s.   As an example can you pass a token from one app to another and have it s=
till work to the same origin, or is it a security feature that it won't work=
 and the token cant be stolen from a app and re-used by a different app on t=
he same device.
>>>=20
>>> There was a middle view by the MDM people that the ID/key would be scope=
d to a profile/container so that all the enterprise apps would share ID/key b=
ut that social apps would not.
>>>=20
>>> I don't know that there is necessarily one right answer, as this may be i=
mplementation dependent. =20
>>>=20
>>> I don't think there is anything in the specs on this.   One reason I ask=
 is that some of those differences impact on Native Application Single Signo=
n and OAuth.
>>>=20
>>> In some of the Windows prototypes I understand there is a Key Storage AP=
I that is used to share keys between the browser and apps. =20
>>>=20
>>> Android, Windows, iOS , OSX may all have implementation specific details=
, but people building applications that leverage token binding will need to u=
nderstand the differences as they may have security and privacy implications=
.
>>>=20
>>> Figuring out where some of these issues get documented might be useful.
>>>=20
>>> John B.
>>>=20
>>> _______________________________________________
>>> Unbearable mailing list
>>> Unbearable@ietf.org
>>> https://www.ietf.org/mailman/listinfo/unbearable
>=20
> _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org
> https://www.ietf.org/mailman/listinfo/unbearable


From rdom54@gmail.com  Sun Mar 15 03:16:47 2015
Return-Path: <rdom54@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2FFF21A00BE for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 03:16:47 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.163
X-Spam-Level: 
X-Spam-Status: No, score=0.163 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, FREEMAIL_ENVFROM_END_DIGIT=0.25, FREEMAIL_FROM=0.001, MALFORMED_FREEMAIL=1.813, SPF_PASS=-0.001] autolearn=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 VM2nyWz0xZ10 for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 03:16:45 -0700 (PDT)
Received: from mail-wi0-x234.google.com (mail-wi0-x234.google.com [IPv6:2a00:1450:400c:c05::234]) (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 71F6A1A00B8 for <unbearable@ietf.org>; Sun, 15 Mar 2015 03:16:45 -0700 (PDT)
Received: by wifj2 with SMTP id j2so19872984wif.1 for <unbearable@ietf.org>; Sun, 15 Mar 2015 03:16:44 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=sender:from:to:date:mime-version:subject:reply-to:cc:message-id :priority:in-reply-to:references:content-type :content-transfer-encoding:content-description; bh=SmYom/4cdPP6kHpWFCVbcCJ7ZCPzxDWecqbrWZcj4hw=; b=IXoYyONZEggOXGCtsk0hgF2p7uxLM+AqFVxhxGeytivKkzhhrl0+JBvMgME5Nyzdko 7pqlmuWMpFeMWVDUOOmA52ZqgNGpVD/VscYhAyBD7kaBtdTurQ4q9hE/RKri5xiBfE61 2bcpcepgThgzpa0M4cPqvZT5ovUzFmIcglMT3jQgJhw5t1RjAYFeSJAkimM1jMyL358P bZ36ZkO6j0bB6F/Ft6IDeN1CpcjhThRujoBDHRZ3lhzEOYcqKjWspLUkwA1kaAh/FgN6 4jRjQbZNCC307+VW1hSmWz3q7fOiUoxGdCb6OVxFKlgcQFIjWtIsc2dN9dGQRxb4RhUI M8wA==
X-Received: by 10.194.201.164 with SMTP id kb4mr111091991wjc.32.1426414604212;  Sun, 15 Mar 2015 03:16:44 -0700 (PDT)
Received: from [192.168.1.224] (host220-75-dynamic.1-87-r.retail.telecomitalia.it. [87.1.75.220]) by mx.google.com with ESMTPSA id uc18sm10428135wib.0.2015.03.15.03.16.42 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Sun, 15 Mar 2015 03:16:43 -0700 (PDT)
Sender: Domenico Rotondi <rdom54@gmail.com>
From: Domenico Rotondi <d.rotondi@computer.org>
X-Google-Original-From: "Domenico Rotondi" <D.Rotondi@Computer.Org>
To: John Bradley <ve7jtb@ve7jtb.com>
Date: Sun, 15 Mar 2015 11:16:37 +0100
MIME-Version: 1.0
Message-ID: <55055C05.17717.9366AE@D.Rotondi.Computer.Org>
Priority: normal
In-reply-to: <B1944375-14CE-4870-9191-D7CD6E2FD434@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com>, <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com>, <B1944375-14CE-4870-9191-D7CD6E2FD434@ve7jtb.com>
X-mailer: Pegasus Mail for Windows (4.70)
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Content-description: Mail message body
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/mKFy3ckZNSVM46bf0qMQ49L132I>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: D.Rotondi@computer.org
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 11:15:52 -0000

On 13 Mar 2015 at 14:43, John Bradley wrote:
Hi,
OAuth token is a kind of capability-based security token. A capability-based security 
token can be "ownership based" where having (even a copy of) a token is equivalent 
to being authorized to perform the operation on the resource  the token is associated 
to, as well as more complex things like having the token usable only by entities who 
can prove to be the owner.
Among the possible features tehre is the possibility to grant the right to delegate all 
or part of the rights granted by the token.
So, to answer your question, having a token usable only by a subject being able to 
prove it owns the token restrcits the scope of its usage; while adding the possibility to 
delegate the right to use the token can allow the use by another app to which the 
owner app gratns a delegated token.
Another alternative could be to intelligently use the Google proppsed macaroons 
mechanism.
Ciao
   Domenico
> Certainly in some potential OAuth use-cases where we want o encourage people to use the system browser vs a embedded web view, we may want to use the same key to bind a code in the browser and a native app.    
> 
> I think understanding how applications outside the browser or between browsers are going to share keys is worth putting some time into so that at a minimum developers will understand the issues.
> 
> In the Win10 prototype do you have some way to have a browser or app restrict the sharing of the key?  
> Sort of a non exportable flag, though in principal the keys themselves are never exportable but the use of them is shared.
> 
> John B.
> 
> 
> > On Mar 13, 2015, at 2:18 PM, Andrei Popov <Andrei.Popov@microsoft.com> wrote:
> > 
> > Thanks for raising this topic John.
> > 
> >> I found some people that believe all apps sharing a common TLS library would be able to share the same ID/key for a origin.
> > In our prototypes so far (in Win10), the Token Binding Protocol implementation, the Token Binding key storage APIs, and the TLS library are three separate components. Among other things, this allows 3-rd-party apps share Token Binding keys and achieve single sign-on regardless of the TLS library they might be using (if they choose to do so).
> > 
> >> The different views impact security considerations in native environments.   As an example can you pass a token from one app to another and have it still work to the same origin, or is it a security feature that it won't work and the token cant be stolen from a app and re-used by a different app on the same device.
> > I think that single sign-on between apps and browsers is at least a useful option.
> > 
> >> I don't think there is anything in the specs on this.
> >> Figuring out where some of these issues get documented might be useful.
> > Documenting key scoping is indeed useful; on the other hand standardizing the specifics might require standardizing user and application isolation models across platforms, which I don't think would be in scope for this forum. Perhaps we could agree on some general principles and include implementation guidance on Token Binding key scoping?
> > 
> > Cheers,
> > 
> > Andrei
> > 
> > -----Original Message-----
> > From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John Bradley
> > Sent: Friday, March 13, 2015 9:57 AM
> > To: unbearable@ietf.org
> > Subject: [Unbearable] Scoping of keys on the client.
> > 
> > Now that we are officially a WG I have a question (no hat)
> > 
> > One thing that I identified as a potential issue talking to people who are working on early deployments, is the expectations around how keys are shared between applications.
> > 
> > I found some people that believe all apps sharing a common TLS library would be able to share the same ID/key for a origin.
> > 
> > Other folks were under the impression that each application would have its own key for an origin.
> > 
> > The different views impact security considerations in native environments.   As an example can you pass a token from one app to another and have it still work to the same origin, or is it a security feature that it won't work and the token cant be stolen from a app and re-used by a different app on the same device.
> > 
> > There was a middle view by the MDM people that the ID/key would be scoped to a profile/container so that all the enterprise apps would share ID/key but that social apps would not.
> > 
> > I don't know that there is necessarily one right answer, as this may be implementation dependent.  
> > 
> > I don't think there is anything in the specs on this.   One reason I ask is that some of those differences impact on Native Application Single Signon and OAuth.
> > 
> > In some of the Windows prototypes I understand there is a Key Storage API that is used to share keys between the browser and apps.  
> > 
> > Android, Windows, iOS , OSX may all have implementation specific details, but people building applications that leverage token binding will need to understand the differences as they may have security and privacy implications.
> > 
> > Figuring out where some of these issues get documented might be useful.
> > 
> > John B.
> 
> 



From nobody Sun Mar 15 05:19:31 2015
Return-Path: <tonynad@microsoft.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 75B5A1A079A for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 05:19:29 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
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 80lwRcrj2SAd for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 05:19:27 -0700 (PDT)
Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0141.outbound.protection.outlook.com [65.55.169.141]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D73F41A0430 for <unbearable@ietf.org>; Sun, 15 Mar 2015 05:19:26 -0700 (PDT)
Received: from BN3PR0301MB1234.namprd03.prod.outlook.com (25.161.207.22) by BN3PR0301MB1249.namprd03.prod.outlook.com (25.161.207.25) with Microsoft SMTP Server (TLS) id 15.1.112.19; Sun, 15 Mar 2015 12:19:24 +0000
Received: from BN3PR0301MB1234.namprd03.prod.outlook.com ([25.161.207.22]) by BN3PR0301MB1234.namprd03.prod.outlook.com ([25.161.207.22]) with mapi id 15.01.0112.000; Sun, 15 Mar 2015 12:19:24 +0000
From: Anthony Nadalin <tonynad@microsoft.com>
To: John Bradley <ve7jtb@ve7jtb.com>
Thread-Topic: [Unbearable] Scoping of keys on the client.
Thread-Index: AQHQXa7T8BKibP80kEWurpiwHcJtZZ0aqC8AgAAH14CAACS5gIAB325ggAAN+ACAALZrYA==
Date: Sun, 15 Mar 2015 12:19:23 +0000
Message-ID: <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com>
In-Reply-To: <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [37.205.61.206]
authentication-results: ve7jtb.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0301MB1249;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(24454002)(51444003)(13464003)(52604005)(377454003)(51704005)(46102003)(86362001)(106116001)(110136001)(33656002)(99286002)(93886004)(40100003)(122556002)(92566002)(19580395003)(19580405001)(102836002)(2900100001)(2950100001)(15975445007)(77156002)(62966003)(50986999)(74316001)(87936001)(2656002)(66066001)(54356999)(76176999)(76576001)(42262002); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0301MB1249; H:BN3PR0301MB1234.namprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BN3PR0301MB1249922E96685EBE24A66DD0A6050@BN3PR0301MB1249.namprd03.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BN3PR0301MB1249; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0301MB1249; 
x-forefront-prvs: 05168A3970
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: microsoft.onmicrosoft.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 15 Mar 2015 12:19:23.7456 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0301MB1249
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/M15YcUA5_Mm6L27dIIUGIrqCt60>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, Phil Hunt <phil.hunt@oracle.com>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 12:19:29 -0000

How the browsers store keys or devices store keys seems totally out of scop=
e of this charter along with key management=20

-----Original Message-----
From: John Bradley [mailto:ve7jtb@ve7jtb.com]=20
Sent: Saturday, March 14, 2015 6:24 PM
To: Anthony Nadalin
Cc: Phil Hunt; Andrei Popov; unbearable@ietf.org
Subject: Re: [Unbearable] Scoping of keys on the client.

I was and until told otherwise under the impression that the private keys a=
re explicitly not exposed to the application or JS in the browser.  That on=
ly the public keys are.  =20

If the private key is exposed then it can be stolen along with the token, s=
ort of defeating the purpose of the specs.=20

I don't know if there is an intent to have a way for applications to contro=
l what other apps have access (not or extract) to crate tls sessions with t=
hose private keys.=20

It seems like some environments are making the tls session keys available g=
lobally to any app on the device,  and some like Microsoft in Win10 prototy=
pes may have a API for controlling access.  Nothing is documented on this t=
hat has been submitted to the WG to this point.=20

This may not be something that we want to standardize,  but some understand=
ing needs to happen around this so that people can understand how to use to=
ken binding safely.=20

John B.=20

Sent from my iPhone

> On Mar 14, 2015, at 9:37 PM, Anthony Nadalin <tonynad@microsoft.com> wrot=
e:
>=20
> Understanding how keys are shared between apps, is an app thing as there =
can be countless ways, the app that puts the key(s) there needs to be able =
to document how it expects the sharing to occur.
>=20
> -----Original Message-----
> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John B=
radley
> Sent: Friday, March 13, 2015 12:58 PM
> To: Phil Hunt
> Cc: Andrei Popov; unbearable@ietf.org
> Subject: Re: [Unbearable] Scoping of keys on the client.
>=20
> I think it is important to think through how the choices may impact how p=
rotocols can use token binding.=20
>=20
> The NAAPS profile of connect is a possible candidate for using token bind=
ing but that is highly dependent on how keys are shared on the client.=20
>=20
> A simple OAuth use case is using token binding to secure a refresh token.=
 =20
>=20
> How secure that is would depend on what other apps on the device may be a=
ble to use that token if it leaked.=20
>=20
> Understanding how keys are shared between apps is I think important.=20
>=20
> John B
>=20
>=20
> Sent from my iPhone
>=20
>> On Mar 13, 2015, at 2:46 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>>=20
>> The problem being that this just becomes device binding and app to app t=
oken binding still isn't possible.=20
>>=20
>> Do we care?
>>=20
>> Phil
>>=20
>>> On Mar 13, 2015, at 10:18, Andrei Popov <Andrei.Popov@microsoft.com> wr=
ote:
>>>=20
>>> Thanks for raising this topic John.
>>>=20
>>>> I found some people that believe all apps sharing a common TLS library=
 would be able to share the same ID/key for a origin.
>>> In our prototypes so far (in Win10), the Token Binding Protocol impleme=
ntation, the Token Binding key storage APIs, and the TLS library are three =
separate components. Among other things, this allows 3-rd-party apps share =
Token Binding keys and achieve single sign-on regardless of the TLS library=
 they might be using (if they choose to do so).
>>>=20
>>>> The different views impact security considerations in native environme=
nts.   As an example can you pass a token from one app to another and have =
it still work to the same origin, or is it a security feature that it won't=
 work and the token cant be stolen from a app and re-used by a different ap=
p on the same device.
>>> I think that single sign-on between apps and browsers is at least a use=
ful option.
>>>=20
>>>> I don't think there is anything in the specs on this.
>>>> Figuring out where some of these issues get documented might be useful=
.
>>> Documenting key scoping is indeed useful; on the other hand standardizi=
ng the specifics might require standardizing user and application isolation=
 models across platforms, which I don't think would be in scope for this fo=
rum. Perhaps we could agree on some general principles and include implemen=
tation guidance on Token Binding key scoping?
>>>=20
>>> Cheers,
>>>=20
>>> Andrei
>>>=20
>>> -----Original Message-----
>>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John=
 Bradley
>>> Sent: Friday, March 13, 2015 9:57 AM
>>> To: unbearable@ietf.org
>>> Subject: [Unbearable] Scoping of keys on the client.
>>>=20
>>> Now that we are officially a WG I have a question (no hat)
>>>=20
>>> One thing that I identified as a potential issue talking to people who =
are working on early deployments, is the expectations around how keys are s=
hared between applications.
>>>=20
>>> I found some people that believe all apps sharing a common TLS library =
would be able to share the same ID/key for a origin.
>>>=20
>>> Other folks were under the impression that each application would have =
its own key for an origin.
>>>=20
>>> The different views impact security considerations in native environmen=
ts.   As an example can you pass a token from one app to another and have i=
t still work to the same origin, or is it a security feature that it won't =
work and the token cant be stolen from a app and re-used by a different app=
 on the same device.
>>>=20
>>> There was a middle view by the MDM people that the ID/key would be scop=
ed to a profile/container so that all the enterprise apps would share ID/ke=
y but that social apps would not.
>>>=20
>>> I don't know that there is necessarily one right answer, as this may be=
 implementation dependent. =20
>>>=20
>>> I don't think there is anything in the specs on this.   One reason I as=
k is that some of those differences impact on Native Application Single Sig=
non and OAuth.
>>>=20
>>> In some of the Windows prototypes I understand there is a Key Storage A=
PI that is used to share keys between the browser and apps. =20
>>>=20
>>> Android, Windows, iOS , OSX may all have implementation specific detail=
s, but people building applications that leverage token binding will need t=
o understand the differences as they may have security and privacy implicat=
ions.
>>>=20
>>> Figuring out where some of these issues get documented might be useful.
>>>=20
>>> John B.
>>>=20
>>> _______________________________________________
>>> Unbearable mailing list
>>> Unbearable@ietf.org
>>> https://www.ietf.org/mailman/listinfo/unbearable
>=20
> _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org
> https://www.ietf.org/mailman/listinfo/unbearable


From nobody Sun Mar 15 08:52:55 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8511D1A1B30 for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 08:52:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 FdRMVG7TOlXT for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 08:52:52 -0700 (PDT)
Received: from mail-qg0-f51.google.com (mail-qg0-f51.google.com [209.85.192.51]) (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 BB1C01A1B2E for <unbearable@ietf.org>; Sun, 15 Mar 2015 08:52:51 -0700 (PDT)
Received: by qgg60 with SMTP id 60so22409445qgg.3 for <unbearable@ietf.org>; Sun, 15 Mar 2015 08:52:51 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:message-id:references:to; bh=BPPiTiieZ41zL4IhcqmadkkBQokB928cRHoHtbpK33Y=; b=jIUgzuf93rsiAS7SkEP8o1SQd8uA0kGIhdqCcF8EQnQ0sG36QyLa934U1NNC6iu86o BpslY6Gb8wdIWMBjAvhDwmhsm1zfTBdAJWDVn1G/N1rRxHRdg3zvSiwZMR6p9yonHd/R X/C642ucOoKbb97Nk+ZeiZ8Q5unpjpukTWE7vpXfJobWpKpbdh75YSQN20m5K7Ex5YAV EgdwhiiOphe0rWvGaNe982YBYVNGJ+o8cgwBZhwZGO8otGQg7y57nmwO1evKSLy5eYFx vag/s/O56z94ziEIoOFU6njA/8nZddNe30B/ckIdAoHoFn2olbhS9MFo8mBg8yMXZCes KpSQ==
X-Gm-Message-State: ALoCoQnTOtGJ3SS1mudJkJUQn2Yt6Ya7MLK3FkONNv2Tol/sQLmEgWmO1NLBOZBBRbKhgXTlw9S7
X-Received: by 10.140.233.78 with SMTP id e75mr71321038qhc.92.1426434770853; Sun, 15 Mar 2015 08:52:50 -0700 (PDT)
Received: from [192.168.8.101] ([181.202.26.213]) by mx.google.com with ESMTPSA id 201sm5626188qhr.24.2015.03.15.08.52.48 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Sun, 15 Mar 2015 08:52:49 -0700 (PDT)
Content-Type: multipart/signed; boundary="Apple-Mail=_7EBBFA07-A638-4144-A48A-E4A75A5F8C38"; protocol="application/pkcs7-signature"; micalg=sha1
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: John Bradley <ve7jtb@ve7jtb.com>
In-Reply-To: <55055C05.17717.9366AE@D.Rotondi.Computer.Org>
Date: Sun, 15 Mar 2015 12:52:43 -0300
Message-Id: <272B4855-4B06-4F95-BA68-9492C0724623@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <, > <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <, > <B1944375-14CE-4870-9191-D7CD6E2FD434@ve7jtb.com> <55055C05.17717.9366AE@D.Rotondi.Computer.Org>
To: D.Rotondi@computer.org
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/XTOUyEsVUM_Wv1RoF3_5_TpO6RE>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 15:52:54 -0000

--Apple-Mail=_7EBBFA07-A638-4144-A48A-E4A75A5F8C38
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

I was thinking of something simpler.

Some applications use system web browsers for user interactions with a =
server.

If=20
1: app A invokes a page in a system browser.
2: some interaction with a server B results in a token that is bound to =
the tls session.
3: Token is passed back to app via a custom scheme URI (or some other =
way)=20
4: can the app on the same device now use that token to interact with =
server B.

This is probably an important question for protocols/developers wanting =
to reference token binding.
However it may not be something this WG can answer with other than it is=20=

"implementation specific and out of scope for these specs".

However even if api's for controlling how the TLS sessions are shared on =
the device are out of scope,
there will probably need to be some security considerations around what =
to expect with respect to=20
key extractability and TLS session sharing between applications.

John B.
=20
> On Mar 15, 2015, at 7:16 AM, Domenico Rotondi <d.rotondi@computer.org> =
wrote:
>=20
> On 13 Mar 2015 at 14:43, John Bradley wrote:
> Hi,
> OAuth token is a kind of capability-based security token. A =
capability-based security=20
> token can be "ownership based" where having (even a copy of) a token =
is equivalent=20
> to being authorized to perform the operation on the resource  the =
token is associated=20
> to, as well as more complex things like having the token usable only =
by entities who=20
> can prove to be the owner.
> Among the possible features tehre is the possibility to grant the =
right to delegate all=20
> or part of the rights granted by the token.
> So, to answer your question, having a token usable only by a subject =
being able to=20
> prove it owns the token restrcits the scope of its usage; while adding =
the possibility to=20
> delegate the right to use the token can allow the use by another app =
to which the=20
> owner app gratns a delegated token.
> Another alternative could be to intelligently use the Google proppsed =
macaroons=20
> mechanism.
> Ciao
>   Domenico
>> Certainly in some potential OAuth use-cases where we want o encourage =
people to use the system browser vs a embedded web view, we may want to =
use the same key to bind a code in the browser and a native app.   =20
>>=20
>> I think understanding how applications outside the browser or between =
browsers are going to share keys is worth putting some time into so that =
at a minimum developers will understand the issues.
>>=20
>> In the Win10 prototype do you have some way to have a browser or app =
restrict the sharing of the key? =20
>> Sort of a non exportable flag, though in principal the keys =
themselves are never exportable but the use of them is shared.
>>=20
>> John B.
>>=20
>>=20
>>> On Mar 13, 2015, at 2:18 PM, Andrei Popov =
<Andrei.Popov@microsoft.com> wrote:
>>>=20
>>> Thanks for raising this topic John.
>>>=20
>>>> I found some people that believe all apps sharing a common TLS =
library would be able to share the same ID/key for a origin.
>>> In our prototypes so far (in Win10), the Token Binding Protocol =
implementation, the Token Binding key storage APIs, and the TLS library =
are three separate components. Among other things, this allows =
3-rd-party apps share Token Binding keys and achieve single sign-on =
regardless of the TLS library they might be using (if they choose to do =
so).
>>>=20
>>>> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
>>> I think that single sign-on between apps and browsers is at least a =
useful option.
>>>=20
>>>> I don't think there is anything in the specs on this.
>>>> Figuring out where some of these issues get documented might be =
useful.
>>> Documenting key scoping is indeed useful; on the other hand =
standardizing the specifics might require standardizing user and =
application isolation models across platforms, which I don't think would =
be in scope for this forum. Perhaps we could agree on some general =
principles and include implementation guidance on Token Binding key =
scoping?
>>>=20
>>> Cheers,
>>>=20
>>> Andrei
>>>=20
>>> -----Original Message-----
>>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of =
John Bradley
>>> Sent: Friday, March 13, 2015 9:57 AM
>>> To: unbearable@ietf.org
>>> Subject: [Unbearable] Scoping of keys on the client.
>>>=20
>>> Now that we are officially a WG I have a question (no hat)
>>>=20
>>> One thing that I identified as a potential issue talking to people =
who are working on early deployments, is the expectations around how =
keys are shared between applications.
>>>=20
>>> I found some people that believe all apps sharing a common TLS =
library would be able to share the same ID/key for a origin.
>>>=20
>>> Other folks were under the impression that each application would =
have its own key for an origin.
>>>=20
>>> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
>>>=20
>>> There was a middle view by the MDM people that the ID/key would be =
scoped to a profile/container so that all the enterprise apps would =
share ID/key but that social apps would not.
>>>=20
>>> I don't know that there is necessarily one right answer, as this may =
be implementation dependent. =20
>>>=20
>>> I don't think there is anything in the specs on this.   One reason I =
ask is that some of those differences impact on Native Application =
Single Signon and OAuth.
>>>=20
>>> In some of the Windows prototypes I understand there is a Key =
Storage API that is used to share keys between the browser and apps. =20
>>>=20
>>> Android, Windows, iOS , OSX may all have implementation specific =
details, but people building applications that leverage token binding =
will need to understand the differences as they may have security and =
privacy implications.
>>>=20
>>> Figuring out where some of these issues get documented might be =
useful.
>>>=20
>>> John B.
>>=20
>>=20
>=20
>=20


--Apple-Mail=_7EBBFA07-A638-4144-A48A-E4A75A5F8C38
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64

MIAGCSqGSIb3DQEHAqCAMIACAQExCzAJBgUrDgMCGgUAMIAGCSqGSIb3DQEHAQAAoIINPDCCBjQw
ggQcoAMCAQICASAwDQYJKoZIhvcNAQEFBQAwfTELMAkGA1UEBhMCSUwxFjAUBgNVBAoTDVN0YXJ0
Q29tIEx0ZC4xKzApBgNVBAsTIlNlY3VyZSBEaWdpdGFsIENlcnRpZmljYXRlIFNpZ25pbmcxKTAn
BgNVBAMTIFN0YXJ0Q29tIENlcnRpZmljYXRpb24gQXV0aG9yaXR5MB4XDTA3MTAyNDIxMDI1NVoX
DTE3MTAyNDIxMDI1NVowgYwxCzAJBgNVBAYTAklMMRYwFAYDVQQKEw1TdGFydENvbSBMdGQuMSsw
KQYDVQQLEyJTZWN1cmUgRGlnaXRhbCBDZXJ0aWZpY2F0ZSBTaWduaW5nMTgwNgYDVQQDEy9TdGFy
dENvbSBDbGFzcyAyIFByaW1hcnkgSW50ZXJtZWRpYXRlIENsaWVudCBDQTCCASIwDQYJKoZIhvcN
AQEBBQADggEPADCCAQoCggEBAMsohUWcASz7GfKrpTOMKqANy9BV7V0igWdGxA8IU77L3aTxErQ+
fcxtDYZ36Z6GH0YFn7fq5RADteP0AYzrCA+EQTfi8q1+kA3m0nwtwXG94M5sIqsvs7lRP1aycBke
/s5g9hJHryZ2acScnzczjBCAo7X1v5G3yw8MDP2m2RCye0KfgZ4nODerZJVzhAlOD9YejvAXZqHk
sw56HzElVIoYSZ3q4+RJuPXXfIoyby+Y2m1E+YzX5iCZXBx05gk6MKAW1vaw4/v2OOLy6FZH3XHH
tOkzUreG//CsFnB9+uaYSlR65cdGzTsmoIK8WH1ygoXhRBm98SD7Hf/r3FELNvUCAwEAAaOCAa0w
ggGpMA8GA1UdEwEB/wQFMAMBAf8wDgYDVR0PAQH/BAQDAgEGMB0GA1UdDgQWBBSuVYNv7DHKufcd
+q9rMfPIHeOsuzAfBgNVHSMEGDAWgBROC+8apEBbpRdphzDKNGhD0EGu8jBmBggrBgEFBQcBAQRa
MFgwJwYIKwYBBQUHMAGGG2h0dHA6Ly9vY3NwLnN0YXJ0c3NsLmNvbS9jYTAtBggrBgEFBQcwAoYh
aHR0cDovL3d3dy5zdGFydHNzbC5jb20vc2ZzY2EuY3J0MFsGA1UdHwRUMFIwJ6AloCOGIWh0dHA6
Ly93d3cuc3RhcnRzc2wuY29tL3Nmc2NhLmNybDAnoCWgI4YhaHR0cDovL2NybC5zdGFydHNzbC5j
b20vc2ZzY2EuY3JsMIGABgNVHSAEeTB3MHUGCysGAQQBgbU3AQIBMGYwLgYIKwYBBQUHAgEWImh0
dHA6Ly93d3cuc3RhcnRzc2wuY29tL3BvbGljeS5wZGYwNAYIKwYBBQUHAgEWKGh0dHA6Ly93d3cu
c3RhcnRzc2wuY29tL2ludGVybWVkaWF0ZS5wZGYwDQYJKoZIhvcNAQEFBQADggIBADqpJw3I07QW
ke9plNBpxUxcffc7nUrIQpJHDci91DFG7fVhHRkMZ1J+BKg5UNUxIFJ2Z9B90Micc/NXcs7kPBRd
n6XGO/vPc87Y6R+cWS9Nc9+fp3Enmsm94OxOwI9wn8qnr/6o3mD4noP9JphwUPTXwHovjavRnhUQ
HLfo/i2NG0XXgTHXS2Xm0kVUozXqpYpAdumMiB/vezj1QHQJDmUdPYMcp+reg9901zkyT3fDW/iv
JVv6pWtkh6Pw2ytZT7mvg7YhX3V50Nv860cV11mocUVcqBLv0gcT+HBDYtbuvexNftwNQKD5193A
7zN4vG7CTYkXxytSjKuXrpEatEiFPxWgb84nVj25SU5q/r1Xhwby6mLhkbaXslkVtwEWT3Van49r
KjlK4XrUKYYWtnfzq6aSak5u0Vpxd1rY79tWhD3EdCvOhNz/QplNa+VkIsrcp7+8ZhP1l1b2U6Ma
xIVteuVMD3X0vziIwr7jxYae9FZjbxlpUemqXjcC0QaFfN7qI0JsQMALL7iGRBg7K0CoOBzECdD3
fuZil5kU/LP9cr1BK31U0Uy651bFnAMMMkqhAChIbn0ei72VnbpSsrrSdF0BAGYQ8vyHae5aCg+H
75dVCV33K6FuxZrf09yTz+Vx/PkdRUYkXmZz/OTfyJXsUOUXrym6KvI2rYpccSk5MIIHADCCBeig
AwIBAgICSAcwDQYJKoZIhvcNAQEFBQAwgYwxCzAJBgNVBAYTAklMMRYwFAYDVQQKEw1TdGFydENv
bSBMdGQuMSswKQYDVQQLEyJTZWN1cmUgRGlnaXRhbCBDZXJ0aWZpY2F0ZSBTaWduaW5nMTgwNgYD
VQQDEy9TdGFydENvbSBDbGFzcyAyIFByaW1hcnkgSW50ZXJtZWRpYXRlIENsaWVudCBDQTAeFw0x
NDAzMjQyMzU2MjNaFw0xNjAzMjUwOTM5MzFaMIGfMRkwFwYDVQQNExBxekYwMVhZQ1pNTDM4N2hE
MQswCQYDVQQGEwJDTDEiMCAGA1UECBMZTWV0cm9wb2xpdGFuYSBkZSBTYW50aWFnbzEWMBQGA1UE
BxMNSXNsYSBkZSBNYWlwbzEVMBMGA1UEAxMMSm9obiBCcmFkbGV5MSIwIAYJKoZIhvcNAQkBFhNq
YnJhZGxleUBpY2xvdWQuY29tMIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAtTL0o4QG
WC+jnmYa7xEjcBTAeIOt7ILy40qsnJHNedVaTH0EU5yHzoaEOGHuOuwJUz/C7r2TvXpJ/Ud4w6VO
HdOUGnnKUiH5MV/kIysZ7DpN5D1f+yEast00oKsEbf/D6flzfex2JFV9rT7AQ+FQaTdf3S9K7gM2
F5kODFg805BMYTGT+haw9VOMXju5s93VEjUQcnGrLy0RtoN76GM6ItxqNnEt/Ln+2GNq8JvPyUKe
JsAxfIlTyqIbw32VlusKXL4+jmgFi+LY6bsfg3VHLvy58QsQnCwHg15uARvy5X6owyGcG7xHwNml
fNWtBZ3DHNPh37HC9lmAy4iqw4PvNwIDAQABo4IDVTCCA1EwCQYDVR0TBAIwADALBgNVHQ8EBAMC
BLAwHQYDVR0lBBYwFAYIKwYBBQUHAwIGCCsGAQUFBwMEMB0GA1UdDgQWBBSUDb6BlJD7FIYgWj1w
4z+GsOXs7zAfBgNVHSMEGDAWgBSuVYNv7DHKufcd+q9rMfPIHeOsuzCBmQYDVR0RBIGRMIGOgRNq
YnJhZGxleUBpY2xvdWQuY29tgRNqYnJhZGxleUBpY2xvdWQuY29tgRdqb2huLmJyYWRsZXlAd2lu
Z2FhLmNvbYERdmU3anRiQHZlN2p0Yi5jb22BD2picmFkbGV5QG1lLmNvbYEQamJyYWRsZXlAbWFj
LmNvbYETamJyYWRsZXlAd2luZ2FhLmNvbTCCAUwGA1UdIASCAUMwggE/MIIBOwYLKwYBBAGBtTcB
AgMwggEqMC4GCCsGAQUFBwIBFiJodHRwOi8vd3d3LnN0YXJ0c3NsLmNvbS9wb2xpY3kucGRmMIH3
BggrBgEFBQcCAjCB6jAnFiBTdGFydENvbSBDZXJ0aWZpY2F0aW9uIEF1dGhvcml0eTADAgEBGoG+
VGhpcyBjZXJ0aWZpY2F0ZSB3YXMgaXNzdWVkIGFjY29yZGluZyB0byB0aGUgQ2xhc3MgMiBWYWxp
ZGF0aW9uIHJlcXVpcmVtZW50cyBvZiB0aGUgU3RhcnRDb20gQ0EgcG9saWN5LCByZWxpYW5jZSBv
bmx5IGZvciB0aGUgaW50ZW5kZWQgcHVycG9zZSBpbiBjb21wbGlhbmNlIG9mIHRoZSByZWx5aW5n
IHBhcnR5IG9ibGlnYXRpb25zLjA2BgNVHR8ELzAtMCugKaAnhiVodHRwOi8vY3JsLnN0YXJ0c3Ns
LmNvbS9jcnR1Mi1jcmwuY3JsMIGOBggrBgEFBQcBAQSBgTB/MDkGCCsGAQUFBzABhi1odHRwOi8v
b2NzcC5zdGFydHNzbC5jb20vc3ViL2NsYXNzMi9jbGllbnQvY2EwQgYIKwYBBQUHMAKGNmh0dHA6
Ly9haWEuc3RhcnRzc2wuY29tL2NlcnRzL3N1Yi5jbGFzczIuY2xpZW50LmNhLmNydDAjBgNVHRIE
HDAahhhodHRwOi8vd3d3LnN0YXJ0c3NsLmNvbS8wDQYJKoZIhvcNAQEFBQADggEBALscEldbrgeF
B1WC/hMdYxFT4Lc8ALtErgJryRozTdeMlzpsncIKyy8M54HhxQAMOqFe2HR+R9H7WeIzmkV95yJn
JY3bd4bxnnemhLrDyi1VlNjEjkK5kgegI8JavahFXl4FwJHHv8TOh71Wf3fiy0Do7d7TQmVDRrzt
1k/2w4CXKweQ2mdFw7fskiYoPGEK7pFiicGMFBzLiKRm61CqojS4IYShiP0nCZZWPwNJYs5lstxD
SSMaD+KccZVxkL7X2Qj9PJ+PCAQ6dMhvwTXrdcnrE7fI8PhFvHWrERjg7yIu1WI4Fgviy0u7437v
WzufSnfqMwbfz20fucO0chYq+tkxggNsMIIDaAIBATCBkzCBjDELMAkGA1UEBhMCSUwxFjAUBgNV
BAoTDVN0YXJ0Q29tIEx0ZC4xKzApBgNVBAsTIlNlY3VyZSBEaWdpdGFsIENlcnRpZmljYXRlIFNp
Z25pbmcxODA2BgNVBAMTL1N0YXJ0Q29tIENsYXNzIDIgUHJpbWFyeSBJbnRlcm1lZGlhdGUgQ2xp
ZW50IENBAgJIBzAJBgUrDgMCGgUAoIIBrTAYBgkqhkiG9w0BCQMxCwYJKoZIhvcNAQcBMBwGCSqG
SIb3DQEJBTEPFw0xNTAzMTUxNTUyNDRaMCMGCSqGSIb3DQEJBDEWBBRd1KAN0pfyytLqGLrjgoBA
fnB2dDCBpAYJKwYBBAGCNxAEMYGWMIGTMIGMMQswCQYDVQQGEwJJTDEWMBQGA1UEChMNU3RhcnRD
b20gTHRkLjErMCkGA1UECxMiU2VjdXJlIERpZ2l0YWwgQ2VydGlmaWNhdGUgU2lnbmluZzE4MDYG
A1UEAxMvU3RhcnRDb20gQ2xhc3MgMiBQcmltYXJ5IEludGVybWVkaWF0ZSBDbGllbnQgQ0ECAkgH
MIGmBgsqhkiG9w0BCRACCzGBlqCBkzCBjDELMAkGA1UEBhMCSUwxFjAUBgNVBAoTDVN0YXJ0Q29t
IEx0ZC4xKzApBgNVBAsTIlNlY3VyZSBEaWdpdGFsIENlcnRpZmljYXRlIFNpZ25pbmcxODA2BgNV
BAMTL1N0YXJ0Q29tIENsYXNzIDIgUHJpbWFyeSBJbnRlcm1lZGlhdGUgQ2xpZW50IENBAgJIBzAN
BgkqhkiG9w0BAQEFAASCAQCBHwZfTtazXrrYPA7paCkWTcmspEUTh3dQ74yIBderi4O/dQpUVqFg
0KTNH6Dh+nn8ZujNbhn1Tx6aw89yqXzGdcEWmlhX51IjRAHMVAwdH4B4COQgC3S4ZNoeqVRllaCZ
eVo624A0MTFtCKiYqb6XahsrsF5Vh9pt0KBGXREmyRoceQAIjal8tndMIU7xMPKChdmbCtgJ263F
MS7cabLy4Kq6R5ZOWCEDniudsfI5EixZpLmy59hLVO7zSQv40Qicny+4n7GefCIlL5GB4m60Kt0L
nQJ4Mhtk/sRXtUS1vvkLr0WzNVwntsCMvxHKb3VtZt33OLjgwEdpRfx0iaLrAAAAAAAA
--Apple-Mail=_7EBBFA07-A638-4144-A48A-E4A75A5F8C38--


From nobody Sun Mar 15 10:13:08 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3402C1A1B51 for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 10:13:07 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 KSuSOG2BIFgp for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 10:13:04 -0700 (PDT)
Received: from mail-qc0-f174.google.com (mail-qc0-f174.google.com [209.85.216.174]) (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 4EF611A0358 for <unbearable@ietf.org>; Sun, 15 Mar 2015 10:13:04 -0700 (PDT)
Received: by qcyi15 with SMTP id i15so26306204qcy.0 for <unbearable@ietf.org>; Sun, 15 Mar 2015 10:13:03 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:message-id:references:to; bh=oyhf4EJEQ3wI1lNKmo9pitv/g2iLykYIEXDzqCSrGpY=; b=I7KDKU9Op3ownxN/mM5rvXMcsc9TektFk6tlOH2XCKlogjqyt5Vh+0gY0NbJavGwsu kDYYtuRZXCfr6+zfyAXNk+K/oJkWIasUyOHCdFo2b2vQAlZVOY6sXelNo0GjTsDwJBG2 Q4yeRG3Lk6n/53MuVDzrczf2YrNzqJb9EtMSmMp5HLViIj1K5ej6PP7lp2o0JBQiEFbB +e5Qzq/YHq4GdsVHOAviYEwFS5nrcOBaR/D+uQgdEKH7DjJmKUIOMZ31cPIM7+e8knsK YZJP9aIpIOL2o9wr4DHClaNks7BOYAa4qVFLCO1WEwHyi/9N2KsZzQQvUV60mrPzOccl OSOg==
X-Gm-Message-State: ALoCoQkbfIfXHsTHvlv0180hq5fZXZzsr9i5xpX/SZS28yNL0f14BKkDYNpT/BowaXjc8OLxPqf5
X-Received: by 10.140.145.9 with SMTP id 9mr70709289qhr.104.1426439583486; Sun, 15 Mar 2015 10:13:03 -0700 (PDT)
Received: from [192.168.8.101] ([181.202.26.213]) by mx.google.com with ESMTPSA id k195sm5741739qhk.27.2015.03.15.10.13.00 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Sun, 15 Mar 2015 10:13:02 -0700 (PDT)
Content-Type: multipart/signed; boundary="Apple-Mail=_1A691752-4307-4F33-96C0-6504B233D7F1"; protocol="application/pkcs7-signature"; micalg=sha1
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: John Bradley <ve7jtb@ve7jtb.com>
In-Reply-To: <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com>
Date: Sun, 15 Mar 2015 14:12:57 -0300
Message-Id: <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com> <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com>
To: Anthony Nadalin <tonynad@microsoft.com>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/SpUQlSxstXxENp-VFxRVAZCAvJY>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, Phil Hunt <phil.hunt@oracle.com>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 17:13:07 -0000

--Apple-Mail=_1A691752-4307-4F33-96C0-6504B233D7F1
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

Hi Tony,

(no hat)

I don't see in the charter =
https://datatracker.ietf.org/doc/charter-ietf-tokbind/ that it is =
explicitly out of scope.

Understanding what applications can share the session keys may be a =
requirement to securely do objective 1. Allow applications and services =
to prevent unauthorized replay of security tokens.

Is it possible for a server to have any expectations around what is =
going to happen on the client device when a cookie is token bound?   Is =
that to a tab, browser, MDM container or the whole device.
Anything is better than what we have now.  That I agree with completely. =
 =20

However at some point in our work some clever person will ask Token =
Binding to what?    At this point the answer to that seems a bit =
hand-wavy.

The other thing in the charter would be objective 2 "Allow strong key =
protection, e.g. using hardware-bound keys."

Is that going to be exposed in the API anyway?   The session is bound to =
a single app and the private key is non exportable stored in hardware vs =
stored in software and can be exported by any app and sent off device.

Currently the two specs under consideration for adoption by the WG as =
starting points don't consider that.

The WG can decide that nothing will be exposed to the server in the =
protocol and it needs to use some sort of out of band method to =
determine the client environment.
If the WG decides something else needs to be in the message to help the =
server make security decisions then the earlier we capture it the =
better.

I had been thinking of this as being close to/part of the TLS layer, =
with the client keys being ephemeral and not exportable.   I had assumed =
that any app sharing the TLS lib would have access to share the same TLS =
sessions.
I didn't think any key management would be part of that, as the keys =
would be non exportable and automatically used with a given server.

What my initial question turned up is that implementers are seeing this =
as a layer on top of TLS and applications would need to call some =
abstraction of the TLS API that contains some extensions to start token =
binding on a connection and perhaps do key management.

=46rom a simple HTML in a web browser point of view the two things might =
look indistinguishable.

I am guessing that when tokbind is done in a browser, the browser would =
make token binding available on all the connections, and the JS wouldn't =
need to do anything special.
Even in this simple case are we going to in privacy considerations going =
to say something about segmenting the sessions that are in "private" =
tabs to prevent correlation?

At this point I am trying to educate myself on the issues around the =
specs and the use of the specs, and not advocating for or against any =
particular API modification.

We do have a real security issue now, and having this WG take a long =
time to produce something is not in the public interest.

I agree that the core task is to get the core spec and the HTTP binding =
done in a timely way.

John B.



> On Mar 15, 2015, at 9:19 AM, Anthony Nadalin <tonynad@microsoft.com> =
wrote:
>=20
> How the browsers store keys or devices store keys seems totally out of =
scope of this charter along with key management=20
>=20
> -----Original Message-----
> From: John Bradley [mailto:ve7jtb@ve7jtb.com]=20
> Sent: Saturday, March 14, 2015 6:24 PM
> To: Anthony Nadalin
> Cc: Phil Hunt; Andrei Popov; unbearable@ietf.org
> Subject: Re: [Unbearable] Scoping of keys on the client.
>=20
> I was and until told otherwise under the impression that the private =
keys are explicitly not exposed to the application or JS in the browser. =
 That only the public keys are.  =20
>=20
> If the private key is exposed then it can be stolen along with the =
token, sort of defeating the purpose of the specs.=20
>=20
> I don't know if there is an intent to have a way for applications to =
control what other apps have access (not or extract) to crate tls =
sessions with those private keys.=20
>=20
> It seems like some environments are making the tls session keys =
available globally to any app on the device,  and some like Microsoft in =
Win10 prototypes may have a API for controlling access.  Nothing is =
documented on this that has been submitted to the WG to this point.=20
>=20
> This may not be something that we want to standardize,  but some =
understanding needs to happen around this so that people can understand =
how to use token binding safely.=20
>=20
> John B.=20
>=20
> Sent from my iPhone
>=20
>> On Mar 14, 2015, at 9:37 PM, Anthony Nadalin <tonynad@microsoft.com> =
wrote:
>>=20
>> Understanding how keys are shared between apps, is an app thing as =
there can be countless ways, the app that puts the key(s) there needs to =
be able to document how it expects the sharing to occur.
>>=20
>> -----Original Message-----
>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of =
John Bradley
>> Sent: Friday, March 13, 2015 12:58 PM
>> To: Phil Hunt
>> Cc: Andrei Popov; unbearable@ietf.org
>> Subject: Re: [Unbearable] Scoping of keys on the client.
>>=20
>> I think it is important to think through how the choices may impact =
how protocols can use token binding.=20
>>=20
>> The NAAPS profile of connect is a possible candidate for using token =
binding but that is highly dependent on how keys are shared on the =
client.=20
>>=20
>> A simple OAuth use case is using token binding to secure a refresh =
token. =20
>>=20
>> How secure that is would depend on what other apps on the device may =
be able to use that token if it leaked.=20
>>=20
>> Understanding how keys are shared between apps is I think important.=20=

>>=20
>> John B
>>=20
>>=20
>> Sent from my iPhone
>>=20
>>> On Mar 13, 2015, at 2:46 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>>>=20
>>> The problem being that this just becomes device binding and app to =
app token binding still isn't possible.=20
>>>=20
>>> Do we care?
>>>=20
>>> Phil
>>>=20
>>>> On Mar 13, 2015, at 10:18, Andrei Popov =
<Andrei.Popov@microsoft.com> wrote:
>>>>=20
>>>> Thanks for raising this topic John.
>>>>=20
>>>>> I found some people that believe all apps sharing a common TLS =
library would be able to share the same ID/key for a origin.
>>>> In our prototypes so far (in Win10), the Token Binding Protocol =
implementation, the Token Binding key storage APIs, and the TLS library =
are three separate components. Among other things, this allows =
3-rd-party apps share Token Binding keys and achieve single sign-on =
regardless of the TLS library they might be using (if they choose to do =
so).
>>>>=20
>>>>> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
>>>> I think that single sign-on between apps and browsers is at least a =
useful option.
>>>>=20
>>>>> I don't think there is anything in the specs on this.
>>>>> Figuring out where some of these issues get documented might be =
useful.
>>>> Documenting key scoping is indeed useful; on the other hand =
standardizing the specifics might require standardizing user and =
application isolation models across platforms, which I don't think would =
be in scope for this forum. Perhaps we could agree on some general =
principles and include implementation guidance on Token Binding key =
scoping?
>>>>=20
>>>> Cheers,
>>>>=20
>>>> Andrei
>>>>=20
>>>> -----Original Message-----
>>>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of =
John Bradley
>>>> Sent: Friday, March 13, 2015 9:57 AM
>>>> To: unbearable@ietf.org
>>>> Subject: [Unbearable] Scoping of keys on the client.
>>>>=20
>>>> Now that we are officially a WG I have a question (no hat)
>>>>=20
>>>> One thing that I identified as a potential issue talking to people =
who are working on early deployments, is the expectations around how =
keys are shared between applications.
>>>>=20
>>>> I found some people that believe all apps sharing a common TLS =
library would be able to share the same ID/key for a origin.
>>>>=20
>>>> Other folks were under the impression that each application would =
have its own key for an origin.
>>>>=20
>>>> The different views impact security considerations in native =
environments.   As an example can you pass a token from one app to =
another and have it still work to the same origin, or is it a security =
feature that it won't work and the token cant be stolen from a app and =
re-used by a different app on the same device.
>>>>=20
>>>> There was a middle view by the MDM people that the ID/key would be =
scoped to a profile/container so that all the enterprise apps would =
share ID/key but that social apps would not.
>>>>=20
>>>> I don't know that there is necessarily one right answer, as this =
may be implementation dependent. =20
>>>>=20
>>>> I don't think there is anything in the specs on this.   One reason =
I ask is that some of those differences impact on Native Application =
Single Signon and OAuth.
>>>>=20
>>>> In some of the Windows prototypes I understand there is a Key =
Storage API that is used to share keys between the browser and apps. =20
>>>>=20
>>>> Android, Windows, iOS , OSX may all have implementation specific =
details, but people building applications that leverage token binding =
will need to understand the differences as they may have security and =
privacy implications.
>>>>=20
>>>> Figuring out where some of these issues get documented might be =
useful.
>>>>=20
>>>> John B.
>>>>=20
>>>> _______________________________________________
>>>> Unbearable mailing list
>>>> Unbearable@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/unbearable
>>=20
>> _______________________________________________
>> Unbearable mailing list
>> Unbearable@ietf.org
>> https://www.ietf.org/mailman/listinfo/unbearable


--Apple-Mail=_1A691752-4307-4F33-96C0-6504B233D7F1
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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--Apple-Mail=_1A691752-4307-4F33-96C0-6504B233D7F1--


From nobody Sun Mar 15 11:11:36 2015
Return-Path: <ilari.liusvaara@elisanet.fi>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2AB8C1A1B60 for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 11:11:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level: 
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham
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 3G9qsfuCX8bE for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 11:11:33 -0700 (PDT)
Received: from emh06.mail.saunalahti.fi (emh06.mail.saunalahti.fi [62.142.5.116]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D30711A1B5E for <unbearable@ietf.org>; Sun, 15 Mar 2015 11:11:32 -0700 (PDT)
Received: from LK-Perkele-VII (a88-112-44-140.elisa-laajakaista.fi [88.112.44.140]) by emh06.mail.saunalahti.fi (Postfix) with ESMTP id 95C8B69994; Sun, 15 Mar 2015 20:11:30 +0200 (EET)
Date: Sun, 15 Mar 2015 20:11:30 +0200
From: Ilari Liusvaara <ilari.liusvaara@elisanet.fi>
To: John Bradley <ve7jtb@ve7jtb.com>
Message-ID: <20150315181130.GA16994@LK-Perkele-VII>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com> <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: Ilari Liusvaara <ilari.liusvaara@elisanet.fi>
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/ZsHPk1PdyhOVrp2rRBErn9bEtqw>
Cc: Anthony Nadalin <tonynad@microsoft.com>, Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, Phil Hunt <phil.hunt@oracle.com>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 18:11:35 -0000

On Sun, Mar 15, 2015 at 02:12:57PM -0300, John Bradley wrote:


[On what should be exposed]

I say there is very little server end can sanely do with information
about the client environment (besides just denying service if it
doesn't think client environment is "strong" enough, where defintion
of "strong" can be highly variable and might not have anything to do
with actual security).

Additionally, every piece of information exposed besides the public
key (where type is part of public key) is potential extra privacy
risk. Some kinds of information (including device identity) are
especially sensitive.


[On what layer to bind]

I think the deployment model web browsers have in mind is either:
1) End-to-end TLS connection between client and server.
2) Connection from client to gateway trusted by origin (CDN, load-
   balancer, etc...), which can extract the identifier for server
   use.

Both can be done fine by binding to TLS layer.

It seems that OAuth folks are wanting to do end-to-end binding at
layer above TLS (as TLS lacks enough reach according them).


[On scope of secrets]

Even if secrets are at TLS layer, it does not follow that keys are
scoped to TLS library.

My view is that the secrets should be associated with tokens they
protect and pinned only by those tokens and active sessions. That
is, once all tokens and sessions using secret go away, that secret
is deleted.

Also, obviously any application accessing the token needs access to
the corresponding secret.

This kind of management is obviously greatly above TLS layer.



-Ilari


From nobody Sun Mar 15 12:14:26 2015
Return-Path: <bascule@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 714EC1A1B74 for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 12:14:25 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 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, SPF_PASS=-0.001] autolearn=ham
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 bkp5o3wIfFcy for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 12:14:23 -0700 (PDT)
Received: from mail-yk0-x232.google.com (mail-yk0-x232.google.com [IPv6:2607:f8b0:4002:c07::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 7B0EA1A1B63 for <unbearable@ietf.org>; Sun, 15 Mar 2015 12:14:23 -0700 (PDT)
Received: by ykcn8 with SMTP id n8so10605791ykc.3 for <unbearable@ietf.org>; Sun, 15 Mar 2015 12:14:22 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=gE6OS875N3pGo+N5UOpxP9QHXkE8bxC+F0hmxGRv4Hc=; b=r1+NW6dyOtpN7IG5bTXzYTRKTdj3tIldkZyAXb7fTgW0El24Rz5KoM67PWZtlVyNV4 Fw+zcKn0zHfwEijp1pMYqS2QEwHKFIbzpu0fm7Z7YRemQJmsjENtY/mn0nyqWqdbf0p3 rxU5GCN5ffKIPN6u4g1Hmf50kLXLQsG5XnI2bDGeceV3OEfs5Rx5yAelWaEEeSdiyl7J /6LSkTTrLmPyyQPHcghHRA9PnWAK6jUX7AunknKrP+GsZPkK5EuEtLvskZugOJSo/Lgd WdlAWqnUQX0HEQ85OXoYcIacFSm4ocs78gVUJ2+DrX3z5y4WR2Uk+Py7yINxdH7uprmm mPEA==
X-Received: by 10.236.65.227 with SMTP id f63mr59381528yhd.74.1426446862889; Sun, 15 Mar 2015 12:14:22 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.170.35.203 with HTTP; Sun, 15 Mar 2015 12:14:02 -0700 (PDT)
In-Reply-To: <55055C05.17717.9366AE@D.Rotondi.Computer.Org>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <B1944375-14CE-4870-9191-D7CD6E2FD434@ve7jtb.com> <55055C05.17717.9366AE@D.Rotondi.Computer.Org>
From: Tony Arcieri <bascule@gmail.com>
Date: Sun, 15 Mar 2015 12:14:02 -0700
Message-ID: <CAHOTMVJd-_Xi+HonYHEPrKkoK5GEpZGOpmrR36tsOF1MGYNOkA@mail.gmail.com>
To: D.Rotondi@computer.org
Content-Type: multipart/alternative; boundary=001a11c30d9a8db9a70511588a83
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/Loz9ekpGyrw1C7-hG0fj4R6HsOs>
Cc: John Bradley <ve7jtb@ve7jtb.com>, "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 19:14:25 -0000

--001a11c30d9a8db9a70511588a83
Content-Type: text/plain; charset=UTF-8

On Sun, Mar 15, 2015 at 3:16 AM, Domenico Rotondi <d.rotondi@computer.org>
wrote:

> OAuth token is a kind of capability-based security token.


I think you need to be very careful when using the term "capability". It's
a term that doesn't mean a whole lot to most people but has incredible
significance in systems like CapTP/E.


> A capability-based security token can be "ownership based" where having
> (even a copy of) a token is equivalent to being authorized to perform the
> operation on the resource


Yes, however the capability must be entangled and inseparable from the
resource it acts on. For HTTP requests, I would consider that encoding both
an HTTP method and a full path to a resource in addition to the hostname.


> the token is associated to, as well as more complex things like having the
> token usable only by entities who
> can prove to be the owner.
> Among the possible features tehre is the possibility to grant the right to
> delegate all
> or part of the rights granted by the token.
>

These are properties that I think are necessary of true capability systems,
but not sufficient. I'd look to the Capability Myths Demolished paper for
that:

http://zesty.ca/capmyths/usenix.pdf


> Another alternative could be to intelligently use the Google proppsed
> macaroons
> mechanism.


Note that the Macaroons paper quite carefully distinguishes between
capabilities and credentials, and suggests that Macaroons are credentials
and *not* capabilities.

https://static.googleusercontent.com/media/research.google.com/en/us/pubs/archive/41892.pdf

--
Tony Arcieri

--001a11c30d9a8db9a70511588a83
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_extra"><div class=3D"gmail_quote">On S=
un, Mar 15, 2015 at 3:16 AM, Domenico Rotondi <span dir=3D"ltr">&lt;<a href=
=3D"mailto:d.rotondi@computer.org" target=3D"_blank" onclick=3D"window.open=
(&#39;https://mail.google.com/mail/?view=3Dcm&amp;tf=3D1&amp;to=3Dd.rotondi=
@computer.org&amp;cc=3D&amp;bcc=3D&amp;su=3D&amp;body=3D&#39;,&#39;_blank&#=
39;);return false;">d.rotondi@computer.org</a>&gt;</span> wrote:<br><blockq=
uote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-wi=
dth:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-=
left:1ex">OAuth token is a kind of capability-based security token.</blockq=
uote><div><br></div><div>I think you need to be very careful when using the=
 term &quot;capability&quot;. It&#39;s a term that doesn&#39;t mean a whole=
 lot to most people but has incredible significance in systems like CapTP/E=
.</div><div>=C2=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0=
px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);b=
order-left-style:solid;padding-left:1ex"> A capability-based security=C2=A0=
token can be &quot;ownership based&quot; where having (even a copy of) a to=
ken is equivalent=C2=A0to being authorized to perform the operation on the =
resource</blockquote><div><br></div><div>Yes, however the capability must b=
e entangled and inseparable from the resource it acts on. For HTTP requests=
, I would consider that encoding both an HTTP method and a full path to a r=
esource in addition to the hostname.</div><div>=C2=A0</div><blockquote clas=
s=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;b=
order-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"=
>the token is associated=C2=A0to, as well as more complex things like havin=
g the token usable only by entities who<br>
can prove to be the owner.<br>
Among the possible features tehre is the possibility to grant the right to =
delegate all<br>
or part of the rights granted by the token.<br></blockquote><div><br></div>=
<div>These are properties that I think are necessary of true capability sys=
tems, but not sufficient. I&#39;d look to the Capability Myths Demolished p=
aper for that:</div><div><br></div><div><a href=3D"http://zesty.ca/capmyths=
/usenix.pdf">http://zesty.ca/capmyths/usenix.pdf</a><br></div><div>=C2=A0</=
div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;bor=
der-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:sol=
id;padding-left:1ex">
Another alternative could be to intelligently use the Google proppsed macar=
oons<br>
mechanism.</blockquote><div><br></div><div>Note that the Macaroons paper qu=
ite carefully distinguishes between capabilities and credentials, and sugge=
sts that Macaroons are credentials and *not* capabilities.</div><div><br></=
div><div><a href=3D"https://static.googleusercontent.com/media/research.goo=
gle.com/en/us/pubs/archive/41892.pdf">https://static.googleusercontent.com/=
media/research.google.com/en/us/pubs/archive/41892.pdf</a><br></div><div><b=
r></div><div>--</div><div>Tony Arcieri</div></div>
</div></div>

--001a11c30d9a8db9a70511588a83--


From nobody Sun Mar 15 12:34:35 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 418711A1B74 for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 12:34:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 VNS7bp66V5-9 for <unbearable@ietfa.amsl.com>; Sun, 15 Mar 2015 12:34:32 -0700 (PDT)
Received: from mail-qg0-f50.google.com (mail-qg0-f50.google.com [209.85.192.50]) (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 277AB1A0113 for <unbearable@ietf.org>; Sun, 15 Mar 2015 12:34:32 -0700 (PDT)
Received: by qgez64 with SMTP id z64so24873301qge.2 for <unbearable@ietf.org>; Sun, 15 Mar 2015 12:34:31 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:message-id:references:to; bh=sDUuFvg1X3+ZBUa0VGkcKQ941CmhVrzJ0S+ZuNRycDc=; b=iENCSutWrkTSa956n5HXAh2MtMBgcvKusb/6vSIsztbvxLwV2x+wZrXj1AhzSxLhqJ mCXkEnMHsj9PTf31YA6SyGZXT8IWz+K4AShqdjYsJWpC+QxEDesd1cG4nPeHumsb0tBL RP8+ps+5AFZ5efe3QZTSyMy+5CBE5Q5smcShMSuXbweMLAVBYJ84kCzCMydh1vg/EExV el/VHeW0RfKslOmzswudrOWcN+MNuZkw0m0JBM1v7rKBQPtJfq8XuTC3Wv5K+5BDd9JJ QP1g+sPcbU+2csERgfPpWODO9HgaL8ZoQd5R43PI39T0CiOW41/DegW+nuZ5C3aRQ45X O05Q==
X-Gm-Message-State: ALoCoQm27GdtFwcA9TSKqPocONfqYfhDpJfPVwdJ1zzORi79MDg0xjqHa6HW95rapRVA1gGUNv1G
X-Received: by 10.229.61.134 with SMTP id t6mr49845395qch.7.1426448071316; Sun, 15 Mar 2015 12:34:31 -0700 (PDT)
Received: from [192.168.8.101] ([181.202.26.213]) by mx.google.com with ESMTPSA id j94sm5989202qgd.47.2015.03.15.12.34.28 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Sun, 15 Mar 2015 12:34:30 -0700 (PDT)
Content-Type: multipart/signed; boundary="Apple-Mail=_55DB3D02-1913-4A41-B530-9ECF08C5FB9F"; protocol="application/pkcs7-signature"; micalg=sha1
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: John Bradley <ve7jtb@ve7jtb.com>
In-Reply-To: <20150315181130.GA16994@LK-Perkele-VII>
Date: Sun, 15 Mar 2015 16:34:24 -0300
Message-Id: <2EFCA3AA-D611-4DAE-AB28-799327B08292@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com> <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com> <20150315181130.GA16994@LK-Perkele-VII>
To: Ilari Liusvaara <ilari.liusvaara@elisanet.fi>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/OEqJ2fuVFGA4B-2Fwgd925hI26g>
Cc: Anthony Nadalin <tonynad@microsoft.com>, Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, Phil Hunt <phil.hunt@oracle.com>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Sun, 15 Mar 2015 19:34:34 -0000

--Apple-Mail=_55DB3D02-1913-4A41-B530-9ECF08C5FB9F
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

I am mostly trying to get discussion started on the list, and people =
thinking about what questions need to bb worked on:)

Yes privacy is important and minimizing disclosure seems like a =
important principal.

The Server knowing something about the client may allow it to take other =
precautions or restrict the access of a token to match the security of =
the environment.

The counter argument would be that to trust what the client is saying =
may lead to the question of attestation keys and other things that are =
in the FIDO specs.
That is probably more complicated than we want.   Though if people think =
it is important get it on the table early.

OAuth is working on two binding/presentment specs for access tokens.
One that is signed HTTP requests =
https://tools.ietf.org/html/draft-ietf-oauth-signed-http-request-01 that =
is defiantly a layer above tokbind.
The other is binding the token to the TLS session talked about in =
http://tools.ietf.org/html/draft-tschofenig-oauth-hotk-03#section-3.2.2

The POP security architecture doc may be a good place to look for people =
that want to find out what OAuth is thinking =
https://tools.ietf.org/html/draft-ietf-oauth-pop-architecture-01

OAuth can define cannel binding of tokens for native apps with plain =
TLS.   I suppose that could also be done using tokbind, though depending =
on key management that may or may not be a good idea compared to
regular mutual TLS.

Tying the TLS keys to sessions is perhaps the most privacy preserving =
way to do it but may loose many of the other benefits. =20
Some people believe that question should be out of scope and left to =
implementations.

If that is the case then nailing that down early will save a lot of =
discussion, though I think it is something the WG will need to =
understand even if it is only represented in our security and privacy =
considerations.

John B.

> On Mar 15, 2015, at 3:11 PM, Ilari Liusvaara =
<ilari.liusvaara@elisanet.fi> wrote:
>=20
> On Sun, Mar 15, 2015 at 02:12:57PM -0300, John Bradley wrote:
>=20
>=20
> [On what should be exposed]
>=20
> I say there is very little server end can sanely do with information
> about the client environment (besides just denying service if it
> doesn't think client environment is "strong" enough, where defintion
> of "strong" can be highly variable and might not have anything to do
> with actual security).
>=20
> Additionally, every piece of information exposed besides the public
> key (where type is part of public key) is potential extra privacy
> risk. Some kinds of information (including device identity) are
> especially sensitive.
>=20
>=20
> [On what layer to bind]
>=20
> I think the deployment model web browsers have in mind is either:
> 1) End-to-end TLS connection between client and server.
> 2) Connection from client to gateway trusted by origin (CDN, load-
>   balancer, etc...), which can extract the identifier for server
>   use.
>=20
> Both can be done fine by binding to TLS layer.
>=20
> It seems that OAuth folks are wanting to do end-to-end binding at
> layer above TLS (as TLS lacks enough reach according them).
>=20
>=20
> [On scope of secrets]
>=20
> Even if secrets are at TLS layer, it does not follow that keys are
> scoped to TLS library.
>=20
> My view is that the secrets should be associated with tokens they
> protect and pinned only by those tokens and active sessions. That
> is, once all tokens and sessions using secret go away, that secret
> is deleted.
>=20
> Also, obviously any application accessing the token needs access to
> the corresponding secret.
>=20
> This kind of management is obviously greatly above TLS layer.
>=20
>=20
>=20
> -Ilari


--Apple-Mail=_55DB3D02-1913-4A41-B530-9ECF08C5FB9F
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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--Apple-Mail=_55DB3D02-1913-4A41-B530-9ECF08C5FB9F--


From nobody Mon Mar 16 04:05:57 2015
Return-Path: <ilari.liusvaara@elisanet.fi>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DA3DE1A8701 for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 04:05:56 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level: 
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001] autolearn=ham
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 de4lUskcM5GI for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 04:05:52 -0700 (PDT)
Received: from emh01.mail.saunalahti.fi (emh01.mail.saunalahti.fi [62.142.5.107]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 438021A8700 for <unbearable@ietf.org>; Mon, 16 Mar 2015 04:05:51 -0700 (PDT)
Received: from LK-Perkele-VII (a88-112-44-140.elisa-laajakaista.fi [88.112.44.140]) by emh01.mail.saunalahti.fi (Postfix) with ESMTP id 8A85790076; Mon, 16 Mar 2015 13:05:48 +0200 (EET)
Date: Mon, 16 Mar 2015 13:05:48 +0200
From: Ilari Liusvaara <ilari.liusvaara@elisanet.fi>
To: John Bradley <ve7jtb@ve7jtb.com>
Message-ID: <20150316110548.GA2214@LK-Perkele-VII>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com> <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com> <20150315181130.GA16994@LK-Perkele-VII> <2EFCA3AA-D611-4DAE-AB28-799327B08292@ve7jtb.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <2EFCA3AA-D611-4DAE-AB28-799327B08292@ve7jtb.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: Ilari Liusvaara <ilari.liusvaara@elisanet.fi>
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/BlSb2_B-woSCdD1lzAWuBpD6-54>
Cc: Anthony Nadalin <tonynad@microsoft.com>, Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, Phil Hunt <phil.hunt@oracle.com>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 16 Mar 2015 11:05:57 -0000

On Sun, Mar 15, 2015 at 04:34:24PM -0300, John Bradley wrote:

Also, it occurs to me that scope of keys can be even smaller than
an application. What if the application is a browser that support
profiles? Each profile has its own cookie jar and thus should have
its own keys.

It gets even more fine-grained if "private browsing" is considered
as each private browsing context needs its own keys.
 
> Yes privacy is important and minimizing disclosure seems like a
> important principal.

For some, "important" about privacy is a major understatement.

> The Server knowing something about the client may allow it to take
> other precautions or restrict the access of a token to match the
> security of the environment.

Well, I don't see great amounts of difference. By the time one can
dump the keys from well-made SW-only solution, one is very likely
in position to do great deal of damage without dumping the keys
anyway.

> The counter argument would be that to trust what the client is saying
> may lead to the question of attestation keys and other things that are
> in the FIDO specs. That is probably more complicated than we want. 
> Though if people think it is important get it on the table early.

The reason why attestation is so complicated in FIDO is because
of they need to do it in a way that doesn't break the web (e.g. by
allowing cross-domain tracking).

> OAuth is working on two binding/presentment specs for access tokens.
> One that is signed HTTP requests
> https://tools.ietf.org/html/draft-ietf-oauth-signed-http-request-01
> that is defiantly a layer above tokbind. The other is binding the token
> to the TLS session talked about in
> http://tools.ietf.org/html/draft-tschofenig-oauth-hotk-03#section-3.2.2

I would say that if one has choice between binding to TLS session and
HTTP-level request signing, one should go to first. It is MUCH simpler.

> The POP security architecture doc may be a good place to look for people
> that want to find out what OAuth is thinking
> https://tools.ietf.org/html/draft-ietf-oauth-pop-architecture-01

Reading that document, I see:

"3. Use Cases"
"3.4. Offering Application Layer End-to-End Security"

That is precisely "TLS does not have enough reach".

Whereas the two starting documents (implicitly representing two major
browser vendors) implicitly consider TLS to have enough reach.

> OAuth can define cannel binding of tokens for native apps with plain TLS.
> I suppose that could also be done using tokbind, though depending on key
> management that may or may not be a good idea compared to regular mutual
> TLS.

I consider mutual TLS much more secure than messing with bearer tokens
and signature keys protecting those (of course, mutual TLS has usability
problems in practice).

> Tying the TLS keys to sessions is perhaps the most privacy preserving
> way to do it but may loose many of the other benefits. Some people
> believe that question should be out of scope and left to implementations.

It was tying to both sessions and tokens. Basically, it seems to me that
doing anything beyond that opens up new distinct ways of tracking and
such.

TLS sessions alone are not long-lived enough (otherwise this whole effort
would not be needed). And pinning to tokens alone just plain does not
work (keys would immediately be lost).


-Ilari


From nobody Mon Mar 16 09:52:52 2015
Return-Path: <wmills_92105@yahoo.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 598AC1A88BF for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 09:52:50 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.509
X-Spam-Level: 
X-Spam-Status: No, score=-1.509 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_ENVFROM_END_DIGIT=0.25, FREEMAIL_FROM=0.001, FREEMAIL_REPLYTO_END_DIGIT=0.25, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 66vGew_Mlba2 for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 09:52:48 -0700 (PDT)
Received: from nm1-vm1.bullet.mail.bf1.yahoo.com (nm1-vm1.bullet.mail.bf1.yahoo.com [98.139.213.163]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A143F1A88BC for <unbearable@ietf.org>; Mon, 16 Mar 2015 09:52:47 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s2048; t=1426524766; bh=esnfzUT9tnReOWYKJ2YNgPuBuNH+sHz2U/p2HpiLfZw=; h=Date:From:Reply-To:To:Cc:In-Reply-To:References:Subject:From:Subject; b=OdI0py9vyFzLBB1IK0QXmFRcjMTpGpc2cHepxZoL0Eakc1oxS68CKAdsR0f960p8JNaRXoimlpNG9gZh8rFPbORas6uIHMdG6MJ6d5v5cy1jF0gb4GY+Mf2LHAyZS9iAcNobtYwGum1M6Nap5L7VKebLEosCrp2JJ5+vDc4zRI4JJfYZ2lHz5P29npoq0VPWYiKhZ3jvK1Jg8iIYmOUmhlozGToOfpRB7/i7gz123uZKJ+G/2d8kw/xEEW3RtD4trFi53rfVElfx9NQQXNt2mW+HRwamsXyuhrPyQduZ+rm71wi8LNqYAd918+R4nNdVFPC0Fs4FNWZLOvPXkWqRYQ==
Received: from [98.139.214.32] by nm1.bullet.mail.bf1.yahoo.com with NNFMP; 16 Mar 2015 16:52:46 -0000
Received: from [98.139.212.217] by tm15.bullet.mail.bf1.yahoo.com with NNFMP;  16 Mar 2015 16:52:46 -0000
Received: from [127.0.0.1] by omp1026.mail.bf1.yahoo.com with NNFMP; 16 Mar 2015 16:52:46 -0000
X-Yahoo-Newman-Property: ymail-3
X-Yahoo-Newman-Id: 774982.69636.bm@omp1026.mail.bf1.yahoo.com
X-YMail-OSG: HOGdPzAVM1mHyuiWWyc1BeobvDIJ_r6Euag5ZMEeDlCXOqxK1zyYqgphgtyS3aP f_hG7N6puX0MIwAQHqwubHhxneBqZKmkGJCe0mS4oSzzcjjqE.lrWKCbq1_x60r6Lw16tQ_nO6Ww VakaQ1HQkch0bwk4kr4ay7vmVL7vOg7ywsVywm5URrcyQip_nk8CLTersm1FcJgo7osj9HfBVopc ehTcceG97d9KBw.IHSgv66mzVX0O5GFHjD02cQMwV0B7fGGDhazWbVQ.PSN7XBApIWxGM7ttVNEO WexFVZonEBFR8fLjBF8FdEVUcFbUy2uZQUJYzGZFJ.AK36CrXtcmTLoWhorb0YNU4uIBrUKpgvqt qsoXP.mG1Fx.pR.zlEvopKAtFOEM_VK4Gq3d1ITpKwLuW9KSJUgkFTN2XNmYaHDJyD_QhciGRuPJ wrt.eY_58s1XLpgfEqed6N6_t_TgwDzW_RBD_yNIkgmWL3V9.gRf0H4mIhp1BFwRbvvvQ8NbnPVN 0ke9FoAfYLLwLkK33eL_GnM_tLUzfoNP4bV30ayhZXLhW4gSkKg--
Received: by 76.13.26.143; Mon, 16 Mar 2015 16:52:46 +0000 
Date: Mon, 16 Mar 2015 16:52:45 +0000 (UTC)
From: Bill Mills <wmills_92105@yahoo.com>
To: Anthony Nadalin <tonynad@microsoft.com>,  John Bradley <ve7jtb@ve7jtb.com>, Phil Hunt <phil.hunt@oracle.com>
Message-ID: <481847549.568021.1426524765968.JavaMail.yahoo@mail.yahoo.com>
In-Reply-To: <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com>
References: <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;  boundary="----=_Part_568020_1405232540.1426524765962"
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/jXnAzNuC5alH9xg3_IlBfsfyvvA>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: Bill Mills <wmills_92105@yahoo.com>
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 16 Mar 2015 16:52:50 -0000

------=_Part_568020_1405232540.1426524765962
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

If there is a standard that determines how secrets are handled, such as TLS=
, I think it's reasonable to use/reference that but it may limit you to bei=
ng able to use only that if it's MTI. =C2=A0Specifying how keys are managed=
 and stored in a spec is very very likely to just make the spec less adopta=
ble unless that's the main purpose of the spec and it's intended to be used=
 on multiple platforms.
I agree with Tony,key management should be out of scope here. =C2=A0I'd go =
so far as to say it should be explicitly out of scope. =C2=A0I think this c=
onversation is similar to the question in the POP token drafts where there =
is a push to mandate how secrets are managed/shared/populated as an MTI ite=
m, and I think this is a big mistake.
If there is really a need for a new key distribution/management/sharing sta=
ndard that's probably worth it's own WG. =C2=A0It's a rich space though. =
=C2=A0Lots of solutions out there.=20


     On Saturday, March 14, 2015 5:37 PM, Anthony Nadalin <tonynad@microsof=
t.com> wrote:
  =20

 Understanding how keys are shared between apps, is an app thing as there c=
an be countless ways, the app that puts the key(s) there needs to be able t=
o document how it expects the sharing to occur.

-----Original Message-----
From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John Bra=
dley
Sent: Friday, March 13, 2015 12:58 PM
To: Phil Hunt
Cc: Andrei Popov; unbearable@ietf.org
Subject: Re: [Unbearable] Scoping of keys on the client.

I think it is important to think through how the choices may impact how pro=
tocols can use token binding.=20

The NAAPS profile of connect is a possible candidate for using token bindin=
g but that is highly dependent on how keys are shared on the client.=20

A simple OAuth use case is using token binding to secure a refresh token.=
=C2=A0=20

How secure that is would depend on what other apps on the device may be abl=
e to use that token if it leaked.=20

Understanding how keys are shared between apps is I think important.=20

John B


Sent from my iPhone

> On Mar 13, 2015, at 2:46 PM, Phil Hunt <phil.hunt@oracle.com> wrote:
>=20
> The problem being that this just becomes device binding and app to app to=
ken binding still isn't possible.=20
>=20
> Do we care?
>=20
> Phil
>=20
>> On Mar 13, 2015, at 10:18, Andrei Popov <Andrei.Popov@microsoft.com> wro=
te:
>>=20
>> Thanks for raising this topic John.
>>=20
>>> I found some people that believe all apps sharing a common TLS library =
would be able to share the same ID/key for a origin.
>> In our prototypes so far (in Win10), the Token Binding Protocol implemen=
tation, the Token Binding key storage APIs, and the TLS library are three s=
eparate components. Among other things, this allows 3-rd-party apps share T=
oken Binding keys and achieve single sign-on regardless of the TLS library =
they might be using (if they choose to do so).
>>=20
>>> The different views impact security considerations in native environmen=
ts.=C2=A0 As an example can you pass a token from one app to another and ha=
ve it still work to the same origin, or is it a security feature that it wo=
n't work and the token cant be stolen from a app and re-used by a different=
 app on the same device.
>> I think that single sign-on between apps and browsers is at least a usef=
ul option.
>>=20
>>> I don't think there is anything in the specs on this.
>>> Figuring out where some of these issues get documented might be useful.
>> Documenting key scoping is indeed useful; on the other hand standardizin=
g the specifics might require standardizing user and application isolation =
models across platforms, which I don't think would be in scope for this for=
um. Perhaps we could agree on some general principles and include implement=
ation guidance on Token Binding key scoping?
>>=20
>> Cheers,
>>=20
>> Andrei
>>=20
>> -----Original Message-----
>> From: Unbearable [mailto:unbearable-bounces@ietf.org] On Behalf Of John =
Bradley
>> Sent: Friday, March 13, 2015 9:57 AM
>> To: unbearable@ietf.org
>> Subject: [Unbearable] Scoping of keys on the client.
>>=20
>> Now that we are officially a WG I have a question (no hat)
>>=20
>> One thing that I identified as a potential issue talking to people who a=
re working on early deployments, is the expectations around how keys are sh=
ared between applications.
>>=20
>> I found some people that believe all apps sharing a common TLS library w=
ould be able to share the same ID/key for a origin.
>>=20
>> Other folks were under the impression that each application would have i=
ts own key for an origin.
>>=20
>> The different views impact security considerations in native environment=
s.=C2=A0 As an example can you pass a token from one app to another and hav=
e it still work to the same origin, or is it a security feature that it won=
't work and the token cant be stolen from a app and re-used by a different =
app on the same device.
>>=20
>> There was a middle view by the MDM people that the ID/key would be scope=
d to a profile/container so that all the enterprise apps would share ID/key=
 but that social apps would not.
>>=20
>> I don't know that there is necessarily one right answer, as this may be =
implementation dependent.=C2=A0=20
>>=20
>> I don't think there is anything in the specs on this.=C2=A0 One reason I=
 ask is that some of those differences impact on Native Application Single =
Signon and OAuth.
>>=20
>> In some of the Windows prototypes I understand there is a Key Storage AP=
I that is used to share keys between the browser and apps.=C2=A0=20
>>=20
>> Android, Windows, iOS , OSX may all have implementation specific details=
, but people building applications that leverage token binding will need to=
 understand the differences as they may have security and privacy implicati=
ons.
>>=20
>> Figuring out where some of these issues get documented might be useful.
>>=20
>> John B.
>>=20
>> _______________________________________________
>> Unbearable mailing list
>> Unbearable@ietf.org
>> https://www.ietf.org/mailman/listinfo/unbearable

_______________________________________________
Unbearable mailing list
Unbearable@ietf.org
https://www.ietf.org/mailman/listinfo/unbearable

_______________________________________________
Unbearable mailing list
Unbearable@ietf.org
https://www.ietf.org/mailman/listinfo/unbearable


  
------=_Part_568020_1405232540.1426524765962
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<html><body><div style=3D"color:#000; background-color:#fff; font-family:He=
lveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif;fo=
nt-size:12px"><div id=3D"yui_3_16_0_1_1426272734568_297678" dir=3D"ltr"><sp=
an id=3D"yui_3_16_0_1_1426272734568_298050">If there is a standard that det=
ermines how secrets are handled, such as TLS, I think it's reasonable to us=
e/reference that but it may limit you to being able to use only that if it'=
s MTI. &nbsp;</span><span style=3D"font-size: 12px;" id=3D"yui_3_16_0_1_142=
6272734568_300297">Specifying how keys are managed and stored in a spec is =
very very likely to just make the spec less adoptable unless that's the mai=
n purpose of the spec and it's intended to be used on multiple platforms.</=
span></div><div id=3D"yui_3_16_0_1_1426272734568_297678" dir=3D"ltr"><span =
style=3D"font-size: 12px;"><br></span></div><div id=3D"yui_3_16_0_1_1426272=
734568_297678" dir=3D"ltr"><span style=3D"font-size: 12px;" id=3D"yui_3_16_=
0_1_1426272734568_301601">I agree with Tony,key management should be out of=
 scope here. &nbsp;I'd go so far as to say it should be explicitly out of s=
cope. &nbsp;I think this conversation is similar to the question in the POP=
 token drafts where there is a push to mandate how secrets are managed/shar=
ed/populated as an MTI item, and I think this is a big mistake.</span></div=
><div id=3D"yui_3_16_0_1_1426272734568_297678" dir=3D"ltr"><span style=3D"f=
ont-size: 12px;"><br></span></div><div id=3D"yui_3_16_0_1_1426272734568_297=
678" dir=3D"ltr"><span style=3D"font-size: 12px;" id=3D"yui_3_16_0_1_142627=
2734568_304709">If there is really a need for a new key distribution/manage=
ment/sharing standard that's probably worth it's own WG. &nbsp;It's a rich =
space though. &nbsp;Lots of solutions out there.</span></div>  <br><div cla=
ss=3D"qtdSeparateBR"><br><br></div><div class=3D"yahoo_quoted" style=3D"dis=
play: block;"> <div style=3D"font-family: HelveticaNeue, Helvetica Neue, He=
lvetica, Arial, Lucida Grande, sans-serif; font-size: 12px;"> <div style=3D=
"font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grand=
e, sans-serif; font-size: 16px;"> <div dir=3D"ltr"> <font size=3D"2" face=
=3D"Arial"> On Saturday, March 14, 2015 5:37 PM, Anthony Nadalin &lt;tonyna=
d@microsoft.com&gt; wrote:<br> </font> </div>  <br><br> <div class=3D"y_msg=
_container">Understanding how keys are shared between apps, is an app thing=
 as there can be countless ways, the app that puts the key(s) there needs t=
o be able to document how it expects the sharing to occur.<br clear=3D"none=
"><br clear=3D"none">-----Original Message-----<br clear=3D"none">From: Unb=
earable [mailto:<a shape=3D"rect" ymailto=3D"mailto:unbearable-bounces@ietf=
.org" href=3D"mailto:unbearable-bounces@ietf.org">unbearable-bounces@ietf.o=
rg</a>] On Behalf Of John Bradley<br clear=3D"none">Sent: Friday, March 13,=
 2015 12:58 PM<br clear=3D"none">To: Phil Hunt<br clear=3D"none">Cc: Andrei=
 Popov; <a shape=3D"rect" ymailto=3D"mailto:unbearable@ietf.org" href=3D"ma=
ilto:unbearable@ietf.org">unbearable@ietf.org</a><br clear=3D"none">Subject=
: Re: [Unbearable] Scoping of keys on the client.<br clear=3D"none"><br cle=
ar=3D"none">I think it is important to think through how the choices may im=
pact how protocols can use token binding. <br clear=3D"none"><br clear=3D"n=
one">The NAAPS profile of connect is a possible candidate for using token b=
inding but that is highly dependent on how keys are shared on the client. <=
br clear=3D"none"><br clear=3D"none">A simple OAuth use case is using token=
 binding to secure a refresh token.&nbsp; <br clear=3D"none"><br clear=3D"n=
one">How secure that is would depend on what other apps on the device may b=
e able to use that token if it leaked. <br clear=3D"none"><br clear=3D"none=
">Understanding how keys are shared between apps is I think important. <br =
clear=3D"none"><br clear=3D"none">John B<br clear=3D"none"><br clear=3D"non=
e"><br clear=3D"none">Sent from my iPhone<br clear=3D"none"><br clear=3D"no=
ne">&gt; On Mar 13, 2015, at 2:46 PM, Phil Hunt &lt;<a shape=3D"rect" ymail=
to=3D"mailto:phil.hunt@oracle.com" href=3D"mailto:phil.hunt@oracle.com">phi=
l.hunt@oracle.com</a>&gt; wrote:<br clear=3D"none">&gt; <br clear=3D"none">=
&gt; The problem being that this just becomes device binding and app to app=
 token binding still isn't possible. <br clear=3D"none">&gt; <br clear=3D"n=
one">&gt; Do we care?<br clear=3D"none">&gt; <br clear=3D"none">&gt; Phil<b=
r clear=3D"none">&gt; <br clear=3D"none">&gt;&gt; On Mar 13, 2015, at 10:18=
, Andrei Popov &lt;<a shape=3D"rect" ymailto=3D"mailto:Andrei.Popov@microso=
ft.com" href=3D"mailto:Andrei.Popov@microsoft.com">Andrei.Popov@microsoft.c=
om</a>&gt; wrote:<br clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;&gt; Th=
anks for raising this topic John.<br clear=3D"none">&gt;&gt; <br clear=3D"n=
one">&gt;&gt;&gt; I found some people that believe all apps sharing a commo=
n TLS library would be able to share the same ID/key for a origin.<br clear=
=3D"none">&gt;&gt; In our prototypes so far (in Win10), the Token Binding P=
rotocol implementation, the Token Binding key storage APIs, and the TLS lib=
rary are three separate components. Among other things, this allows 3-rd-pa=
rty apps share Token Binding keys and achieve single sign-on regardless of =
the TLS library they might be using (if they choose to do so).<br clear=3D"=
none">&gt;&gt; <br clear=3D"none">&gt;&gt;&gt; The different views impact s=
ecurity considerations in native environments.&nbsp;  As an example can you=
 pass a token from one app to another and have it still work to the same or=
igin, or is it a security feature that it won't work and the token cant be =
stolen from a app and re-used by a different app on the same device.<br cle=
ar=3D"none">&gt;&gt; I think that single sign-on between apps and browsers =
is at least a useful option.<br clear=3D"none">&gt;&gt; <br clear=3D"none">=
&gt;&gt;&gt; I don't think there is anything in the specs on this.<br clear=
=3D"none">&gt;&gt;&gt; Figuring out where some of these issues get document=
ed might be useful.<br clear=3D"none">&gt;&gt; Documenting key scoping is i=
ndeed useful; on the other hand standardizing the specifics might require s=
tandardizing user and application isolation models across platforms, which =
I don't think would be in scope for this forum. Perhaps we could agree on s=
ome general principles and include implementation guidance on Token Binding=
 key scoping?<br clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;&gt; Cheers=
,<br clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;&gt; Andrei<br clear=3D=
"none">&gt;&gt; <br clear=3D"none">&gt;&gt; -----Original Message-----<br c=
lear=3D"none">&gt;&gt; From: Unbearable [mailto:<a shape=3D"rect" ymailto=
=3D"mailto:unbearable-bounces@ietf.org" href=3D"mailto:unbearable-bounces@i=
etf.org">unbearable-bounces@ietf.org</a>] On Behalf Of John Bradley<br clea=
r=3D"none">&gt;&gt; Sent: Friday, March 13, 2015 9:57 AM<br clear=3D"none">=
&gt;&gt; To: <a shape=3D"rect" ymailto=3D"mailto:unbearable@ietf.org" href=
=3D"mailto:unbearable@ietf.org">unbearable@ietf.org</a><br clear=3D"none">&=
gt;&gt; Subject: [Unbearable] Scoping of keys on the client.<br clear=3D"no=
ne">&gt;&gt; <br clear=3D"none">&gt;&gt; Now that we are officially a WG I =
have a question (no hat)<br clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;=
&gt; One thing that I identified as a potential issue talking to people who=
 are working on early deployments, is the expectations around how keys are =
shared between applications.<br clear=3D"none">&gt;&gt; <br clear=3D"none">=
&gt;&gt; I found some people that believe all apps sharing a common TLS lib=
rary would be able to share the same ID/key for a origin.<br clear=3D"none"=
>&gt;&gt; <br clear=3D"none">&gt;&gt; Other folks were under the impression=
 that each application would have its own key for an origin.<br clear=3D"no=
ne">&gt;&gt; <br clear=3D"none">&gt;&gt; The different views impact securit=
y considerations in native environments.&nbsp;  As an example can you pass =
a token from one app to another and have it still work to the same origin, =
or is it a security feature that it won't work and the token cant be stolen=
 from a app and re-used by a different app on the same device.<br clear=3D"=
none">&gt;&gt; <br clear=3D"none">&gt;&gt; There was a middle view by the M=
DM people that the ID/key would be scoped to a profile/container so that al=
l the enterprise apps would share ID/key but that social apps would not.<br=
 clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;&gt; I don't know that ther=
e is necessarily one right answer, as this may be implementation dependent.=
&nbsp; <br clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;&gt; I don't thin=
k there is anything in the specs on this.&nbsp;  One reason I ask is that s=
ome of those differences impact on Native Application Single Signon and OAu=
th.<br clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;&gt; In some of the W=
indows prototypes I understand there is a Key Storage API that is used to s=
hare keys between the browser and apps.&nbsp; <br clear=3D"none">&gt;&gt; <=
br clear=3D"none">&gt;&gt; Android, Windows, iOS , OSX may all have impleme=
ntation specific details, but people building applications that leverage to=
ken binding will need to understand the differences as they may have securi=
ty and privacy implications.<br clear=3D"none">&gt;&gt; <br clear=3D"none">=
&gt;&gt; Figuring out where some of these issues get documented might be us=
eful.<br clear=3D"none">&gt;&gt; <br clear=3D"none">&gt;&gt; John B.<br cle=
ar=3D"none">&gt;&gt; <br clear=3D"none">&gt;&gt; __________________________=
_____________________<br clear=3D"none">&gt;&gt; Unbearable mailing list<br=
 clear=3D"none">&gt;&gt; <a shape=3D"rect" ymailto=3D"mailto:Unbearable@iet=
f.org" href=3D"mailto:Unbearable@ietf.org">Unbearable@ietf.org</a><br clear=
=3D"none">&gt;&gt; <a shape=3D"rect" href=3D"https://www.ietf.org/mailman/l=
istinfo/unbearable" target=3D"_blank">https://www.ietf.org/mailman/listinfo=
/unbearable</a><div class=3D"yqt9066052574" id=3D"yqtfd72020"><br clear=3D"=
none"><br clear=3D"none">_______________________________________________<br=
 clear=3D"none">Unbearable mailing list<br clear=3D"none"><a shape=3D"rect"=
 ymailto=3D"mailto:Unbearable@ietf.org" href=3D"mailto:Unbearable@ietf.org"=
>Unbearable@ietf.org</a><br clear=3D"none"><a shape=3D"rect" href=3D"https:=
//www.ietf.org/mailman/listinfo/unbearable" target=3D"_blank">https://www.i=
etf.org/mailman/listinfo/unbearable</a><br clear=3D"none"><br clear=3D"none=
">_______________________________________________<br clear=3D"none">Unbeara=
ble mailing list<br clear=3D"none"><a shape=3D"rect" ymailto=3D"mailto:Unbe=
arable@ietf.org" href=3D"mailto:Unbearable@ietf.org">Unbearable@ietf.org</a=
><br clear=3D"none"><a shape=3D"rect" href=3D"https://www.ietf.org/mailman/=
listinfo/unbearable" target=3D"_blank">https://www.ietf.org/mailman/listinf=
o/unbearable</a><br clear=3D"none"></div><br><br></div>  </div> </div>  </d=
iv></div></body></html>
------=_Part_568020_1405232540.1426524765962--


From nobody Mon Mar 16 12:26:15 2015
Return-Path: <squid3@treenet.co.nz>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C76511A8BB3 for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 12:26:10 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.923
X-Spam-Level: 
X-Spam-Status: No, score=0.923 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FH_HOST_EQ_D_D_D_D=0.765, FH_HOST_EQ_D_D_D_DB=0.888, HOST_EQ_STATIC=1.172, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=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 IpODxuePcBln for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 12:26:08 -0700 (PDT)
Received: from treenet.co.nz (121-99-228-82.static.orcon.net.nz [121.99.228.82]) by ietfa.amsl.com (Postfix) with ESMTP id E391F1A8BB6 for <unbearable@ietf.org>; Mon, 16 Mar 2015 12:26:02 -0700 (PDT)
Received: from [192.168.20.33] (121-99-59-16.bng1.tvc.orcon.net.nz [121.99.59.16]) by treenet.co.nz (Postfix) with ESMTP id 6CEB2E6F07 for <unbearable@ietf.org>; Tue, 17 Mar 2015 07:25:30 +1200 (NZST)
Message-ID: <55072E27.5080001@treenet.co.nz>
Date: Tue, 17 Mar 2015 08:25:27 +1300
From: Amos Jeffries <squid3@treenet.co.nz>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: unbearable@ietf.org
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com> <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com> <20150315181130.GA16994@LK-Perkele-VII> <2EFCA3AA-D611-4DAE-AB28-799327B08292@ve7jtb.com> <20150316110548.GA2214@LK-Perkele-VII>
In-Reply-To: <20150316110548.GA2214@LK-Perkele-VII>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/AnCWQvb6QmGQDl90DKxyBRsKzFU>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 16 Mar 2015 19:26:10 -0000

On 17/03/2015 12:05 a.m., Ilari Liusvaara wrote:
> On Sun, Mar 15, 2015 at 04:34:24PM -0300, John Bradley wrote:
> 
> Also, it occurs to me that scope of keys can be even smaller than
> an application. What if the application is a browser that support
> profiles? Each profile has its own cookie jar and thus should have
> its own keys.
> 
> It gets even more fine-grained if "private browsing" is considered
> as each private browsing context needs its own keys.
>  
>> Yes privacy is important and minimizing disclosure seems like a
>> important principal.
> 
> For some, "important" about privacy is a major understatement.
> 
>> The Server knowing something about the client may allow it to take
>> other precautions or restrict the access of a token to match the
>> security of the environment.
> 
> Well, I don't see great amounts of difference. By the time one can
> dump the keys from well-made SW-only solution, one is very likely
> in position to do great deal of damage without dumping the keys
> anyway.
> 
>> The counter argument would be that to trust what the client is saying
>> may lead to the question of attestation keys and other things that are
>> in the FIDO specs. That is probably more complicated than we want. 
>> Though if people think it is important get it on the table early.
> 
> The reason why attestation is so complicated in FIDO is because
> of they need to do it in a way that doesn't break the web (e.g. by
> allowing cross-domain tracking).

Sigh. Please dont use that phase "break the web". Its been over inflated
with meanings and political posturing.

Your example is even one situation in which "break" implies a wrong
meaning. Cross-domain tracking breaks *Trust*, even while the web at a
protocol/specification level continues to work as specified with regular
users not aware of anything.

> 
>> OAuth is working on two binding/presentment specs for access tokens.
>> One that is signed HTTP requests
>> https://tools.ietf.org/html/draft-ietf-oauth-signed-http-request-01
>> that is defiantly a layer above tokbind. The other is binding the token
>> to the TLS session talked about in
>> http://tools.ietf.org/html/draft-tschofenig-oauth-hotk-03#section-3.2.2
> 
> I would say that if one has choice between binding to TLS session and
> HTTP-level request signing, one should go to first. It is MUCH simpler.

No, they are approximately the same in both complexity and difficulty.

With TLS one has to account for multiple HTTP layer hops with different
TLS connections, even duifferent types of TLS, and messages in each
connection being multiplexed between different applications and/or clients.

With HTTP signing one has to account for information leakage, what
headers to include in the signature, which to omit, how to do the
signing securely, and cached responses.

Both gives token binding from point A to B, where either A or B could be
endpoints or midpoints.

> 
>> The POP security architecture doc may be a good place to look for people
>> that want to find out what OAuth is thinking
>> https://tools.ietf.org/html/draft-ietf-oauth-pop-architecture-01
> 
> Reading that document, I see:
> 
> "3. Use Cases"
> "3.4. Offering Application Layer End-to-End Security"
> 
> That is precisely "TLS does not have enough reach".
> 
> Whereas the two starting documents (implicitly representing two major
> browser vendors) implicitly consider TLS to have enough reach.

Warning to the wise; this is the same situation faced by HTTP WG
developing HTTP/2 out of SPDY. The browser vendors driving the
specification showed a bias towards the belief that TLS was end-to-end
across the entire Interent. The existence of that e2e assumption in the
very core concepts of the SPDY draft and carried over to HTTP/2 draft
have lead to some very, very long debates and division within the WG.

Having accurate information and WG consensus on what exactly is the
reach of TLS (or any protocol being bound to) would help avoid a lot of
that. If nothing else it would help to explicitly prohibit binding use
in cases where required reach is not being guaranteed and/or supplied.

> 
>> OAuth can define cannel binding of tokens for native apps with plain TLS.
>> I suppose that could also be done using tokbind, though depending on key
>> management that may or may not be a good idea compared to regular mutual
>> TLS.
> 
> I consider mutual TLS much more secure than messing with bearer tokens
> and signature keys protecting those (of course, mutual TLS has usability
> problems in practice).

Ditto. Also its worth noting that mutual TLS authentication may exist
between any two parties in the end-to-end messaging pathway. There are
numerous permutation possibilities where the application layer is fully
disconnected from the TLS auth.

The following HTTP structure is quite popular:
 client -> ISP proxy -> CDN -> origin

where 5 of the 6 permutations and point-to-point, and only 2 have
client->origin contact in any way.

> 
>> Tying the TLS keys to sessions is perhaps the most privacy preserving
>> way to do it but may loose many of the other benefits. Some people
>> believe that question should be out of scope and left to implementations.
> 
> It was tying to both sessions and tokens. Basically, it seems to me that
> doing anything beyond that opens up new distinct ways of tracking and
> such.
> 
> TLS sessions alone are not long-lived enough (otherwise this whole effort
> would not be needed). And pinning to tokens alone just plain does not
> work (keys would immediately be lost).

Agreed. IMO, leaving it to implemetations without any guidance is a mistake.

At the very least implementors need a set of pros/cons and security
considerations to base any such decisions on. Without that there are
guaranteed interoperability and security flaws.


Amos


From nobody Mon Mar 16 12:50:52 2015
Return-Path: <vanupam@google.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3205A1A9073 for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 12:50:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.388
X-Spam-Level: 
X-Spam-Status: No, score=-1.388 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 CruivbLKOpuM for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 12:50:48 -0700 (PDT)
Received: from mail-qg0-x232.google.com (mail-qg0-x232.google.com [IPv6:2607:f8b0:400d:c04::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 944471A905E for <unbearable@ietf.org>; Mon, 16 Mar 2015 12:50:43 -0700 (PDT)
Received: by qgfa8 with SMTP id a8so50332063qgf.0 for <unbearable@ietf.org>; Mon, 16 Mar 2015 12:50:42 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=qmDF8PDVlPbaa8x3i4SJ85cTvsLjzBoXAzSSSxXCK8c=; b=QCsrehv64jrsUexoTp1arpDO1815bqy4pNDR5DSYatcaTqgE52lGpjQsZBK7t+f1zR Jt7CBCJIZHPX4ZAq2iLmvdv1HIrpDstbmSpQCp3bcJfonv1MST7ZlRhs1HjLSZALuV2O UktU5wbzjXjNHflD/Dogn7Gkm8JrZRYAHUxoL7WAn0OAfO11RsQrkSCA/ndJ6e3GpBDZ Hf6Jx/GNttyzFodrkkvl2v1Y9vmK+BKciHpMMfe4yWKAotuGuW4VhwgWaAhccOCFJF9f o6gpbpmKKyiOhIiQM9htxtyt9lcuztxSvYyIWaaUOrf0RNpxaTy8fetfcq6X1Wh8gevH hpAw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc:content-type; bh=qmDF8PDVlPbaa8x3i4SJ85cTvsLjzBoXAzSSSxXCK8c=; b=fa5gbgHjZEZJZQ0p+nZ3v2MjASpDhBybglEpyNmEcuu09OYx5aV6p5DtH4RSfmRmfZ /KGW7R4faGJi/JRS9F5WfBUBCx09WUgeuet8I2V532ZIOzmyk0+e6OctRzEMN5WgAsMm esclpH8WXTX5SlFFBWltNuER2M4Bn5TkIVjk1qnE9JQYlx7hvRPoAVe9tWG64N0GrgS9 0sn4XCnTcHOUwPuD2bFEQfEtoZFnQJahv5Jkb5Zxln9OrFpQm2kRGilRhZIYDuWV9nfc mcTHxcnCWfkBmthODavR2jKKFPqbGqTbG1eMVDRTuSG6fs6zea9TSrqBo1rSxal/sov8 uAjQ==
X-Gm-Message-State: ALoCoQmOgUHjflJQYfMurSBiItkMUJuX1UDV+MroBYE+uUYTnfXqinSD5hISpKNZShh/93rbj345
MIME-Version: 1.0
X-Received: by 10.55.55.81 with SMTP id e78mr124373548qka.107.1426535442776; Mon, 16 Mar 2015 12:50:42 -0700 (PDT)
Received: by 10.229.109.72 with HTTP; Mon, 16 Mar 2015 12:50:42 -0700 (PDT)
In-Reply-To: <55072E27.5080001@treenet.co.nz>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com> <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com> <20150315181130.GA16994@LK-Perkele-VII> <2EFCA3AA-D611-4DAE-AB28-799327B08292@ve7jtb.com> <20150316110548.GA2214@LK-Perkele-VII> <55072E27.5080001@treenet.co.nz>
Date: Mon, 16 Mar 2015 12:50:42 -0700
Message-ID: <CAOVPt=sj_UPHhU80OgC04GOHX4yyjnW8c30Wpuc-in2Rtdk_4A@mail.gmail.com>
From: Vinod Anupam <vanupam@google.com>
To: Amos Jeffries <squid3@treenet.co.nz>
Content-Type: multipart/alternative; boundary=001a11490ac853bcd505116d2afe
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/A-JI9b8uj2BmsO1AFnSrzVd_a_8>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 16 Mar 2015 19:50:51 -0000

--001a11490ac853bcd505116d2afe
Content-Type: text/plain; charset=UTF-8

On Mon, Mar 16, 2015 at 12:25 PM, Amos Jeffries <squid3@treenet.co.nz>
wrote:

> On 17/03/2015 12:05 a.m., Ilari Liusvaara wrote:
> > On Sun, Mar 15, 2015 at 04:34:24PM -0300, John Bradley wrote:
> >
> > Also, it occurs to me that scope of keys can be even smaller than
> > an application. What if the application is a browser that support
> > profiles? Each profile has its own cookie jar and thus should have
> > its own keys.
> >
> > It gets even more fine-grained if "private browsing" is considered
> > as each private browsing context needs its own keys.
> >
> >> Yes privacy is important and minimizing disclosure seems like a
> >> important principal.
> >
> > For some, "important" about privacy is a major understatement.
> >
> >> The Server knowing something about the client may allow it to take
> >> other precautions or restrict the access of a token to match the
> >> security of the environment.
> >
> > Well, I don't see great amounts of difference. By the time one can
> > dump the keys from well-made SW-only solution, one is very likely
> > in position to do great deal of damage without dumping the keys
> > anyway.
> >
> >> The counter argument would be that to trust what the client is saying
> >> may lead to the question of attestation keys and other things that are
> >> in the FIDO specs. That is probably more complicated than we want.
> >> Though if people think it is important get it on the table early.
> >
> > The reason why attestation is so complicated in FIDO is because
> > of they need to do it in a way that doesn't break the web (e.g. by
> > allowing cross-domain tracking).
>
> Sigh. Please dont use that phase "break the web". Its been over inflated
> with meanings and political posturing.
>
> Your example is even one situation in which "break" implies a wrong
> meaning. Cross-domain tracking breaks *Trust*, even while the web at a
> protocol/specification level continues to work as specified with regular
> users not aware of anything.
>
> >
> >> OAuth is working on two binding/presentment specs for access tokens.
> >> One that is signed HTTP requests
> >> https://tools.ietf.org/html/draft-ietf-oauth-signed-http-request-01
> >> that is defiantly a layer above tokbind. The other is binding the token
> >> to the TLS session talked about in
> >> http://tools.ietf.org/html/draft-tschofenig-oauth-hotk-03#section-3.2.2
> >
> > I would say that if one has choice between binding to TLS session and
> > HTTP-level request signing, one should go to first. It is MUCH simpler.
>
> No, they are approximately the same in both complexity and difficulty.
>
> With TLS one has to account for multiple HTTP layer hops with different
> TLS connections, even duifferent types of TLS, and messages in each
> connection being multiplexed between different applications and/or clients.
>
> With HTTP signing one has to account for information leakage, what
> headers to include in the signature, which to omit, how to do the
> signing securely, and cached responses.
>
> Both gives token binding from point A to B, where either A or B could be
> endpoints or midpoints.
>
> >
> >> The POP security architecture doc may be a good place to look for people
> >> that want to find out what OAuth is thinking
> >> https://tools.ietf.org/html/draft-ietf-oauth-pop-architecture-01
> >
> > Reading that document, I see:
> >
> > "3. Use Cases"
> > "3.4. Offering Application Layer End-to-End Security"
> >
> > That is precisely "TLS does not have enough reach".
> >
> > Whereas the two starting documents (implicitly representing two major
> > browser vendors) implicitly consider TLS to have enough reach.
>
> Warning to the wise; this is the same situation faced by HTTP WG
> developing HTTP/2 out of SPDY. The browser vendors driving the
> specification showed a bias towards the belief that TLS was end-to-end
> across the entire Interent. The existence of that e2e assumption in the
> very core concepts of the SPDY draft and carried over to HTTP/2 draft
> have lead to some very, very long debates and division within the WG.
>
> Having accurate information and WG consensus on what exactly is the
> reach of TLS (or any protocol being bound to) would help avoid a lot of
> that. If nothing else it would help to explicitly prohibit binding use
> in cases where required reach is not being guaranteed and/or supplied.
>
> >
> >> OAuth can define cannel binding of tokens for native apps with plain
> TLS.
> >> I suppose that could also be done using tokbind, though depending on key
> >> management that may or may not be a good idea compared to regular mutual
> >> TLS.
> >
> > I consider mutual TLS much more secure than messing with bearer tokens
> > and signature keys protecting those (of course, mutual TLS has usability
> > problems in practice).
>
> Ditto. Also its worth noting that mutual TLS authentication may exist
> between any two parties in the end-to-end messaging pathway. There are
> numerous permutation possibilities where the application layer is fully
> disconnected from the TLS auth.
>
> The following HTTP structure is quite popular:
>  client -> ISP proxy -> CDN -> origin
>
> where 5 of the 6 permutations and point-to-point, and only 2 have
> client->origin contact in any way.
>
> >
> >> Tying the TLS keys to sessions is perhaps the most privacy preserving
> >> way to do it but may loose many of the other benefits. Some people
> >> believe that question should be out of scope and left to
> implementations.
> >
> > It was tying to both sessions and tokens. Basically, it seems to me that
> > doing anything beyond that opens up new distinct ways of tracking and
> > such.
> >
> > TLS sessions alone are not long-lived enough (otherwise this whole effort
> > would not be needed). And pinning to tokens alone just plain does not
> > work (keys would immediately be lost).
>
> Agreed. IMO, leaving it to implemetations without any guidance is a
> mistake.
>
> At the very least implementors need a set of pros/cons and security
> considerations to base any such decisions on. Without that there are
> guaranteed interoperability and security flaws.
>
>
Agree.

At a high level, here's the typical anticipated Token Binding flow:


On a client:
keyPairForServerA = createKeyPairForServerIfNeeded(serverA);
keyPairForServerB = createKeyPairForServerIfNeeded(serverB);



First party scenario, with ServerA:

Client to ServerA-

proofOfKeyForServerA = signToProvePossession(privateKeyForServerA);
tokenForServerA = requestBoundToken(proofOfKeyForServerA, pubKeyForServerA);
presentBoundToken(proofOfKeyForServerA, tokenForServerA);



Federation scenario, with ServerA issuing token for use with ServerB:

Client to ServerA-
proofOfKeyForServerA = signToProvePossession(privateKeyForServerA);
proofOfKeyForServerA = signToProvePossession(privateKeyForServerB);
tokenForServerB = requestBoundToken(proofOfKeyOfServerA, pubKeyForServerA,
                                    proofOfKeyOfServerB, pubKeyForServerB);

Client to ServerB-
proofOfKeyForServerB = signToProvePossession(privateKeyForServerB);
presentBoundToken(proofOfKeyForServerB, tokenForServerB);




Main points:
- Tokens are bound to 'keys' (both in First Party and Federation scenarios)

- Key management is orthogonal to Token Binding, but governs "security"
since
  o- In general, any app that can prove possession of a specific key can
ask for tokens to be bound to that key, and can present tokens issued to
that key
  o- In general, any app that can prove possession of a pair of specific
keys can ask one of the servers to issue tokens bound to the other key and
present those tokens to the other server


Implications:
- An app that does not want to share keys with any other app should use a
key management solution that allows only that app to access and use the
keys.

- An app (from a family of apps, for some notion of family) that wants to
share keys with other apps in that family should use a key management
solution that allows that app family to access and use the keys (including
providing a way for an app to prove that it belongs to the family)

- An app that wants to use keys scoped to a machine and/or user and/or
profile should use a key management solution that allows any apps on that
machine and/or user and/or profile to access and use the keys.

The key management solution will define overall security (e.g. can malware
on client side access keys)
And there will be dependencies on the app-platform interface with regard to
hardware-backed keys.

I think we will see different key management solutions (though the first
and the third can be though of as special cases of the second)

Implementers need to determine what makes sense in their case.

We need to make sure that we are not blocking out important use cases.



>
> Amos
>
> _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org
> https://www.ietf.org/mailman/listinfo/unbearable
>

--001a11490ac853bcd505116d2afe
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div><br></div><div><br></div><div class=3D"gmail_extra"><=
br><div class=3D"gmail_quote">On Mon, Mar 16, 2015 at 12:25 PM, Amos Jeffri=
es <span dir=3D"ltr">&lt;<a href=3D"mailto:squid3@treenet.co.nz" target=3D"=
_blank">squid3@treenet.co.nz</a>&gt;</span> wrote:<br><blockquote class=3D"=
gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border=
-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div=
 class=3D""><div class=3D"h5">On 17/03/2015 12:05 a.m., Ilari Liusvaara wro=
te:<br>
&gt; On Sun, Mar 15, 2015 at 04:34:24PM -0300, John Bradley wrote:<br>
&gt;<br>
&gt; Also, it occurs to me that scope of keys can be even smaller than<br>
&gt; an application. What if the application is a browser that support<br>
&gt; profiles? Each profile has its own cookie jar and thus should have<br>
&gt; its own keys.<br>
&gt;<br>
&gt; It gets even more fine-grained if &quot;private browsing&quot; is cons=
idered<br>
&gt; as each private browsing context needs its own keys.<br>
&gt;<br>
&gt;&gt; Yes privacy is important and minimizing disclosure seems like a<br=
>
&gt;&gt; important principal.<br>
&gt;<br>
&gt; For some, &quot;important&quot; about privacy is a major understatemen=
t.<br>
&gt;<br>
&gt;&gt; The Server knowing something about the client may allow it to take=
<br>
&gt;&gt; other precautions or restrict the access of a token to match the<b=
r>
&gt;&gt; security of the environment.<br>
&gt;<br>
&gt; Well, I don&#39;t see great amounts of difference. By the time one can=
<br>
&gt; dump the keys from well-made SW-only solution, one is very likely<br>
&gt; in position to do great deal of damage without dumping the keys<br>
&gt; anyway.<br>
&gt;<br>
&gt;&gt; The counter argument would be that to trust what the client is say=
ing<br>
&gt;&gt; may lead to the question of attestation keys and other things that=
 are<br>
&gt;&gt; in the FIDO specs. That is probably more complicated than we want.=
<br>
&gt;&gt; Though if people think it is important get it on the table early.<=
br>
&gt;<br>
&gt; The reason why attestation is so complicated in FIDO is because<br>
&gt; of they need to do it in a way that doesn&#39;t break the web (e.g. by=
<br>
&gt; allowing cross-domain tracking).<br>
<br>
</div></div>Sigh. Please dont use that phase &quot;break the web&quot;. Its=
 been over inflated<br>
with meanings and political posturing.<br>
<br>
Your example is even one situation in which &quot;break&quot; implies a wro=
ng<br>
meaning. Cross-domain tracking breaks *Trust*, even while the web at a<br>
protocol/specification level continues to work as specified with regular<br=
>
users not aware of anything.<br>
<span class=3D""><br>
&gt;<br>
&gt;&gt; OAuth is working on two binding/presentment specs for access token=
s.<br>
&gt;&gt; One that is signed HTTP requests<br>
&gt;&gt; <a href=3D"https://tools.ietf.org/html/draft-ietf-oauth-signed-htt=
p-request-01" target=3D"_blank">https://tools.ietf.org/html/draft-ietf-oaut=
h-signed-http-request-01</a><br>
&gt;&gt; that is defiantly a layer above tokbind. The other is binding the =
token<br>
&gt;&gt; to the TLS session talked about in<br>
&gt;&gt; <a href=3D"http://tools.ietf.org/html/draft-tschofenig-oauth-hotk-=
03#section-3.2.2" target=3D"_blank">http://tools.ietf.org/html/draft-tschof=
enig-oauth-hotk-03#section-3.2.2</a><br>
&gt;<br>
&gt; I would say that if one has choice between binding to TLS session and<=
br>
&gt; HTTP-level request signing, one should go to first. It is MUCH simpler=
.<br>
<br>
</span>No, they are approximately the same in both complexity and difficult=
y.<br>
<br>
With TLS one has to account for multiple HTTP layer hops with different<br>
TLS connections, even duifferent types of TLS, and messages in each<br>
connection being multiplexed between different applications and/or clients.=
<br>
<br>
With HTTP signing one has to account for information leakage, what<br>
headers to include in the signature, which to omit, how to do the<br>
signing securely, and cached responses.<br>
<br>
Both gives token binding from point A to B, where either A or B could be<br=
>
endpoints or midpoints.<br>
<span class=3D""><br>
&gt;<br>
&gt;&gt; The POP security architecture doc may be a good place to look for =
people<br>
&gt;&gt; that want to find out what OAuth is thinking<br>
&gt;&gt; <a href=3D"https://tools.ietf.org/html/draft-ietf-oauth-pop-archit=
ecture-01" target=3D"_blank">https://tools.ietf.org/html/draft-ietf-oauth-p=
op-architecture-01</a><br>
&gt;<br>
&gt; Reading that document, I see:<br>
&gt;<br>
&gt; &quot;3. Use Cases&quot;<br>
&gt; &quot;3.4. Offering Application Layer End-to-End Security&quot;<br>
&gt;<br>
&gt; That is precisely &quot;TLS does not have enough reach&quot;.<br>
&gt;<br>
&gt; Whereas the two starting documents (implicitly representing two major<=
br>
&gt; browser vendors) implicitly consider TLS to have enough reach.<br>
<br>
</span>Warning to the wise; this is the same situation faced by HTTP WG<br>
developing HTTP/2 out of SPDY. The browser vendors driving the<br>
specification showed a bias towards the belief that TLS was end-to-end<br>
across the entire Interent. The existence of that e2e assumption in the<br>
very core concepts of the SPDY draft and carried over to HTTP/2 draft<br>
have lead to some very, very long debates and division within the WG.<br>
<br>
Having accurate information and WG consensus on what exactly is the<br>
reach of TLS (or any protocol being bound to) would help avoid a lot of<br>
that. If nothing else it would help to explicitly prohibit binding use<br>
in cases where required reach is not being guaranteed and/or supplied.<br>
<span class=3D""><br>
&gt;<br>
&gt;&gt; OAuth can define cannel binding of tokens for native apps with pla=
in TLS.<br>
&gt;&gt; I suppose that could also be done using tokbind, though depending =
on key<br>
&gt;&gt; management that may or may not be a good idea compared to regular =
mutual<br>
&gt;&gt; TLS.<br>
&gt;<br>
&gt; I consider mutual TLS much more secure than messing with bearer tokens=
<br>
&gt; and signature keys protecting those (of course, mutual TLS has usabili=
ty<br>
&gt; problems in practice).<br>
<br>
</span>Ditto. Also its worth noting that mutual TLS authentication may exis=
t<br>
between any two parties in the end-to-end messaging pathway. There are<br>
numerous permutation possibilities where the application layer is fully<br>
disconnected from the TLS auth.<br>
<br>
The following HTTP structure is quite popular:<br>
=C2=A0client -&gt; ISP proxy -&gt; CDN -&gt; origin<br>
<br>
where 5 of the 6 permutations and point-to-point, and only 2 have<br>
client-&gt;origin contact in any way.<br>
<span class=3D""><br>
&gt;<br>
&gt;&gt; Tying the TLS keys to sessions is perhaps the most privacy preserv=
ing<br>
&gt;&gt; way to do it but may loose many of the other benefits. Some people=
<br>
&gt;&gt; believe that question should be out of scope and left to implement=
ations.<br>
&gt;<br>
&gt; It was tying to both sessions and tokens. Basically, it seems to me th=
at<br>
&gt; doing anything beyond that opens up new distinct ways of tracking and<=
br>
&gt; such.<br>
&gt;<br>
&gt; TLS sessions alone are not long-lived enough (otherwise this whole eff=
ort<br>
&gt; would not be needed). And pinning to tokens alone just plain does not<=
br>
&gt; work (keys would immediately be lost).<br>
<br>
</span>Agreed. IMO, leaving it to implemetations without any guidance is a =
mistake.<br>
<br>
At the very least implementors need a set of pros/cons and security<br>
considerations to base any such decisions on. Without that there are<br>
guaranteed interoperability and security flaws.<br>
<span class=3D""><font color=3D"#888888"><br></font></span></blockquote><di=
v><br></div><div>Agree.</div><div><br></div><div>At a high level, here&#39;=
s the typical anticipated Token Binding flow:</div><div><br></div><div><br>=
</div><div>On a client:</div><div>keyPairForServerA =3D createKeyPairForSer=
verIfNeeded(serverA);</div><div>keyPairForServerB =3D createKeyPairForServe=
rIfNeeded(serverB);</div><div><br></div><div><br></div><div><br></div><div>=
First party scenario, with ServerA:</div><div><br></div><div>Client to Serv=
erA-</div><div><br></div><div>proofOfKeyForServerA =3D signToProvePossessio=
n(privateKeyForServerA);</div><div>tokenForServerA =3D requestBoundToken(pr=
oofOfKeyForServerA, pubKeyForServerA);</div><div>presentBoundToken(proofOfK=
eyForServerA, tokenForServerA);</div><div><br></div><div><br></div><div><br=
></div><div>Federation scenario, with ServerA issuing token for use with Se=
rverB:</div><div><br></div><div>Client to ServerA-</div><div>proofOfKeyForS=
erverA =3D signToProvePossession(privateKeyForServerA);</div><div>proofOfKe=
yForServerA =3D signToProvePossession(privateKeyForServerB);</div><div>toke=
nForServerB =3D requestBoundToken(proofOfKeyOfServerA, pubKeyForServerA,</d=
iv><div>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 proofOfKeyOfSer=
verB, pubKeyForServerB);</div><div><br></div><div>Client to ServerB-</div><=
div>proofOfKeyForServerB =3D signToProvePossession(privateKeyForServerB);</=
div><div>presentBoundToken(proofOfKeyForServerB, tokenForServerB);</div><di=
v><br></div><div><br></div><div><br></div><div><br></div><div>Main points:<=
/div><div>- Tokens are bound to &#39;keys&#39; (both in First Party and Fed=
eration scenarios)</div><div><br></div><div>- Key management is orthogonal =
to Token Binding, but governs &quot;security&quot; since</div><div>=C2=A0 o=
- In general, any app that can prove possession of a specific key can ask f=
or tokens to be bound to that key, and can present tokens issued to that ke=
y</div><div>=C2=A0 o- In general, any app that can prove possession of a pa=
ir of specific keys can ask one of the servers to issue tokens bound to the=
 other key and present those tokens to the other server</div><div><br></div=
><div><br></div><div>Implications:</div><div>- An app that does not want to=
 share keys with any other app should use a key management solution that al=
lows only that app to access and use the keys.</div><div><br></div><div>- A=
n app (from a family of apps, for some notion of family) that wants to shar=
e keys with other apps in that family should use a key management solution =
that allows that app family to access and use the keys (including providing=
 a way for an app to prove that it belongs to the family)</div><div><br></d=
iv><div>- An app that wants to use keys scoped to a machine and/or user and=
/or profile should use a key management solution that allows any apps on th=
at machine and/or user and/or profile to access and use the keys.</div><div=
><br></div><div>The key management solution will define overall security (e=
.g. can malware on client side access keys)</div><div>And there will be dep=
endencies on the app-platform interface with regard to hardware-backed keys=
.</div><div><br></div><div>I think we will see different key management sol=
utions (though the first and the third can be though of as special cases of=
 the second)</div><div><br></div><div>Implementers need to determine what m=
akes sense in their case.</div><div><br></div><div>We need to make sure tha=
t we are not blocking out important use cases.</div><div><br></div><div>=C2=
=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8e=
x;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-styl=
e:solid;padding-left:1ex"><span class=3D""><font color=3D"#888888">
<br>
Amos<br>
</font></span><div class=3D""><div class=3D"h5"><br>
_______________________________________________<br>
Unbearable mailing list<br>
<a href=3D"mailto:Unbearable@ietf.org">Unbearable@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/unbearable" target=3D"_bla=
nk">https://www.ietf.org/mailman/listinfo/unbearable</a><br>
</div></div></blockquote></div><br></div></div>

--001a11490ac853bcd505116d2afe--


From nobody Mon Mar 16 13:12:42 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9490F1A90AF for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 13:12:40 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.6
X-Spam-Level: 
X-Spam-Status: No, score=-2.6 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 oLOiMqbcLcI3 for <unbearable@ietfa.amsl.com>; Mon, 16 Mar 2015 13:12:36 -0700 (PDT)
Received: from mail-qg0-f44.google.com (mail-qg0-f44.google.com [209.85.192.44]) (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 4B9961A1B77 for <unbearable@ietf.org>; Mon, 16 Mar 2015 13:12:27 -0700 (PDT)
Received: by qgez64 with SMTP id z64so50965471qge.2 for <unbearable@ietf.org>; Mon, 16 Mar 2015 13:12:26 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:message-id:references:to; bh=8pOsjzZOMPVPJ4ustkTvVNFJrw1ARZXbf84u0RoOjHE=; b=CmhKvfxx5rF4ieEB2T79eaqtGc3uDlPqF2Tw4kPj+FfITtENtbTDrhLCYIY4/LSJlD mcBt0T1BLN+oCe73y+xMG8VEo/D90rrmpi3kPnrCpRQKVQuKm8VJlgbiv84lpNlIEGlX h2ffKEpVMLm1ZLkjPReY+xDC4Hguevo7Q6Yd5I7gfbe1Pt5Qnj5ZZljMV6n8QGIYFsK6 qTWfDdbvLCl/tcZgOCLCmRIyx+ge91UJjLSbbs3IdF8tx8Hg0xr8G9aWjP9HYXoyHI1Z BfYbxwKbXCyxcBHp6qy7ACIUNsgZd6p6KS+caSwvmMol0hG/ouafuDXOivngSWcJW2ng jzrw==
X-Gm-Message-State: ALoCoQmgUfIa46X4WDAk7MpqvCE66MQaTW0DvtW/uLWnREAqF2L9a6HBivuyWvuu0urqgduU2o1P
X-Received: by 10.55.31.10 with SMTP id f10mr90861445qkf.58.1426536746462; Mon, 16 Mar 2015 13:12:26 -0700 (PDT)
Received: from [192.168.8.100] ([181.202.31.7]) by mx.google.com with ESMTPSA id u194sm8111077qhd.36.2015.03.16.13.12.23 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Mon, 16 Mar 2015 13:12:25 -0700 (PDT)
Content-Type: multipart/signed; boundary="Apple-Mail=_E2C22B70-E435-463B-B7BE-242B30197069"; protocol="application/pkcs7-signature"; micalg=sha1
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: John Bradley <ve7jtb@ve7jtb.com>
In-Reply-To: <CAOVPt=sj_UPHhU80OgC04GOHX4yyjnW8c30Wpuc-in2Rtdk_4A@mail.gmail.com>
Date: Mon, 16 Mar 2015 17:12:18 -0300
Message-Id: <48EB9F86-4E5B-48B4-A9F2-887FCB6EE538@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com> <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com> <20150315181130.GA16994@LK-Perkele-VII> <2EFCA3AA-D611-4DAE-AB28-799327B08292@ve7jtb.com> <20150316110548.GA2214@LK-Perkele-VII> <55072E27.5080001@treenet.co.nz> <CAOVPt=sj_UPHhU80OgC04GOHX4yyjnW8c30Wpuc-in2Rtdk_4A@mail.gmail.com>
To: Vinod Anupam <vanupam@google.com>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/nol0e_1W1oC4Ss2x59tRG7p-ZU8>
Cc: Amos Jeffries <squid3@treenet.co.nz>, "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 16 Mar 2015 20:12:40 -0000

--Apple-Mail=_E2C22B70-E435-463B-B7BE-242B30197069
Content-Type: multipart/alternative;
	boundary="Apple-Mail=_B4A3104D-907B-4E40-9692-5ECFA3D0303D"


--Apple-Mail=_B4A3104D-907B-4E40-9692-5ECFA3D0303D
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=us-ascii

Thanks

That was more or less what I was looking for with my question.

John B.

> On Mar 16, 2015, at 4:50 PM, Vinod Anupam <vanupam@google.com> wrote:
>=20
>=20
>=20
>=20
> On Mon, Mar 16, 2015 at 12:25 PM, Amos Jeffries <squid3@treenet.co.nz =
<mailto:squid3@treenet.co.nz>> wrote:
> On 17/03/2015 12:05 a.m., Ilari Liusvaara wrote:
> > On Sun, Mar 15, 2015 at 04:34:24PM -0300, John Bradley wrote:
> >
> > Also, it occurs to me that scope of keys can be even smaller than
> > an application. What if the application is a browser that support
> > profiles? Each profile has its own cookie jar and thus should have
> > its own keys.
> >
> > It gets even more fine-grained if "private browsing" is considered
> > as each private browsing context needs its own keys.
> >
> >> Yes privacy is important and minimizing disclosure seems like a
> >> important principal.
> >
> > For some, "important" about privacy is a major understatement.
> >
> >> The Server knowing something about the client may allow it to take
> >> other precautions or restrict the access of a token to match the
> >> security of the environment.
> >
> > Well, I don't see great amounts of difference. By the time one can
> > dump the keys from well-made SW-only solution, one is very likely
> > in position to do great deal of damage without dumping the keys
> > anyway.
> >
> >> The counter argument would be that to trust what the client is =
saying
> >> may lead to the question of attestation keys and other things that =
are
> >> in the FIDO specs. That is probably more complicated than we want.
> >> Though if people think it is important get it on the table early.
> >
> > The reason why attestation is so complicated in FIDO is because
> > of they need to do it in a way that doesn't break the web (e.g. by
> > allowing cross-domain tracking).
>=20
> Sigh. Please dont use that phase "break the web". Its been over =
inflated
> with meanings and political posturing.
>=20
> Your example is even one situation in which "break" implies a wrong
> meaning. Cross-domain tracking breaks *Trust*, even while the web at a
> protocol/specification level continues to work as specified with =
regular
> users not aware of anything.
>=20
> >
> >> OAuth is working on two binding/presentment specs for access =
tokens.
> >> One that is signed HTTP requests
> >> https://tools.ietf.org/html/draft-ietf-oauth-signed-http-request-01 =
<https://tools.ietf.org/html/draft-ietf-oauth-signed-http-request-01>
> >> that is defiantly a layer above tokbind. The other is binding the =
token
> >> to the TLS session talked about in
> >> =
http://tools.ietf.org/html/draft-tschofenig-oauth-hotk-03#section-3.2.2 =
<http://tools.ietf.org/html/draft-tschofenig-oauth-hotk-03#section-3.2.2>
> >
> > I would say that if one has choice between binding to TLS session =
and
> > HTTP-level request signing, one should go to first. It is MUCH =
simpler.
>=20
> No, they are approximately the same in both complexity and difficulty.
>=20
> With TLS one has to account for multiple HTTP layer hops with =
different
> TLS connections, even duifferent types of TLS, and messages in each
> connection being multiplexed between different applications and/or =
clients.
>=20
> With HTTP signing one has to account for information leakage, what
> headers to include in the signature, which to omit, how to do the
> signing securely, and cached responses.
>=20
> Both gives token binding from point A to B, where either A or B could =
be
> endpoints or midpoints.
>=20
> >
> >> The POP security architecture doc may be a good place to look for =
people
> >> that want to find out what OAuth is thinking
> >> https://tools.ietf.org/html/draft-ietf-oauth-pop-architecture-01 =
<https://tools.ietf.org/html/draft-ietf-oauth-pop-architecture-01>
> >
> > Reading that document, I see:
> >
> > "3. Use Cases"
> > "3.4. Offering Application Layer End-to-End Security"
> >
> > That is precisely "TLS does not have enough reach".
> >
> > Whereas the two starting documents (implicitly representing two =
major
> > browser vendors) implicitly consider TLS to have enough reach.
>=20
> Warning to the wise; this is the same situation faced by HTTP WG
> developing HTTP/2 out of SPDY. The browser vendors driving the
> specification showed a bias towards the belief that TLS was end-to-end
> across the entire Interent. The existence of that e2e assumption in =
the
> very core concepts of the SPDY draft and carried over to HTTP/2 draft
> have lead to some very, very long debates and division within the WG.
>=20
> Having accurate information and WG consensus on what exactly is the
> reach of TLS (or any protocol being bound to) would help avoid a lot =
of
> that. If nothing else it would help to explicitly prohibit binding use
> in cases where required reach is not being guaranteed and/or supplied.
>=20
> >
> >> OAuth can define cannel binding of tokens for native apps with =
plain TLS.
> >> I suppose that could also be done using tokbind, though depending =
on key
> >> management that may or may not be a good idea compared to regular =
mutual
> >> TLS.
> >
> > I consider mutual TLS much more secure than messing with bearer =
tokens
> > and signature keys protecting those (of course, mutual TLS has =
usability
> > problems in practice).
>=20
> Ditto. Also its worth noting that mutual TLS authentication may exist
> between any two parties in the end-to-end messaging pathway. There are
> numerous permutation possibilities where the application layer is =
fully
> disconnected from the TLS auth.
>=20
> The following HTTP structure is quite popular:
>  client -> ISP proxy -> CDN -> origin
>=20
> where 5 of the 6 permutations and point-to-point, and only 2 have
> client->origin contact in any way.
>=20
> >
> >> Tying the TLS keys to sessions is perhaps the most privacy =
preserving
> >> way to do it but may loose many of the other benefits. Some people
> >> believe that question should be out of scope and left to =
implementations.
> >
> > It was tying to both sessions and tokens. Basically, it seems to me =
that
> > doing anything beyond that opens up new distinct ways of tracking =
and
> > such.
> >
> > TLS sessions alone are not long-lived enough (otherwise this whole =
effort
> > would not be needed). And pinning to tokens alone just plain does =
not
> > work (keys would immediately be lost).
>=20
> Agreed. IMO, leaving it to implemetations without any guidance is a =
mistake.
>=20
> At the very least implementors need a set of pros/cons and security
> considerations to base any such decisions on. Without that there are
> guaranteed interoperability and security flaws.
>=20
>=20
> Agree.
>=20
> At a high level, here's the typical anticipated Token Binding flow:
>=20
>=20
> On a client:
> keyPairForServerA =3D createKeyPairForServerIfNeeded(serverA);
> keyPairForServerB =3D createKeyPairForServerIfNeeded(serverB);
>=20
>=20
>=20
> First party scenario, with ServerA:
>=20
> Client to ServerA-
>=20
> proofOfKeyForServerA =3D signToProvePossession(privateKeyForServerA);
> tokenForServerA =3D requestBoundToken(proofOfKeyForServerA, =
pubKeyForServerA);
> presentBoundToken(proofOfKeyForServerA, tokenForServerA);
>=20
>=20
>=20
> Federation scenario, with ServerA issuing token for use with ServerB:
>=20
> Client to ServerA-
> proofOfKeyForServerA =3D signToProvePossession(privateKeyForServerA);
> proofOfKeyForServerA =3D signToProvePossession(privateKeyForServerB);
> tokenForServerB =3D requestBoundToken(proofOfKeyOfServerA, =
pubKeyForServerA,
>                                     proofOfKeyOfServerB, =
pubKeyForServerB);
>=20
> Client to ServerB-
> proofOfKeyForServerB =3D signToProvePossession(privateKeyForServerB);
> presentBoundToken(proofOfKeyForServerB, tokenForServerB);
>=20
>=20
>=20
>=20
> Main points:
> - Tokens are bound to 'keys' (both in First Party and Federation =
scenarios)
>=20
> - Key management is orthogonal to Token Binding, but governs =
"security" since
>   o- In general, any app that can prove possession of a specific key =
can ask for tokens to be bound to that key, and can present tokens =
issued to that key
>   o- In general, any app that can prove possession of a pair of =
specific keys can ask one of the servers to issue tokens bound to the =
other key and present those tokens to the other server
>=20
>=20
> Implications:
> - An app that does not want to share keys with any other app should =
use a key management solution that allows only that app to access and =
use the keys.
>=20
> - An app (from a family of apps, for some notion of family) that wants =
to share keys with other apps in that family should use a key management =
solution that allows that app family to access and use the keys =
(including providing a way for an app to prove that it belongs to the =
family)
>=20
> - An app that wants to use keys scoped to a machine and/or user and/or =
profile should use a key management solution that allows any apps on =
that machine and/or user and/or profile to access and use the keys.
>=20
> The key management solution will define overall security (e.g. can =
malware on client side access keys)
> And there will be dependencies on the app-platform interface with =
regard to hardware-backed keys.
>=20
> I think we will see different key management solutions (though the =
first and the third can be though of as special cases of the second)
>=20
> Implementers need to determine what makes sense in their case.
>=20
> We need to make sure that we are not blocking out important use cases.
>=20
> =20
>=20
> Amos
>=20
> _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org <mailto:Unbearable@ietf.org>
> https://www.ietf.org/mailman/listinfo/unbearable =
<https://www.ietf.org/mailman/listinfo/unbearable>
>=20
> _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org <mailto:Unbearable@ietf.org>
> https://www.ietf.org/mailman/listinfo/unbearable =
<https://www.ietf.org/mailman/listinfo/unbearable>

--Apple-Mail=_B4A3104D-907B-4E40-9692-5ECFA3D0303D
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
	charset=us-ascii

<html><head><meta http-equiv=3D"Content-Type" content=3D"text/html =
charset=3Dus-ascii"></head><body style=3D"word-wrap: break-word; =
-webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" =
class=3D"">Thanks<div class=3D""><br class=3D""></div><div class=3D"">That=
 was more or less what I was looking for with my question.</div><div =
class=3D""><br class=3D""></div><div class=3D"">John B.</div><div =
class=3D""><br class=3D""><div><blockquote type=3D"cite" class=3D""><div =
class=3D"">On Mar 16, 2015, at 4:50 PM, Vinod Anupam &lt;<a =
href=3D"mailto:vanupam@google.com" class=3D"">vanupam@google.com</a>&gt; =
wrote:</div><br class=3D"Apple-interchange-newline"><div class=3D""><div =
dir=3D"ltr" style=3D"font-family: Helvetica; font-size: 12px; =
font-style: normal; font-variant: normal; font-weight: normal; =
letter-spacing: normal; line-height: normal; orphans: auto; text-align: =
start; text-indent: 0px; text-transform: none; white-space: normal; =
widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;" =
class=3D""><div class=3D""><br class=3D"Apple-interchange-newline"><br =
class=3D""></div><div class=3D"gmail_extra"><br class=3D""><div =
class=3D"gmail_quote">On Mon, Mar 16, 2015 at 12:25 PM, Amos =
Jeffries<span class=3D"Apple-converted-space">&nbsp;</span><span =
dir=3D"ltr" class=3D"">&lt;<a href=3D"mailto:squid3@treenet.co.nz" =
target=3D"_blank" class=3D"">squid3@treenet.co.nz</a>&gt;</span><span =
class=3D"Apple-converted-space">&nbsp;</span>wrote:<br =
class=3D""><blockquote class=3D"gmail_quote" style=3D"margin: 0px 0px =
0px 0.8ex; border-left-width: 1px; border-left-color: rgb(204, 204, =
204); border-left-style: solid; padding-left: 1ex;"><div class=3D""><div =
class=3D"h5">On 17/03/2015 12:05 a.m., Ilari Liusvaara wrote:<br =
class=3D"">&gt; On Sun, Mar 15, 2015 at 04:34:24PM -0300, John Bradley =
wrote:<br class=3D"">&gt;<br class=3D"">&gt; Also, it occurs to me that =
scope of keys can be even smaller than<br class=3D"">&gt; an =
application. What if the application is a browser that support<br =
class=3D"">&gt; profiles? Each profile has its own cookie jar and thus =
should have<br class=3D"">&gt; its own keys.<br class=3D"">&gt;<br =
class=3D"">&gt; It gets even more fine-grained if "private browsing" is =
considered<br class=3D"">&gt; as each private browsing context needs its =
own keys.<br class=3D"">&gt;<br class=3D"">&gt;&gt; Yes privacy is =
important and minimizing disclosure seems like a<br class=3D"">&gt;&gt; =
important principal.<br class=3D"">&gt;<br class=3D"">&gt; For some, =
"important" about privacy is a major understatement.<br class=3D"">&gt;<br=
 class=3D"">&gt;&gt; The Server knowing something about the client may =
allow it to take<br class=3D"">&gt;&gt; other precautions or restrict =
the access of a token to match the<br class=3D"">&gt;&gt; security of =
the environment.<br class=3D"">&gt;<br class=3D"">&gt; Well, I don't see =
great amounts of difference. By the time one can<br class=3D"">&gt; dump =
the keys from well-made SW-only solution, one is very likely<br =
class=3D"">&gt; in position to do great deal of damage without dumping =
the keys<br class=3D"">&gt; anyway.<br class=3D"">&gt;<br =
class=3D"">&gt;&gt; The counter argument would be that to trust what the =
client is saying<br class=3D"">&gt;&gt; may lead to the question of =
attestation keys and other things that are<br class=3D"">&gt;&gt; in the =
FIDO specs. That is probably more complicated than we want.<br =
class=3D"">&gt;&gt; Though if people think it is important get it on the =
table early.<br class=3D"">&gt;<br class=3D"">&gt; The reason why =
attestation is so complicated in FIDO is because<br class=3D"">&gt; of =
they need to do it in a way that doesn't break the web (e.g. by<br =
class=3D"">&gt; allowing cross-domain tracking).<br class=3D""><br =
class=3D""></div></div>Sigh. Please dont use that phase "break the web". =
Its been over inflated<br class=3D"">with meanings and political =
posturing.<br class=3D""><br class=3D"">Your example is even one =
situation in which "break" implies a wrong<br class=3D"">meaning. =
Cross-domain tracking breaks *Trust*, even while the web at a<br =
class=3D"">protocol/specification level continues to work as specified =
with regular<br class=3D"">users not aware of anything.<br =
class=3D""><span class=3D""><br class=3D"">&gt;<br class=3D"">&gt;&gt; =
OAuth is working on two binding/presentment specs for access tokens.<br =
class=3D"">&gt;&gt; One that is signed HTTP requests<br =
class=3D"">&gt;&gt;<span class=3D"Apple-converted-space">&nbsp;</span><a =
href=3D"https://tools.ietf.org/html/draft-ietf-oauth-signed-http-request-0=
1" target=3D"_blank" =
class=3D"">https://tools.ietf.org/html/draft-ietf-oauth-signed-http-reques=
t-01</a><br class=3D"">&gt;&gt; that is defiantly a layer above tokbind. =
The other is binding the token<br class=3D"">&gt;&gt; to the TLS session =
talked about in<br class=3D"">&gt;&gt;<span =
class=3D"Apple-converted-space">&nbsp;</span><a =
href=3D"http://tools.ietf.org/html/draft-tschofenig-oauth-hotk-03#section-=
3.2.2" target=3D"_blank" =
class=3D"">http://tools.ietf.org/html/draft-tschofenig-oauth-hotk-03#secti=
on-3.2.2</a><br class=3D"">&gt;<br class=3D"">&gt; I would say that if =
one has choice between binding to TLS session and<br class=3D"">&gt; =
HTTP-level request signing, one should go to first. It is MUCH =
simpler.<br class=3D""><br class=3D""></span>No, they are approximately =
the same in both complexity and difficulty.<br class=3D""><br =
class=3D"">With TLS one has to account for multiple HTTP layer hops with =
different<br class=3D"">TLS connections, even duifferent types of TLS, =
and messages in each<br class=3D"">connection being multiplexed between =
different applications and/or clients.<br class=3D""><br class=3D"">With =
HTTP signing one has to account for information leakage, what<br =
class=3D"">headers to include in the signature, which to omit, how to do =
the<br class=3D"">signing securely, and cached responses.<br =
class=3D""><br class=3D"">Both gives token binding from point A to B, =
where either A or B could be<br class=3D"">endpoints or midpoints.<br =
class=3D""><span class=3D""><br class=3D"">&gt;<br class=3D"">&gt;&gt; =
The POP security architecture doc may be a good place to look for =
people<br class=3D"">&gt;&gt; that want to find out what OAuth is =
thinking<br class=3D"">&gt;&gt;<span =
class=3D"Apple-converted-space">&nbsp;</span><a =
href=3D"https://tools.ietf.org/html/draft-ietf-oauth-pop-architecture-01" =
target=3D"_blank" =
class=3D"">https://tools.ietf.org/html/draft-ietf-oauth-pop-architecture-0=
1</a><br class=3D"">&gt;<br class=3D"">&gt; Reading that document, I =
see:<br class=3D"">&gt;<br class=3D"">&gt; "3. Use Cases"<br =
class=3D"">&gt; "3.4. Offering Application Layer End-to-End Security"<br =
class=3D"">&gt;<br class=3D"">&gt; That is precisely "TLS does not have =
enough reach".<br class=3D"">&gt;<br class=3D"">&gt; Whereas the two =
starting documents (implicitly representing two major<br class=3D"">&gt; =
browser vendors) implicitly consider TLS to have enough reach.<br =
class=3D""><br class=3D""></span>Warning to the wise; this is the same =
situation faced by HTTP WG<br class=3D"">developing HTTP/2 out of SPDY. =
The browser vendors driving the<br class=3D"">specification showed a =
bias towards the belief that TLS was end-to-end<br class=3D"">across the =
entire Interent. The existence of that e2e assumption in the<br =
class=3D"">very core concepts of the SPDY draft and carried over to =
HTTP/2 draft<br class=3D"">have lead to some very, very long debates and =
division within the WG.<br class=3D""><br class=3D"">Having accurate =
information and WG consensus on what exactly is the<br class=3D"">reach =
of TLS (or any protocol being bound to) would help avoid a lot of<br =
class=3D"">that. If nothing else it would help to explicitly prohibit =
binding use<br class=3D"">in cases where required reach is not being =
guaranteed and/or supplied.<br class=3D""><span class=3D""><br =
class=3D"">&gt;<br class=3D"">&gt;&gt; OAuth can define cannel binding =
of tokens for native apps with plain TLS.<br class=3D"">&gt;&gt; I =
suppose that could also be done using tokbind, though depending on =
key<br class=3D"">&gt;&gt; management that may or may not be a good idea =
compared to regular mutual<br class=3D"">&gt;&gt; TLS.<br =
class=3D"">&gt;<br class=3D"">&gt; I consider mutual TLS much more =
secure than messing with bearer tokens<br class=3D"">&gt; and signature =
keys protecting those (of course, mutual TLS has usability<br =
class=3D"">&gt; problems in practice).<br class=3D""><br =
class=3D""></span>Ditto. Also its worth noting that mutual TLS =
authentication may exist<br class=3D"">between any two parties in the =
end-to-end messaging pathway. There are<br class=3D"">numerous =
permutation possibilities where the application layer is fully<br =
class=3D"">disconnected from the TLS auth.<br class=3D""><br =
class=3D"">The following HTTP structure is quite popular:<br =
class=3D"">&nbsp;client -&gt; ISP proxy -&gt; CDN -&gt; origin<br =
class=3D""><br class=3D"">where 5 of the 6 permutations and =
point-to-point, and only 2 have<br class=3D"">client-&gt;origin contact =
in any way.<br class=3D""><span class=3D""><br class=3D"">&gt;<br =
class=3D"">&gt;&gt; Tying the TLS keys to sessions is perhaps the most =
privacy preserving<br class=3D"">&gt;&gt; way to do it but may loose =
many of the other benefits. Some people<br class=3D"">&gt;&gt; believe =
that question should be out of scope and left to implementations.<br =
class=3D"">&gt;<br class=3D"">&gt; It was tying to both sessions and =
tokens. Basically, it seems to me that<br class=3D"">&gt; doing anything =
beyond that opens up new distinct ways of tracking and<br class=3D"">&gt; =
such.<br class=3D"">&gt;<br class=3D"">&gt; TLS sessions alone are not =
long-lived enough (otherwise this whole effort<br class=3D"">&gt; would =
not be needed). And pinning to tokens alone just plain does not<br =
class=3D"">&gt; work (keys would immediately be lost).<br class=3D""><br =
class=3D""></span>Agreed. IMO, leaving it to implemetations without any =
guidance is a mistake.<br class=3D""><br class=3D"">At the very least =
implementors need a set of pros/cons and security<br =
class=3D"">considerations to base any such decisions on. Without that =
there are<br class=3D"">guaranteed interoperability and security =
flaws.<br class=3D""><span class=3D""><font color=3D"#888888" =
class=3D""><br class=3D""></font></span></blockquote><div class=3D""><br =
class=3D""></div><div class=3D"">Agree.</div><div class=3D""><br =
class=3D""></div><div class=3D"">At a high level, here's the typical =
anticipated Token Binding flow:</div><div class=3D""><br =
class=3D""></div><div class=3D""><br class=3D""></div><div class=3D"">On =
a client:</div><div class=3D"">keyPairForServerA =3D =
createKeyPairForServerIfNeeded(serverA);</div><div =
class=3D"">keyPairForServerB =3D =
createKeyPairForServerIfNeeded(serverB);</div><div class=3D""><br =
class=3D""></div><div class=3D""><br class=3D""></div><div class=3D""><br =
class=3D""></div><div class=3D"">First party scenario, with =
ServerA:</div><div class=3D""><br class=3D""></div><div class=3D"">Client =
to ServerA-</div><div class=3D""><br class=3D""></div><div =
class=3D"">proofOfKeyForServerA =3D =
signToProvePossession(privateKeyForServerA);</div><div =
class=3D"">tokenForServerA =3D requestBoundToken(proofOfKeyForServerA, =
pubKeyForServerA);</div><div =
class=3D"">presentBoundToken(proofOfKeyForServerA, =
tokenForServerA);</div><div class=3D""><br class=3D""></div><div =
class=3D""><br class=3D""></div><div class=3D""><br class=3D""></div><div =
class=3D"">Federation scenario, with ServerA issuing token for use with =
ServerB:</div><div class=3D""><br class=3D""></div><div class=3D"">Client =
to ServerA-</div><div class=3D"">proofOfKeyForServerA =3D =
signToProvePossession(privateKeyForServerA);</div><div =
class=3D"">proofOfKeyForServerA =3D =
signToProvePossession(privateKeyForServerB);</div><div =
class=3D"">tokenForServerB =3D requestBoundToken(proofOfKeyOfServerA, =
pubKeyForServerA,</div><div class=3D"">&nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; =
&nbsp; &nbsp; &nbsp; &nbsp; proofOfKeyOfServerB, =
pubKeyForServerB);</div><div class=3D""><br class=3D""></div><div =
class=3D"">Client to ServerB-</div><div class=3D"">proofOfKeyForServerB =
=3D signToProvePossession(privateKeyForServerB);</div><div =
class=3D"">presentBoundToken(proofOfKeyForServerB, =
tokenForServerB);</div><div class=3D""><br class=3D""></div><div =
class=3D""><br class=3D""></div><div class=3D""><br class=3D""></div><div =
class=3D""><br class=3D""></div><div class=3D"">Main points:</div><div =
class=3D"">- Tokens are bound to 'keys' (both in First Party and =
Federation scenarios)</div><div class=3D""><br class=3D""></div><div =
class=3D"">- Key management is orthogonal to Token Binding, but governs =
"security" since</div><div class=3D"">&nbsp; o- In general, any app that =
can prove possession of a specific key can ask for tokens to be bound to =
that key, and can present tokens issued to that key</div><div =
class=3D"">&nbsp; o- In general, any app that can prove possession of a =
pair of specific keys can ask one of the servers to issue tokens bound =
to the other key and present those tokens to the other server</div><div =
class=3D""><br class=3D""></div><div class=3D""><br class=3D""></div><div =
class=3D"">Implications:</div><div class=3D"">- An app that does not =
want to share keys with any other app should use a key management =
solution that allows only that app to access and use the keys.</div><div =
class=3D""><br class=3D""></div><div class=3D"">- An app (from a family =
of apps, for some notion of family) that wants to share keys with other =
apps in that family should use a key management solution that allows =
that app family to access and use the keys (including providing a way =
for an app to prove that it belongs to the family)</div><div =
class=3D""><br class=3D""></div><div class=3D"">- An app that wants to =
use keys scoped to a machine and/or user and/or profile should use a key =
management solution that allows any apps on that machine and/or user =
and/or profile to access and use the keys.</div><div class=3D""><br =
class=3D""></div><div class=3D"">The key management solution will define =
overall security (e.g. can malware on client side access keys)</div><div =
class=3D"">And there will be dependencies on the app-platform interface =
with regard to hardware-backed keys.</div><div class=3D""><br =
class=3D""></div><div class=3D"">I think we will see different key =
management solutions (though the first and the third can be though of as =
special cases of the second)</div><div class=3D""><br =
class=3D""></div><div class=3D"">Implementers need to determine what =
makes sense in their case.</div><div class=3D""><br class=3D""></div><div =
class=3D"">We need to make sure that we are not blocking out important =
use cases.</div><div class=3D""><br class=3D""></div><div =
class=3D"">&nbsp;</div><blockquote class=3D"gmail_quote" style=3D"margin: =
0px 0px 0px 0.8ex; border-left-width: 1px; border-left-color: rgb(204, =
204, 204); border-left-style: solid; padding-left: 1ex;"><span =
class=3D""><font color=3D"#888888" class=3D""><br class=3D"">Amos<br =
class=3D""></font></span><div class=3D""><div class=3D"h5"><br =
class=3D"">_______________________________________________<br =
class=3D"">Unbearable mailing list<br class=3D""><a =
href=3D"mailto:Unbearable@ietf.org" class=3D"">Unbearable@ietf.org</a><br =
class=3D""><a href=3D"https://www.ietf.org/mailman/listinfo/unbearable" =
target=3D"_blank" =
class=3D"">https://www.ietf.org/mailman/listinfo/unbearable</a><br =
class=3D""></div></div></blockquote></div><br class=3D""></div></div><span=
 style=3D"font-family: Helvetica; font-size: 12px; font-style: normal; =
font-variant: normal; font-weight: normal; letter-spacing: normal; =
line-height: normal; orphans: auto; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; widows: auto; word-spacing: =
0px; -webkit-text-stroke-width: 0px; float: none; display: inline =
!important;" =
class=3D"">_______________________________________________</span><br =
style=3D"font-family: Helvetica; font-size: 12px; font-style: normal; =
font-variant: normal; font-weight: normal; letter-spacing: normal; =
line-height: normal; orphans: auto; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; widows: auto; word-spacing: =
0px; -webkit-text-stroke-width: 0px;" class=3D""><span =
style=3D"font-family: Helvetica; font-size: 12px; font-style: normal; =
font-variant: normal; font-weight: normal; letter-spacing: normal; =
line-height: normal; orphans: auto; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; widows: auto; word-spacing: =
0px; -webkit-text-stroke-width: 0px; float: none; display: inline =
!important;" class=3D"">Unbearable mailing list</span><br =
style=3D"font-family: Helvetica; font-size: 12px; font-style: normal; =
font-variant: normal; font-weight: normal; letter-spacing: normal; =
line-height: normal; orphans: auto; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; widows: auto; word-spacing: =
0px; -webkit-text-stroke-width: 0px;" class=3D""><a =
href=3D"mailto:Unbearable@ietf.org" style=3D"font-family: Helvetica; =
font-size: 12px; font-style: normal; font-variant: normal; font-weight: =
normal; letter-spacing: normal; line-height: normal; orphans: auto; =
text-align: start; text-indent: 0px; text-transform: none; white-space: =
normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: =
0px;" class=3D"">Unbearable@ietf.org</a><br style=3D"font-family: =
Helvetica; font-size: 12px; font-style: normal; font-variant: normal; =
font-weight: normal; letter-spacing: normal; line-height: normal; =
orphans: auto; text-align: start; text-indent: 0px; text-transform: =
none; white-space: normal; widows: auto; word-spacing: 0px; =
-webkit-text-stroke-width: 0px;" class=3D""><a =
href=3D"https://www.ietf.org/mailman/listinfo/unbearable" =
style=3D"font-family: Helvetica; font-size: 12px; font-style: normal; =
font-variant: normal; font-weight: normal; letter-spacing: normal; =
line-height: normal; orphans: auto; text-align: start; text-indent: 0px; =
text-transform: none; white-space: normal; widows: auto; word-spacing: =
0px; -webkit-text-stroke-width: 0px;" =
class=3D"">https://www.ietf.org/mailman/listinfo/unbearable</a></div></blo=
ckquote></div><br class=3D""></div></body></html>=

--Apple-Mail=_B4A3104D-907B-4E40-9692-5ECFA3D0303D--

--Apple-Mail=_E2C22B70-E435-463B-B7BE-242B30197069
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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--Apple-Mail=_E2C22B70-E435-463B-B7BE-242B30197069--


From nobody Tue Mar 17 13:49:06 2015
Return-Path: <rdom54@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F2D601A88E4 for <unbearable@ietfa.amsl.com>; Tue, 17 Mar 2015 13:49:03 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 1.564
X-Spam-Level: *
X-Spam-Status: No, score=1.564 tagged_above=-999 required=5 tests=[BAYES_05=-0.5, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, FREEMAIL_ENVFROM_END_DIGIT=0.25, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, MALFORMED_FREEMAIL=1.813, SPF_PASS=-0.001] autolearn=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 vXIzEPPMh_Eb for <unbearable@ietfa.amsl.com>; Tue, 17 Mar 2015 13:49:02 -0700 (PDT)
Received: from mail-wg0-x235.google.com (mail-wg0-x235.google.com [IPv6:2a00:1450:400c:c00::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 BD1231A88C2 for <unbearable@ietf.org>; Tue, 17 Mar 2015 13:49:01 -0700 (PDT)
Received: by wggv3 with SMTP id v3so18488554wgg.1 for <unbearable@ietf.org>; Tue, 17 Mar 2015 13:49:00 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=sender:from:to:date:mime-version:subject:reply-to:cc:message-id :priority:in-reply-to:references:content-type; bh=hZxc51g5IsI17GArL9E95OmhHDFvWuyE0OyaP1ejNqw=; b=FkoGkswqBTYnEVfCdVAKhCAyOZO1Fm9+KSaGKXmsWeYGDXfkusf5KfXowdjA64a4bz /eA//2JpecrUzd0smExYhgy1Au7iUnQq1yW4yqJ4gO88MiDkBrLlhlgYkdsNJxARN7ok kg2J4Rs8L2tQQ4KjjitdYjLTKT8y8hUIXr+edfP0/uS9nzN8Qy9luBMkAhYOuc8itNxn HE02SwJfQAC4G4T7gd7wXG38gVjiHyMjN56Edp7ssOziNC2C4zGdhtU0wfjmRuRcPQuD lACC2UizE2UMQqmOx3BN+nXkdPi9afs+LKq2/NMK/epXKSFby53KFWrlpiDWX/b2jx50 uvCQ==
X-Received: by 10.180.72.98 with SMTP id c2mr837982wiv.87.1426625340377; Tue, 17 Mar 2015 13:49:00 -0700 (PDT)
Received: from [192.168.1.224] (host154-72-dynamic.32-79-r.retail.telecomitalia.it. [79.32.72.154]) by mx.google.com with ESMTPSA id q6sm208875wix.3.2015.03.17.13.48.58 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 17 Mar 2015 13:48:59 -0700 (PDT)
Sender: Domenico Rotondi <rdom54@gmail.com>
From: Domenico Rotondi <d.rotondi@computer.org>
X-Google-Original-From: "Domenico Rotondi" <D.Rotondi@Computer.Org>
To: Tony Arcieri <bascule@gmail.com>
Date: Tue, 17 Mar 2015 21:48:53 +0100
MIME-Version: 1.0
Message-ID: <55089335.47.D22F963@D.Rotondi.Computer.Org>
Priority: normal
In-reply-to: <CAHOTMVJd-_Xi+HonYHEPrKkoK5GEpZGOpmrR36tsOF1MGYNOkA@mail.gmail.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com>, <55055C05.17717.9366AE@D.Rotondi.Computer.Org>, <CAHOTMVJd-_Xi+HonYHEPrKkoK5GEpZGOpmrR36tsOF1MGYNOkA@mail.gmail.com>
X-mailer: Pegasus Mail for Windows (4.70)
Content-type: Multipart/Related; boundary="Message-Boundary-31420"
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/lpv_s9v4olfR8bChOExnx0LI-OM>
Cc: John Bradley <ve7jtb@ve7jtb.com>, "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: D.Rotondi@computer.org
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 17 Mar 2015 20:49:04 -0000

--Message-Boundary-31420
Content-type: Multipart/Alternative; boundary="Alt-Boundary-18520.220395875"

--Alt-Boundary-18520.220395875
Content-type: text/plain; charset=UTF-8
Content-transfer-encoding: Quoted-printable
Content-description: Mail message body

Hi,
see inline my comments.
Ciao
   Domenico



On 15 Mar 2015 at 12:14, Tony Arcieri wrote:

On Sun, Mar 15, 2015 at 3:16 AM, Domenico Rotondi <d.rotondi@computer.org>=
 wrote:
    OAuth token is a kind of capability-based security token.

I think you need to be very careful when using the term "capability". It's=
 a term that doesn't mean a 
whole lot to most people but has incredible significance in systems like C=
apTP/E.
[Dom] I know it's a term many people doesn't know, as I know its meaning i=
ts quite wide. Anyway, its 
most synthetic defintion can be summed up as "its a way to manage access t=
o a resource based on the 
possession of a token that states what 'capability' you have on that resou=
rce"
=C2=A0
    A capability-based security=C2=A0token can be "ownership based" where =
having (even 
    a copy of) a token is equivalent=C2=A0to being authorized to perform t=
he operation on 
    the resource

Yes, however the capability must be entangled and inseparable from the res=
ource it acts on. For HTTP 
requests, I would consider that encoding both an HTTP method and a full pa=
th to a resource in 
addition to the hostname.
[Dom] exactly. A capability token identifies the resource(s) you are entit=
led to act on, as well as what 
'action' can you do on it. So for an HTTP request you have to clearly dein=
tify the resurce (using its 
URL) and the HTTP 'verbs' you can use on that resoure. To avoid security i=
ssues, like confused deputy 
problem, the capability token has to specify only 1 verb.
    the token is associated=C2=A0to, as well as more complex things like h=
aving the token 
    usable only by entities who
    can prove to be the owner.
    Among the possible features tehre is the possibility to grant the righ=
t to delegate 
    all
    or part of the rights granted by the token.

These are properties that I think are necessary of true capability systems=
, but not sufficient. I'd look to 
the Capability Myths Demolished paper for that:

http://zesty.ca/capmyths/usenix.pdf
=C2=A0
[Dom] in same capability based system, capabilities are directly associate=
d by the OS to each object's 
instance; so in these systems there is no need for object's instances to p=
rove they are the owner of a 
capability token. In more distributed and open systems the proff of owners=
hip becomes a critical 
aspect especially when rights delegation has to be supported.
A big adavantage of capability tokens is that you decouple Identity Manage=
ment from Access Control 
Management, which improves scalability and can be essential in some contex=
ts.

Another alternative could be to intelligently use the Google proppsed maca=
roons
mechanism.

Note that the Macaroons paper quite carefully distinguishes between capabi=
lities and credentials, and 
suggests that Macaroons are credentials and *not* capabilities.

https://static.googleusercontent.com/media/research.google.com/en/us/pubs/=
archive/41892.pdf

[Dom] the Google paper doesn't use the term "credentials" as we normally u=
se it (i.e., to identify 
somethings that is tied to authentication); the meaning they give to "cred=
entials" is the one the 
diplomacy gives to this term: a set of statements that entitle you to have=
 a role (e.g., an ambassador 
representing country X toward country Y).
This is the meaning of a capability token (as is the meaning of an OAuth t=
oken): the token states that 
you are entitled to do a specific action on a specific reosurce.
--
Tony Arcieri

--Alt-Boundary-18520.220395875
Content-type: text/html; charset=UTF-8
Content-transfer-encoding: Quoted-printable
Content-description: Mail message body

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
          "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html  xmlns=3D"http://www.w3.org/1999/xhtml" xml:lang=3D"en" lang=3D"en">=
<head>
<title></title>
<meta http-equiv=3D"content-type" content=3D"text/html;charset=3Dutf-8"/>
<meta http-equiv=3D"Content-Style-Type" content=3D"text/css"/>
</head>
<body>
<div align=3D"left"><font face=3D"Times New Roman"><span style=3D" font-si=
ze:12pt">Hi,</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman"><span style=3D" font-si=
ze:12pt">see inline my comments.</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman"><span style=3D" font-si=
ze:12pt">Ciao</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman"><span style=3D" font-si=
ze:12pt">&#160;&#160; Domenico</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman"><span style=3D" font-si=
ze:12pt"><br />
</span></font></div>
<div align=3D"left">
<hr width=3D"100%" align=3D"left"/>
<font face=3D"Times New Roman"><span style=3D" font-size:12pt"><br />
</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman"><span style=3D" font-si=
ze:12pt">On 15 Mar 2015 at 12:14, Tony Arcieri wrote:</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman"><span style=3D" font-si=
ze:12pt"><br />
</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt">On Sun, Mar 15, 2015 at 3:16 AM, Domenico Rotondi </span=
>
<span style=3D" font-size:10pt">&lt;</span></font><a href=3D"mailto:d.roto=
ndi@computer.org"><font face=3D"Times New Roman" color=3D"#0000ff" size=3D=
"2">
<span style=3D" font-size:10pt"><u>d.rotondi@computer.org</u></span></font=
></a><font face=3D"Times New Roman" size=3D"2"><span style=3D" font-size:1=
0pt">&gt;</span>
<span style=3D" font-size:10pt"> wrote:</span></font></div>
<p style=3D"margin-left:9mm; margin-right:0mm; text-indent:0mm; margin-top=
:1.76mm; margin-bottom:1.76mm;"><font face=3D"Times New Roman">
<span style=3D" font-size:12pt">OAuth token is a kind of capability-based =
security token.</span></font></p>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt"><br />
</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt">I think you need to be very careful when using the term =
&quot;capability&quot;. It's a term that doesn't mean a 
whole lot to most people but has incredible significance in systems like C=
apTP/E.</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" color=3D"#0000ff" size=3D=
"2"><span style=3D" font-size:10pt">[Dom] I know it's a term many people d=
oesn't know, as I know its meaning its quite wide. Anyway, its 
most synthetic defintion can be summed up as &quot;its a way to manage acc=
ess to a resource based on the 
possession of a token that states what 'capability' you have on that resou=
rce&quot;</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt">&#160;</span></font></div>
<p style=3D"margin-left:9mm; margin-right:0mm; text-indent:0mm; margin-top=
:1.76mm; margin-bottom:1.76mm;"><font face=3D"Times New Roman">
<span style=3D" font-size:12pt">A capability-based security&#160;token can=
 be &quot;ownership based&quot; where having (even 
a copy of) a token is equivalent&#160;to being authorized to perform the o=
peration on 
the resource</span></font></p>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt"><br />
</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt">Yes, however the capability must be entangled and insepa=
rable from the resource it acts on. For HTTP 
requests, I would consider that encoding both an HTTP method and a full pa=
th to a resource in 
addition to the hostname.</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" color=3D"#0000ff" size=3D=
"2"><span style=3D" font-size:10pt">[Dom] exactly. A capability token iden=
tifies the resource(s) you are entitled to act on, as well as what 
'action' can you do on it. So for an HTTP request you have to clearly dein=
tify the resurce (using its 
URL) and the HTTP 'verbs' you can use on that resoure. To avoid security i=
ssues, like confused deputy 
problem, the capability token has to specify only 1 verb.</span></font></d=
iv>
<div align=3D"left" style=3D"margin-left:9mm; margin-right:0mm; text-inden=
t:0mm; margin-top:0.00mm; margin-bottom:0.00mm;"><font face=3D"Times New R=
oman">
<span style=3D" font-size:12pt">the token is associated&#160;to, as well a=
s more complex things like having the token 
usable only by entities who<br />
can prove to be the owner.<br />
Among the possible features tehre is the possibility to grant the right to=
 delegate 
all<br />
or part of the rights granted by the token.</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt"><br />
</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt">These are properties that I think are necessary of true =
capability systems, but not sufficient. I'd look to 
the Capability Myths Demolished paper for that:</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt"><br />
</span></font></div>
<div align=3D"left"><a href=3D"http://zesty.ca/capmyths/usenix.pdf"><font =
face=3D"Times New Roman" color=3D"#0000ff" size=3D"2"><span style=3D" font=
-size:10pt">
<u>http://zesty.ca/capmyths/usenix.pdf</u></span></font></a></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt">&#160;</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" color=3D"#0000ff" size=3D=
"2"><span style=3D" font-size:10pt">[Dom] in same capability based system,=
 capabilities are directly associated by the OS to each object's 
instance; so in these systems there is no need for object's instances to p=
rove they are the owner of a 
capability token. In more distributed and open systems the proff of owners=
hip becomes a critical 
aspect especially when rights delegation has to be supported.</span></font=
></div>
<div align=3D"left"><font face=3D"Times New Roman" color=3D"#0000ff" size=3D=
"2"><span style=3D" font-size:10pt">A big adavantage of capability tokens =
is that you decouple Identity Management from Access Control 
Management, which improves scalability and can be essential in some contex=
ts.</span></font></div>
<p style=3D"margin-left:9mm; margin-right:0mm; text-indent:0mm; margin-top=
:1.76mm; margin-bottom:1.76mm;"><font face=3D"Times New Roman">
<span style=3D" font-size:12pt"><br />
</span></font></p>
<div align=3D"left"><font face=3D"Times New Roman"><span style=3D" font-si=
ze:12pt">Another alternative could be to intelligently use the Google prop=
psed macaroons
<br />
mechanism.</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt"><br />
</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt">Note that the Macaroons paper quite carefully distinguis=
hes between capabilities and credentials, and 
suggests that Macaroons are credentials and *not* capabilities.</span></fo=
nt></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt"><br />
</span></font></div>
<div align=3D"left"><a href=3D"https://static.googleusercontent.com/media/=
research.google.com/en/us/pubs/archive/41892.pdf"><font face=3D"Times New =
Roman" color=3D"#0000ff" size=3D"2">
<span style=3D" font-size:10pt"><u>https://static.googleusercontent.com/me=
dia/research.google.com/en/us/pubs/archive/41892.pdf</u></span></font></a>=
</div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt"><br />
</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" color=3D"#0000ff" size=3D=
"2"><span style=3D" font-size:10pt">[Dom] the Google paper doesn't use the=
 term &quot;credentials&quot; as we normally use it (i.e., to identify 
somethings that is tied to authentication); the meaning they give to &quot=
;credentials&quot; is the one the 
diplomacy gives to this term: a set of statements that entitle you to have=
 a role (e.g., an ambassador 
representing country X toward country Y).</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" color=3D"#0000ff" size=3D=
"2"><span style=3D" font-size:10pt">This is the meaning of a capability to=
ken (as is the meaning of an OAuth token): the token states that 
you are entitled to do a specific action on a specific reosurce.</span></f=
ont></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt">--</span></font></div>
<div align=3D"left"><font face=3D"Times New Roman" size=3D"2"><span style=3D=
" font-size:10pt">Tony Arcieri</span></font></div>
</body>
</html>

--Alt-Boundary-18520.220395875--

--Message-Boundary-31420--


From nobody Tue Mar 17 19:00:35 2015
Return-Path: <dkg@fifthhorseman.net>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A5BB71A89C4 for <unbearable@ietfa.amsl.com>; Tue, 17 Mar 2015 19:00:33 -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
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 NTGdPuzCbJlJ for <unbearable@ietfa.amsl.com>; Tue, 17 Mar 2015 19:00:32 -0700 (PDT)
Received: from che.mayfirst.org (che.mayfirst.org [209.234.253.108]) by ietfa.amsl.com (Postfix) with ESMTP id 399E11A89B3 for <unbearable@ietf.org>; Tue, 17 Mar 2015 19:00:32 -0700 (PDT)
Received: from fifthhorseman.net (unknown [38.109.115.130]) by che.mayfirst.org (Postfix) with ESMTPSA id 6814CF984; Tue, 17 Mar 2015 22:00:28 -0400 (EDT)
Received: by fifthhorseman.net (Postfix, from userid 1000) id DCD7F20296; Tue, 17 Mar 2015 19:00:13 -0700 (PDT)
From: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
To: John Bradley <ve7jtb@ve7jtb.com>, Ilari Liusvaara <ilari.liusvaara@elisanet.fi>
In-Reply-To: <2EFCA3AA-D611-4DAE-AB28-799327B08292@ve7jtb.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com> <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com> <20150315181130.GA16994@LK-Perkele-VII> <2EFCA3AA-D611-4DAE-AB28-799327B08292@ve7jtb.com>
User-Agent: Notmuch/0.18.2 (http://notmuchmail.org) Emacs/24.4.1 (x86_64-pc-linux-gnu)
Date: Tue, 17 Mar 2015 22:00:13 -0400
Message-ID: <87oanrkqtu.fsf@alice.fifthhorseman.net>
MIME-Version: 1.0
Content-Type: text/plain
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/geAvBftHt4zjcDrHdpOZ-HcyGdA>
Cc: Anthony Nadalin <tonynad@microsoft.com>, Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, Phil Hunt <phil.hunt@oracle.com>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 18 Mar 2015 02:00:33 -0000

On Sun 2015-03-15 15:34:24 -0400, John Bradley wrote:
> Tying the TLS keys to sessions is perhaps the most privacy preserving
> way to do it but may loose many of the other benefits.

There are lots of sessions involved at different layers the stack at
this point ;) Do you mean "tying the keys to TLS sessions"?

Even in this case, the above suggestion isn't clear.  The only way that
a TLS session is guaranteed to be terminated by both peers is in the
case of a fatal alert, so we still don't really know when the key should
be invalidated.

Even failure of a server to accept a proposed TLS session resumption
doesn't necessarily invalidate a given TLS session ID (though many TLS
clients will treat a server response rejecting the session ID as a
signal to discard it).

       --dkg


From nobody Wed Mar 18 10:17:44 2015
Return-Path: <Andrei.Popov@microsoft.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B532D1A8AAF for <unbearable@ietfa.amsl.com>; Wed, 18 Mar 2015 10:17:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
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 QlkjnmgZvmsc for <unbearable@ietfa.amsl.com>; Wed, 18 Mar 2015 10:17:32 -0700 (PDT)
Received: from na01-by2-obe.outbound.protection.outlook.com (mail-by2on0126.outbound.protection.outlook.com [207.46.100.126]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1C4D81A8771 for <unbearable@ietf.org>; Wed, 18 Mar 2015 10:17:32 -0700 (PDT)
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com (25.161.207.26) by BN3PR0301MB1235.namprd03.prod.outlook.com (25.161.207.23) with Microsoft SMTP Server (TLS) id 15.1.112.19; Wed, 18 Mar 2015 17:17:24 +0000
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) by BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) with mapi id 15.01.0112.000; Wed, 18 Mar 2015 17:17:24 +0000
From: Andrei Popov <Andrei.Popov@microsoft.com>
To: Daniel Kahn Gillmor <dkg@fifthhorseman.net>, John Bradley <ve7jtb@ve7jtb.com>, Ilari Liusvaara <ilari.liusvaara@elisanet.fi>
Thread-Topic: [Unbearable] Scoping of keys on the client.
Thread-Index: AQHQXa7TliKvYzf3qUabRBSpT6xKBZ0ao/aQgAAMEICAACS5gIAB4DQAgAANMgCAALcHgIAAUgaAgAAQXACAABcpAIADkHaAgAD9jbA=
Date: Wed, 18 Mar 2015 17:17:24 +0000
Message-ID: <BN3PR0301MB12501DB261CFE459F71348798C000@BN3PR0301MB1250.namprd03.prod.outlook.com>
References: <23715BE4-920C-497D-9C15-BF4CF49211E6@ve7jtb.com> <BN3PR0301MB1250B72035FC7C3D24211CA28C070@BN3PR0301MB1250.namprd03.prod.outlook.com> <C520EF34-655A-49FA-B7A2-801BF5E4E5A6@oracle.com> <723006D8-4253-451E-BBCE-214FF9441706@ve7jtb.com> <BN3PR0301MB123468821DDE0EDB724D0C8BA6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <F21E8D7C-AE3A-41B0-84FE-D85374F851CD@ve7jtb.com> <BN3PR0301MB12340B5FA0ADED9ADBBB9E82A6050@BN3PR0301MB1234.namprd03.prod.outlook.com> <67C21A00-EFA2-4F25-8F94-9F74B5FC40E1@ve7jtb.com> <20150315181130.GA16994@LK-Perkele-VII> <2EFCA3AA-D611-4DAE-AB28-799327B08292@ve7jtb.com> <87oanrkqtu.fsf@alice.fifthhorseman.net>
In-Reply-To: <87oanrkqtu.fsf@alice.fifthhorseman.net>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [2001:4898:80e8:ed31::2]
authentication-results: fifthhorseman.net; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0301MB1235;
x-microsoft-antispam-prvs: <BN3PR0301MB12356D2B188037ABA43851E18C000@BN3PR0301MB1235.namprd03.prod.outlook.com>
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(24454002)(377424004)(13464003)(377454003)(46102003)(19580405001)(50986999)(54356999)(76576001)(76176999)(106116001)(33656002)(87936001)(2656002)(99286002)(2900100001)(93886004)(2950100001)(102836002)(62966003)(92566002)(19580395003)(86612001)(86362001)(40100003)(77156002)(74316001)(122556002)(3826002); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0301MB1235; H:BN3PR0301MB1250.namprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5005006)(5002010); SRVR:BN3PR0301MB1235; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0301MB1235; 
x-forefront-prvs: 051900244E
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginatorOrg: microsoft.onmicrosoft.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 18 Mar 2015 17:17:24.7060 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0301MB1235
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/faFyTikpP5dBWyILOHRc7ZFEcX8>
Cc: Anthony Nadalin <tonynad@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, Phil Hunt <phil.hunt@oracle.com>
Subject: Re: [Unbearable] Scoping of keys on the client.
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 18 Mar 2015 17:17:34 -0000

Hi Daniel,

In the current proposals, the client proves possession of the Token Binding=
 key for each TLS connection (not session).

For privacy reasons, the user must have an option of deleting Token Binding=
 keys (e.g. when clearing cookies).

Cheers,

Andrei

-----Original Message-----
From: Daniel Kahn Gillmor [mailto:dkg@fifthhorseman.net]=20
Sent: Tuesday, March 17, 2015 7:00 PM
To: John Bradley; Ilari Liusvaara
Cc: Anthony Nadalin; Andrei Popov; unbearable@ietf.org; Phil Hunt
Subject: Re: [Unbearable] Scoping of keys on the client.

On Sun 2015-03-15 15:34:24 -0400, John Bradley wrote:
> Tying the TLS keys to sessions is perhaps the most privacy preserving=20
> way to do it but may loose many of the other benefits.

There are lots of sessions involved at different layers the stack at this p=
oint ;) Do you mean "tying the keys to TLS sessions"?

Even in this case, the above suggestion isn't clear.  The only way that a T=
LS session is guaranteed to be terminated by both peers is in the case of a=
 fatal alert, so we still don't really know when the key should be invalida=
ted.

Even failure of a server to accept a proposed TLS session resumption doesn'=
t necessarily invalidate a given TLS session ID (though many TLS clients wi=
ll treat a server response rejecting the session ID as a signal to discard =
it).

       --dkg


From nobody Mon Mar 23 08:02:22 2015
Return-Path: <leifj@sunet.se>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A76501A8F50 for <unbearable@ietfa.amsl.com>; Mon, 23 Mar 2015 08:02:19 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.661
X-Spam-Level: 
X-Spam-Status: No, score=-1.661 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HELO_EQ_SE=0.35, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 B0MJKngrhmQx for <unbearable@ietfa.amsl.com>; Mon, 23 Mar 2015 08:02:18 -0700 (PDT)
Received: from e-mailfilter01.sunet.se (e-mailfilter01.sunet.se [IPv6:2001:6b0:8:2::201]) (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 7340E1A8F51 for <unbearable@ietf.org>; Mon, 23 Mar 2015 08:01:40 -0700 (PDT)
Received: from smtp1.sunet.se (smtp1.sunet.se [192.36.171.214]) by e-mailfilter01.sunet.se (8.14.4/8.14.4/Debian-4) with ESMTP id t2NF1cI4004701 (version=TLSv1/SSLv3 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK) for <unbearable@ietf.org>; Mon, 23 Mar 2015 16:01:38 +0100
Received: from kerio.sunet.se (kerio.sunet.se [192.36.171.210]) by smtp1.sunet.se (8.14.9/8.14.7) with ESMTP id t2NF1ZEM021000 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for <unbearable@ietf.org>; Mon, 23 Mar 2015 16:01:38 +0100 (CET)
VBR-Info: md=sunet.se; mc=all; mv=swamid.se
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=sunet.se; s=default; t=1427122898; bh=K7PK9DZtJMx0aKguMNoFfTeky2lfJN8IZkAWfgeq/Hg=; h=Date:From:To:Subject; b=zkSKFIzMPeKuNkLp+Lc+FePLMIgl+3W9M+npeJl/QX7xbOnW6L380Yc9FksIzsntl LKcAfoLYdrE3MxyGQJgb/gDxTTpyAvZknqLRt1a1qEJXPPrcbapbduFo6QQOnJ/AWa HNfOQyvDJqBmB+0yEUm0ntocAtA4hPWCHfal+Dck=
X-Footer: c3VuZXQuc2U=
Received: from [31.133.152.59] ([31.133.152.59]) (authenticated user leifj@sunet.se) by kerio.sunet.se (Kerio Connect 8.3.4 patch 1) (using TLSv1.2 with cipher DHE-RSA-AES256-SHA (256 bits)) for unbearable@ietf.org; Mon, 23 Mar 2015 16:01:34 +0100
Message-ID: <55102ACC.7060907@sunet.se>
Date: Mon, 23 Mar 2015 16:01:32 +0100
From: Leif Johansson <leifj@sunet.se>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: "unbearable@ietf.org" <unbearable@ietf.org>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
X-Bayes-Prob: 0.0001 (Score 0, tokens from: outbound, outbound-sunet-se:default, sunet-se:default, base:default, @@RPTN)
X-CanIt-Geo: ip=192.36.171.210; country=SE; latitude=59.3294; longitude=18.0686; http://maps.google.com/maps?q=59.3294,18.0686&z=6
X-CanItPRO-Stream: outbound-sunet-se:outbound (inherits from outbound-sunet-se:default, sunet-se:default, base:default)
X-Canit-Stats-ID: 09O731CJM - 08cca8f38eb9 - 20150323
X-CanIt-Archive-Cluster: PfMRe/vJWMiXwM2YIH5BVExnUnw
Received-SPF: neutral (e-mailfilter01.sunet.se: 192.36.171.210 is neither permitted nor denied by domain leifj@sunet.se) receiver=e-mailfilter01.sunet.se; client-ip=192.36.171.210; envelope-from=<leifj@sunet.se>; helo=smtp1.sunet.se; identity=mailfrom
X-Scanned-By: CanIt (www . roaringpenguin . com) on 192.36.171.201
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/tkZoPlG4LmG4fYuZuQ3JNrQ5ZO4>
Subject: [Unbearable] slides & agenda for Dallas
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Mon, 23 Mar 2015 15:02:19 -0000

are up on the agenda page - see you all (or some of you at least)
tomorrow in Dallas!

	Cheers Leif


From nobody Tue Mar 24 10:18:46 2015
Return-Path: <leifj@mnt.se>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D47051AC3A1 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 10:18:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.6
X-Spam-Level: 
X-Spam-Status: No, score=-2.6 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7] autolearn=ham
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 xsTXfMw_ghQX for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 10:18:33 -0700 (PDT)
Received: from mail-oi0-f47.google.com (mail-oi0-f47.google.com [209.85.218.47]) (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 D3F6D1A9218 for <unbearable@ietf.org>; Tue, 24 Mar 2015 10:18:22 -0700 (PDT)
Received: by oifl3 with SMTP id l3so143644993oif.0 for <unbearable@ietf.org>; Tue, 24 Mar 2015 10:18:22 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:message-id:date:from:user-agent:mime-version:to :subject:content-type:content-transfer-encoding; bh=d/KWGauudd58GG1Z7+pwbxxhIGmRQi7XIu13CcxPbcA=; b=En53WZFIjU2PsXstgZCQEFJXn4UedGhEn2th3KFGcVAb6UiRw9szgTm5SIMlDUfJ0v C8YdLK0I6SjIzqaBA/ZuwaG/UoSKB3oupXTF0VMfudcxZG6yUcdPxfZ5xwA32xx11bxg i+47rtc/2kM2wGGuF1b/IT0ZSl61zJHeLrQ4GfhtfpPVliByYkvJi0uAeaI8X9bmSckD hGCjdeqrVzZ3Su6XJjniwZt4nQbiinF05OHuY8issPlvl73jg0DotORgu/VHXwnDFSzI ufn8TK0EijlDuOCz3an41JKz6+iiEDob2wOqkwjmWopb0Tl3Ot80/vjSy+1U1KVOODeh oGnA==
X-Gm-Message-State: ALoCoQkn3vL6edpZUfgTh4n9/3KltqBpYzhhxJh9Fo4+Hik6UE6dTPPz66BNZ8KqwCgwIckaNbtB
X-Received: by 10.182.24.202 with SMTP id w10mr4210134obf.9.1427217502421; Tue, 24 Mar 2015 10:18:22 -0700 (PDT)
Received: from [172.16.86.24] ([199.227.110.154]) by mx.google.com with ESMTPSA id dc10sm2507181obb.11.2015.03.24.10.18.20 for <unbearable@ietf.org> (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 24 Mar 2015 10:18:21 -0700 (PDT)
Message-ID: <55119C5C.5060104@mnt.se>
Date: Tue, 24 Mar 2015 18:18:20 +0100
From: Leif Johansson <leifj@mnt.se>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.5.0
MIME-Version: 1.0
To: "unbearable@ietf.org" <unbearable@ietf.org>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/BZ281nhROcxK9PJ2mvZI0ApqvXU>
Subject: [Unbearable] housekeeping: issue tracking
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 24 Mar 2015 17:18:39 -0000

Today in Dallas the WG decided to use the github site

https://github.com/TokenBinding/Internet-Drafts

to track issues for the documents.

Please send a message to the wg list as you open issues and lets apply
common sense when opening issues so the wg won't get inundated.

	Cheers Leif


From nobody Tue Mar 24 10:36:10 2015
Return-Path: <ynir.ietf@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5443B1AC3F9 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 10:36:08 -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, SPF_PASS=-0.001] autolearn=ham
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 U3_lJOQMvOlj for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 10:36:06 -0700 (PDT)
Received: from mail-wi0-x232.google.com (mail-wi0-x232.google.com [IPv6:2a00:1450:400c:c05::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 8DB021AC40E for <unbearable@ietf.org>; Tue, 24 Mar 2015 10:35:57 -0700 (PDT)
Received: by wibdy8 with SMTP id dy8so81233521wib.0 for <unbearable@ietf.org>; Tue, 24 Mar 2015 10:35:56 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=from:content-type:content-transfer-encoding:subject:message-id:date :to:mime-version; bh=d83e8mrKnbrHuK3JXMWZzuqNFNMUaeZcUivBPewHEis=; b=TXCZeLPvo7pcCNUrZx9qzGMzhxKSBC/7O+OpjsxWw53HzirQcNpBSygCcVnWpGKP3V mp5Cg/XWdGqd0JM2o1DIFWh3GP/TC+1fWWaZ9DDLWDUaORRGnBqw1NWVNvckQdgqXUsi 0cy505Vy0UNh/3qI2Ezoscd5AoNx9HbXXPNCjdKP9EfMNe7Fzgfc7qPkeP1V/tadDFOf qHgGRw4hk2Xe0uBvKuffkFIQMfCTTNYIfdXpR68HxYG6RVBlq+t15a1t7gxdGcFhAx+K rKy6oTCyBTvY4VvQp18tPxUC7ArSnyt3O4iuC/CO+hg/SCWimYtIcfI+5nyTLQGTWDXU zePw==
X-Received: by 10.180.89.210 with SMTP id bq18mr30981915wib.28.1427218556379;  Tue, 24 Mar 2015 10:35:56 -0700 (PDT)
Received: from ?IPv6:2001:67c:370:176:8c6e:7a0a:ced9:2267? ([2001:67c:370:176:8c6e:7a0a:ced9:2267]) by mx.google.com with ESMTPSA id fo9sm16831380wib.16.2015.03.24.10.35.55 for <unbearable@ietf.org> (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Tue, 24 Mar 2015 10:35:55 -0700 (PDT)
From: Yoav Nir <ynir.ietf@gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Message-Id: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com>
Date: Tue, 24 Mar 2015 12:35:53 -0500
To: unbearable@ietf.org
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/hoSLhFxH51ylX0qRuPyX-G_oNOQ>
Subject: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 24 Mar 2015 17:36:08 -0000

Hi

To continue the conversation from today=E2=80=99s session, I=E2=80=99m =
not talking about server-side proxies such as load balancers, but about =
client-side TLS MitM.

The issue here is that in the presence of such a proxy, there is one TLS =
session between the client and proxy, and another between the proxy and =
the server. Different servers mean different tls_unique values, and with =
different values, the signature validation fails.

One suggested solution in today=E2=80=99s meeting was for the proxy to =
manage the sessions on behalf of the UA. Makes sense to trust the proxy =
with session management, since you=E2=80=99re anyway trusting it with =
the cleartext. So your client may or may not have a token bound with the =
proxy, but the proxy does have a token bound with the server. This means =
that =E2=80=9Cwe=E2=80=9D trust the proxy to impersonate UAs to servers =
and push some of its own requests authenticated with the users=E2=80=99s =
key.

In practice today, the browsers know when they are being proxied. This =
is usually done with some heuristic. Google=E2=80=99s is to identify all =
CAs that do NOT come bundled with the OS but were added by the user, and =
consider all such CAs to be intercepting proxies. So if the server =
certificate is signed by such a CA, the connection is considered to be =
proxied. We could solve the issue by having the browser indicate no =
support for token binding when faced with a proxied connection. In such =
a case the server may revert to a legacy session management mechanism, =
probably cookies.

I don=E2=80=99t think that this is a desirable outcome. Ideally there =
wouldn=E2=80=99t be TLS proxies at all, but given that they are, it =
would be good if we could have end-to-end session management even in the =
absence of end-to-end confidentiality. So if we could use some kind of =
mechanism to do this in the presence of a proxy. It can be done in =
several ways. Someone (I think Martin) mentioned a challenge-response =
idea. Such an idea should be fleshed out to make sure we neither slow =
down every request nor create a new bearer token.

Yoav




From nobody Tue Mar 24 11:37:56 2015
Return-Path: <ietf-secretariat-reply@ietf.org>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EE94F1AC3F3 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 10:37:09 -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
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 BdyoxEvbjEZX for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 10:37:08 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 1CBA61AC411 for <unbearable@ietf.org>; Tue, 24 Mar 2015 10:37:05 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
To: <unbearable@ietf.org>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.12.3
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150324173705.12387.5960.idtracker@ietfa.amsl.com>
Date: Tue, 24 Mar 2015 10:37:05 -0700
From: IETF Secretariat <ietf-secretariat-reply@ietf.org>
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/ezoRCpe6iTVgPSfuwFwRZ0kxQj8>
X-Mailman-Approved-At: Tue, 24 Mar 2015 11:37:54 -0700
Subject: [Unbearable] Milestones changed for tokbind WG
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 24 Mar 2015 17:37:10 -0000

Changed milestone "WG document for the Token Binding Protocol v1.0.",
set due date to March 2015 from January 2016.

Changed milestone "WG document for HTTPS Token Binding.", set due date
to March 2015 from January 2016.

Changed milestone "HTTPS Token Binding to IESG.", set due date to
December 2015 from January 2016.

Changed milestone "Token Binding Protocol v1.0 to IESG.", set due date
to December 2015 from January 2016.

URL: http://datatracker.ietf.org/wg/tokbind/charter/


From nobody Tue Mar 24 14:45:47 2015
Return-Path: <martin.thomson@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9DB011A905F for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 14:45:45 -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, SPF_PASS=-0.001] autolearn=ham
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 xCif0f23Gi2S for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 14:45:44 -0700 (PDT)
Received: from mail-ob0-x232.google.com (mail-ob0-x232.google.com [IPv6:2607:f8b0:4003:c01::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 B54C91A90D0 for <unbearable@ietf.org>; Tue, 24 Mar 2015 14:45:30 -0700 (PDT)
Received: by obcxo2 with SMTP id xo2so5269745obc.0 for <unbearable@ietf.org>; Tue, 24 Mar 2015 14:45:30 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=BezMKKEj9N7aV/wpENJpazap6G7/NU79M6rzQP2m9Ok=; b=ldv0dwMpjk/VQ3hT0sGD5feQbaJ9cGY2GZ+lmFmtPEPjVXioEMtWmdcdpQ9xSNW+7c 2VS+3G4TBMDOTv2QDkf7TlEOGaGsAthxIdQOym3fYwAefjA+c8xyfgW9P9r46RKmnCFZ os3M69jr1xjjZb6dgOVDzaMnxMIqodmDsqJnJSIGOsdJsOIfPzqCTpVNe4vSDy2rxTYY 08p254pChkPZF0hLtD/X6NkOUJXGmfhJFTdn1X9FMtEyl/R1klBYnRVSspIpHssxJa4V mfG33+BHg78PeDoLke6OoDdrJusAAu4pjJsKR/Evo4ftm92ZjFngfMZkXt9fAnQsTGPu +4dA==
MIME-Version: 1.0
X-Received: by 10.60.56.39 with SMTP id x7mr4941996oep.65.1427233530290; Tue, 24 Mar 2015 14:45:30 -0700 (PDT)
Received: by 10.202.48.151 with HTTP; Tue, 24 Mar 2015 14:45:30 -0700 (PDT)
In-Reply-To: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com>
References: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com>
Date: Tue, 24 Mar 2015 16:45:30 -0500
Message-ID: <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com>
From: Martin Thomson <martin.thomson@gmail.com>
To: Yoav Nir <ynir.ietf@gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/nng3E3hzBeAnJvuHV8zWA3kEzL4>
Cc: unbearable@ietf.org
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 24 Mar 2015 21:45:45 -0000

On 24 March 2015 at 12:35, Yoav Nir <ynir.ietf@gmail.com> wrote:
> I don=E2=80=99t think that this is a desirable outcome. Ideally there wou=
ldn=E2=80=99t be TLS proxies at all, but given that they are, it would be g=
ood if we could have end-to-end session management even in the absence of e=
nd-to-end confidentiality. So if we could use some kind of mechanism to do =
this in the presence of a proxy. It can be done in several ways. Someone (I=
 think Martin) mentioned a challenge-response idea. Such an idea should be =
fleshed out to make sure we neither slow down every request nor create a ne=
w bearer token.


I think that for the most part, the only reason that we have a
proposal to use TLS unique is that it is a) present at the right time,
and b) hard to move (i.e., replay).  A challenge from the server would
accomplish the same end, but only at the cost of a round trip.

I agree that the situation with MitM is potentially bad.  If the MitM
supports this mechanism, the client will go through the
re-authorization process and thereby bless a private key controlled by
the MitM.  I think that runs counter to the goals of the mechanism.

But I don't know how to solve that problem, either by preventing the
impersonation, or by detecting the MitM at the server and suppressing
the binding code.  At this point, I'm inclined to leave as-is.  That
implies some sort of backup protection mechanism if the MitM strips
the binding and maybe even a hard failure if the MitM breaks it.  A
backup mechanism will be required for clients that don't support token
binding.


From nobody Tue Mar 24 15:06:26 2015
Return-Path: <Andrei.Popov@microsoft.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1B4341AC439 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 15:06:19 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
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 hWxFkZx1ILQc for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 15:06:13 -0700 (PDT)
Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0144.outbound.protection.outlook.com [65.55.169.144]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 66C351ACC83 for <unbearable@ietf.org>; Tue, 24 Mar 2015 15:05:35 -0700 (PDT)
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com (25.161.207.26) by BN3PR0301MB1249.namprd03.prod.outlook.com (25.161.207.25) with Microsoft SMTP Server (TLS) id 15.1.112.19; Tue, 24 Mar 2015 22:05:33 +0000
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) by BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) with mapi id 15.01.0112.000; Tue, 24 Mar 2015 22:05:33 +0000
From: Andrei Popov <Andrei.Popov@microsoft.com>
To: Martin Thomson <martin.thomson@gmail.com>, Yoav Nir <ynir.ietf@gmail.com>
Thread-Topic: [Unbearable] Some further thoughts about TLS proxies and tokbind
Thread-Index: AQHQZlkHCxbU1dZyK0uqsbXnlnCGq50sKv8AgAABy5A=
Date: Tue, 24 Mar 2015 22:05:32 +0000
Message-ID: <BN3PR0301MB1250E8DC62A65CDBA697288D8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com>
References: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com> <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com>
In-Reply-To: <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [69.15.241.124]
authentication-results: gmail.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0301MB1249;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(377454003)(51444003)(13464003)(54356999)(74316001)(50986999)(76576001)(2900100001)(102836002)(15975445007)(2950100001)(62966003)(77156002)(2656002)(87936001)(76176999)(19580405001)(66066001)(86362001)(33656002)(561944003)(106116001)(99286002)(40100003)(92566002)(46102003)(122556002)(19580395003); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0301MB1249; H:BN3PR0301MB1250.namprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BN3PR0301MB1249B5B554AC7D49704667C18C0A0@BN3PR0301MB1249.namprd03.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BN3PR0301MB1249; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0301MB1249; 
x-forefront-prvs: 0525BB0ADF
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-OriginatorOrg: microsoft.onmicrosoft.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 24 Mar 2015 22:05:32.8105 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0301MB1249
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/VqV0RP9OMJjSx9v4eM_OLidQQDQ>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 24 Mar 2015 22:06:19 -0000
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From nobody Tue Mar 24 16:19:54 2015
Return-Path: <ynir.ietf@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6365D1ACCE7 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 16:19:51 -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, SPF_PASS=-0.001] autolearn=ham
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 cpnLODn0sLdA for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 16:19:50 -0700 (PDT)
Received: from mail-wi0-x231.google.com (mail-wi0-x231.google.com [IPv6:2a00:1450:400c:c05::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 A32FD1ACCE5 for <unbearable@ietf.org>; Tue, 24 Mar 2015 16:19:49 -0700 (PDT)
Received: by wixw10 with SMTP id w10so13839255wix.0 for <unbearable@ietf.org>; Tue, 24 Mar 2015 16:19:48 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=content-type:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=2CyCW6Aha8GnBZvj8wIFQWen3MTEKTfHtOw2Zd9lFU0=; b=CRtSfW5sWoRp4G9p7ZY9lTiJtrtlYdjaNDPGHvVGRNU2ElGrvVfdvSZjlm4+A470E3 /qMUnTEJSjC0Xj5gAdVGwzffxQ9QZZu6lRFO8mfjFgFeJNGcUbu7y2aKChGaPLjraj1r m7fBPa0FqJ2o7fEYalb4Y7L8L20Th7m+TJ/Ar2/83cB9M7i8tJS+nDufiVelJYeGG9E4 pvz98TUfv2Ec3SRluW8FqGEU6ydaG8CI+GcJ7PRp5QgApo9/G4C3p35b7XeX8JiN5v05 7ex88Op1J/jRTBU0ZZJ/6Oabfv8TKODstdnXZ3BqCTB/x0zU0JNgBcN0uFg3gfzihB7D jm3Q==
X-Received: by 10.194.110.38 with SMTP id hx6mr12154349wjb.128.1427239188460;  Tue, 24 Mar 2015 16:19:48 -0700 (PDT)
Received: from ?IPv6:2001:67c:370:176:1081:78d6:c8ef:faa2? ([2001:67c:370:176:1081:78d6:c8ef:faa2]) by mx.google.com with ESMTPSA id bx3sm983336wjc.21.2015.03.24.16.19.47 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Tue, 24 Mar 2015 16:19:47 -0700 (PDT)
Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: Yoav Nir <ynir.ietf@gmail.com>
In-Reply-To: <BN3PR0301MB1250E8DC62A65CDBA697288D8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com>
Date: Tue, 24 Mar 2015 18:19:45 -0500
Content-Transfer-Encoding: quoted-printable
Message-Id: <3C2FF331-C587-4BB2-9299-3A82FA3DE613@gmail.com>
References: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com> <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com> <BN3PR0301MB1250E8DC62A65CDBA697288D8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com>
To: Andrei Popov <Andrei.Popov@microsoft.com>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/Dj5SSMf-GOkDhHFv1vWRPwKPoW0>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Martin Thomson <martin.thomson@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 24 Mar 2015 23:19:51 -0000

> On Mar 24, 2015, at 5:05 PM, Andrei Popov <Andrei.Popov@microsoft.com> =
wrote:
>=20
> If the client allows MITM (e.g. by installing a MITM root cert), the =
MITM can modify any application data, suppress the client's offer of =
Token Binding, or cause the server to bind tokens to a MITM-controlled =
key.=20
>=20
> Yoav: do you have a suggestion as to how we can achieve end-to-end =
session management that works in the presence of a TLS proxy?

I can think of a few.

When we were coming up with ideas in websec, one suggestion was for the =
client to sign the request (or rather a selection of the fields from the =
request that the proxy was not likely to mess with). That way you are on =
the one hand not bound to TLS, and on the other hand not making a bearer =
token, because a signature over =E2=80=9CGET / HTTP/1.1=E2=80=9D works =
only for requests getting =E2=80=9C/=E2=80=9C.  The downside of this is =
that every single request needs a new signature, and the signatures can =
still be re-used for the same resources unless you add some URL =
randomization similar to anti-CSRF measures.

Another way is to add to HTTP (or whatever the protocol is) some way of =
getting a unique identifier for this connection and using that. For =
example, =E2=80=9CGET /.well-known/sessionid=E2=80=9D returning two =
values: identifier and challenge. After that the client would include =
both "Token-Binding: EncodedTokenBindingMessage=E2=80=9D and =
=E2=80=9CToken-Binding-Identifier: identifier=E2=80=9D in all subsequent =
requests. This does create a constant token that lasts for the duration =
of the connection, and the proxy has to be trusted to not make use of =
it.

These two can be combined so that the client signs both the challenge =
and parts of the request. That only has the downside of needing to sign =
every request.

I don=E2=80=99t know how to make a method that will both work in the =
presence of a malicious proxy *and* not require a signature on every =
request. Perhaps something can be designed with a MAC that uses a =
per-session key, but that stinks of changing HTTP into TLS.

Yoav


From nobody Tue Mar 24 16:33:39 2015
Return-Path: <ynir.ietf@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F114E1A0113 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 16:33:38 -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, SPF_PASS=-0.001] autolearn=ham
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 duGkXDGITYb4 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 16:33:35 -0700 (PDT)
Received: from mail-wg0-x233.google.com (mail-wg0-x233.google.com [IPv6:2a00:1450:400c:c00::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 3028E1A01AE for <unbearable@ietf.org>; Tue, 24 Mar 2015 16:33:35 -0700 (PDT)
Received: by wgra20 with SMTP id a20so7722184wgr.3 for <unbearable@ietf.org>; Tue, 24 Mar 2015 16:33:34 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=content-type:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=y6o+UTvo9f4W8ZyRAoEC0RKpJOHPRdscLGIpbhE82Cg=; b=ku//nn3u3j2dEIsPG1+TY3CPp6lzGDSxXpbpr8YZ1nn+5OGNHbXrdhZhvseRcx0+4t t3i/QF4JVzJZQuP3ugS7LKC96y6R4j4wB9K9bGotFxFwA4BltjgPKPDOOeUSzCAThhON 8zIvuvLy2ffop4c5oWksP0J+psK8V9WqFIt8GmW211qzfSCRNfsp5Nz1KV6KIXJy4Vq+ 0RI3j23LZdvq2ykUFUhN/9J5gwAnLlTFsUHQSgqShk9TXVqV4cr8r4eLKIJAAGaA9iWG wYNm+iUair7VqSEoWvI+nqw0DOYSvN2zm5GGYVa4oL7w3FEkUlSNVLCNec3VNmMm6TJa Ze7Q==
X-Received: by 10.194.61.244 with SMTP id t20mr11748761wjr.83.1427240013978; Tue, 24 Mar 2015 16:33:33 -0700 (PDT)
Received: from ?IPv6:2001:67c:370:176:1081:78d6:c8ef:faa2? ([2001:67c:370:176:1081:78d6:c8ef:faa2]) by mx.google.com with ESMTPSA id ps4sm1008132wjc.31.2015.03.24.16.33.32 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Tue, 24 Mar 2015 16:33:33 -0700 (PDT)
Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: Yoav Nir <ynir.ietf@gmail.com>
In-Reply-To: <3C2FF331-C587-4BB2-9299-3A82FA3DE613@gmail.com>
Date: Tue, 24 Mar 2015 18:33:31 -0500
Content-Transfer-Encoding: quoted-printable
Message-Id: <54DCF616-3D00-4A49-A326-B44D30327F76@gmail.com>
References: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com> <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com> <BN3PR0301MB1250E8DC62A65CDBA697288D8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com> <3C2FF331-C587-4BB2-9299-3A82FA3DE613@gmail.com>
To: Andrei Popov <Andrei.Popov@microsoft.com>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/wgthzm65OlAgFw0HlKtU_DPvOWA>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Martin Thomson <martin.thomson@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 24 Mar 2015 23:33:39 -0000

> On Mar 24, 2015, at 6:19 PM, Yoav Nir <ynir.ietf@gmail.com> wrote:
>=20
>=20
>> On Mar 24, 2015, at 5:05 PM, Andrei Popov =
<Andrei.Popov@microsoft.com> wrote:
>>=20
>> If the client allows MITM (e.g. by installing a MITM root cert), the =
MITM can modify any application data, suppress the client's offer of =
Token Binding, or cause the server to bind tokens to a MITM-controlled =
key.=20
>>=20
>> Yoav: do you have a suggestion as to how we can achieve end-to-end =
session management that works in the presence of a TLS proxy?
>=20
> I can think of a few.
>=20
> When we were coming up with ideas in websec, one suggestion was for =
the client to sign the request (or rather a selection of the fields from =
the request that the proxy was not likely to mess with). That way you =
are on the one hand not bound to TLS, and on the other hand not making a =
bearer token, because a signature over =E2=80=9CGET / HTTP/1.1=E2=80=9D =
works only for requests getting =E2=80=9C/=E2=80=9C.  The downside of =
this is that every single request needs a new signature, and the =
signatures can still be re-used for the same resources unless you add =
some URL randomization similar to anti-CSRF measures.
>=20
> Another way is to add to HTTP (or whatever the protocol is) some way =
of getting a unique identifier for this connection and using that. For =
example, =E2=80=9CGET /.well-known/sessionid=E2=80=9D returning two =
values: identifier and challenge. After that the client would include =
both "Token-Binding: EncodedTokenBindingMessage=E2=80=9D and =
=E2=80=9CToken-Binding-Identifier: identifier=E2=80=9D in all subsequent =
requests. This does create a constant token that lasts for the duration =
of the connection, and the proxy has to be trusted to not make use of =
it.
>=20
> These two can be combined so that the client signs both the challenge =
and parts of the request. That only has the downside of needing to sign =
every request.
>=20
> I don=E2=80=99t know how to make a method that will both work in the =
presence of a malicious proxy *and* not require a signature on every =
request. Perhaps something can be designed with a MAC that uses a =
per-session key, but that stinks of changing HTTP into TLS.

To be clear, I=E2=80=99m not suggesting text for the draft at this =
point. I think we=E2=80=99re at the spitball stage.=20

I=E2=80=99m not entirely opposed to Martin=E2=80=99s suggestion to leave =
things as they are, if we can have advice for a UA what to do if a =
connection is detected (by heuristic) to be proxied, and have a good =
story on what happens in such a case. I guess this sort-of depends on =
how this is negotiated.

Yoav


From nobody Tue Mar 24 16:39:53 2015
Return-Path: <Andrei.Popov@microsoft.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 16E851A0016 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 16:39:52 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
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 upDI9VWO27Uv for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 16:39:50 -0700 (PDT)
Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0116.outbound.protection.outlook.com [65.55.169.116]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4804A1ACCEE for <unbearable@ietf.org>; Tue, 24 Mar 2015 16:39:50 -0700 (PDT)
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com (25.161.207.26) by BN3PR0301MB1249.namprd03.prod.outlook.com (25.161.207.25) with Microsoft SMTP Server (TLS) id 15.1.112.19; Tue, 24 Mar 2015 23:39:48 +0000
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) by BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) with mapi id 15.01.0112.000; Tue, 24 Mar 2015 23:39:48 +0000
From: Andrei Popov <Andrei.Popov@microsoft.com>
To: Yoav Nir <ynir.ietf@gmail.com>
Thread-Topic: [Unbearable] Some further thoughts about TLS proxies and tokbind
Thread-Index: AQHQZlkHCxbU1dZyK0uqsbXnlnCGq50sKv8AgAABy5CAABiLgIAAAyBw
Date: Tue, 24 Mar 2015 23:39:48 +0000
Message-ID: <BN3PR0301MB1250CC6E00173D055E1F942A8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com>
References: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com> <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com> <BN3PR0301MB1250E8DC62A65CDBA697288D8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com> <3C2FF331-C587-4BB2-9299-3A82FA3DE613@gmail.com>
In-Reply-To: <3C2FF331-C587-4BB2-9299-3A82FA3DE613@gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [69.15.241.124]
authentication-results: gmail.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0301MB1249;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(377454003)(51704005)(13464003)(54356999)(74316001)(50986999)(76576001)(102836002)(2900100001)(2950100001)(93886004)(62966003)(77156002)(2656002)(87936001)(76176999)(19580405001)(66066001)(86362001)(33656002)(106116001)(99286002)(40100003)(92566002)(46102003)(110136001)(122556002)(19580395003); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0301MB1249; H:BN3PR0301MB1250.namprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BN3PR0301MB124919A60F0F023AB02B8F678C0A0@BN3PR0301MB1249.namprd03.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BN3PR0301MB1249; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0301MB1249; 
x-forefront-prvs: 0525BB0ADF
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-OriginatorOrg: microsoft.onmicrosoft.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 24 Mar 2015 23:39:48.3254 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0301MB1249
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/xTIMZLZIj4UtbstfHjQiNO4tTug>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Martin Thomson <martin.thomson@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 24 Mar 2015 23:39:52 -0000
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From nobody Tue Mar 24 17:07:59 2015
Return-Path: <ynir.ietf@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 20A051A1B0E for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 17:07:58 -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, SPF_PASS=-0.001] autolearn=ham
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 MbxYW6K_ZnQg for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 17:07:56 -0700 (PDT)
Received: from mail-ob0-x22e.google.com (mail-ob0-x22e.google.com [IPv6:2607:f8b0:4003:c01::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 702081A1A97 for <unbearable@ietf.org>; Tue, 24 Mar 2015 17:07:56 -0700 (PDT)
Received: by obcjt1 with SMTP id jt1so7218726obc.2 for <unbearable@ietf.org>; Tue, 24 Mar 2015 17:07:56 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=content-type:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=sptckGqavJgVMZX5vif2KeSazADhtZJx4aCHcdKJISA=; b=kDAqGQYXOqcyiZkq6hNY0t9Fk79bR5oYJ2JL2bK2Gy2YkHlULU/xmlWkopcqmwXz5N VkS/N/k9VbVa7spbcTt2lpejq9gt6zoHxKl2mZHftqIauFM6PWvajqdkspo+G7Voz/TO hXC0iaM8yneizjnIiKuPVvShThyY8OZrF1SnSfHAV3keZQMokXff32syr9ZQ0I2vpqSf r8y34w5L9JpJ3hhofWWoOoEl+CRvdHzeaimG59OvUhsaXGg2Mr/t8vIjzFJpGrFZy4kf Gs3EosF664t2IKNYrgBFCHSqMwB77XlPX7KuPH0Ua4OjgOyYDrDjb6L0YIH2fv0RkcA6 LJDQ==
X-Received: by 10.202.200.83 with SMTP id y80mr4921394oif.122.1427242076004; Tue, 24 Mar 2015 17:07:56 -0700 (PDT)
Received: from [10.111.111.8] ([173.200.48.34]) by mx.google.com with ESMTPSA id hc7sm679452obb.16.2015.03.24.17.07.54 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Tue, 24 Mar 2015 17:07:55 -0700 (PDT)
Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: Yoav Nir <ynir.ietf@gmail.com>
In-Reply-To: <BN3PR0301MB1250CC6E00173D055E1F942A8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com>
Date: Tue, 24 Mar 2015 19:07:53 -0500
Content-Transfer-Encoding: quoted-printable
Message-Id: <F989DCB4-0753-411C-99A2-ABA413DB4A3A@gmail.com>
References: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com> <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com> <BN3PR0301MB1250E8DC62A65CDBA697288D8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com> <3C2FF331-C587-4BB2-9299-3A82FA3DE613@gmail.com> <BN3PR0301MB1250CC6E00173D055E1F942A8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com>
To: Andrei Popov <Andrei.Popov@microsoft.com>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/uATUa0Ue7cmPwlPZNvyk4rbiHvg>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Martin Thomson <martin.thomson@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 00:07:58 -0000

> On Mar 24, 2015, at 6:39 PM, Andrei Popov <Andrei.Popov@microsoft.com> =
wrote:
>=20
>> When we were coming up with ideas in websec, one suggestion was for =
the client to sign the request (or rather a selection of the fields from =
the request that the proxy was not likely to mess with). That way you =
are on the one hand not bound to TLS, and on the other hand not making a =
bearer token, because a signature over =E2=80=9CGET / HTTP/1.1=E2=80=9D =
works only for requests getting =E2=80=9C/=E2=80=9C.
>=20
> But what prevents the MITM from removing the signature, or from =
signing with a key of MITM's choosing?
>=20
>> Another way is to add to HTTP (or whatever the protocol is) some way =
of getting a unique identifier for this connection and using that. For =
example, =E2=80=9CGET /.well-known/sessionid=E2=80=9D returning two =
values: identifier and challenge. After that the client would include =
both "Token-Binding: EncodedTokenBindingMessage=E2=80=9D and =
=E2=80=9CToken-Binding-Identifier: identifier=E2=80=9D in all subsequent =
requests.
>=20
> Why can't the MITM use the identifier and challenge in the same way?
>=20
> Or do you mean that these approaches will work until the MITM is =
updated to detect and circumvent these mechanisms?

The MITM is not malicious. It=E2=80=99s performing either caching or =
content filtering or anti-malware. In a sense it is trusted by the user, =
because the user has installed the proxy=E2=80=99s CA certificate to =
make it work.

If the UA is always behind the proxy, then the proxy can strip the =
client=E2=80=99s signature and sign with its own key, and then sign some =
fake requests with the same key. I don=E2=80=99t see how we can prevent =
that. But if the client is sometimes not behind the proxy, then there =
will be two keys and the UA won=E2=80=99t get the same authorizations =
behind the proxy as it gets when not behind the proxy, unless the proxy =
plays nice.  Perhaps the client should not generate the key pair until =
it has an unproxied connection.

Yoav


From nobody Tue Mar 24 17:44:28 2015
Return-Path: <Andrei.Popov@microsoft.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9A4661A1B9E for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 17:44:26 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
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 9OUbKL-39Z_5 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 17:44:24 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1on0760.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::760]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 79AB71A1B85 for <unbearable@ietf.org>; Tue, 24 Mar 2015 17:44:24 -0700 (PDT)
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com (25.161.207.26) by BN3PR0301MB1252.namprd03.prod.outlook.com (25.161.207.28) with Microsoft SMTP Server (TLS) id 15.1.112.19; Wed, 25 Mar 2015 00:44:03 +0000
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) by BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) with mapi id 15.01.0112.000; Wed, 25 Mar 2015 00:44:03 +0000
From: Andrei Popov <Andrei.Popov@microsoft.com>
To: Yoav Nir <ynir.ietf@gmail.com>
Thread-Topic: [Unbearable] Some further thoughts about TLS proxies and tokbind
Thread-Index: AQHQZlkHCxbU1dZyK0uqsbXnlnCGq50sKv8AgAABy5CAABiLgIAAAyBwgAAKUoCAAASt8A==
Date: Wed, 25 Mar 2015 00:44:03 +0000
Message-ID: <BN3PR0301MB1250083227EBF9A1761B3A438C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
References: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com> <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com> <BN3PR0301MB1250E8DC62A65CDBA697288D8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com> <3C2FF331-C587-4BB2-9299-3A82FA3DE613@gmail.com> <BN3PR0301MB1250CC6E00173D055E1F942A8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com> <F989DCB4-0753-411C-99A2-ABA413DB4A3A@gmail.com>
In-Reply-To: <F989DCB4-0753-411C-99A2-ABA413DB4A3A@gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [69.15.241.124]
authentication-results: gmail.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0301MB1252;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(51704005)(13464003)(377454003)(51914003)(106116001)(92566002)(102836002)(66066001)(93886004)(2656002)(62966003)(2900100001)(77156002)(110136001)(99286002)(2950100001)(46102003)(19580405001)(74316001)(122556002)(76576001)(33656002)(86362001)(50986999)(40100003)(19580395003)(87936001)(76176999)(54356999); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0301MB1252; H:BN3PR0301MB1250.namprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BN3PR0301MB1252C8DDB6618DE13D0940998C0B0@BN3PR0301MB1252.namprd03.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BN3PR0301MB1252; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0301MB1252; 
x-forefront-prvs: 052670E5A4
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-OriginatorOrg: microsoft.onmicrosoft.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 25 Mar 2015 00:44:03.3989 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0301MB1252
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/T36ZUjQaWbzoUwHCk_egCa-XqUw>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Martin Thomson <martin.thomson@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 00:44:26 -0000
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From nobody Tue Mar 24 18:20:24 2015
Return-Path: <martin.thomson@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9D39C1A1DBD for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 18:20:23 -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, SPF_PASS=-0.001] autolearn=ham
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 J1JAXP3Z5h2D for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 18:20:22 -0700 (PDT)
Received: from mail-oi0-x22a.google.com (mail-oi0-x22a.google.com [IPv6:2607:f8b0:4003:c06::22a]) (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 3C35B1A0027 for <unbearable@ietf.org>; Tue, 24 Mar 2015 18:20:22 -0700 (PDT)
Received: by oiag65 with SMTP id g65so9410450oia.2 for <unbearable@ietf.org>; Tue, 24 Mar 2015 18:20:21 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=O6wpquS2/EBDwrKdtbxp+htxa/l7Jxd5B4ORF7VryEQ=; b=ijYmhH4PSQIDbt9Nn7CmJhZtjlDO68XSUUx74sxnuTjzvJ6ZDFQbhTVPqtFoUcPrzp +f9DtKPxFEdtpUsNlPlE6ccSZhp2UOQEyzOsbHuabhqRlZiFBddXPLMjAZlkzpGbBmRW hYGlq5XWKVujFUHtTrfYVeSo7xYHyOKsQbRLxxiwu+BS+BedOMsAD8kxwLpPgyGvXj+p zKzdENQcgeqeHD4awbChMi8jDHdjHwTCn8dYL/bmMH4NHH0HJkjM2koR9iUa0PTjbBiV kbHgCTSbKGkPAbY34+1e6713tgSvq85r0ieZlvyZoNePa+yD5bhlvXSRgIVlshOgvhbg JiQg==
MIME-Version: 1.0
X-Received: by 10.60.155.135 with SMTP id vw7mr5533556oeb.62.1427246421769; Tue, 24 Mar 2015 18:20:21 -0700 (PDT)
Received: by 10.202.48.151 with HTTP; Tue, 24 Mar 2015 18:20:21 -0700 (PDT)
In-Reply-To: <BN3PR0301MB1250083227EBF9A1761B3A438C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
References: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com> <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com> <BN3PR0301MB1250E8DC62A65CDBA697288D8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com> <3C2FF331-C587-4BB2-9299-3A82FA3DE613@gmail.com> <BN3PR0301MB1250CC6E00173D055E1F942A8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com> <F989DCB4-0753-411C-99A2-ABA413DB4A3A@gmail.com> <BN3PR0301MB1250083227EBF9A1761B3A438C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
Date: Tue, 24 Mar 2015 20:20:21 -0500
Message-ID: <CABkgnnU+=rds4t3xFTOmXwg0ZN9s3GrzVx5j3HQa7psuObSeKQ@mail.gmail.com>
From: Martin Thomson <martin.thomson@gmail.com>
To: Andrei Popov <Andrei.Popov@microsoft.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/-5QQrKTTvsVR5m480_rSxMnIlwY>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Yoav Nir <ynir.ietf@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 01:20:23 -0000

On 24 March 2015 at 19:44, Andrei Popov <Andrei.Popov@microsoft.com> wrote:
> Thanks for the clarification Yoav; now I see that you're talking about a =
MITM that inspects application data, and is willing to collaborate with the=
 client in preventing token theft. Yes, this would work with a form of Toke=
n Binding protocol that does not use the tls_unique. Or the other options y=
ou've described.

Another method would be to have an explicit delegation from the client
to the MitM.  Which would allow clients to avoid being affected by
moves between networks that have the proxy or not.  What concerns me
there is the complexity that introduces, so...

> If this is an important enough scenario and folks are passionate about su=
pporting it, I'd prefer that it be documented in a separate I-D.

Yes, this seems best.


From nobody Tue Mar 24 20:43:22 2015
Return-Path: <wmills_92105@yahoo.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E83611ACD79 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 20:43:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.175
X-Spam-Level: 
X-Spam-Status: No, score=-1.175 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_ENVFROM_END_DIGIT=0.25, FREEMAIL_FROM=0.001, FREEMAIL_REPLYTO_END_DIGIT=0.25, HTML_MESSAGE=0.001, IP_NOT_FRIENDLY=0.334, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 GjAuw1KegW6G for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 20:43:19 -0700 (PDT)
Received: from nm49-vm2.bullet.mail.gq1.yahoo.com (nm49-vm2.bullet.mail.gq1.yahoo.com [67.195.87.232]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A811F1A876B for <unbearable@ietf.org>; Tue, 24 Mar 2015 20:43:19 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s2048; t=1427254999; bh=cwRnku5kC3N4YKPT9MbcPeOqHP0ItjB8+MmhEh505uA=; h=Date:From:Reply-To:To:Cc:In-Reply-To:References:Subject:From:Subject; b=SIvU7QcV3+K8PpLRmXYa6UtxsXuqFwU9l7oDxbCaLlIT+FOBxAlHhZhPPIe9Jz2kv4bxpPDqPd5iN96OWQsJjIsV1Ax1sc39GjcUaStJJin8bxx6BdkcNOIUfyeLTu8jhHMeuMgZpHlU5XAI7BN+bJsSvb8LF6O1vxqq9TroiIVSd8s5rMRmAvQi84lC4IJmpQwweYeNFeA/qg+fZH/4+dTFbcRz/ynZrNV5edNBRlMvoFPKpnxXGrwFYAwz/2rB7OukOu1QLlhLHmwvq4s5Qyujxal+5VkXl55APXqbzjfuxOsg9Os7Te9UxuwVRpfLqM0SJG4YPpNzJKIA+iQmOg==
Received: from [127.0.0.1] by nm49.bullet.mail.gq1.yahoo.com with NNFMP; 25 Mar 2015 03:43:19 -0000
Received: from [98.137.12.55] by nm49.bullet.mail.gq1.yahoo.com with NNFMP; 25 Mar 2015 03:40:28 -0000
Received: from [98.139.170.182] by tm15.bullet.mail.gq1.yahoo.com with NNFMP;  25 Mar 2015 03:40:25 -0000
Received: from [98.139.212.208] by tm25.bullet.mail.bf1.yahoo.com with NNFMP;  25 Mar 2015 03:40:25 -0000
Received: from [127.0.0.1] by omp1017.mail.bf1.yahoo.com with NNFMP; 25 Mar 2015 03:40:25 -0000
X-Yahoo-Newman-Property: ymail-4
X-Yahoo-Newman-Id: 352031.12899.bm@omp1017.mail.bf1.yahoo.com
X-YMail-OSG: uzB3hy4VM1mj6e8hqrG5npUOUL1xMnjXyBn.BHMy2N5xYwRSGWI8l3yVSsKpbyG yJVWUKmcBTLrydZ6QK1VRvMJxN_CbjvZ36VI_INA3ktmBy2x60saUBX9y4CMioCv.XS.JScM8YwG uKmd7zuDjMHnXKX9IUbTwDwKJnC0ibIfyB_A19cC4wNc6eQWTm2lMCtkHXcoyFeyeCwJnxWIQP3T gC_eg1ysjq08cfEepJYs.XHAzJKcAR8PFt.V._ADTN47xo30fN0PPhX_q3rCNpIqYmpIjPU1875N QfuQR_Gy7Qb2pdF5CmPot5xv9KSqCTjTiVHHOeeSl0sFw5jaJjkx8le_6SwCLHpu7EzOE5qIheQq brv8VjmzHtAz18yQbllkaORPR.df.6lvwb33UoGjEQZUSFcJ27AtJRJhIvdLISJkocVwH4wZVPOa jpSE3WxIXZ4eRX_mLXvGNZ8YhCR7KkB7iaPHQmy7A4_70Dm4k8qH1nqB_iTfIa3FPC6OiSsUWKnM G_VHg9fOm1xhMPFx79qnqOoDGsw--
Received: by 66.196.81.116; Wed, 25 Mar 2015 03:40:24 +0000 
Date: Wed, 25 Mar 2015 03:40:24 +0000 (UTC)
From: Bill Mills <wmills_92105@yahoo.com>
To: Martin Thomson <martin.thomson@gmail.com>,  Andrei Popov <Andrei.Popov@microsoft.com>
Message-ID: <1697090784.1242767.1427254824371.JavaMail.yahoo@mail.yahoo.com>
In-Reply-To: <CABkgnnU+=rds4t3xFTOmXwg0ZN9s3GrzVx5j3HQa7psuObSeKQ@mail.gmail.com>
References: <CABkgnnU+=rds4t3xFTOmXwg0ZN9s3GrzVx5j3HQa7psuObSeKQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;  boundary="----=_Part_1242766_737019033.1427254824366"
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/0itOLR7wCUw5xJMOF1keVtefk-g>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Yoav Nir <ynir.ietf@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: Bill Mills <wmills_92105@yahoo.com>
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 03:43:21 -0000

------=_Part_1242766_737019033.1427254824366
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

There are many possible TLS proxy types that can cause problems:
1) Local host based proxy like Fiddler2) Corporate egress proxy, almost the=
 same as above but not under user control3) Great FW of China with mandator=
y trusted certs installed on many machines.=C2=A0 Same class as 2 but more =
evil.4) TOR which we don't need to solve for.5) Akami type CDN proxies6) Lo=
ad balancer/accellerators

I think we can solve for #6, and perhaps #5, the rest are out.=20

-bill



     On Tuesday, March 24, 2015 6:23 PM, Martin Thomson <martin.thomson@gma=
il.com> wrote:
  =20

 On 24 March 2015 at 19:44, Andrei Popov <Andrei.Popov@microsoft.com> wrote=
:
> Thanks for the clarification Yoav; now I see that you're talking about a =
MITM that inspects application data, and is willing to collaborate with the=
 client in preventing token theft. Yes, this would work with a form of Toke=
n Binding protocol that does not use the tls_unique. Or the other options y=
ou've described.

Another method would be to have an explicit delegation from the client
to the MitM.=C2=A0 Which would allow clients to avoid being affected by
moves between networks that have the proxy or not.=C2=A0 What concerns me
there is the complexity that introduces, so...

> If this is an important enough scenario and folks are passionate about su=
pporting it, I'd prefer that it be documented in a separate I-D.

Yes, this seems best.

_______________________________________________
Unbearable mailing list
Unbearable@ietf.org
https://www.ietf.org/mailman/listinfo/unbearable


  
------=_Part_1242766_737019033.1427254824366
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<html><body><div style=3D"color:#000; background-color:#fff; font-family:He=
lveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif;fo=
nt-size:12px"><div dir=3D"ltr" id=3D"yui_3_16_0_1_1427230764497_40314">Ther=
e are many possible TLS proxy types that can cause problems:</div><div id=
=3D"yui_3_16_0_1_1427230764497_40313"><br></div><div dir=3D"ltr" id=3D"yui_=
3_16_0_1_1427230764497_39541">1) Local host based proxy like Fiddler</div><=
div id=3D"yui_3_16_0_1_1427230764497_41302" dir=3D"ltr">2) Corporate egress=
 proxy, almost the same as above but not under user control</div><div id=3D=
"yui_3_16_0_1_1427230764497_44300" dir=3D"ltr">3) Great FW of China with ma=
ndatory trusted certs installed on many machines.&nbsp; Same class as 2 but=
 more evil.</div><div id=3D"yui_3_16_0_1_1427230764497_48261" dir=3D"ltr">4=
) TOR which we don't need to solve for.</div><div id=3D"yui_3_16_0_1_142723=
0764497_50248" dir=3D"ltr">5) Akami type CDN proxies</div><div id=3D"yui_3_=
16_0_1_1427230764497_50249" dir=3D"ltr">6) Load balancer/accellerators<br><=
/div><div id=3D"yui_3_16_0_1_1427230764497_39540"><br><span></span></div><d=
iv>I think we can solve for #6, and perhaps #5, the rest are out. <br></div=
><div><br></div><div dir=3D"ltr">-bill<br></div><br><div class=3D"qtdSepara=
teBR"><br><br></div><div style=3D"display: block;" class=3D"yahoo_quoted"> =
<div style=3D"font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial,=
 Lucida Grande, sans-serif; font-size: 12px;"> <div style=3D"font-family: H=
elveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; =
font-size: 16px;"> <div dir=3D"ltr"> <font size=3D"2" face=3D"Arial"> On Tu=
esday, March 24, 2015 6:23 PM, Martin Thomson &lt;martin.thomson@gmail.com&=
gt; wrote:<br> </font> </div>  <br><br> <div class=3D"y_msg_container">On 2=
4 March 2015 at 19:44, Andrei Popov &lt;<a shape=3D"rect" ymailto=3D"mailto=
:Andrei.Popov@microsoft.com" href=3D"mailto:Andrei.Popov@microsoft.com">And=
rei.Popov@microsoft.com</a>&gt; wrote:<br clear=3D"none">&gt; Thanks for th=
e clarification Yoav; now I see that you're talking about a MITM that inspe=
cts application data, and is willing to collaborate with the client in prev=
enting token theft. Yes, this would work with a form of Token Binding proto=
col that does not use the tls_unique. Or the other options you've described=
.<br clear=3D"none"><br clear=3D"none">Another method would be to have an e=
xplicit delegation from the client<br clear=3D"none">to the MitM.&nbsp; Whi=
ch would allow clients to avoid being affected by<br clear=3D"none">moves b=
etween networks that have the proxy or not.&nbsp; What concerns me<br clear=
=3D"none">there is the complexity that introduces, so...<br clear=3D"none">=
<br clear=3D"none">&gt; If this is an important enough scenario and folks a=
re passionate about supporting it, I'd prefer that it be documented in a se=
parate I-D.<br clear=3D"none"><br clear=3D"none">Yes, this seems best.<div =
class=3D"yqt7700890581" id=3D"yqtfd17678"><br clear=3D"none"><br clear=3D"n=
one">_______________________________________________<br clear=3D"none">Unbe=
arable mailing list<br clear=3D"none"><a shape=3D"rect" ymailto=3D"mailto:U=
nbearable@ietf.org" href=3D"mailto:Unbearable@ietf.org">Unbearable@ietf.org=
</a><br clear=3D"none"><a shape=3D"rect" href=3D"https://www.ietf.org/mailm=
an/listinfo/unbearable" target=3D"_blank">https://www.ietf.org/mailman/list=
info/unbearable</a><br clear=3D"none"></div><br><br></div>  </div> </div>  =
</div></div></body></html>
------=_Part_1242766_737019033.1427254824366--


From nobody Tue Mar 24 21:19:43 2015
Return-Path: <wmills_92105@yahoo.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4E51B1ACD86 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 21:19:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 1.191
X-Spam-Level: *
X-Spam-Status: No, score=1.191 tagged_above=-999 required=5 tests=[BAYES_50=0.8, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_ENVFROM_END_DIGIT=0.25, FREEMAIL_FROM=0.001, FREEMAIL_REPLYTO_END_DIGIT=0.25, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 xVwvCOsWqDzH for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 21:19:42 -0700 (PDT)
Received: from nm8-vm0.bullet.mail.bf1.yahoo.com (nm8-vm0.bullet.mail.bf1.yahoo.com [98.139.213.95]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 7F4B31ACD7B for <unbearable@ietf.org>; Tue, 24 Mar 2015 21:19:42 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s2048; t=1427257181; bh=gjfA0opJHmDq3zy4kW/DGKesmK2Z0R7hizt4kIYND/8=; h=Date:From:Reply-To:To:Subject:From:Subject; b=Jfh7/pMAEdIveqGUtdT0iAPhyEafNayh8/flZxqeX6IyhOeSgBfKm9DNBhUROIlqhRGz0NLPmvESDJZclEjFm2NJC722aKI87/Br+St+V1WBHh1W+PfOmpwej+pMNCnyD6yWdG9hzjEGkh4Ayucp7R8OlSSLugL1EZ5wNol8QaO3KHbibCagaWXynDgidhpFUmP6/2RWQ5OsUSWMWBVdIseDSham0qrDjOUAX1qhMyD04URJORrZjomHUc3SclF7gYp7QGpcvigs4xllY8hfmrIbiv+cXTI4cdCGyJ9xWdz1LY+QpCIAdEGdiafOPWD7Jq9RPxv3r6ilgMseFFH0lA==
Received: from [98.139.170.178] by nm8.bullet.mail.bf1.yahoo.com with NNFMP; 25 Mar 2015 04:19:41 -0000
Received: from [98.139.212.197] by tm21.bullet.mail.bf1.yahoo.com with NNFMP;  25 Mar 2015 04:19:41 -0000
Received: from [127.0.0.1] by omp1006.mail.bf1.yahoo.com with NNFMP; 25 Mar 2015 04:19:41 -0000
X-Yahoo-Newman-Property: ymail-3
X-Yahoo-Newman-Id: 770188.54833.bm@omp1006.mail.bf1.yahoo.com
X-YMail-OSG: NJBFOW0VM1mxyw82rhF7ga1FdXrNgaa5E9K8jUmqcJcB3PTtc6fQs29xOvqzQph fxRz8NmXARcQY00LdpW8PkH96tiIawy93XfsaWLHjWCj.sx9CPSxcJv1yAn6uRxB5sc5zy1GBCll yDUWaZ6dpuYQxx9hyZY3QVlIdLXJqz_chc8GxKgNqUbWTM5msvIJ83HsJ6i_heW2NTtSxsL6XZYa Mxuk9GCK4k7vhibq83ZQHJuOatwY3hN.YX1FR3bRxqhS5oBA0.SVNiiU_C7x8wUBQXe5QAVLT0IX pecJusP92TYPy9Ym4DBl9DZQZnzEUHo4tleWx7PRLiI13Nrdm2WXODCMLxxgAV9jgPiWala74UR_ TPFtcdv2WJDk5icD4GpvRq_VhtaGr4CsG92j7e5pfHHMlpjG8MQIj7Z3XkE7whEP2gmQxGDF1rXb Z2j2fAUWagpsaq0CvPG4ssiCGhvhibTG38UBytZvT3kT_.u6rE2HRrUMiM1OuHlyh1VgIOoV8LXX palhsByUI.lEuh_MIM9bL1_b4JTCX7wGkGgw9xlTldTI-
Received: by 66.196.80.149; Wed, 25 Mar 2015 04:19:41 +0000 
Date: Wed, 25 Mar 2015 04:19:41 +0000 (UTC)
From: Bill Mills <wmills_92105@yahoo.com>
To: Tokbind WG <unbearable@ietf.org>
Message-ID: <1458551009.1219485.1427257181088.JavaMail.yahoo@mail.yahoo.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;  boundary="----=_Part_1219484_695399761.1427257181087"
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/zT1JNQ3nGDzaJhcZiQlFSDY1zgY>
Subject: [Unbearable] Auth in the HTTP server
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
Reply-To: Bill Mills <wmills_92105@yahoo.com>
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 04:19:43 -0000

------=_Part_1219484_695399761.1427257181087
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit

If the HTTP server is not the TLS server how does the tls-unique value get communicated to the HTTP server and how/when does the HTTP server validate the transaction?
Simple string comparison of a header inserted by the TLS proxy isn't as simple as it looks.
-bill
------=_Part_1219484_695399761.1427257181087
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 7bit

<html><body><div style="color:#000; background-color:#fff; font-family:HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif;font-size:12px"><div dir="ltr" id="yui_3_16_0_1_1427254277374_42718">If the HTTP server is not the TLS server how does the tls-unique value get communicated to the HTTP server and how/when does the HTTP server validate the transaction?</div><div dir="ltr" id="yui_3_16_0_1_1427254277374_42718"><br></div><div dir="ltr" id="yui_3_16_0_1_1427254277374_42718">Simple string comparison of a header inserted by the TLS proxy isn't as simple as it looks.</div><div dir="ltr" id="yui_3_16_0_1_1427254277374_42718"><br></div><div dir="ltr" id="yui_3_16_0_1_1427254277374_42718">-bill</div></div></body></html>
------=_Part_1219484_695399761.1427257181087--


From nobody Tue Mar 24 21:38:55 2015
Return-Path: <Andrei.Popov@microsoft.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BD15F1ACD94 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 21:38:53 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level: 
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
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 fN-v1EjS1vmY for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 21:38:51 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1bon0795.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::1:795]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A1DA31ACD8A for <unbearable@ietf.org>; Tue, 24 Mar 2015 21:38:51 -0700 (PDT)
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com (25.161.207.26) by BN3PR0301MB1250.namprd03.prod.outlook.com (25.161.207.26) with Microsoft SMTP Server (TLS) id 15.1.112.19; Wed, 25 Mar 2015 04:38:29 +0000
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) by BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) with mapi id 15.01.0112.000; Wed, 25 Mar 2015 04:38:29 +0000
From: Andrei Popov <Andrei.Popov@microsoft.com>
To: Bill Mills <wmills_92105@yahoo.com>, Tokbind WG <unbearable@ietf.org>
Thread-Topic: [Unbearable] Auth in the HTTP server
Thread-Index: AQHQZrLtakCOKJGaDkukCEAwTMxWr50smgww
Date: Wed, 25 Mar 2015 04:38:29 +0000
Message-ID: <BN3PR0301MB1250702BD4DE14A42D8567288C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
References: <1458551009.1219485.1427257181088.JavaMail.yahoo@mail.yahoo.com>
In-Reply-To: <1458551009.1219485.1427257181088.JavaMail.yahoo@mail.yahoo.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [24.173.67.179]
authentication-results: yahoo.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0301MB1250;
x-microsoft-antispam-prvs: <BN3PR0301MB12504CDCD4D1F14F614C56ED8C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(6009001)(377454003)(2950100001)(102836002)(86362001)(86612001)(77156002)(66066001)(76576001)(106116001)(2900100001)(76176999)(15975445007)(122556002)(99286002)(50986999)(107886001)(54356999)(551934003)(40100003)(33656002)(19625215002)(19580395003)(46102003)(2656002)(2521001)(74316001)(87936001)(16236675004)(19300405004)(19580405001)(92566002)(62966003); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0301MB1250; H:BN3PR0301MB1250.namprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5005006)(5002010); SRVR:BN3PR0301MB1250; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0301MB1250; 
x-forefront-prvs: 052670E5A4
Content-Type: multipart/alternative; boundary="_000_BN3PR0301MB1250702BD4DE14A42D8567288C0B0BN3PR0301MB1250_"
MIME-Version: 1.0
X-OriginatorOrg: microsoft.onmicrosoft.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 25 Mar 2015 04:38:29.7160 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0301MB1250
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/f1iyQTvatI7lUCQvS8IalY_gPZI>
Subject: Re: [Unbearable] Auth in the HTTP server
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 04:38:53 -0000

--_000_BN3PR0301MB1250702BD4DE14A42D8567288C0B0BN3PR0301MB1250_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BN3PR0301MB1250702BD4DE14A42D8567288C0B0BN3PR0301MB1250_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BN3PR0301MB1250702BD4DE14A42D8567288C0B0BN3PR0301MB1250_--


From nobody Tue Mar 24 21:49:05 2015
Return-Path: <martin.thomson@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E2E681ACD97 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 21:49:03 -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, SPF_PASS=-0.001] autolearn=ham
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 nvZgb3B-LcN6 for <unbearable@ietfa.amsl.com>; Tue, 24 Mar 2015 21:49:02 -0700 (PDT)
Received: from mail-oi0-x22d.google.com (mail-oi0-x22d.google.com [IPv6:2607:f8b0:4003:c06::22d]) (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 603DA1ACD95 for <unbearable@ietf.org>; Tue, 24 Mar 2015 21:49:02 -0700 (PDT)
Received: by oier21 with SMTP id r21so12310196oie.1 for <unbearable@ietf.org>; Tue, 24 Mar 2015 21:49:01 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=PorUFoEZ+/jPu3cUbto4rWwEuCwhqcmbON8n9ci4ObA=; b=j/pxV2jXPF5cro17nhE0jQWS9WAS5GBWjKRemRDaxqEDnkFK7yPPsrKaEX5JUbZOR4 y1P8+06t7QEKAGn4qFwZ6vCOqnW/GyGpNaYmOOAaEg9ZNO84w036G7eqcdPw83M+WxoJ /SsszszC3hs776vcQj2s8pe/mJZfOg4r++bITtgRNpPMsNmJHLB9aL8pk0BprED7LZeQ z5LeLzyp9T8N8yEqDu6TwIrxuK2Qmuq52xoSYt6ZcHsKSHUvtpXsDeLNIgKKnOFRkYEA vKCLXgOtYH3qJ1BE62Sdc/JzAL4Cmu+IgfFdLydhntJXfr/nBe4uKGEKXWy0Pjca2drw sl3w==
MIME-Version: 1.0
X-Received: by 10.182.34.131 with SMTP id z3mr6075592obi.4.1427258941916; Tue, 24 Mar 2015 21:49:01 -0700 (PDT)
Received: by 10.202.48.151 with HTTP; Tue, 24 Mar 2015 21:49:01 -0700 (PDT)
In-Reply-To: <1697090784.1242767.1427254824371.JavaMail.yahoo@mail.yahoo.com>
References: <CABkgnnU+=rds4t3xFTOmXwg0ZN9s3GrzVx5j3HQa7psuObSeKQ@mail.gmail.com> <1697090784.1242767.1427254824371.JavaMail.yahoo@mail.yahoo.com>
Date: Tue, 24 Mar 2015 23:49:01 -0500
Message-ID: <CABkgnnUcxa6SVyaCBtYByNDXy5Rtqpb0e89cUFWkPuup7PaLOw@mail.gmail.com>
From: Martin Thomson <martin.thomson@gmail.com>
To: Bill Mills <wmills_92105@yahoo.com>
Content-Type: text/plain; charset=UTF-8
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/V5N8iIKsuPJP3AEUhwp_zgQJM_I>
Cc: Andrei Popov <Andrei.Popov@microsoft.com>, "unbearable@ietf.org" <unbearable@ietf.org>, Yoav Nir <ynir.ietf@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 04:49:04 -0000

On 24 March 2015 at 22:40, Bill Mills <wmills_92105@yahoo.com> wrote:
> 4) TOR which we don't need to solve for.

TOR doesn't MitM TLS connections.

If someone wants to build a protocol that allows a server to pull the
necessary information from a trusted gateway, or for a client to
delegate to a trusted proxy, then that might be worthy of
consideration.  My preference would be to do a) end-to-end, then maybe
b) gateway communication, but not c) proxy impersonation assistance.


From nobody Wed Mar 25 01:17:42 2015
Return-Path: <bascule@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 05B4C1ACE12 for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 01:17:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 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, SPF_PASS=-0.001] autolearn=ham
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 zXZfaWLEQ4mg for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 01:17:32 -0700 (PDT)
Received: from mail-oi0-x22e.google.com (mail-oi0-x22e.google.com [IPv6:2607:f8b0:4003: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 CDD7D1ACE18 for <unbearable@ietf.org>; Wed, 25 Mar 2015 01:17:31 -0700 (PDT)
Received: by oier21 with SMTP id r21so15193263oie.1 for <unbearable@ietf.org>; Wed, 25 Mar 2015 01:17:31 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=1DCZchuNdMjxS8NXoYBxLg+GfXf47eRhflW1m58a1ns=; b=ZMtXJb7tyWNEnzPq/1QxSrT8jLUcqk2fe5U5bed5hytsKJ0uQPPzBKmzg5xcGKZz8y Q1TIWRxBEMRvXgZEQhOR217QpmoErQ6HwlnqEfwX5VMars7nVzGB94gyXW+wzRoC9/H7 9CcBjtQIr7lhngpwZfP39y1eMEu6IUgOeSRpwr9qV3F94oJ9b1QRQRS01AgiWS8hnn64 wYZkdhY95VHYPtwh2L75CXK5A/TyMF/LYVUc1hyD+eU1wVZURjp+9q3LkZyOr/NoEU/T 4p/h/VLdD/v6MbKTDPOPWRoUQQ2CSOmoiRfavhKPaPSbCT3151lGvP6aqgfgyDsqsWeU /plg==
X-Received: by 10.182.215.229 with SMTP id ol5mr6509334obc.49.1427271451118; Wed, 25 Mar 2015 01:17:31 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.76.166.232 with HTTP; Wed, 25 Mar 2015 01:17:10 -0700 (PDT)
In-Reply-To: <CABkgnnUcxa6SVyaCBtYByNDXy5Rtqpb0e89cUFWkPuup7PaLOw@mail.gmail.com>
References: <CABkgnnU+=rds4t3xFTOmXwg0ZN9s3GrzVx5j3HQa7psuObSeKQ@mail.gmail.com> <1697090784.1242767.1427254824371.JavaMail.yahoo@mail.yahoo.com> <CABkgnnUcxa6SVyaCBtYByNDXy5Rtqpb0e89cUFWkPuup7PaLOw@mail.gmail.com>
From: Tony Arcieri <bascule@gmail.com>
Date: Wed, 25 Mar 2015 01:17:10 -0700
Message-ID: <CAHOTMV+cPMFq+10LbCExMB-8hiSTqHdSDxwL=jF2-OrCokG+gA@mail.gmail.com>
To: Martin Thomson <martin.thomson@gmail.com>
Content-Type: multipart/alternative; boundary=001a11c25534d792a70512188708
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/3m1zCDlHNqdzrx-KjGWhhnYGkc4>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Andrei Popov <Andrei.Popov@microsoft.com>, Bill Mills <wmills_92105@yahoo.com>, Yoav Nir <ynir.ietf@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 08:17:37 -0000

--001a11c25534d792a70512188708
Content-Type: text/plain; charset=UTF-8

On Tue, Mar 24, 2015 at 9:49 PM, Martin Thomson <martin.thomson@gmail.com>
wrote:

> TOR doesn't MitM TLS connections.


Tor exit nodes are conveniently positioned to MitM traffic*

*That said, I am a fan of Tor**
**As a fan of Tor, I am rather dubious of the opinions people have about
Tor when they describe it as "TOR". It's not TOR, it's Tor***
***Also all of the discussions on this list scare me, but that's a
conversation for a different thread

-- 
Tony Arcieri

--001a11c25534d792a70512188708
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_extra"><div class=3D"gmail_quote">On T=
ue, Mar 24, 2015 at 9:49 PM, Martin Thomson <span dir=3D"ltr">&lt;<a href=
=3D"mailto:martin.thomson@gmail.com" target=3D"_blank" onclick=3D"window.op=
en(&#39;https://mail.google.com/mail/?view=3Dcm&amp;tf=3D1&amp;to=3Dmartin.=
thomson@gmail.com&amp;cc=3D&amp;bcc=3D&amp;su=3D&amp;body=3D&#39;,&#39;_bla=
nk&#39;);return false;">martin.thomson@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">TOR doesn&#39;t MitM TLS connections.</blockq=
uote><div><br></div><div>Tor exit nodes are conveniently positioned to MitM=
 traffic*</div><div><br></div><div>*That said, I am a fan of Tor**</div><di=
v>**As a fan of Tor, I am rather dubious of the opinions people have about =
Tor when they describe it as &quot;TOR&quot;. It&#39;s not TOR, it&#39;s To=
r***</div><div>***Also all of the discussions on this list scare me, but th=
at&#39;s a conversation for a different thread</div></div><div><br></div>--=
 <br><div class=3D"gmail_signature">Tony Arcieri<br></div>
</div></div>

--001a11c25534d792a70512188708--


From nobody Wed Mar 25 01:26:32 2015
Return-Path: <bascule@gmail.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5BA021ACE17 for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 01:26:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level: 
X-Spam-Status: No, score=-1.999 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, SPF_PASS=-0.001] autolearn=ham
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 sBPh9St1UraK for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 01:26:28 -0700 (PDT)
Received: from mail-oi0-x231.google.com (mail-oi0-x231.google.com [IPv6:2607:f8b0:4003:c06::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 4871F1ACE12 for <unbearable@ietf.org>; Wed, 25 Mar 2015 01:26:28 -0700 (PDT)
Received: by oiag65 with SMTP id g65so15280937oia.2 for <unbearable@ietf.org>; Wed, 25 Mar 2015 01:26:27 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=T05RG6GmzmYs7RyCHS8oSKAPGAqu2wAie8g3tGCtGaU=; b=PdIW59XnsdpkC0OdtOiyivoUqfbcJGjtk3Lc+3wmwuTvgYC+tzoksuU2R5oOQTYPKo 5/ktqQVeDyRBkUpsKJnDTEpcHrOUIpiV7FkAqEG1wh2fFlYWuwgGH2Nc7FiQtGW4sYWb XGPRJM+yG1RwWQsTgwXG4+7L08Y72cSL8HiRjrC8Vv3Ks/rqbzVJPj5vyVi+jb2py9ln CeJK5TKNBed4QiKYT8fOFTC06YC+26T91jY0VsK5HHyW00oGZkToj8iPksKrnU3lVzD7 xGWev2Mzvtn0HoKSzKL0I63g1L1Mnm+R5liQwqf+D5jOmjvRRIMADnvSGX03xsCAsb2b dhVQ==
X-Received: by 10.202.178.134 with SMTP id b128mr6127725oif.80.1427271987812;  Wed, 25 Mar 2015 01:26:27 -0700 (PDT)
MIME-Version: 1.0
Received: by 10.76.166.232 with HTTP; Wed, 25 Mar 2015 01:26:07 -0700 (PDT)
In-Reply-To: <BN3PR0301MB1250702BD4DE14A42D8567288C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
References: <1458551009.1219485.1427257181088.JavaMail.yahoo@mail.yahoo.com> <BN3PR0301MB1250702BD4DE14A42D8567288C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
From: Tony Arcieri <bascule@gmail.com>
Date: Wed, 25 Mar 2015 01:26:07 -0700
Message-ID: <CAHOTMVKvqwpXvUOG2RELZRoyG80tdEua3rghPgiJ-y-NcjnbaQ@mail.gmail.com>
To: Andrei Popov <Andrei.Popov@microsoft.com>
Content-Type: multipart/alternative; boundary=001a113b60fed4e7c5051218a7c2
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/sk4mN_UiFkcMjvbYiY48wPIg-3Q>
Cc: Tokbind WG <unbearable@ietf.org>, Bill Mills <wmills_92105@yahoo.com>
Subject: Re: [Unbearable] Auth in the HTTP server
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 08:26:30 -0000

--001a113b60fed4e7c5051218a7c2
Content-Type: text/plain; charset=UTF-8

I've still been too lazy to scrutinize this group's charter, but can
someone explain to me in a sentence or less why this group isn't massively
duplicating already deployed efforts like TLS channel ID? It seems like
these are solved problems that you're trying to throw ALPN at for dubious
gains.

On Tue, Mar 24, 2015 at 9:38 PM, Andrei Popov <Andrei.Popov@microsoft.com>
wrote:

>  There are existing mechanisms for HTTP servers to retrieve TLS client
> cert info, and in some implementations, the tls-unique is similarly
> available  to the HTTP stack. Where this functionality is missing, it can
> be added.
>
>
>
> There are two validation stages:
>
> 1.      Signature verification. This can be performed at the TLS layer or
> HTTP layer. In the former case, the Token Binding ID has to be plumbed
> through the HTTP stack to the server app. In the latter case, the
> tls-unique has to be provided by the TLS layer to the HTTP stack, and the
> Token Binding ID has to be provided by the HTTP stack to the server app.
> There may be other architectural options as well.
>
> 2.      Bound security token verification (comparison of Token Binding ID
> in the token with the one established for the connection). This can be
> performed by the server application (possibly, with the help of a framework
> such as ASP.NET).
>
>
>
> Cheers,
>
>
>
> Andrei
>
>
>
> *From:* Unbearable [mailto:unbearable-bounces@ietf.org] *On Behalf Of *Bill
> Mills
> *Sent:* Tuesday, March 24, 2015 11:20 PM
> *To:* Tokbind WG
> *Subject:* [Unbearable] Auth in the HTTP server
>
>
>
> If the HTTP server is not the TLS server how does the tls-unique value get
> communicated to the HTTP server and how/when does the HTTP server validate
> the transaction?
>
>
>
> Simple string comparison of a header inserted by the TLS proxy isn't as
> simple as it looks.
>
>
>
> -bill
>
> _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org
> https://www.ietf.org/mailman/listinfo/unbearable
>
>


-- 
Tony Arcieri

--001a113b60fed4e7c5051218a7c2
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">I&#39;ve still been too lazy to scrutinize this group&#39;=
s charter, but can someone explain to me in a sentence or less why this gro=
up isn&#39;t massively duplicating already deployed efforts like TLS channe=
l ID? It seems like these are solved problems that you&#39;re trying to thr=
ow ALPN at for dubious gains.</div><div class=3D"gmail_extra"><br><div clas=
s=3D"gmail_quote">On Tue, Mar 24, 2015 at 9:38 PM, Andrei Popov <span dir=
=3D"ltr">&lt;<a href=3D"mailto:Andrei.Popov@microsoft.com" target=3D"_blank=
">Andrei.Popov@microsoft.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">





<div lang=3D"EN-US" link=3D"#0563C1" vlink=3D"#954F72">
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">There are existing mechanisms for HTT=
P servers to retrieve TLS client cert info, and in some implementations, th=
e tls-unique is similarly available=C2=A0 to the HTTP
 stack. Where this functionality is missing, it can be added.<u></u><u></u>=
</span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d"><u></u>=C2=A0<u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">There are two validation stages:<u></=
u><u></u></span></p>
<p><u></u><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,s=
ans-serif;color:#1f497d"><span>1.<span style=3D"font:7.0pt &quot;Times New =
Roman&quot;">=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
</span></span></span><u></u><span style=3D"font-size:11.0pt;font-family:&qu=
ot;Calibri&quot;,sans-serif;color:#1f497d">Signature verification. This can=
 be performed at the TLS layer or HTTP layer. In the former case, the Token=
 Binding ID has to be plumbed through the
 HTTP stack to the server app. In the latter case, the tls-unique has to be=
 provided by the TLS layer to the HTTP stack, and the Token Binding ID has =
to be provided by the HTTP stack to the server app. There may be other arch=
itectural options as well.<u></u><u></u></span></p>
<p><u></u><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,s=
ans-serif;color:#1f497d"><span>2.<span style=3D"font:7.0pt &quot;Times New =
Roman&quot;">=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
</span></span></span><u></u><span style=3D"font-size:11.0pt;font-family:&qu=
ot;Calibri&quot;,sans-serif;color:#1f497d">Bound security token verificatio=
n (comparison of Token Binding ID in the token with the one established for=
 the connection). This can be performed by
 the server application (possibly, with the help of a framework such as <a =
href=3D"http://ASP.NET" target=3D"_blank">ASP.NET</a>).<u></u><u></u></span=
></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d"><u></u>=C2=A0<u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">Cheers,<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d"><u></u>=C2=A0<u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">Andrei<u></u><u></u></span></p>
<p class=3D"MsoNormal"><a name=3D"14c4f39e7e6a50ad__MailEndCompose"><span s=
tyle=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,sans-serif;color:#=
1f497d"><u></u>=C2=A0<u></u></span></a></p>
<div>
<div style=3D"border:none;border-top:solid #e1e1e1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif">From:</span></b><span style=3D"font-size:11.0pt;=
font-family:&quot;Calibri&quot;,sans-serif"> Unbearable [mailto:<a href=3D"=
mailto:unbearable-bounces@ietf.org" target=3D"_blank">unbearable-bounces@ie=
tf.org</a>]
<b>On Behalf Of </b>Bill Mills<br>
<b>Sent:</b> Tuesday, March 24, 2015 11:20 PM<br>
<b>To:</b> Tokbind WG<br>
<b>Subject:</b> [Unbearable] Auth in the HTTP server<u></u><u></u></span></=
p>
</div>
</div>
<p class=3D"MsoNormal"><u></u>=C2=A0<u></u></p>
<div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
9.0pt;font-family:&quot;Helvetica&quot;,sans-serif;color:black">If the HTTP=
 server is not the TLS server how does the tls-unique value get communicate=
d to the HTTP server and how/when does the HTTP
 server validate the transaction?<u></u><u></u></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
9.0pt;font-family:&quot;Helvetica&quot;,sans-serif;color:black"><u></u>=C2=
=A0<u></u></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
9.0pt;font-family:&quot;Helvetica&quot;,sans-serif;color:black">Simple stri=
ng comparison of a header inserted by the TLS proxy isn&#39;t as simple as =
it looks.<u></u><u></u></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
9.0pt;font-family:&quot;Helvetica&quot;,sans-serif;color:black"><u></u>=C2=
=A0<u></u></span></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white"><span style=3D"font-size:=
9.0pt;font-family:&quot;Helvetica&quot;,sans-serif;color:black">-bill<u></u=
><u></u></span></p>
</div>
</div>
</div>
</div>

<br>_______________________________________________<br>
Unbearable mailing list<br>
<a href=3D"mailto:Unbearable@ietf.org">Unbearable@ietf.org</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/unbearable" target=3D"_bla=
nk">https://www.ietf.org/mailman/listinfo/unbearable</a><br>
<br></blockquote></div><br><br clear=3D"all"><div><br></div>-- <br><div cla=
ss=3D"gmail_signature">Tony Arcieri<br></div>
</div>

--001a113b60fed4e7c5051218a7c2--


From nobody Wed Mar 25 04:30:33 2015
Return-Path: <ve7jtb@ve7jtb.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A05FA1A1F1D for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 04:30:31 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.601
X-Spam-Level: 
X-Spam-Status: No, score=-2.601 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham
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 W09E1Xw4pAHH for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 04:30:29 -0700 (PDT)
Received: from mail-wi0-f171.google.com (mail-wi0-f171.google.com [209.85.212.171]) (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 C2F431A1EEC for <unbearable@ietf.org>; Wed, 25 Mar 2015 04:30:28 -0700 (PDT)
Received: by wibgn9 with SMTP id gn9so33988763wib.1 for <unbearable@ietf.org>; Wed, 25 Mar 2015 04:30:27 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:content-type:mime-version:subject:from :in-reply-to:date:cc:message-id:references:to; bh=tah6NgX/+ri+8cXPN6I3ioEaG2WEOAqDrNKuSMKofng=; b=AF40puCVlvREHouZaoAYKoEsFOJjeK3dtRdEOIUSRawyQXW+V9tlLVjEieXCPNj39J PeF1HRx8bvdPrIK5aSe9C8TUQTn4fUtFPFi1n8RQCo9FFPJgIwMctc8IIYDkKUCECF7v 5o5Jt7PQfJtpSsdVehivIE87m2GBWFPQ5e9t5NyBKL/+TpljhqYQPqfidsx6npP1G3Bx dMlwW1GTQeMxHNv4OsG+PCyzFrs/Q1jZg9zAc3eq8UEZHY6Bp26O8uwKciMNaIps+tjA 6GbsrfqKyrZnm3vYjeKWch7jhxRLy50oxNezkSYrO0HBvup/+/XqI9YHgC5ki/naqQqp tO4g==
X-Gm-Message-State: ALoCoQkz+elWyKMWGKfcifm40UAKgDlaCG6CUCZMY+4JE53jXpOJ/NqCrTvsijr9ByHO59SQV4nv
X-Received: by 10.180.84.3 with SMTP id u3mr37463918wiy.38.1427283027363; Wed, 25 Mar 2015 04:30:27 -0700 (PDT)
Received: from dhcp-90f1.meeting.ietf.org (dhcp-90f1.meeting.ietf.org. [31.133.144.241]) by mx.google.com with ESMTPSA id fo9sm20182060wib.16.2015.03.25.04.30.24 (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Wed, 25 Mar 2015 04:30:26 -0700 (PDT)
Content-Type: multipart/signed; boundary="Apple-Mail=_F296EDB5-D392-457A-9A2D-947F2AFC8E6E"; protocol="application/pkcs7-signature"; micalg=sha1
Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2070.6\))
From: John Bradley <ve7jtb@ve7jtb.com>
In-Reply-To: <BN3PR0301MB1250083227EBF9A1761B3A438C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
Date: Wed, 25 Mar 2015 06:30:21 -0500
Message-Id: <14C18B6F-E75D-499B-ADF0-EE166F0C4A1C@ve7jtb.com>
References: <C3FDEA79-E6EA-4549-997C-804999A27CCD@gmail.com> <CABkgnnV86AtxE4jKgQZz-9FAuLeGxLZKPLx2XFZVYhPifx2rOg@mail.gmail.com> <BN3PR0301MB1250E8DC62A65CDBA697288D8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com> <3C2FF331-C587-4BB2-9299-3A82FA3DE613@gmail.com> <BN3PR0301MB1250CC6E00173D055E1F942A8C0A0@BN3PR0301MB1250.namprd03.prod.outlook.com> <F989DCB4-0753-411C-99A2-ABA413DB4A3A@gmail.com> <BN3PR0301MB1250083227EBF9A1761B3A438C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
To: Andrei Popov <Andrei.Popov@microsoft.com>
X-Mailer: Apple Mail (2.2070.6)
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/4xfOG3PK2PLSQ13U2wDFO0hdG5w>
Cc: "unbearable@ietf.org" <unbearable@ietf.org>, Martin Thomson <martin.thomson@gmail.com>, Yoav Nir <ynir.ietf@gmail.com>
Subject: Re: [Unbearable] Some further thoughts about TLS proxies and tokbind
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 11:30:31 -0000

--Apple-Mail=_F296EDB5-D392-457A-9A2D-947F2AFC8E6E
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8

My personal opinion is that guaranteeing end to end integrity and =
confidentiality of the channel between the endpoints is a feature of =
token binding.

Intermediate transport boxes that inspect the content should prevent =
token binding from being established.

At that point the server can decide if it should use unbound tokens or =
perhaps not allow the user to even enter credentials because they are =
being compromised.

I think trying to differentiate between "good" and "bad" Man in the =
middle is problematic, the bad people will always try to look like the =
good people.  =
http://www.motherjones.com/politics/2013/09/flying-pig-nsa-impersonates-go=
ogle

Don't get me wrong, I happen to think the NSA/GCHQ/CSE is doing a fine =
job, but they are not the only ones playing this game.

Having a way for a proxy that the end system knows about (perhaps not =
transparent) also proxy token binding may be reasonable.=20
Having a fallback that doesn't use tlsunique and by having a challenge =
can establish keys over a channel that is not end to end is I think =
almost a different protocol.  Perhaps worth considering separately, but =
not something that should be the default.

The trend in the IETF is to not have transparent interception and =
tampering with TLS connections.

The SPUD https://datatracker.ietf.org/meeting/92/agenda/spud/ session =
this morning is looking at some of these issues.

John B.

> On Mar 24, 2015, at 7:44 PM, Andrei Popov <Andrei.Popov@microsoft.com> =
wrote:
>=20
> Thanks for the clarification Yoav; now I see that you're talking about =
a MITM that inspects application data, and is willing to collaborate =
with the client in preventing token theft. Yes, this would work with a =
form of Token Binding protocol that does not use the tls_unique. Or the =
other options you've described.
>=20
> If this is an important enough scenario and folks are passionate about =
supporting it, I'd prefer that it be documented in a separate I-D.
>=20
> Cheers,
>=20
> Andrei
>=20
> -----Original Message-----
> From: Yoav Nir [mailto:ynir.ietf@gmail.com]=20
> Sent: Tuesday, March 24, 2015 7:08 PM
> To: Andrei Popov
> Cc: Martin Thomson; unbearable@ietf.org
> Subject: Re: [Unbearable] Some further thoughts about TLS proxies and =
tokbind
>=20
>=20
>> On Mar 24, 2015, at 6:39 PM, Andrei Popov =
<Andrei.Popov@microsoft.com> wrote:
>>=20
>>> When we were coming up with ideas in websec, one suggestion was for =
the client to sign the request (or rather a selection of the fields from =
the request that the proxy was not likely to mess with). That way you =
are on the one hand not bound to TLS, and on the other hand not making a =
bearer token, because a signature over =E2=80=9CGET / HTTP/1.1=E2=80=9D =
works only for requests getting =E2=80=9C/=E2=80=9C.
>>=20
>> But what prevents the MITM from removing the signature, or from =
signing with a key of MITM's choosing?
>>=20
>>> Another way is to add to HTTP (or whatever the protocol is) some way =
of getting a unique identifier for this connection and using that. For =
example, =E2=80=9CGET /.well-known/sessionid=E2=80=9D returning two =
values: identifier and challenge. After that the client would include =
both "Token-Binding: EncodedTokenBindingMessage=E2=80=9D and =
=E2=80=9CToken-Binding-Identifier: identifier=E2=80=9D in all subsequent =
requests.
>>=20
>> Why can't the MITM use the identifier and challenge in the same way?
>>=20
>> Or do you mean that these approaches will work until the MITM is =
updated to detect and circumvent these mechanisms?
>=20
> The MITM is not malicious. It=E2=80=99s performing either caching or =
content filtering or anti-malware. In a sense it is trusted by the user, =
because the user has installed the proxy=E2=80=99s CA certificate to =
make it work.
>=20
> If the UA is always behind the proxy, then the proxy can strip the =
client=E2=80=99s signature and sign with its own key, and then sign some =
fake requests with the same key. I don=E2=80=99t see how we can prevent =
that. But if the client is sometimes not behind the proxy, then there =
will be two keys and the UA won=E2=80=99t get the same authorizations =
behind the proxy as it gets when not behind the proxy, unless the proxy =
plays nice.  Perhaps the client should not generate the key pair until =
it has an unproxied connection.
>=20
> Yoav
>=20
> _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org
> https://www.ietf.org/mailman/listinfo/unbearable


--Apple-Mail=_F296EDB5-D392-457A-9A2D-947F2AFC8E6E
Content-Disposition: attachment;
	filename=smime.p7s
Content-Type: application/pkcs7-signature;
	name=smime.p7s
Content-Transfer-Encoding: base64
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--Apple-Mail=_F296EDB5-D392-457A-9A2D-947F2AFC8E6E--


From nobody Wed Mar 25 07:27:19 2015
Return-Path: <Andrei.Popov@microsoft.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A9F401A8033 for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 07:27:17 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.901
X-Spam-Level: 
X-Spam-Status: No, score=-1.901 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham
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 Ts-DB1NTCifN for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 07:27:14 -0700 (PDT)
Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1on0788.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::788]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8520D1A6FF9 for <unbearable@ietf.org>; Wed, 25 Mar 2015 07:27:14 -0700 (PDT)
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com (25.161.207.26) by BN3PR0301MB1251.namprd03.prod.outlook.com (25.161.207.27) with Microsoft SMTP Server (TLS) id 15.1.118.21; Wed, 25 Mar 2015 14:26:52 +0000
Received: from BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) by BN3PR0301MB1250.namprd03.prod.outlook.com ([25.161.207.26]) with mapi id 15.01.0112.000; Wed, 25 Mar 2015 14:26:52 +0000
From: Andrei Popov <Andrei.Popov@microsoft.com>
To: Tony Arcieri <bascule@gmail.com>
Thread-Topic: [Unbearable] Auth in the HTTP server
Thread-Index: AQHQZrLtakCOKJGaDkukCEAwTMxWr50smgwwgABDPICAAGLwoA==
Date: Wed, 25 Mar 2015 14:26:52 +0000
Message-ID: <BN3PR0301MB12509DE949E69291ADF8E80D8C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
References: <1458551009.1219485.1427257181088.JavaMail.yahoo@mail.yahoo.com> <BN3PR0301MB1250702BD4DE14A42D8567288C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com> <CAHOTMVKvqwpXvUOG2RELZRoyG80tdEua3rghPgiJ-y-NcjnbaQ@mail.gmail.com>
In-Reply-To: <CAHOTMVKvqwpXvUOG2RELZRoyG80tdEua3rghPgiJ-y-NcjnbaQ@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [69.15.241.124]
authentication-results: gmail.com; dkim=none (message not signed) header.d=none;
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:BN3PR0301MB1251;
x-forefront-antispam-report: BMV:1; SFV:NSPM; SFS:(10019020)(24454002)(377454003)(87936001)(66066001)(2656002)(19300405004)(551934003)(50986999)(76176999)(86362001)(19625215002)(33656002)(122556002)(40100003)(2900100001)(77156002)(19609705001)(46102003)(19617315012)(19580395003)(19580405001)(106116001)(99286002)(54356999)(16236675004)(62966003)(102836002)(2950100001)(92566002)(15975445007)(1411001)(16601075003)(74316001)(76576001)(86612001)(110136001); DIR:OUT; SFP:1102; SCL:1; SRVR:BN3PR0301MB1251; H:BN3PR0301MB1250.namprd03.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
x-microsoft-antispam-prvs: <BN3PR0301MB12518D686193B6DB43744B9F8C0B0@BN3PR0301MB1251.namprd03.prod.outlook.com>
x-exchange-antispam-report-test: UriScan:;
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(5002010)(5005006); SRVR:BN3PR0301MB1251; BCL:0; PCL:0; RULEID:; SRVR:BN3PR0301MB1251; 
x-forefront-prvs: 052670E5A4
Content-Type: multipart/alternative; boundary="_000_BN3PR0301MB12509DE949E69291ADF8E80D8C0B0BN3PR0301MB1250_"
MIME-Version: 1.0
X-OriginatorOrg: microsoft.onmicrosoft.com
X-MS-Exchange-CrossTenant-originalarrivaltime: 25 Mar 2015 14:26:52.2182 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN3PR0301MB1251
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/9AN2rTBqGNeuPzdPbjeVxFjbVz8>
Cc: Tokbind WG <unbearable@ietf.org>, Bill Mills <wmills_92105@yahoo.com>
Subject: Re: [Unbearable] Auth in the HTTP server
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 14:27:17 -0000

--_000_BN3PR0301MB12509DE949E69291ADF8E80D8C0B0BN3PR0301MB1250_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BN3PR0301MB12509DE949E69291ADF8E80D8C0B0BN3PR0301MB1250_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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==

--_000_BN3PR0301MB12509DE949E69291ADF8E80D8C0B0BN3PR0301MB1250_--


From nobody Wed Mar 25 08:26:00 2015
Return-Path: <balfanz@google.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 12D841A877A for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 08:25:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.388
X-Spam-Level: 
X-Spam-Status: No, score=-1.388 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 KNXmeNbPhuKF for <unbearable@ietfa.amsl.com>; Wed, 25 Mar 2015 08:25:56 -0700 (PDT)
Received: from mail-ie0-x236.google.com (mail-ie0-x236.google.com [IPv6:2607:f8b0:4001:c03::236]) (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 21B4F1A8761 for <unbearable@ietf.org>; Wed, 25 Mar 2015 08:25:56 -0700 (PDT)
Received: by iedm5 with SMTP id m5so24678442ied.3 for <unbearable@ietf.org>; Wed, 25 Mar 2015 08:25:55 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc:content-type; bh=RDRg6owF2sFhwyF7x3xaijPuqblcdr7FVbytfzwyrzg=; b=EpsIeFob4332rjNWQGeuqvFDNDTDjYe/yIMsgnUs74UuNcimsScIrx2aYHM5QwF18I yPlRJD3+fghk5TfWnawVZKWB4+swPu+CV/Qn+ZVNAs8KVnyMgxmu34NP1aLgBPjAyhfD fq5tgE35AWRL5eDmiVzBAE8jrKhcybA398Ua/oibT6bnyVqS6uy647PjDT+HIx0AOsDr ulAO/k7H4eOC4bSYItO1hgF2JD9Utoij1AEKblZcCJ1YalVEtVc1mkeM0aAAxddhOPe6 R8G+i9c2+6V2DIwRnYdT9IBheLbiucp6rn8OeAnvCihKmaK4K3FgMVveFYVnIYCQLHAO ZxtA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc:content-type; bh=RDRg6owF2sFhwyF7x3xaijPuqblcdr7FVbytfzwyrzg=; b=iIxqu8moNoYTajF1kT6U2fnfYonHbXIHPXMxesen1MiyWO/Hx1BcWYvpIFnQOLf0o9 1JcldlYbMT0o2vtBpalNgIg5/by+9nIZlcIWN9Rc+aI7svHGKdS4kOsQTjee28asvzhb mpfKwdq+p0gYytS3ENKQys4wMJnFLlfwqopCHzWOu/Z5TSDl2uaFwuagj/dHJMgHTcrl CAnSt7uQKTNzNoH06aUVnzXrjucXQ/OR+lvJxyqpCAjaOVg1+q4/VY83Czc7gGBi69BB g9cQy3Tl20sFQUPdf2dvvOfdmx0CC2ptcatl+ofTv8KtO1/BR6pYkXxQ6Ft+LP/vsqLw tt7A==
X-Gm-Message-State: ALoCoQnHbx+N+20uzDO9bj5hdy29E2IwOSnQInxr5Q5n10+B3rjn1JmPv+xKSY7ZEbjxHlOZrXHB
X-Received: by 10.42.68.204 with SMTP id y12mr32312133ici.84.1427297155377; Wed, 25 Mar 2015 08:25:55 -0700 (PDT)
MIME-Version: 1.0
References: <1458551009.1219485.1427257181088.JavaMail.yahoo@mail.yahoo.com> <BN3PR0301MB1250702BD4DE14A42D8567288C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com> <CAHOTMVKvqwpXvUOG2RELZRoyG80tdEua3rghPgiJ-y-NcjnbaQ@mail.gmail.com> <BN3PR0301MB12509DE949E69291ADF8E80D8C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
In-Reply-To: <BN3PR0301MB12509DE949E69291ADF8E80D8C0B0@BN3PR0301MB1250.namprd03.prod.outlook.com>
From: Dirk Balfanz <balfanz@google.com>
Date: Wed, 25 Mar 2015 15:25:53 +0000
Message-ID: <CADHfa2CcKtNcR-PoEbyQbu82NauDU3arCUeSYFnDfebcSVt+rw@mail.gmail.com>
To: Andrei Popov <Andrei.Popov@microsoft.com>, Tony Arcieri <bascule@gmail.com>
Content-Type: multipart/alternative; boundary=20cf303345f7ef9e3a05121e8342
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/-BexEb0ayIRx9sWS7O0hs1iOi6U>
Cc: Tokbind WG <unbearable@ietf.org>, Bill Mills <wmills_92105@yahoo.com>
Subject: Re: [Unbearable] Auth in the HTTP server
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 25 Mar 2015 15:25:58 -0000

--20cf303345f7ef9e3a05121e8342
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

To echo what Andrei said: we (Google) view Token Binding as the next step
in the Origin-Bound Certificates->Channel ID->Token Binding evolution.

Dirk.

On Wed, Mar 25, 2015 at 7:27 AM Andrei Popov <Andrei.Popov@microsoft.com>
wrote:

>  Hi Tony,
>
>
>
> It=E2=80=99s probably correct to think of this as the next incarnation of=
 Google=E2=80=99s
> Channel ID, with negotiable signature algorithms and key lengths, and (at
> least so far) not requiring TLS protocol changes. The use of ALPN is not
> really essential to Token Binding, and may change based on the discussion
> in this WG.
>
>
>
> Cheers,
>
>
>
> Andrei
>
>
>
> *From:* Tony Arcieri [mailto:bascule@gmail.com]
> *Sent:* Wednesday, March 25, 2015 3:26 AM
> *To:* Andrei Popov
> *Cc:* Bill Mills; Tokbind WG
> *Subject:* Re: [Unbearable] Auth in the HTTP server
>
>
>
> I've still been too lazy to scrutinize this group's charter, but can
> someone explain to me in a sentence or less why this group isn't massivel=
y
> duplicating already deployed efforts like TLS channel ID? It seems like
> these are solved problems that you're trying to throw ALPN at for dubious
> gains.
>
>
>
> On Tue, Mar 24, 2015 at 9:38 PM, Andrei Popov <Andrei.Popov@microsoft.com=
>
> wrote:
>
>  There are existing mechanisms for HTTP servers to retrieve TLS client
> cert info, and in some implementations, the tls-unique is similarly
> available  to the HTTP stack. Where this functionality is missing, it can
> be added.
>
>
>
> There are two validation stages:
>
> 1.      Signature verification. This can be performed at the TLS layer or
> HTTP layer. In the former case, the Token Binding ID has to be plumbed
> through the HTTP stack to the server app. In the latter case, the
> tls-unique has to be provided by the TLS layer to the HTTP stack, and the
> Token Binding ID has to be provided by the HTTP stack to the server app.
> There may be other architectural options as well.
>
> 2.      Bound security token verification (comparison of Token Binding ID
> in the token with the one established for the connection). This can be
> performed by the server application (possibly, with the help of a framewo=
rk
> such as ASP.NET).
>
>
>
> Cheers,
>
>
>
> Andrei
>
>
>
> *From:* Unbearable [mailto:unbearable-bounces@ietf.org] *On Behalf Of *Bi=
ll
> Mills
> *Sent:* Tuesday, March 24, 2015 11:20 PM
> *To:* Tokbind WG
> *Subject:* [Unbearable] Auth in the HTTP server
>
>
>
> If the HTTP server is not the TLS server how does the tls-unique value ge=
t
> communicated to the HTTP server and how/when does the HTTP server validat=
e
> the transaction?
>
>
>
> Simple string comparison of a header inserted by the TLS proxy isn't as
> simple as it looks.
>
>
>
> -bill
>
>
> _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org
> https://www.ietf.org/mailman/listinfo/unbearable
>
>
>
>
>
> --
>
> Tony Arcieri
>   _______________________________________________
> Unbearable mailing list
> Unbearable@ietf.org
> https://www.ietf.org/mailman/listinfo/unbearable
>

--20cf303345f7ef9e3a05121e8342
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">To echo what Andrei said: we (Google) view Token Binding a=
s the next step in the Origin-Bound Certificates-&gt;Channel ID-&gt;Token B=
inding evolution.<br><div><br></div><div>Dirk.</div></div><br><div class=3D=
"gmail_quote">On Wed, Mar 25, 2015 at 7:27 AM Andrei Popov &lt;<a href=3D"m=
ailto:Andrei.Popov@microsoft.com">Andrei.Popov@microsoft.com</a>&gt; wrote:=
<br><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-lef=
t:1px #ccc solid;padding-left:1ex">





<div lang=3D"EN-US" link=3D"blue" vlink=3D"purple">
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">Hi Tony,<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d"><u></u>=C2=A0<u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">It=E2=80=99s probably correct to thin=
k of this as the next incarnation of Google=E2=80=99s Channel ID, with nego=
tiable signature algorithms and key lengths, and (at least so
 far) not requiring TLS protocol changes. The use of ALPN is not really ess=
ential to Token Binding, and may change based on the discussion in this WG.=
<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d"><u></u>=C2=A0<u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">Cheers,<u></u><u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d"><u></u>=C2=A0<u></u></span></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">Andrei<u></u><u></u></span></p>
<p class=3D"MsoNormal"><a name=3D"msg-f:1496625856520325722__MailEndCompose=
"><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,sans-seri=
f;color:#1f497d"><u></u>=C2=A0<u></u></span></a></p>
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif">From:</span></b><span style=3D"font-size:11.0pt;=
font-family:&quot;Calibri&quot;,sans-serif"> Tony Arcieri [mailto:<a href=
=3D"mailto:bascule@gmail.com" target=3D"_blank">bascule@gmail.com</a>]
<br>
<b>Sent:</b> Wednesday, March 25, 2015 3:26 AM<br>
<b>To:</b> Andrei Popov<br>
<b>Cc:</b> Bill Mills; Tokbind WG<br>
<b>Subject:</b> Re: [Unbearable] Auth in the HTTP server<u></u><u></u></spa=
n></p></div></div><div lang=3D"EN-US" link=3D"blue" vlink=3D"purple"><div>
<p class=3D"MsoNormal"><u></u>=C2=A0<u></u></p>
<div>
<p class=3D"MsoNormal">I&#39;ve still been too lazy to scrutinize this grou=
p&#39;s charter, but can someone explain to me in a sentence or less why th=
is group isn&#39;t massively duplicating already deployed efforts like TLS =
channel ID? It seems like these are solved problems
 that you&#39;re trying to throw ALPN at for dubious gains.<u></u><u></u></=
p>
</div>
<div>
<p class=3D"MsoNormal"><u></u>=C2=A0<u></u></p>
<div>
<p class=3D"MsoNormal">On Tue, Mar 24, 2015 at 9:38 PM, Andrei Popov &lt;<a=
 href=3D"mailto:Andrei.Popov@microsoft.com" target=3D"_blank">Andrei.Popov@=
microsoft.com</a>&gt; wrote:<u></u><u></u></p>
<blockquote style=3D"border:none;border-left:solid #cccccc 1.0pt;padding:0i=
n 0in 0in 6.0pt;margin-left:4.8pt;margin-right:0in">
<div>
<div>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">There are existing mechanisms for HTT=
P servers to retrieve TLS client cert info, and in some implementations,
 the tls-unique is similarly available=C2=A0 to the HTTP stack. Where this =
functionality is missing, it can be added.</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">=C2=A0</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">There are two validation stages:</spa=
n><u></u><u></u></p>
<p><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,sans-ser=
if;color:#1f497d">1.</span><span style=3D"font-size:7.0pt;color:#1f497d">=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
</span><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,sans=
-serif;color:#1f497d">Signature verification. This can be performed at the =
TLS layer or HTTP layer. In the former case, the Token Binding ID has to be=
 plumbed through the HTTP stack to the server
 app. In the latter case, the tls-unique has to be provided by the TLS laye=
r to the HTTP stack, and the Token Binding ID has to be provided by the HTT=
P stack to the server app. There may be other architectural options as well=
.</span><u></u><u></u></p>
<p><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,sans-ser=
if;color:#1f497d">2.</span><span style=3D"font-size:7.0pt;color:#1f497d">=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
</span><span style=3D"font-size:11.0pt;font-family:&quot;Calibri&quot;,sans=
-serif;color:#1f497d">Bound security token verification (comparison of Toke=
n Binding ID in the token with the one established for the connection). Thi=
s can be performed by the server application
 (possibly, with the help of a framework such as <a href=3D"http://ASP.NET"=
 target=3D"_blank">
ASP.NET</a>).</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">=C2=A0</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">Cheers,</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">=C2=A0</span><u></u><u></u></p>
<p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;font-family:&quot;Ca=
libri&quot;,sans-serif;color:#1f497d">Andrei</span><u></u><u></u></p>
<p class=3D"MsoNormal"><a name=3D"msg-f:1496625856520325722_14c4f39e7e6a50a=
d__MailEndCompose"><span style=3D"font-size:11.0pt;font-family:&quot;Calibr=
i&quot;,sans-serif;color:#1f497d">=C2=A0</span></a><u></u><u></u></p>
<div>
<div style=3D"border:none;border-top:solid #e1e1e1 1.0pt;padding:3.0pt 0in =
0in 0in">
<p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:&quot=
;Calibri&quot;,sans-serif">From:</span></b><span style=3D"font-size:11.0pt;=
font-family:&quot;Calibri&quot;,sans-serif"> Unbearable [mailto:<a href=3D"=
mailto:unbearable-bounces@ietf.org" target=3D"_blank">unbearable-bounces@ie=
tf.org</a>]
<b>On Behalf Of </b>Bill Mills<br>
<b>Sent:</b> Tuesday, March 24, 2015 11:20 PM<br>
<b>To:</b> Tokbind WG<br>
<b>Subject:</b> [Unbearable] Auth in the HTTP server</span><u></u><u></u></=
p>
</div>
</div>
<p class=3D"MsoNormal">=C2=A0<u></u><u></u></p>
<div>
<div>
<p class=3D"MsoNormal" style=3D"background:white">
<span style=3D"font-size:9.0pt;font-family:&quot;Helvetica&quot;,sans-serif=
;color:black">If the HTTP server is not the TLS server how does the tls-uni=
que value get communicated to the HTTP server and how/when does the HTTP se=
rver validate the transaction?</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white">
<span style=3D"font-size:9.0pt;font-family:&quot;Helvetica&quot;,sans-serif=
;color:black">=C2=A0</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white">
<span style=3D"font-size:9.0pt;font-family:&quot;Helvetica&quot;,sans-serif=
;color:black">Simple string comparison of a header inserted by the TLS prox=
y isn&#39;t as simple as it looks.</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white">
<span style=3D"font-size:9.0pt;font-family:&quot;Helvetica&quot;,sans-serif=
;color:black">=C2=A0</span><u></u><u></u></p>
</div>
<div>
<p class=3D"MsoNormal" style=3D"background:white">
<span style=3D"font-size:9.0pt;font-family:&quot;Helvetica&quot;,sans-serif=
;color:black">-bill</span><u></u><u></u></p>
</div>
</div>
</div>
</div>
<p class=3D"MsoNormal" style=3D"margin-bottom:12.0pt"><br>
_______________________________________________<br>
Unbearable mailing list<br>
<a href=3D"mailto:Unbearable@ietf.org" target=3D"_blank">Unbearable@ietf.or=
g</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/unbearable" target=3D"_bla=
nk">https://www.ietf.org/mailman/listinfo/unbearable</a><u></u><u></u></p>
</blockquote>
</div>
<p class=3D"MsoNormal"><br>
<br clear=3D"all">
<u></u><u></u></p>
<div>
<p class=3D"MsoNormal"><u></u>=C2=A0<u></u></p>
</div>
<p class=3D"MsoNormal">-- <u></u><u></u></p>
<div>
<p class=3D"MsoNormal">Tony Arcieri<u></u><u></u></p>
</div>
</div>
</div></div>

______________________________<u></u>_________________<br>
Unbearable mailing list<br>
<a href=3D"mailto:Unbearable@ietf.org" target=3D"_blank">Unbearable@ietf.or=
g</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/unbearable" target=3D"_bla=
nk">https://www.ietf.org/mailman/<u></u>listinfo/unbearable</a><br>
</blockquote></div>

--20cf303345f7ef9e3a05121e8342--


From nobody Thu Mar 26 09:07:57 2015
Return-Path: <balfanz@google.com>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6E39E1A00F9 for <unbearable@ietfa.amsl.com>; Thu, 26 Mar 2015 09:07:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.388
X-Spam-Level: 
X-Spam-Status: No, score=-1.388 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=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 zBNH77_KR9NB for <unbearable@ietfa.amsl.com>; Thu, 26 Mar 2015 09:07:53 -0700 (PDT)
Received: from mail-ie0-x235.google.com (mail-ie0-x235.google.com [IPv6:2607:f8b0:4001:c03::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 1116F1A7034 for <unbearable@ietf.org>; Thu, 26 Mar 2015 09:07:52 -0700 (PDT)
Received: by iedfl3 with SMTP id fl3so56312431ied.1 for <unbearable@ietf.org>; Thu, 26 Mar 2015 09:07:51 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:from:date:message-id:subject:to:cc:content-type; bh=MMawJBRQfJ+gJhaZ9Pskvs1nWrfq2j7LzCaqe9SHIdI=; b=KJlCwrYoD9ta2WT7NmLXzV1lllt7NmchNE6I9Hdy6j7Plx5PO7h8ay0mWpocScd9wh +Xk5fWQ9qCPkgFpUKmTb62F3PeL9E0kwPoglHPGik981rVS0vbGUeQ5Vd/xBWLEb3KwA GrNydBqc8+pKHzRi8wuMOdsdH4jgBrBzljGa/X7rWqqaeOpvrFeGw6ckb1zz2yBLg88i p7eE4FPQ0xrYfmUjLY1AsvgWrqTiL98+mXfhBj8KJ2PWW/O7AowSmFrT2vUgIJhESi+f T2CLWeZfszeWPECZrxjJDunEBopf9KiIZOubLY372v9DGlTXUeY9W8MzgBllIWBswc1o M5UQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:from:date:message-id:subject:to:cc :content-type; bh=MMawJBRQfJ+gJhaZ9Pskvs1nWrfq2j7LzCaqe9SHIdI=; b=ehirhCI9b/Ik9RYsd/LOxdlcgMEjHEjJSvfuEC34HQL1Eh+qBjlsdAFnRD+E0At9b/ Nb+R8jgTMvM4A/3G+U792PXqriq4ndwTSAHJNwHDnXRzITkqykjc3/bs8591rCKDsKqe 7WfvjbpfY/vyTC5oBf/EpSfVrmVjhnTZEpxJGztSaK8AptqIpucJlNDBylgsNw+oMUoo FHUGAhby8NSQm7DzvAGg2DgqFcXN9+Ti5b1x17qlsXGlXwfx0rokpIGe16qKDuhBWNKH NJ0+P0RFlRmys3bPLUbBUd1qDLnlbaNbyX+ZnQiEQR+zz1CShnU9LHbJ9wvsZVKtGKGB 5ZfA==
X-Gm-Message-State: ALoCoQktN6e7umjVFRUa9eNvjc4JqLvynYG55HVmd8UH+yXSBKcs7gAmhgUs2603ya0eY5KSrHLV
X-Received: by 10.50.143.42 with SMTP id sb10mr37805611igb.49.1427386071478; Thu, 26 Mar 2015 09:07:51 -0700 (PDT)
MIME-Version: 1.0
From: Dirk Balfanz <balfanz@google.com>
Date: Thu, 26 Mar 2015 16:07:50 +0000
Message-ID: <CADHfa2AbQmwVVXF-qZPU_cxAuDEoG2+YEtwkm9vFf8td-7M6VA@mail.gmail.com>
To: public-webappsec@w3.org
Content-Type: multipart/alternative; boundary=001a1134b3cebfa6930512333716
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/ObOyP8ZkWBAh00ZMSWEHnty6hYo>
Cc: Tokbind WG <unbearable@ietf.org>, "wseltzer@w3.org" <wseltzer@w3.org>, Martin Thomson <mt@mozilla.com>, Andrei Popov <Andrei.Popov@microsoft.com>
Subject: [Unbearable] Update on IETF Token Binding WG / request for comments
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 16:07:54 -0000

--001a1134b3cebfa6930512333716
Content-Type: text/plain; charset=UTF-8

Hi there,

At IETF 92, we discussed the previously-mentioned
<https://lists.w3.org/Archives/Public/public-webappsec/2015Feb/0202.html> Token
Binding proposals, where they were officially adopted by the IETF Token
Binding WG.

One of the feedbacks we received there was that IETF I-Ds should stay
silent on new DOM APIs, and that those APIs should be developed in other
places. Currently, one of the proposed I-Ds
<http://tools.ietf.org/html/draft-balfanz-https-token-binding> mentions a
new DOM API, and I will remove that mention in the next revision.

Which then brings up the question of what to do about the DOM API, which
will no longer be mentioned in the I-D.

I was asked to gather comments and feedback from this group, and for that
purpose I wrote a doc that explains the problem and summarizes what the DOM
API would have to do:

https://docs.google.com/document/d/1x_3KI_v35NQFEsAbGGjNcyTw3Waoy1mL3YfTaQbyyuM/view?rm=minimal&pli=1

Comments/Feedback welcome, in particular comments on where this API should
be discussed/developed.

Dirk.

--001a1134b3cebfa6930512333716
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi there,=C2=A0</div><div><br></div><div>At IETF 92, =
we discussed the <a href=3D"https://lists.w3.org/Archives/Public/public-web=
appsec/2015Feb/0202.html">previously-mentioned</a>=C2=A0Token Binding propo=
sals, where they were officially adopted by the IETF Token Binding WG.=C2=
=A0</div><div><br></div><div>One of the feedbacks we received there was tha=
t IETF I-Ds should stay silent on new DOM APIs, and that those APIs should =
be developed in other places. Currently, <a href=3D"http://tools.ietf.org/h=
tml/draft-balfanz-https-token-binding">one of the proposed I-Ds</a> mention=
s a new DOM API, and I will remove that mention in the next revision.</div>=
<div><br></div><div>Which then brings up the question of what to do about t=
he DOM API, which will no longer be mentioned in the I-D.</div><div><br></d=
iv><div>I was asked to gather comments and feedback from this group, and fo=
r that purpose I wrote a doc that explains the problem and summarizes what =
the DOM API would have to do:</div><div><br></div><a href=3D"https://docs.g=
oogle.com/document/d/1x_3KI_v35NQFEsAbGGjNcyTw3Waoy1mL3YfTaQbyyuM/view?rm=
=3Dminimal&amp;pli=3D1">https://docs.google.com/document/d/1x_3KI_v35NQFEsA=
bGGjNcyTw3Waoy1mL3YfTaQbyyuM/view?rm=3Dminimal&amp;pli=3D1</a><br><div><br>=
</div><div>Comments/Feedback welcome, in particular comments on where this =
API should be discussed/developed.</div><div><br></div><div>Dirk.</div><div=
><br></div></div>

--001a1134b3cebfa6930512333716--


From nobody Thu Mar 26 11:20:49 2015
Return-Path: <annevk@annevk.nl>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 104BB1A8728 for <unbearable@ietfa.amsl.com>; Thu, 26 Mar 2015 09:38:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.378
X-Spam-Level: 
X-Spam-Status: No, score=-1.378 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_NONE=-0.0001] autolearn=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 C5Xj52_Y5j0S for <unbearable@ietfa.amsl.com>; Thu, 26 Mar 2015 09:38:52 -0700 (PDT)
Received: from homiemail-a60.g.dreamhost.com (homie.mail.dreamhost.com [208.97.132.208]) by ietfa.amsl.com (Postfix) with ESMTP id C3EAE1A7D80 for <unbearable@ietf.org>; Thu, 26 Mar 2015 09:38:52 -0700 (PDT)
Received: from homiemail-a60.g.dreamhost.com (localhost [127.0.0.1]) by homiemail-a60.g.dreamhost.com (Postfix) with ESMTP id 7F3E73BC06E for <unbearable@ietf.org>; Thu, 26 Mar 2015 09:38:52 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=annevk.nl; h=mime-version :in-reply-to:references:date:message-id:subject:from:to:cc: content-type; s=annevk.nl; bh=KTHJFZ1/Aydv/6oxR13HyWrmoIM=; b=sy EofiHyb8hTz9LfGi4k1kGtoTo9ITMyicGknl44nylj/avQ7Ls4z5rCXFVbqlkTBT f8Ie0QR/ApOkL/1IFrGPbHKCL+YRke1zNiU0DZhnpkI8eSwbfAldDiRTE8oMyxBs IWk/UncIBBqtTdCFyosCBmcDFaBT1j5tE2RRX0vk0=
Received: from mail-qg0-f49.google.com (mail-qg0-f49.google.com [209.85.192.49]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) (Authenticated sender: annevk@annevk.nl) by homiemail-a60.g.dreamhost.com (Postfix) with ESMTPSA id 6CDDE3BC063 for <unbearable@ietf.org>; Thu, 26 Mar 2015 09:38:52 -0700 (PDT)
Received: by qgh3 with SMTP id 3so90746094qgh.2 for <unbearable@ietf.org>; Thu, 26 Mar 2015 09:38:51 -0700 (PDT)
MIME-Version: 1.0
X-Received: by 10.55.50.203 with SMTP id y194mr31633052qky.8.1427387920029; Thu, 26 Mar 2015 09:38:40 -0700 (PDT)
Received: by 10.229.61.70 with HTTP; Thu, 26 Mar 2015 09:38:39 -0700 (PDT)
In-Reply-To: <CADHfa2AbQmwVVXF-qZPU_cxAuDEoG2+YEtwkm9vFf8td-7M6VA@mail.gmail.com>
References: <CADHfa2AbQmwVVXF-qZPU_cxAuDEoG2+YEtwkm9vFf8td-7M6VA@mail.gmail.com>
Date: Thu, 26 Mar 2015 17:38:39 +0100
Message-ID: <CADnb78hqCcX1K-nnHunfjT-JorGWQjGX5mjKiryWnCtiPPvPdw@mail.gmail.com>
From: Anne van Kesteren <annevk@annevk.nl>
To: Dirk Balfanz <balfanz@google.com>
Content-Type: text/plain; charset=UTF-8
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/b5aWcP6iW5dbaU-AJn8lCIng9Ps>
X-Mailman-Approved-At: Thu, 26 Mar 2015 11:20:47 -0700
Cc: "wseltzer@w3.org" <wseltzer@w3.org>, Tokbind WG <unbearable@ietf.org>, WebAppSec WG <public-webappsec@w3.org>, Martin Thomson <mt@mozilla.com>, Andrei Popov <Andrei.Popov@microsoft.com>
Subject: Re: [Unbearable] Update on IETF Token Binding WG / request for comments
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 26 Mar 2015 16:38:54 -0000

On Thu, Mar 26, 2015 at 5:07 PM, Dirk Balfanz <balfanz@google.com> wrote:
> I was asked to gather comments and feedback from this group, and for that
> purpose I wrote a doc that explains the problem and summarizes what the DOM
> API would have to do:
>
> https://docs.google.com/document/d/1x_3KI_v35NQFEsAbGGjNcyTw3Waoy1mL3YfTaQbyyuM/view?rm=minimal&pli=1
>
> Comments/Feedback welcome, in particular comments on where this API should
> be discussed/developed.

Thanks Dirk. https://github.com/whatwg/fetch/ would be the best place
to discuss and add this API. fetch() is the "new version" of
XMLHttpRequest and the API we'll actively maintain going forward. That
will also place the necessary hooks for this feature in the Fetch
Standard and would allow us to expose it in other contexts later, if
that is deemed necessary.


-- 
https://annevankesteren.nl/


From nobody Fri Mar 27 15:50:23 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4D9521A0195; Fri, 27 Mar 2015 15:29:17 -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
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 dj4kCI8eqaIv; Fri, 27 Mar 2015 15:29:16 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 312D41A00FF; Fri, 27 Mar 2015 15:29:16 -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>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.13.0
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150327222916.28150.75577.idtracker@ietfa.amsl.com>
Date: Fri, 27 Mar 2015 15:29:16 -0700
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/w35xc9CTKxrIhh7eUfM7IsUkerA>
X-Mailman-Approved-At: Fri, 27 Mar 2015 15:50:21 -0700
Cc: unbearable@ietf.org
Subject: [Unbearable] I-D Action: draft-ietf-tokbind-protocol-00.txt
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 22:29:17 -0000

A New Internet-Draft is available from the on-line Internet-Drafts directories.
 This draft is a work item of the Token Binding Working Group of the IETF.

        Title           : The Token Binding Protocol Version 1.0
        Authors         : Andrei Popov
                          Magnus Nystroem
                          Dirk Balfanz
                          Adam Langley
	Filename        : draft-ietf-tokbind-protocol-00.txt
	Pages           : 13
	Date            : 2015-03-27

Abstract:
   This document specifies Version 1.0 of the Token Binding protocol.
   The Token Binding protocol allows client/server applications to
   create long-lived, uniquely identifiable TLS [RFC5246] bindings
   spanning multiple TLS sessions and connections.  Applications are
   then enabled to cryptographically bind security tokens to the TLS
   layer, preventing token export and replay attacks.  To protect
   privacy, the TLS Token Binding identifiers are only transmitted
   encrypted and can be reset by the user at any time.


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

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-tokbind-protocol-00


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 Fri Mar 27 15:52:09 2015
Return-Path: <internet-drafts@ietf.org>
X-Original-To: unbearable@ietfa.amsl.com
Delivered-To: unbearable@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5E3321A0056; Fri, 27 Mar 2015 15:52:08 -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
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 hkldmjWWEi_S; Fri, 27 Mar 2015 15:52:05 -0700 (PDT)
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 5DA1D1A0027; Fri, 27 Mar 2015 15:52:05 -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>
X-Test-IDTracker: no
X-IETF-IDTracker: 5.13.0
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20150327225205.11852.95863.idtracker@ietfa.amsl.com>
Date: Fri, 27 Mar 2015 15:52:05 -0700
Archived-At: <http://mailarchive.ietf.org/arch/msg/unbearable/AlO3_Tyf1WJpnQIF_b0Lp9E27Zc>
Cc: unbearable@ietf.org
Subject: [Unbearable] I-D Action: draft-ietf-tokbind-https-00.txt
X-BeenThere: unbearable@ietf.org
X-Mailman-Version: 2.1.15
List-Id: "\"This list is for discussion of proposals for doing better than bearer tokens \(e.g. HTTP cookies, OAuth tokens etc.\) for web applications. The specific goal is chartering a WG focused on preventing security token export and replay attacks.\"" <unbearable.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/unbearable>, <mailto:unbearable-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/unbearable/>
List-Post: <mailto:unbearable@ietf.org>
List-Help: <mailto:unbearable-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/unbearable>, <mailto:unbearable-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 27 Mar 2015 22:52:08 -0000

A New Internet-Draft is available from the on-line Internet-Drafts directories.
 This draft is a work item of the Token Binding Working Group of the IETF.

        Title           : Token Binding over HTTP
        Authors         : Andrei Popov
                          Magnus Nystroem
                          Dirk Balfanz
                          Adam Langley
	Filename        : draft-ietf-tokbind-https-00.txt
	Pages           : 9
	Date            : 2015-03-27

Abstract:
   This document describes a collection of mechanisms that allow HTTP
   servers to cryptographically bind authentication tokens (such as
   cookies and OAuth tokens) to a TLS [RFC5246] connection.

   We describe both _first-party_ as well as _federated_ scenarios.  In
   a first-party scenario, an HTTP server issues a security token (such
   as a cookie) to a client, and expects the client to send the security
   token back to the server at a later time in order to authenticate.
   Binding the token to the TLS connection between client and server
   protects the security token from theft, and ensures that the security
   token can only be used by the client that it was issued to.

   Federated token bindings, on the other hand, allow servers to
   cryptographically bind security tokens to a TLS [RFC5246] connection
   that the client has with a _different_ server than the one issuing
   the token.

   This Internet-Draft is a companion document to The Token Binding
   Protocol [TBPROTO]


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

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-tokbind-https-00


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/

