From behave-bounces@ietf.org Mon Sep 03 02:18:23 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IS5GP-0001HR-4q; Mon, 03 Sep 2007 02:18:21 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IS5GO-0001HL-BK
	for behave-confirm+ok@megatron.ietf.org; Mon, 03 Sep 2007 02:18:20 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IS5GN-0001HD-W7
	for behave@ietf.org; Mon, 03 Sep 2007 02:18:19 -0400
Received: from smtp.nokia.com ([131.228.20.172] helo=mgw-ext13.nokia.com)
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IS5GM-0000pf-Fi
	for behave@ietf.org; Mon, 03 Sep 2007 02:18:19 -0400
Received: from esebh108.NOE.Nokia.com (esebh108.ntc.nokia.com [172.21.143.145])
	by mgw-ext13.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l836HRlk015995; Mon, 3 Sep 2007 09:18:03 +0300
Received: from esebh104.NOE.Nokia.com ([172.21.143.34]) by
	esebh108.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 3 Sep 2007 09:17:52 +0300
Received: from esebh101.NOE.Nokia.com ([172.21.138.177]) by
	esebh104.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 3 Sep 2007 09:17:51 +0300
Received: from esdhcp04177.research.nokia.com ([172.21.41.77]) by
	esebh101.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 3 Sep 2007 09:17:51 +0300
From: =?iso-8859-1?q?R=E9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: behave@ietf.org
Subject: Re: [BEHAVE] Comments on draft-ietf-behave-turn-04
Date: Mon, 3 Sep 2007 09:18:01 +0300
User-Agent: KMail/1.9.6 (enterprise 0.20070508.662491)
References: <84CA6498-1F9A-4AAF-A19E-6BDD647B55CE@estacado.net>
In-Reply-To: <84CA6498-1F9A-4AAF-A19E-6BDD647B55CE@estacado.net>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709030918.01870.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 03 Sep 2007 06:17:51.0707 (UTC)
	FILETIME=[2848B2B0:01C7EDF2]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 69a74e02bbee44ab4f8eafdbcedd94a1
Cc: 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Le Saturday 01 September 2007 01:38:53 ext Brian Hibbard, vous avez =E9crit=
=A0:
> TCP
> ------
(...)
> If the intent is that simultaneous open is the sole means by which
> allocations can connect to each other, I have a difficult time
> believing this is feasible.
(...)

The real problem is, will TCP-SO fly at all in the real Internet. My person=
al=20
bet is, the success rate will be way too low, so that ICE-TCP and friends=20
will be of limited practical usability.

(...)
> Therefore, I would be greatly relieved if multiple TCP connections on
> a single allocation were not supported.

I strongly agree.

> Set Active Destination
> ----------------------------------
> For UDP, it would be reasonable to suggest that the client start
> using Send indications shortly before (perhaps as long as one
> second?) sending a Set Active Destination request to clear the
> current active destination.  This would help the case that some of
> the packets sent prior to the SAD are processed by the server after
> the SAD is processed, due to packet ordering.

IMHO, a sane TURN client would only stop using "explicit" Send Indications=
=20
once and only once it has received a successful SAD Response from the serve=
r.=20
There is no need to specify an arbitrary latency delay here.

=2D-=20
R=E9mi Denis-Courmont


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 03 11:39:52 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ISE1m-0002od-FT; Mon, 03 Sep 2007 11:39:50 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ISE1l-0002oY-9D
	for behave-confirm+ok@megatron.ietf.org; Mon, 03 Sep 2007 11:39:49 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1ISE1k-0002lF-Rq
	for behave@ietf.org; Mon, 03 Sep 2007 11:39:48 -0400
Received: from mail-hub-1.cs.cornell.edu ([128.84.103.138]
	helo=exch-hub1.cs.cornell.edu)
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1ISE1j-0006j5-LB
	for behave@ietf.org; Mon, 03 Sep 2007 11:39:48 -0400
Received: from exchfe1.cs.cornell.edu (128.84.97.33) by
	mail-hub.cs.cornell.edu (128.84.103.140) with Microsoft SMTP Server id
	8.0.744.0; Mon, 3 Sep 2007 11:39:46 -0400
Received: from exchfe2.cs.cornell.edu ([128.84.97.34]) by
	exchfe1.cs.cornell.edu with Microsoft SMTPSVC(6.0.3790.3959); Mon, 3 Sep
	2007 11:39:41 -0400
Received: from [127.0.0.1] ([128.84.96.18]) by exchfe2.cs.cornell.edu over TLS
	secured channel with Microsoft SMTPSVC(6.0.3790.3959);
	Mon, 3 Sep 2007 11:39:41 -0400
Subject: Re: [BEHAVE] RE: ICE performance
From: Saikat Guha <saikat@cs.cornell.edu>
To: Philip Matthews <philip_matthews@magma.ca>
In-Reply-To: <A56D7804-00EC-4D09-9B69-D73F27EBA2EE@magma.ca>
References: <049501c7c89d$d34da410$7d646b80@amer.cisco.com>
	<A56D7804-00EC-4D09-9B69-D73F27EBA2EE@magma.ca>
Date: Mon, 3 Sep 2007 11:37:55 -0400
Message-ID: <1188833875.3800.10.camel@seshat.guha.cc>
MIME-Version: 1.0
X-Mailer: Evolution 2.10.3 (2.10.3-2.fc7) 
X-OriginalArrivalTime: 03 Sep 2007 15:39:41.0604 (UTC)
	FILETIME=[A4F2C640:01C7EE40]
Received-SPF: None (mail-hub.cs.cornell.edu: saikat@cs.cornell.edu does not
	designate permitted sender hosts)
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 9ed51c9d1356100bce94f1ae4ec616a9
Cc: Behave WG <behave@ietf.org>, Dan Wing <dwing@cisco.com>,
	Bryan Ford <baford@mit.edu>, Dan Kegel <dank@kegel.com>,
	Sinnreich <hsinnrei@adobe.com>, Henry, henrys@adobe.com
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Content-Type: multipart/mixed; boundary="===============2046291714=="
Errors-To: behave-bounces@ietf.org

--===============2046291714==
Content-Type: multipart/signed; micalg=pgp-sha1;
	protocol="application/pgp-signature"; boundary="=-hs9yZyHqKYUwiIlnufSc"

--=-hs9yZyHqKYUwiIlnufSc
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Sorry for the late response (didn't see this before; thanks Philip for
the reminder)

On Thu, 2007-07-19 at 09:05 -0400, Philip Matthews wrote:
> http://www.guha.cc/saikat/pub/imc05-tcpnat.pdf
>=20
> However, I think his setup was more of
> a test setup rather than a production setup.
>=20
> Perhaps Saikat can tell us more.

We found port prediction worked pretty well in home networks where the
flows/sec through the NAT is a low number.=20

Even then, in one case, a virus/worm on the internal host was SYN
flooding making it impossible to predicts ports accurately. Same is the
case for P2P applications that establish new connections rapidly.
(Although, if there is only one P2P app and it keeps track of the SYNs
it sends out, it can do port pred.)

As for enterprise networks, as Cullen mentioned, 1k f/s would make port
prediction impossible.

--=20
Saikat

--=-hs9yZyHqKYUwiIlnufSc
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)

iD8DBQBG3CpTnFltqi691/oRAtJoAJ94m5IOjBSpXookUSV/2k+Pm56FpwCfWNT7
2xCdV+KG1BcDSyZn1M8FBDE=
=7QpN
-----END PGP SIGNATURE-----

--=-hs9yZyHqKYUwiIlnufSc--



--===============2046291714==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

--===============2046291714==--





From behave-bounces@ietf.org Tue Sep 04 07:18:57 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ISWQn-0004mX-C5; Tue, 04 Sep 2007 07:18:53 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ISWQl-0004kl-R4
	for behave-confirm+ok@megatron.ietf.org; Tue, 04 Sep 2007 07:18:51 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1ISWQl-0004kR-Fm
	for behave@ietf.org; Tue, 04 Sep 2007 07:18:51 -0400
Received: from mailgw3.ericsson.se ([193.180.251.60])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1ISWQk-0005Q1-Fi
	for behave@ietf.org; Tue, 04 Sep 2007 07:18:51 -0400
Received: from mailgw3.ericsson.se (unknown [127.0.0.1])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	90218205F3; Tue,  4 Sep 2007 13:18:49 +0200 (CEST)
X-AuditID: c1b4fb3c-b1681bb0000007e1-f5-46dd3f194267
Received: from esealmw126.eemea.ericsson.se (unknown [153.88.254.123])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	63BF520451; Tue,  4 Sep 2007 13:18:49 +0200 (CEST)
Received: from esealmw126.eemea.ericsson.se ([153.88.254.174]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 13:18:48 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 13:18:48 +0200
Message-ID: <46DD3F15.2030808@ericsson.com>
Date: Tue, 04 Sep 2007 13:18:46 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Pyda Srisuresh <srisuresh@yahoo.com>, Bryan Ford <baford@mit.edu>, 
	Dan Kegel <dank@kegel.com>, Behave WG <behave@ietf.org>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 04 Sep 2007 11:18:48.0717 (UTC)
	FILETIME=[5D8383D0:01C7EEE5]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 6ba8aaf827dcb437101951262f69b3de
Cc: 
Subject: [BEHAVE] Ad comments on draft-ietf-behave-p2p-state-03
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Hi,

I have done my AD review of this document and have the below comments.
If you don't agree with them, please discuss them with me.

1. Section 1, page 3, first paragraph:
"These issues are
   likely to persist even into the IPv6 world, where a NAT may be used
   as an IPv4 compatibility mechanism [NAT-PT]."

I think you should rewrite this considering that NAT-PT was deprecated.

2. Section 1, page 3, second paragraph:

"The
   anonymity and inaccessibility of the internal hosts behind a NAT
   device is not a problem for applications such as web browsers, which
   only need to initiate outgoing connections. This inaccessibility is
   sometimes perceived as a privacy benefit."

Is really "privacy" the main benefit. Isn't it the filtering behavior of
the NAT that blocks incoming requests and thus avoid getting the traffic
onto your local network that people like, i.e. the "firewalling" aspect
of NATs.

3. Section 1, page 3, second to last paragraph:

"  NAT traversal strategies that involve explicit signaling between
   applications and  NAT devices, namely [NAT-PMP], [NSIS-NSLP],
   [SOCKS], [RSIP], [MIDCOM], and [UPNP] are out of the scope of this
   document. [UNSAF] is in scope."

I agree that these are not in scope, but I think there should be some
mention that these techiques if available are a complement to the
techniques mentioned, allowing a client behind a NAT behave as if it was
having a public IP address, or at least an NAT external address.

4. Section 2, first paragraph:
"As
   for the notation for ports usage, all clients use ports in the
   range of 1-1000 and servers use ports in the range of
   20000-21000. NAT devices use ports 30000 and above for endpoint
   mapping."

I think there are an error in this description. To me it appears that
clients are using addresses in the range 1025-2000 in the examples.

5. Section 2.7:

  "A P2P application is an application that uses the same endpoint to
   initiate outgoing sessions to peering hosts as well as accept
   incoming sessions from peering hosts."

Considering the definition of "endpoint" in this document I don't think
this definition is characteristic for peer to peer applications. An peer
to peer application clearly can use different address and ports to
communicate with other peers.

6. Section 2.8:
"  A NAT-friendly P2P application registers with a publicly
   addressable rendezvous server, used for registration and peer
   discovery purposes. Pursuant to registering with rendezvous
   server, the application uses its private endpoint, public
   endpoint, or a combination thereof to establish peering sessions."

I don't think this is a strict requierment for a NAT-friendly P2P
application. It is rather an effect of being one of the simplest way of
accomplishing the rendezvous between two peers.

7. Section 2.9. I think one should be explicit that an EIM-NAT can have
any filtering behavior.

8. Section 3.1, last paragraph:

  "The obvious disadvantages of
   relaying are that it consumes the server's processing power and
   network bandwidth, and communication latency between the peering
   clients is likely to be increased even if the server is
   well-connected."

I think "well-connected" is the wrong word here. I think it is rather
topology wise correctly located.

9. Section 3.1, last paragraph:

"The TURN protocol [TURN] defines a method of
   implementing relaying in a relatively secure fashion."

I think there should be some expansion on what type of relay service
that TURN provides. For certain relay services a more application aware
relay may be the better choice rather then packet payload relaying.

10. Section 3.2, page 10:

"In the first case, traffic directed to IP address
   10.0.0.1 will simply be dropped by the network because 10.0.0.1 is
   not a publicly routable IP address."

I think, this sentence needs to be expanded to cover the issue that the
traffic might hit another host and any application listening on this port.

11. Section 3.3 or 3.3.1:

Somewhere here there need to be a description on why this fails with non
EIM-NATs.

12. Section 3.3.1:

"The application does
   not need to attempt to detect the kind of NAT device it is behind,
   as in [STUN], since the procedure above will establish peer-to-peer
   communication channels equally well if either or both clients do not
   happen to be behind a NAT device."

Well, in most usages of STUN you don't need to detect the bevior of the
NAT. I assume that you refer to the model of  using STUN to find the
external address and port and then advertise it. Not knowing the bevior
of the NAT only results in one knowing if it works or not. I think the
wording is a bit missfortunate, as it seems to say that STUN is really
bad, when it is only a specific usage that is problematic.

13. Section 3.3.2, Figure 4:

I think the example is strange. What I can see if A and B sends its
request starting with Registry, then try1 and then try2, try1 should
succed. Also these messages are not well connected to the text. I think
you should swap the order and then reference them a bit better in the text.

14. Section 3.3.2, last paragraph:
  "[ICE] protocol employs this technique effectively, in that
   multiple candidate endpoints (both private and public) are
   communicated between peering end hosts during offer/response
   exchange."

I think you should use either offer/answer or request/response, not mix it.

15. section 5.1, second paragraph:

"When the NAT device
   enroute is not EIM-NAT, the application is unable to reuse an
   already established endpoint mapping for communication with
   different external destinations and the technique would fail."

I think "enroute" is the wrong word. It is the outermost NAT that isn't
an EIM-NAT for it to matter.

Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 04 08:28:34 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ISXWD-0003a0-HL; Tue, 04 Sep 2007 08:28:33 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ISXWC-0003Va-Lj
	for behave-confirm+ok@megatron.ietf.org; Tue, 04 Sep 2007 08:28:32 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1ISXWC-0003VR-Al
	for behave@ietf.org; Tue, 04 Sep 2007 08:28:32 -0400
Received: from smtp.nokia.com ([131.228.20.173] helo=mgw-ext14.nokia.com)
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1ISXWB-0007Zi-O3
	for behave@ietf.org; Tue, 04 Sep 2007 08:28:32 -0400
Received: from esebh106.NOE.Nokia.com (esebh106.ntc.nokia.com [172.21.138.213])
	by mgw-ext14.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l84CSPQ1029951 for <behave@ietf.org>; Tue, 4 Sep 2007 15:28:29 +0300
Received: from esebh103.NOE.Nokia.com ([172.21.143.33]) by
	esebh106.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 15:28:16 +0300
Received: from esebh102.NOE.Nokia.com ([172.21.138.183]) by
	esebh103.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 15:28:16 +0300
Received: from essapo-nirac253122.europe.nokia.com ([10.162.253.122]) by
	esebh102.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 15:28:15 +0300
From: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: behave@ietf.org
Subject: Re: [BEHAVE] Ad comments on draft-ietf-behave-p2p-state-03
Date: Tue, 4 Sep 2007 15:28:13 +0300
User-Agent: KMail/1.9.6 (enterprise 0.20070824.704228)
References: <46DD3F15.2030808@ericsson.com>
In-Reply-To: <46DD3F15.2030808@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709041528.13613.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 04 Sep 2007 12:28:15.0928 (UTC)
	FILETIME=[115D7380:01C7EEEF]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: c1c65599517f9ac32519d043c37c5336
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Le Tuesday 04 September 2007 14:18:46 ext Magnus Westerlund, vous avez =C3=
=A9crit=C2=A0:
> 1. Section 1, page 3, first paragraph:
> "These issues are
>    likely to persist even into the IPv6 world, where a NAT may be used
>    as an IPv4 compatibility mechanism [NAT-PT]."
>
> I think you should rewrite this considering that NAT-PT was deprecated.

I suppose some of "these issues" would still apply in presence of stateful=
=20
firewall, which almost everybody in Chicago seemed to agree would be rampan=
t=20
in IPv6. I agree that the use of NAT-PT as an illustration is unfortunate=20
though.

> 2. Section 1, page 3, second paragraph:
>
> "The
>    anonymity and inaccessibility of the internal hosts behind a NAT
>    device is not a problem for applications such as web browsers, which
>    only need to initiate outgoing connections. This inaccessibility is
>    sometimes perceived as a privacy benefit."
>
> Is really "privacy" the main benefit. Isn't it the filtering behavior of
> the NAT that blocks incoming requests and thus avoid getting the traffic
> onto your local network that people like, i.e. the "firewalling" aspect
> of NATs.

Agreed (though I would argue that a NAT by itself does NOT block incoming=20
requests - especially if it supports source route).

(...)
> 15. section 5.1, second paragraph:
>
> "When the NAT device
>    enroute is not EIM-NAT, the application is unable to reuse an
>    already established endpoint mapping for communication with
>    different external destinations and the technique would fail."
>
> I think "enroute" is the wrong word. It is the outermost NAT that isn't
> an EIM-NAT for it to matter.

I think any non-EIM NAT en-route would break, not only the outermost one.

=2D-=20
R=C3=A9mi Denis-Courmont


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 04 09:37:53 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ISYbI-0003WE-AF; Tue, 04 Sep 2007 09:37:52 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ISYbH-0003UK-Ht
	for behave-confirm+ok@megatron.ietf.org; Tue, 04 Sep 2007 09:37:51 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1ISYbH-0003UC-77
	for behave@ietf.org; Tue, 04 Sep 2007 09:37:51 -0400
Received: from mailgw3.ericsson.se ([193.180.251.60])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1ISYbG-00016a-EH
	for behave@ietf.org; Tue, 04 Sep 2007 09:37:51 -0400
Received: from mailgw3.ericsson.se (unknown [127.0.0.1])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	9089A2055B; Tue,  4 Sep 2007 15:37:49 +0200 (CEST)
X-AuditID: c1b4fb3c-b0e80bb0000007e1-b9-46dd5fad09f3
Received: from esealmw126.eemea.ericsson.se (unknown [153.88.254.123])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	7021420063; Tue,  4 Sep 2007 15:37:49 +0200 (CEST)
Received: from esealmw126.eemea.ericsson.se ([153.88.254.174]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 15:37:49 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 15:37:48 +0200
Message-ID: <46DD5FAB.5070607@ericsson.com>
Date: Tue, 04 Sep 2007 15:37:47 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Pyda Srisuresh <srisuresh@yahoo.com>, Bryan Ford <baford@mit.edu>, 
	Senthil Sivakumar <ssenthil@cisco.com>,
	Saikat Guha <saikat@cs.cornell.edu>, Behave WG <behave@ietf.org>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 04 Sep 2007 13:37:49.0050 (UTC)
	FILETIME=[C8BD55A0:01C7EEF8]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: cf3becbbd6d1a45acbe2ffd4ab88bdc2
Cc: 
Subject: [BEHAVE] AD comments on draft-ietf-behave-icmp-04
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Hi,

I have now done my AD review. Here are my comments.

1. Section 1, third paragraph:

"Traditional NAT inherently mandates a certain level
   of firewall-like functionality. However, firewall functionality in
   general or any other middlebox functionality is out of the
   scope of this document."

I assume that the firewall-like behavior is the filtering behavior. I
think there might be some benefit to express it that way and to refer to
the terms in UDP NAT RFC.

2. Section 2, last sentence:

"A NAT MAY drop
   or appropriately handle a post-1122 ICMP messages. "

I think a normative statement of this type do not belong in the
terminology section. Please find a better place to locate it.

3. Section 3.2, REQ-2:

  "Setting the ICMP NAT session timeout to a very large duration (say,
   240 seconds) could potentially tie up precious NAT resources such as
   query mappings and NAT Sessions for the whole duration. On the other
   hand, setting the timeout very low can result in premature freeing
   of NAT resources and applications failing to complete gracefully.
   The ICMP Query session timeout needs to be a balance between the two
   extremes. 60 seconds timeout is a balance between the two extremes.
   ICMP query session timer MUST not expire in less than 60 seconds. We
   RECOMMEND however that the administrator(s) be allowed to configure
   the timer.

   REQ-2: An ICMP Query session timer MUST NOT expire in less than 60
   seconds.
   a) It is RECOMMENDED that the ICMP Query session timer be made
   configurable."

Once more we use NATs as a motivation to redefine what our constants
are. I can't understand the motivation behind making the requirement to
anything other than 2xMSL, i.e. 240 s. I fear that people will start use
lower values for MSL, due to the fact that they know that if there are a
NAT there is anyway no possibility for the packet to live. Yes, it is a
long time, but MSL exist for being a real outer boundary for how long
packets can live.

If a NAT get too many querys aren't there better ways to deal with that?
After all we are talking about requests comming from the internal side
of the NAT. Are there ways of providing error response to query style ICMP?

4. Section 4.1,
"This
   requirement is meant primarily for future applications. Current
   applications may not be checking for mismatched checksum."

I don't find these two sentences providing much benefit. I think it is
well motivated that payload validation checking is supported for current
and any future applications.

5. Section 4.2, first paragraph:
"Section 4.3 of RFC 3022 describes the fields of an ICMP error
   message that a NAT device translates. In this section, we describe
   the requirements a NAT device MUST conform to while performing the
   translations. Requirements identified in this section are necessary
   for the current applications to work correctly."

I interpret this as an normative reference. This would be a normative
reference from a BCP to a informative document, which I like to avoid.
Can we please rework this, for example by including how the translation
is done in this document instead.

6. Section 4.2.2, page 12, second paragraph:

   In addition, if the ICMP Error payload contains ICMP extensions
   ([ICMP-EXT]), the NATs are encouraged to support ICMP extension
   objects.

I think this sentence is strangely formulated. Please reword it.


Please comment if there are any unclarities or disagrement. But I do
expect a revision.

Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 04 09:45:53 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ISYj2-0003P2-LH; Tue, 04 Sep 2007 09:45:52 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ISYj2-0003Ob-1e
	for behave-confirm+ok@megatron.ietf.org; Tue, 04 Sep 2007 09:45:52 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1ISYj1-0003OT-OL
	for behave@ietf.org; Tue, 04 Sep 2007 09:45:51 -0400
Received: from mailgw4.ericsson.se ([193.180.251.62])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1ISYj0-0001gy-CQ
	for behave@ietf.org; Tue, 04 Sep 2007 09:45:51 -0400
Received: from mailgw4.ericsson.se (unknown [127.0.0.1])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	066DE212A9; Tue,  4 Sep 2007 15:45:49 +0200 (CEST)
X-AuditID: c1b4fb3e-b0034bb0000007e1-49-46dd618c223c
Received: from esealmw126.eemea.ericsson.se (unknown [153.88.254.123])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	B1273212EE; Tue,  4 Sep 2007 15:45:48 +0200 (CEST)
Received: from esealmw126.eemea.ericsson.se ([153.88.254.174]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 15:45:48 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 15:45:48 +0200
Message-ID: <46DD618A.2000401@ericsson.com>
Date: Tue, 04 Sep 2007 15:45:46 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: =?UTF-8?B?UsOpbWkgRGVuaXMtQ291cm1vbnQ=?=
 <remi.denis-courmont@nokia.com>
Subject: Re: [BEHAVE] Ad comments on draft-ietf-behave-p2p-state-03
References: <46DD3F15.2030808@ericsson.com>
	<200709041528.13613.remi.denis-courmont@nokia.com>
In-Reply-To: <200709041528.13613.remi.denis-courmont@nokia.com>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=UTF-8
X-OriginalArrivalTime: 04 Sep 2007 13:45:48.0309 (UTC)
	FILETIME=[E6667450:01C7EEF9]
X-Brightmail-Tracker: AAAAAA==
Content-Transfer-Encoding: quoted-printable
X-Spam-Score: 0.0 (/)
X-Scan-Signature: b7b9551d71acde901886cc48bfc088a6
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

R=C3=A9mi Denis-Courmont skrev:
> Le Tuesday 04 September 2007 14:18:46 ext Magnus Westerlund, vous avez =
=C3=A9crit :
>> 1. Section 1, page 3, first paragraph:
>> "These issues are
>>    likely to persist even into the IPv6 world, where a NAT may be used
>>    as an IPv4 compatibility mechanism [NAT-PT]."
>>
>> I think you should rewrite this considering that NAT-PT was deprecated=
.
>=20
> I suppose some of "these issues" would still apply in presence of state=
ful=20
> firewall, which almost everybody in Chicago seemed to agree would be ra=
mpant=20
> in IPv6. I agree that the use of NAT-PT as an illustration is unfortuna=
te=20
> though.

Yes, I am not disputing the general motivation. It was simply that some
other formulation should be found that not uses NAT-PT as the reference.

>=20
>> 2. Section 1, page 3, second paragraph:
>>
>> "The
>>    anonymity and inaccessibility of the internal hosts behind a NAT
>>    device is not a problem for applications such as web browsers, whic=
h
>>    only need to initiate outgoing connections. This inaccessibility is
>>    sometimes perceived as a privacy benefit."
>>
>> Is really "privacy" the main benefit. Isn't it the filtering behavior =
of
>> the NAT that blocks incoming requests and thus avoid getting the traff=
ic
>> onto your local network that people like, i.e. the "firewalling" aspec=
t
>> of NATs.
>=20
> Agreed (though I would argue that a NAT by itself does NOT block incomi=
ng=20
> requests - especially if it supports source route).

Well, it is probably a matter of how you look on things. A NAT do not
translate and forward incomming packets unless there exist some mapping
for that port. Thus it has some blocking feature in that regards. Then
address and port dependent filtering may further prevent packets from
being forwarded to the internal side.

>=20
> (...)
>> 15. section 5.1, second paragraph:
>>
>> "When the NAT device
>>    enroute is not EIM-NAT, the application is unable to reuse an
>>    already established endpoint mapping for communication with
>>    different external destinations and the technique would fail."
>>
>> I think "enroute" is the wrong word. It is the outermost NAT that isn'=
t
>> an EIM-NAT for it to matter.
>=20
> I think any non-EIM NAT en-route would break, not only the outermost on=
e.
>=20

Yes, you I think you are correct. Forgot about the issue with a inner
NAT which has address dependent mapping resulting in that the outer EIM-
NAT still ends up mapping two different flows onto different mappings.


Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 04 09:55:19 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ISYs8-0002dc-CH; Tue, 04 Sep 2007 09:55:16 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ISYs7-0002dX-V8
	for behave-confirm+ok@megatron.ietf.org; Tue, 04 Sep 2007 09:55:15 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1ISYs7-0002dP-KS
	for behave@ietf.org; Tue, 04 Sep 2007 09:55:15 -0400
Received: from smtp.nokia.com ([131.228.20.170] helo=mgw-ext11.nokia.com)
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1ISYs7-0001bC-2b
	for behave@ietf.org; Tue, 04 Sep 2007 09:55:15 -0400
Received: from esebh108.NOE.Nokia.com (esebh108.ntc.nokia.com [172.21.143.145])
	by mgw-ext11.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l84DtB49025872; Tue, 4 Sep 2007 16:55:12 +0300
Received: from esebh103.NOE.Nokia.com ([172.21.143.33]) by
	esebh108.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 16:54:56 +0300
Received: from esebh101.NOE.Nokia.com ([172.21.138.177]) by
	esebh103.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 16:54:56 +0300
Received: from essapo-nirac25344.europe.nokia.com ([10.162.253.44]) by
	esebh101.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 4 Sep 2007 16:54:56 +0300
From: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: "ext Magnus Westerlund" <magnus.westerlund@ericsson.com>
Subject: Re: [BEHAVE] Ad comments on draft-ietf-behave-p2p-state-03
Date: Tue, 4 Sep 2007 16:54:54 +0300
User-Agent: KMail/1.9.6 (enterprise 20070825.704286)
References: <46DD3F15.2030808@ericsson.com>
	<200709041528.13613.remi.denis-courmont@nokia.com>
	<46DD618A.2000401@ericsson.com>
In-Reply-To: <46DD618A.2000401@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709041654.55639.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 04 Sep 2007 13:54:56.0435 (UTC)
	FILETIME=[2D1BD830:01C7EEFB]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: b19722fc8d3865b147c75ae2495625f2
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Opening a parenthesis here:

Le Tuesday 04 September 2007 16:45:46 ext Magnus Westerlund, vous avez =C3=
=A9crit=C2=A0:
> >> Isn't it the filtering behavior of
> >> the NAT that blocks incoming requests and thus avoid getting the traff=
ic
> >> onto your local network that people like, i.e. the "firewalling" aspect
> >> of NATs.
> >
> > Agreed (though I would argue that a NAT by itself does NOT block incomi=
ng
> > requests - especially if it supports source route).
>
> Well, it is probably a matter of how you look on things. A NAT do not
> translate and forward incomming packets unless there exist some mapping
> for that port. Thus it has some blocking feature in that regards. Then
> address and port dependent filtering may further prevent packets from
> being forwarded to the internal side.

Does it not? If a NAT receives a packet with an "internal" address as=20
destination IP from its "WAN" interface, why would it not forward it to the=
=20
inside? That's totally implementation dependant. NAT !=3D stateful firewall.

There are two ways to trigger this case. Either you are on the same link-la=
yer=20
segment as the NAT WAN port, or you use source route. Surely that will not=
=20
work all the time, but I would not rely on that to secure my network.

=2D-=20
R=C3=A9mi Denis-Courmont


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 04 18:21:13 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ISgli-0001fZ-6R; Tue, 04 Sep 2007 18:21:10 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ISglh-0001fU-BD
	for behave-confirm+ok@megatron.ietf.org; Tue, 04 Sep 2007 18:21:09 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1ISglh-0001e9-0t
	for behave@ietf.org; Tue, 04 Sep 2007 18:21:09 -0400
Received: from sj-iport-6.cisco.com ([171.71.176.117])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1ISglf-0006sw-PX
	for behave@ietf.org; Tue, 04 Sep 2007 18:21:09 -0400
Received: from sj-dkim-2.cisco.com ([171.71.179.186])
	by sj-iport-6.cisco.com with ESMTP; 04 Sep 2007 15:21:07 -0700
X-IronPort-AV: i="4.20,208,1186383600"; 
	d="scan'208"; a="212095584:sNHT47073024"
Received: from sj-core-1.cisco.com (sj-core-1.cisco.com [171.71.177.237])
	by sj-dkim-2.cisco.com (8.12.11/8.12.11) with ESMTP id l84ML7G1009601; 
	Tue, 4 Sep 2007 15:21:07 -0700
Received: from dwingwxp01 ([10.32.240.198])
	by sj-core-1.cisco.com (8.12.10/8.12.6) with ESMTP id l84MKvM3017153;
	Tue, 4 Sep 2007 22:20:57 GMT
From: "Dan Wing" <dwing@cisco.com>
To: "'Philip Matthews'" <philip_matthews@magma.ca>,
	"'David Barrett'" <dbarrett@quinthar.com>
References: <049501c7c89d$d34da410$7d646b80@amer.cisco.com>
	<A56D7804-00EC-4D09-9B69-D73F27EBA2EE@magma.ca>
Subject: RE: [BEHAVE] RE: ICE performance 
Date: Tue, 4 Sep 2007 15:20:57 -0700
Message-ID: <23d001c7ef41$e3650670$c6f0200a@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Office Outlook 11
In-Reply-To: <A56D7804-00EC-4D09-9B69-D73F27EBA2EE@magma.ca>
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138
Thread-Index: Acfg6w39S0V9eWclTzuc5QVyVEsrrwOTf3nQ
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=986; t=1188944467;
	x=1189808467; c=relaxed/simple; s=sjdkim2002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=dwing@cisco.com;
	z=From:=20=22Dan=20Wing=22=20<dwing@cisco.com>
	|Subject:=20RE=3A=20[BEHAVE]=20RE=3A=20ICE=20performance=20
	|Sender:=20; bh=VoUAJxV6+Eh8ixe++19ax8r0DHFtzQYSP/EYopy23wU=;
	b=nQOWQGwMeJ0e44dbRi7JwHnAFNz+/rwueOmVwNqcyAE8nOiJGOEVq6OvznV3WaEYIVvi4vh+
	A69ZHFWewcKol50B2pEwvHCBJo8TQb8t7wI14AfQvLzsiLXbfK1Wb8bL;
Authentication-Results: sj-dkim-2; header.From=dwing@cisco.com; dkim=pass (s
	ig from cisco.com/sjdkim2002 verified; ); 
X-Spam-Score: -4.0 (----)
X-Scan-Signature: 8abaac9e10c826e8252866cbe6766464
Cc: henrys@adobe.com, 'Bryan Ford' <baford@mit.edu>,
	'Behave WG' <behave@ietf.org>, 'Henry Sinnreich' <hsinnrei@adobe.com>,
	'Dan Kegel' <dank@kegel.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org


> I will just note that Saikat implemented port prediction in his TCP  
> study and found it worked pretty well.

On a quiescent network, with no or few other clients creating new bindings, it
can work alright.  However, a client has no way of knowing if it's on a
quiescent network or behind a massive ISP NAT with thousands of other UDP/TCP
clients creating new connections.  From the paper you cite:

  "In one particular case, the client host behind a
   NB:Connection NAT was infected with a virus that generated
   several randomly-addressed SYN packets per second
   causing all predictions to fail!"

-d


> http://www.guha.cc/saikat/pub/imc05-tcpnat.pdf
> 
> However, I think his setup was more of
> a test setup rather than a production setup.
> 
> Perhaps Saikat can tell us more.
> 
> - Philip
> 
> 
> 
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www1.ietf.org/mailman/listinfo/behave


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 04 19:40:54 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ISi0p-0000qI-5Q; Tue, 04 Sep 2007 19:40:51 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ISi0o-0000q9-8j
	for behave-confirm+ok@megatron.ietf.org; Tue, 04 Sep 2007 19:40:50 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ISi0n-0000q1-TY; Tue, 04 Sep 2007 19:40:49 -0400
Received: from sj-iport-6.cisco.com ([171.71.176.117])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43)
	id 1ISi0n-0000RI-EV; Tue, 04 Sep 2007 19:40:49 -0400
Received: from sj-dkim-2.cisco.com ([171.71.179.186])
	by sj-iport-6.cisco.com with ESMTP; 04 Sep 2007 16:40:49 -0700
X-IronPort-AV: i="4.20,208,1186383600"; 
	d="scan'208"; a="212138490:sNHT52690014"
Received: from sj-core-2.cisco.com (sj-core-2.cisco.com [171.71.177.254])
	by sj-dkim-2.cisco.com (8.12.11/8.12.11) with ESMTP id l84NemYt030248; 
	Tue, 4 Sep 2007 16:40:48 -0700
Received: from dwingwxp01 ([10.32.240.198])
	by sj-core-2.cisco.com (8.12.10/8.12.6) with ESMTP id l84NemmN000941;
	Tue, 4 Sep 2007 23:40:48 GMT
From: "Dan Wing" <dwing@cisco.com>
To: "=?iso-8859-1?Q?'R=E9mi_Denis-Courmont'?=" <remi.denis-courmont@nokia.com>,
	"'ext Jonathan Rosenberg'" <jdrosen@cisco.com>
References: <200708201437.55515.remi.denis-courmont@nokia.com><46CA12C7.6020302@cisco.com>
	<200708210926.45072.remi.denis-courmont@nokia.com>
Subject: RE: [BEHAVE] Re: STUN/ICE: Username length inconstency
Date: Tue, 4 Sep 2007 16:40:47 -0700
Message-ID: <243901c7ef4d$05723700$c6f0200a@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
X-Mailer: Microsoft Office Outlook 11
In-Reply-To: <200708210926.45072.remi.denis-courmont@nokia.com>
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138
Thread-Index: AcfjvFfQAQ1nWRnvQTeK2kTsQGUTVgLkF7BA
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=2299; t=1188949248;
	x=1189813248; c=relaxed/simple; s=sjdkim2002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=dwing@cisco.com;
	z=From:=20=22Dan=20Wing=22=20<dwing@cisco.com>
	|Subject:=20RE=3A=20[BEHAVE]=20Re=3A=20STUN/ICE=3A=20Username=20length=20
	inconstency |Sender:=20;
	bh=WVuo6CjAWmttmQ+EYgKHLKHTQtfWkHi42219rMZTi+A=;
	b=Fisb59Xh7tqicEe9tAT77QS3Jo2asSfq+de0rnlpOAG1IcORsGiZ4astYsSBkHqBYd+4/iFD
	+4X0mqkJU3Rf+L1P0h5TgKtC1Wgd0eK9iDjVd3Ix+Q95ILFBRPJYqeK4;
Authentication-Results: sj-dkim-2; header.From=dwing@cisco.com; dkim=pass (s
	ig from cisco.com/sjdkim2002 verified; ); 
X-Spam-Score: 0.0 (/)
X-Scan-Signature: f607d15ccc2bc4eaf3ade8ffa8af02a0
Cc: behave@ietf.org, mmusic@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

(Sorry for the delay, I just returned from vacation.)

It's useful to have a long username (for things like the now-expired
draft-wing-session-auth-00), but I agree we don't want to exceed the MTU =
with
the ICE connectivity checks.  ICE should probably mention that somewhere =
(it
doesn't contain the string "MTU", but I didn't search elsewhere).

-d


> -----Original Message-----
> From: R=E9mi Denis-Courmont [mailto:remi.denis-courmont@nokia.com]=20
> Sent: Monday, August 20, 2007 11:27 PM
> To: ext Jonathan Rosenberg
> Cc: behave@ietf.org; mmusic@ietf.org
> Subject: [BEHAVE] Re: STUN/ICE: Username length inconstency
>=20
> On Tuesday 21 August 2007 01:16:39 ext Jonathan Rosenberg wrote:
> > This is a good catch. I recently introduced the limit in=20
> STUN based on
> > last call comments, forgetting that we had ICE use cases that could
> > result in values longer than 127. I had picked 127 thinking=20
> about human
> > readable usernames. My bad. I think the limit in STUN should be
> > increased to match ICE, not the other way around.
>=20
> Hmm, I am not sure 1023 character long username are that=20
> great really. For a=20
> start, after concatenation (2047 characters), you'll get=20
> connectivity checks=20
> that exceed the typical Internet MTU. That brings either problem:
> - NATs will not handle fragmented packets, or
> - ICE implementation will refuse to send at all (packet too big).
>=20
> Second, I wonder if if makes sense to have a higher entropy=20
> for the username=20
> (12,000 bits here!) than we have for the message integrity (160 bits=20
> currently) or the shared secret password. But that's=20
> something for security=20
> gurus to judge. Even if we switched to SHA512, we'd still=20
> have more than=20
> enough username entropy with just 63 characters from each=20
> sides (756 bits).
>=20
> Third, we do not want to drain the entropy pool of=20
> ICE-capable device too=20
> badly. There are better stuff to dedicate entropy to that ICE=20
> connectivity=20
> checks, such as generating TLS parameters.
>=20
> --=20
> R=E9mi Denis-Courmont
>=20
>=20
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www1.ietf.org/mailman/listinfo/behave


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Fri Sep 07 08:04:13 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ITcZH-0003FZ-Pg; Fri, 07 Sep 2007 08:04:11 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ITcZG-0003FA-EG
	for behave-confirm+ok@megatron.ietf.org; Fri, 07 Sep 2007 08:04:10 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ITcZF-0003Er-Pr; Fri, 07 Sep 2007 08:04:09 -0400
Received: from mailgw3.ericsson.se ([193.180.251.60])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43)
	id 1ITcZF-00036Y-3C; Fri, 07 Sep 2007 08:04:09 -0400
Received: from mailgw3.ericsson.se (unknown [127.0.0.1])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	EFC0420A32; Fri,  7 Sep 2007 14:04:07 +0200 (CEST)
X-AuditID: c1b4fb3c-b0e80bb0000007e1-4f-46e13e3708af
Received: from esealmw126.eemea.ericsson.se (unknown [153.88.254.123])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	D20EB20A24; Fri,  7 Sep 2007 14:04:07 +0200 (CEST)
Received: from esealmw126.eemea.ericsson.se ([153.88.254.174]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 7 Sep 2007 14:04:07 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 7 Sep 2007 14:04:07 +0200
Message-ID: <46E13E37.7040806@ericsson.com>
Date: Fri, 07 Sep 2007 14:04:07 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Dan Wing <dwing@cisco.com>
Subject: Re: [BEHAVE] WGLC for draft-ietf-behave-multicast-09
References: <020401c7df8c$c7009c70$c6f0200a@cisco.com>
In-Reply-To: <020401c7df8c$c7009c70$c6f0200a@cisco.com>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 07 Sep 2007 12:04:07.0151 (UTC)
	FILETIME=[3110DFF0:01C7F147]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: cd26b070c2577ac175cd3a6d878c6248
Cc: mboned@ietf.org, magma@ietf.org, behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Hi,

As AD and shepherd of this document I am concluding this WG last call.
Some comments was received, but it appears nothing controversial. Editor
please update the draft. When a new draft version is available I will
request publication of this document as BCP.

Best Regards

Magnus

Dan Wing skrev:
> I am starting the 3rd WGLC on draft-ietf-behave-multicast-09, "IP Multicast
> Requirements for a Network Address (and port) Translator NAT".  The WGLC ends
> on September 3.  The previous WGLCs were for -06 and -03 (which also went to
> IESG).
> 
> 
> Draft-ietf-behave-multicast-09 integrates all of the comments received to
> date.  Significant changes from -08 include:
> 
> * Consideration of Gorry Fairhurst's comments in Chicago about a multicast
> source inside of a multi-homed network; unfortunately the best I could
> describe to deal with that problem was the new requirement 5-a:
> 
>            "a:  If a network is multihomed, the NATs or the network
>                 configuration MUST ensure that duplicate instances of
>                 the multicast data traffic does not appear on the public
>                 network.  This can be accomplished by network design (an
>                 access control list) or a protocol between the NATs
>                 (e.g., the NAT with the higher IP address transmits
>                 multicast data towards the public network and other NATs
>                 drop it)."
> 
> * a NAT with multiple public interfaces SHOULD pair for multicast in the same
> way it pairs for unicast (new requirement 4-c).
> 
> * The "application considerations" section has been moved to an appendix, as
> it seems outside the scope of a document describing the behavior for NATs.
> 
> 
> Differences between -09 and -08:
>  
> http://tools.ietf.org/rfcdiff?url2=http://tools.ietf.org/id/draft-ietf-behave-
> multicast-09.txt
>   http://tinyurl.com/yqej5l
> 
> Differences between -09 and -06:
>  
> http://tinyurl.com/create.php?url=http://tools.ietf.org/rfcdiff?url1=http://to
> ols.ietf.org/id/draft-ietf-behave-multicast-06.txt&url2=http://tools.ietf.org/
> id/draft-ietf-behave-multicast-09.txt
>   http://tinyurl.com/yulw5t
> 
> 
> Please reply with any comments.
> -d
> 
> 
> 
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www1.ietf.org/mailman/listinfo/behave
> 


-- 

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Fri Sep 07 10:49:17 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ITf92-0000m4-Ie; Fri, 07 Sep 2007 10:49:16 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ITf90-0000lv-OX
	for behave-confirm+ok@megatron.ietf.org; Fri, 07 Sep 2007 10:49:14 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ITf8z-0000jc-JC; Fri, 07 Sep 2007 10:49:14 -0400
Received: from sj-iport-3-in.cisco.com ([171.71.176.72]
	helo=sj-iport-3.cisco.com) by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1ITf8y-0002Rp-8p; Fri, 07 Sep 2007 10:49:13 -0400
X-IronPort-AV: E=Sophos;i="4.20,221,1186383600"; d="scan'208";a="521577090"
Received: from sj-dkim-3.cisco.com ([171.71.179.195])
	by sj-iport-3.cisco.com with ESMTP; 07 Sep 2007 07:49:12 -0700
Received: from sj-core-2.cisco.com (sj-core-2.cisco.com [171.71.177.254])
	by sj-dkim-3.cisco.com (8.12.11/8.12.11) with ESMTP id l87EnBGv028066; 
	Fri, 7 Sep 2007 07:49:11 -0700
Received: from xbh-sjc-221.amer.cisco.com (xbh-sjc-221.cisco.com
	[128.107.191.63])
	by sj-core-2.cisco.com (8.12.10/8.12.6) with ESMTP id l87Emfxl027579;
	Fri, 7 Sep 2007 14:49:11 GMT
Received: from xfe-sjc-211.amer.cisco.com ([171.70.151.174]) by
	xbh-sjc-221.amer.cisco.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 7 Sep 2007 07:49:06 -0700
Received: from [10.32.241.146] ([10.32.241.146]) by xfe-sjc-211.amer.cisco.com
	with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 7 Sep 2007 07:49:06 -0700
Message-ID: <46E164F2.207@cisco.com>
Date: Fri, 07 Sep 2007 10:49:22 -0400
From: Jonathan Rosenberg <jdrosen@cisco.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Dan Wing <dwing@cisco.com>
Subject: Re: [BEHAVE] Re: STUN/ICE: Username length inconstency
References: <200708201437.55515.remi.denis-courmont@nokia.com><46CA12C7.6020302@cisco.com>
	<200708210926.45072.remi.denis-courmont@nokia.com>
	<243901c7ef4d$05723700$c6f0200a@cisco.com>
In-Reply-To: <243901c7ef4d$05723700$c6f0200a@cisco.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
X-OriginalArrivalTime: 07 Sep 2007 14:49:06.0202 (UTC)
	FILETIME=[3D5C2FA0:01C7F15E]
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=3459; t=1189176551;
	x=1190040551; c=relaxed/simple; s=sjdkim3002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=jdrosen@cisco.com;
	z=From:=20Jonathan=20Rosenberg=20<jdrosen@cisco.com>
	|Subject:=20Re=3A=20[BEHAVE]=20Re=3A=20STUN/ICE=3A=20Username=20length=20
	inconstency |Sender:=20;
	bh=ukhKfiosGWYYX2Zi7gu2w24Tp7TMkdx7jklH76Fpo+8=;
	b=pFJ22zAP/umEJR0PIuI0eOHbGvBq84J5kEZqnPMhTaq7JT3ks1CzadNvUzdJCq/ZkvA8pt4W
	S8iKK2ODAti3cBs7U4vtxl63czKg/pkXZZ7Mk/y540vLHgM1RjsakEp8;
Authentication-Results: sj-dkim-3; header.From=jdrosen@cisco.com; dkim=pass (
	sig from cisco.com/sjdkim3002 verified; ); 
X-Spam-Score: -4.0 (----)
X-Scan-Signature: 10ba05e7e8a9aa6adb025f426bef3a30
Cc: behave@ietf.org, mmusic@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

To be specific, let me propose the following:

rfc3489bis: says the username MUST be less than 513 bytes. It also says 
that the STUN message when sent over UDP MUST be less than the MTU.

ICE: says that the username fragment MUST be less than 256 bytes.


The desire for the ability to have larger usernames is not security 
(i.e., this doesn't mean endpoints compute username fragments with 256 
bytes of entropy), but that cases have been identified ala Dan's draft 
of stuffing useful identifiers in there. So we want to allow that to be 
possible in the future. Length limits always need to be carefully 
managed, since they are nice in terms of implementation but you can 
sometimes regret them in a really big way down the road (think SMS 
message sizes, 640k memory limit, etc.)

-Jonathan R.




Dan Wing wrote:
> (Sorry for the delay, I just returned from vacation.)
> 
> It's useful to have a long username (for things like the now-expired
> draft-wing-session-auth-00), but I agree we don't want to exceed the MTU with
> the ICE connectivity checks.  ICE should probably mention that somewhere (it
> doesn't contain the string "MTU", but I didn't search elsewhere).
> 
> -d
> 
> 
>> -----Original Message-----
>> From: Rémi Denis-Courmont [mailto:remi.denis-courmont@nokia.com] 
>> Sent: Monday, August 20, 2007 11:27 PM
>> To: ext Jonathan Rosenberg
>> Cc: behave@ietf.org; mmusic@ietf.org
>> Subject: [BEHAVE] Re: STUN/ICE: Username length inconstency
>>
>> On Tuesday 21 August 2007 01:16:39 ext Jonathan Rosenberg wrote:
>>> This is a good catch. I recently introduced the limit in 
>> STUN based on
>>> last call comments, forgetting that we had ICE use cases that could
>>> result in values longer than 127. I had picked 127 thinking 
>> about human
>>> readable usernames. My bad. I think the limit in STUN should be
>>> increased to match ICE, not the other way around.
>> Hmm, I am not sure 1023 character long username are that 
>> great really. For a 
>> start, after concatenation (2047 characters), you'll get 
>> connectivity checks 
>> that exceed the typical Internet MTU. That brings either problem:
>> - NATs will not handle fragmented packets, or
>> - ICE implementation will refuse to send at all (packet too big).
>>
>> Second, I wonder if if makes sense to have a higher entropy 
>> for the username 
>> (12,000 bits here!) than we have for the message integrity (160 bits 
>> currently) or the shared secret password. But that's 
>> something for security 
>> gurus to judge. Even if we switched to SHA512, we'd still 
>> have more than 
>> enough username entropy with just 63 characters from each 
>> sides (756 bits).
>>
>> Third, we do not want to drain the entropy pool of 
>> ICE-capable device too 
>> badly. There are better stuff to dedicate entropy to that ICE 
>> connectivity 
>> checks, such as generating TLS parameters.
>>
>> -- 
>> Rémi Denis-Courmont
>>
>>
>> _______________________________________________
>> Behave mailing list
>> Behave@ietf.org
>> https://www1.ietf.org/mailman/listinfo/behave
> 

-- 
Jonathan D. Rosenberg, Ph.D.                   600 Lanidex Plaza
Cisco Fellow                                   Parsippany, NJ 07054-2711
Cisco Systems
jdrosen@cisco.com                              FAX:   (973) 952-5050
http://www.jdrosen.net                         PHONE: (973) 952-5000
http://www.cisco.com


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Fri Sep 07 16:30:55 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1ITkTc-0003bv-ID; Fri, 07 Sep 2007 16:30:52 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1ITkTa-0003bj-Ow
	for behave-confirm+ok@megatron.ietf.org; Fri, 07 Sep 2007 16:30:50 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1ITkTa-0003bb-FJ
	for behave@ietf.org; Fri, 07 Sep 2007 16:30:50 -0400
Received: from sj-iport-5.cisco.com ([171.68.10.87])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1ITkTZ-000389-7v
	for behave@ietf.org; Fri, 07 Sep 2007 16:30:50 -0400
X-IronPort-AV: E=Sophos;i="4.20,222,1186383600"; d="scan'208";a="174989607"
Received: from sj-dkim-3.cisco.com ([171.71.179.195])
	by sj-iport-5.cisco.com with ESMTP; 07 Sep 2007 13:30:48 -0700
Received: from sj-core-5.cisco.com (sj-core-5.cisco.com [171.71.177.238])
	by sj-dkim-3.cisco.com (8.12.11/8.12.11) with ESMTP id l87KUjs4015445
	for <behave@ietf.org>; Fri, 7 Sep 2007 13:30:45 -0700
Received: from dwingwxp01 (dhcp-128-107-163-176.cisco.com [128.107.163.176])
	by sj-core-5.cisco.com (8.12.10/8.12.6) with ESMTP id l87KUjD1015580
	for <behave@ietf.org>; Fri, 7 Sep 2007 20:30:45 GMT
From: "Dan Wing" <dwing@cisco.com>
To: <behave@ietf.org>
Date: Fri, 7 Sep 2007 13:30:45 -0700
Message-ID: <08bf01c7f18d$f7ee97c0$c6f0200a@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Office Outlook 11
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138
Thread-Index: AcfxiK1gV4gOTwAmSfG1OzD8uf2ysAABTHoQ
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=1477; t=1189197048;
	x=1190061048; c=relaxed/simple; s=sjdkim3002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=dwing@cisco.com;
	z=From:=20=22Dan=20Wing=22=20<dwing@cisco.com>
	|Subject:=20FW=3A=20Nomcom=202007-8=3A=20Nominations=20Close=20on=20Sep=2
	010,=202007 |Sender:=20;
	bh=uK302J6AgQgfXKQ71AREC3aItsa7Ibf67rvIGEWfh1A=;
	b=KD7lWl0ypLyPqIDgAbg/R7C8AkG3MzdB4CeekfulDF4t8YQGG1SxCoaxbFo8Zokil06BCY9k
	z4HTdT24EJbi1EEBRtPtMVEGt6PUubhEUQvs3M4ghydclVlSnP2t4l6s;
Authentication-Results: sj-dkim-3; header.From=dwing@cisco.com; dkim=pass (s
	ig from cisco.com/sjdkim3002 verified; ); 
X-Spam-Score: -4.0 (----)
X-Scan-Signature: cab78e1e39c4b328567edb48482b6a69
Subject: [BEHAVE] FW: Nomcom 2007-8: Nominations Close on Sep 10, 2007
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Please provide feedback on IETF leadership (IESG and IAB members), and
volunteer yourself or someone else for positions on IESG/IAB.  Thanks.

-d 

-----Original Message-----
From: Lakshminath Dondeti [mailto:ldondeti@qualcomm.com] 
Sent: Friday, September 07, 2007 12:53 PM
To: ietf-announce@ietf.org; wgchairs@ietf.org; IESG IESG; IAB IAB
Subject: Nomcom 2007-8: Nominations Close on Sep 10, 2007

WGchairs, IESG and IAB members

Please forward this request to the lists you manage and request feedback 
and nominations.

All,

Here is the link to nominate: 
https://tools.ietf.org/group/nomcom/07/nominate

You may also send nominations or comments via email to nomcom07@ietf.org 
or ldondeti@qualcomm.com.

We have received very few nominations (1, 2, 2, 2, 3, 4, 8, 8, 19) and 
even fewer accepted (1-2 people in each area, IAB acceptances is 4 at 
last count).

I request the community to provide feedback on the incumbents (send 
email or send notes via the web page).

1) If you think that the incumbent is doing a good job
     a) provide feedback AND
     b) nominate similar people just in case there is strong negative 
feedback on the incumbent

2) If you think the incumbent can do somethings better
    a) provide feedback AND
    b) nominate someone who you think might do better

Candidates, if time commitment is the only issue, please indicate that 
to the nomcom and accept the nominations.

thanks,
Lakshminath


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Sat Sep 08 15:04:12 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IU5bG-0005CM-Ij; Sat, 08 Sep 2007 15:04:10 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IU5bE-0005C2-L8
	for behave-confirm+ok@megatron.ietf.org; Sat, 08 Sep 2007 15:04:08 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IU5bE-0005Bu-BL
	for behave@ietf.org; Sat, 08 Sep 2007 15:04:08 -0400
Received: from [69.55.225.91] (helo=www.implementers.org)
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IU5bD-00073d-5S
	for behave@ietf.org; Sat, 08 Sep 2007 15:04:08 -0400
Received: by www.implementers.org (Postfix, from userid 1006)
	id 8B57F3A02022; Sat,  8 Sep 2007 12:04:06 -0700 (PDT)
Received: from [192.168.1.3] (localhost.localdomain [127.0.0.1])
	by www.implementers.org (Postfix) with ESMTP id A1E923A02020
	for <behave@ietf.org>; Sat,  8 Sep 2007 12:03:52 -0700 (PDT)
Message-ID: <46E2F217.1080909@acm.org>
Date: Sat, 08 Sep 2007 12:03:51 -0700
From: Marc Petit-Huguenin <petithug@acm.org>
User-Agent: Mozilla-Thunderbird 2.0.0.4 (X11/20070828)
MIME-Version: 1.0
To: behave@ietf.org
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
X-Spam-Score: 0.1 (/)
X-Scan-Signature: a7d6aff76b15f3f56fcb94490e1052e4
Subject: [BEHAVE] Allocation move in TURN
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Sorry for the late comment.

There is an interesting paragraph in section 6.1.2.1 of
draft-ietf-behave-turn-04 that says this:

"If the port behavior requested is for a Specific Port, the server
 MUST attempt to allocate that specific port for the client.  If the
 port is allocated to a different internal 5-tuple associated with the
 same STUN long-term credentials, the client is requesting a move.
 The server SHOULD replace the old internal 5-tuple with the new tuple
 over which this Allocate request arrived.  The server MUST reject the
 move request if any of the attributes other than LIFETIME have
 changed (BANDWIDTH, REQUESTED-TRANSPORT, etc.)."

With some small modifications in the spec, this can be used to
automatically take care of NAT reboot or TCP disconnection.

Two modifications would be needed in section 6.1.1.2:

Instead of sending a subsequent Allocate request with exactly the same
attributes that was used on the initial request, the
REQUESTED-PORT-PROPS is modified to carry in the Specific Port Number
field the port allocated from the initial request.  If a NAT reboots,
the Allocate request will be seen as an initial request, and the
previous allocation will be moved to the new tuple.  The client will see
that the reflexive server transport address changed and can use it
immediately in a reINVITE without having to create a new allocation.
This does not work for a TCP disconnection as this will immediately
remove the allocation so I propose to remove the last sentence of
section 6.1.1.2 so a TCP disconnection does not remove the allocation.
When the client detects the TCP disconnection, it can open a new
connection and send an Allocate request as described above.  The server
will then move the tuple to the new connection.  The allocation for a
TCP connection is then removed by the expiration of the allocation timer
as for UDP.


-- 
Marc Petit-Huguenin           [                                 ]
Home: marc@petit-huguenin.org [RFC1855-compliant space for rent ]
Work: marc@8x8.com            [                                 ]
[                                                               ]


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 10 04:25:19 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUea4-0000pr-N8; Mon, 10 Sep 2007 04:25:16 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IUea3-0000pj-3b
	for behave-confirm+ok@megatron.ietf.org; Mon, 10 Sep 2007 04:25:15 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IUeZz-0000pY-7L
	for behave@ietf.org; Mon, 10 Sep 2007 04:25:11 -0400
Received: from smtp.nokia.com ([131.228.20.172] helo=mgw-ext13.nokia.com)
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IUeZx-0005Df-J8
	for behave@ietf.org; Mon, 10 Sep 2007 04:25:11 -0400
Received: from esebh105.NOE.Nokia.com (esebh105.ntc.nokia.com [172.21.138.211])
	by mgw-ext13.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l8A8Ocq2009039; Mon, 10 Sep 2007 11:25:01 +0300
Received: from esebh103.NOE.Nokia.com ([172.21.143.33]) by
	esebh105.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 11:24:51 +0300
Received: from esebh102.NOE.Nokia.com ([172.21.138.183]) by
	esebh103.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 11:24:42 +0300
Received: from mgw-int01.ntc.nokia.com ([172.21.143.96]) by
	esebh102.NOE.Nokia.com over TLS secured channel with Microsoft
	SMTPSVC(6.0.3790.1830); Mon, 10 Sep 2007 11:24:41 +0300
Received: from esdhcp041174.research.nokia.com
	(esdhcp041174.research.nokia.com [172.21.41.174])
	by mgw-int01.ntc.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l8A8Oefw016185; Mon, 10 Sep 2007 11:24:40 +0300
From: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= <Remi.Denis-Courmont@nokia.com>
Organization: Nokia-TP-MSW Helsinki
To: behave@ietf.org
Subject: Re: [BEHAVE] Allocation move in TURN
Date: Mon, 10 Sep 2007 11:24:57 +0300
User-Agent: KMail/1.9.6
References: <46E2F217.1080909@acm.org>
In-Reply-To: <46E2F217.1080909@acm.org>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709101124.57738.Remi.Denis-Courmont@nokia.com>
X-OriginalArrivalTime: 10 Sep 2007 08:24:41.0633 (UTC)
	FILETIME=[090B8110:01C7F384]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 7d33c50f3756db14428398e2bdedd581
Cc: 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

On Saturday 08 September 2007 22:03:51 ext Marc Petit-Huguenin wrote:
> With some small modifications in the spec, this can be used to
> automatically take care of NAT reboot or TCP disconnection.

I am pretty sure I already commented that this would NOT WORK.

When a TCP connection is reset. you do not know where EXACTLY within the=20
send/recv windows this has happened. As such, even if you moved the TCP=20
allocation, you cannot resync with the other end (the real end. not the TUR=
N=20
server). Unless the application-layer protocol can automagically reset to=20
some sane state.

Unfortunately. most TCP-based protocol are not capable of doing this.=20
Especially, MSRP does not support this, and TCP-framed RTP does not support=
=20
this either.

=2D-=20
R=C3=A9mi Denis-Courmont
Software engineer


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 10 05:42:50 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUfn5-00048Q-5S; Mon, 10 Sep 2007 05:42:47 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IUfn3-00048L-PV
	for behave-confirm+ok@megatron.ietf.org; Mon, 10 Sep 2007 05:42:45 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IUfn3-00048D-FW
	for behave@ietf.org; Mon, 10 Sep 2007 05:42:45 -0400
Received: from server.implementers.org ([69.55.230.26])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IUfn3-0007xj-3M
	for behave@ietf.org; Mon, 10 Sep 2007 05:42:45 -0400
Received: by server.implementers.org (Postfix, from userid 1001)
	id 5BEB8E1CD6D0; Mon, 10 Sep 2007 02:42:44 -0700 (PDT)
Received: from [192.168.1.3] (c-24-4-253-5.hsd1.ca.comcast.net [24.4.253.5])
	by server.implementers.org (Postfix) with ESMTP id 13E8FE1CD6CE;
	Mon, 10 Sep 2007 02:42:42 -0700 (PDT)
Message-ID: <46E51191.6050603@acm.org>
Date: Mon, 10 Sep 2007 02:42:41 -0700
From: Marc Petit-Huguenin <petithug@acm.org>
User-Agent: Mozilla-Thunderbird 2.0.0.4 (X11/20070828)
MIME-Version: 1.0
To: =?UTF-8?B?UsOpbWkgRGVuaXMtQ291cm1vbnQ=?=
 <Remi.Denis-Courmont@nokia.com>
Subject: Re: [BEHAVE] Allocation move in TURN
References: <46E2F217.1080909@acm.org>
	<200709101124.57738.Remi.Denis-Courmont@nokia.com>
In-Reply-To: <200709101124.57738.Remi.Denis-Courmont@nokia.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 69a74e02bbee44ab4f8eafdbcedd94a1
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

R=C3=A9mi Denis-Courmont wrote:
> On Saturday 08 September 2007 22:03:51 ext Marc Petit-Huguenin wrote:
>> With some small modifications in the spec, this can be used to
>> automatically take care of NAT reboot or TCP disconnection.
>=20
> I am pretty sure I already commented that this would NOT WORK.
>=20
> When a TCP connection is reset. you do not know where EXACTLY within th=
e=20
> send/recv windows this has happened. As such, even if you moved the TCP=
=20
> allocation, you cannot resync with the other end (the real end. not the=
 TURN=20
> server). Unless the application-layer protocol can automagically reset =
to=20
> some sane state.
>=20
> Unfortunately. most TCP-based protocol are not capable of doing this.=20
> Especially, MSRP does not support this, and TCP-framed RTP does not sup=
port=20
> this either.
>=20

True.

But what if the TURN client sends the number of bytes received before
the disconnection in a new attribute in the first Allocate request of
the new connection, and the TURN server responds by adding in the
response an attribute containing the number of bytes received from the
TURN client before the connection?  Wouldn't this help (assuming that
the TURN client and server have big enough buffers) to resynchronize the
streams?

--=20
Marc Petit-Huguenin           [                                 ]
Home: marc@petit-huguenin.org [RFC1855-compliant space for rent ]
Work: marc@8x8.com            [                                 ]
[                                                               ]


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 10 06:09:50 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUgDE-0006Kt-M7; Mon, 10 Sep 2007 06:09:48 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IUgDD-0006Kk-W1
	for behave-confirm+ok@megatron.ietf.org; Mon, 10 Sep 2007 06:09:47 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IUgDD-0006Kb-M0
	for behave@ietf.org; Mon, 10 Sep 2007 06:09:47 -0400
Received: from smtp.nokia.com ([131.228.20.171] helo=mgw-ext12.nokia.com)
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IUgDD-0008Ut-0g
	for behave@ietf.org; Mon, 10 Sep 2007 06:09:47 -0400
Received: from esebh106.NOE.Nokia.com (esebh106.ntc.nokia.com [172.21.138.213])
	by mgw-ext12.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l8AA9X2i013629 for <behave@ietf.org>; Mon, 10 Sep 2007 13:09:45 +0300
Received: from esebh104.NOE.Nokia.com ([172.21.143.34]) by
	esebh106.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 13:09:42 +0300
Received: from esebh101.NOE.Nokia.com ([172.21.138.177]) by
	esebh104.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 13:09:41 +0300
Received: from esdhcp04162.research.nokia.com ([172.21.41.62]) by
	esebh101.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 13:09:41 +0300
From: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: behave@ietf.org
Subject: Re: [BEHAVE] Allocation move in TURN
Date: Mon, 10 Sep 2007 13:09:37 +0300
User-Agent: KMail/1.9.6 (enterprise 0.20070831.706792)
References: <46E2F217.1080909@acm.org>
	<200709101124.57738.Remi.Denis-Courmont@nokia.com>
	<46E51191.6050603@acm.org>
In-Reply-To: <46E51191.6050603@acm.org>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709101309.37280.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 10 Sep 2007 10:09:41.0696 (UTC)
	FILETIME=[B42CD400:01C7F392]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: cf4fa59384e76e63313391b70cd0dd25
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Le Monday 10 September 2007 12:42:41 ext Marc Petit-Huguenin, vous avez=20
=C3=A9crit=C2=A0:
> But what if the TURN client sends the number of bytes received before
> the disconnection in a new attribute in the first Allocate request of
> the new connection, and the TURN server responds by adding in the
> response an attribute containing the number of bytes received from the
> TURN client before the connection?  Wouldn't this help (assuming that
> the TURN client and server have big enough buffers) to resynchronize the
> streams?

Nevermind that you would need to rewrite the socket API to access the inter=
nal=20
TCP buffers. At that point you might as well extend TCP rather than TURN.

=2D-=20
R=C3=A9mi Denis-Courmont


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 10 08:52:59 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUil8-0008JS-OH; Mon, 10 Sep 2007 08:52:58 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IUil7-0008Gy-MI
	for behave-confirm+ok@megatron.ietf.org; Mon, 10 Sep 2007 08:52:57 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUil6-0008Et-Pq; Mon, 10 Sep 2007 08:52:57 -0400
Received: from mailgw4.ericsson.se ([193.180.251.62])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43)
	id 1IUil6-0004xA-33; Mon, 10 Sep 2007 08:52:56 -0400
Received: from mailgw4.ericsson.se (unknown [127.0.0.1])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	2F3A8213F4; Mon, 10 Sep 2007 14:52:55 +0200 (CEST)
X-AuditID: c1b4fb3e-b0835bb0000007e1-c4-46e53e27a807
Received: from esealmw127.eemea.ericsson.se (unknown [153.88.254.122])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	02D8020392; Mon, 10 Sep 2007 14:52:55 +0200 (CEST)
Received: from esealmw127.eemea.ericsson.se ([153.88.254.175]) by
	esealmw127.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 14:52:55 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw127.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 14:52:54 +0200
Message-ID: <46E53E26.1090703@ericsson.com>
Date: Mon, 10 Sep 2007 14:52:54 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Jonathan Rosenberg <jdrosen@cisco.com>
Subject: Re: [MMUSIC] Re: [BEHAVE] Re: STUN/ICE: Username length inconstency
References: <200708201437.55515.remi.denis-courmont@nokia.com><46CA12C7.6020302@cisco.com>	<200708210926.45072.remi.denis-courmont@nokia.com>	<243901c7ef4d$05723700$c6f0200a@cisco.com>
	<46E164F2.207@cisco.com>
In-Reply-To: <46E164F2.207@cisco.com>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 10 Sep 2007 12:52:54.0983 (UTC)
	FILETIME=[816DD570:01C7F3A9]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: a7d6aff76b15f3f56fcb94490e1052e4
Cc: behave@ietf.org, mmusic@ietf.org, Dan Wing <dwing@cisco.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Jonathan Rosenberg skrev:
> To be specific, let me propose the following:
> 
> rfc3489bis: says the username MUST be less than 513 bytes. It also says
> that the STUN message when sent over UDP MUST be less than the MTU.
> 
> ICE: says that the username fragment MUST be less than 256 bytes.
> 
> 
> The desire for the ability to have larger usernames is not security
> (i.e., this doesn't mean endpoints compute username fragments with 256
> bytes of entropy), but that cases have been identified ala Dan's draft
> of stuffing useful identifiers in there. So we want to allow that to be
> possible in the future. Length limits always need to be carefully
> managed, since they are nice in terms of implementation but you can
> sometimes regret them in a really big way down the road (think SMS
> message sizes, 640k memory limit, etc.)
> 

So do I interpret this correctly, in that STUN will have byte limit
which is lower than in the previous draft version, but if one uses
characters that has few bytes per character they will be allowed to be
longer than 128? And now ICE and STUN will be aligned.

Please comment on this! We need feedback to determine that there are
consensus on this change so that we can go. I do not plan to do another
WG last call on the complete document, more than give people a few days
to scream about the new update to resolve the issues as soon as there
are consensus on the solution.

Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 10 08:58:26 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUiqP-0006mk-Nu; Mon, 10 Sep 2007 08:58:25 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IUiqO-0006mf-Ur
	for behave-confirm+ok@megatron.ietf.org; Mon, 10 Sep 2007 08:58:24 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IUiqO-0006mX-Kh
	for behave@ietf.org; Mon, 10 Sep 2007 08:58:24 -0400
Received: from mailgw4.ericsson.se ([193.180.251.62])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IUiqO-00055y-5p
	for behave@ietf.org; Mon, 10 Sep 2007 08:58:24 -0400
Received: from mailgw4.ericsson.se (unknown [127.0.0.1])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	28A1B21475
	for <behave@ietf.org>; Mon, 10 Sep 2007 14:57:56 +0200 (CEST)
X-AuditID: c1b4fb3e-ae831bb0000007e1-89-46e53f539517
Received: from esealmw128.eemea.ericsson.se (unknown [153.88.254.121])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	CA2A82144B
	for <behave@ietf.org>; Mon, 10 Sep 2007 14:57:55 +0200 (CEST)
Received: from esealmw128.eemea.ericsson.se ([153.88.254.176]) by
	esealmw128.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 14:57:55 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw128.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 14:57:55 +0200
Message-ID: <46E53F53.4030000@ericsson.com>
Date: Mon, 10 Sep 2007 14:57:55 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: behave@ietf.org
Subject: Re: [BEHAVE] I-D Action:draft-ietf-behave-rfc3489bis-09.txt
References: <E1IM3cH-0005ng-O5@stiedprstage1.ietf.org>
In-Reply-To: <E1IM3cH-0005ng-O5@stiedprstage1.ietf.org>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 10 Sep 2007 12:57:55.0481 (UTC)
	FILETIME=[348A3090:01C7F3AA]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 93238566e09e6e262849b4f805833007
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

This version resolved a most of the comments raised during the WG last
call. However one issue and editorial was not covered by this version.
The plan is to request publication as soon as we have consensus on the
single remaining issue and update is made available. So please check
that this version resolves any WG last call comments you had.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-behave-rfc3489bis-09.txt

DIFF:
http://tools.ietf.org/wg/behave/draft-ietf-behave-rfc3489bis/draft-ietf-behave-rfc3489bis-09-from-08.diff.html

Best Regards

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 10 09:15:35 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUj6y-0001yV-KJ; Mon, 10 Sep 2007 09:15:32 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IUj6x-0001yN-Pr
	for behave-confirm+ok@megatron.ietf.org; Mon, 10 Sep 2007 09:15:31 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUj6x-0001yC-FE; Mon, 10 Sep 2007 09:15:31 -0400
Received: from smtp.nokia.com ([131.228.20.172] helo=mgw-ext13.nokia.com)
	by chiedprmail1.ietf.org with esmtp (Exim 4.43)
	id 1IUj6w-0005Y1-U3; Mon, 10 Sep 2007 09:15:31 -0400
Received: from esebh108.NOE.Nokia.com (esebh108.ntc.nokia.com [172.21.143.145])
	by mgw-ext13.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l8ADF28x013103; Mon, 10 Sep 2007 16:15:26 +0300
Received: from esebh104.NOE.Nokia.com ([172.21.143.34]) by
	esebh108.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 16:15:20 +0300
Received: from esebh101.NOE.Nokia.com ([172.21.138.177]) by
	esebh104.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 16:15:20 +0300
Received: from esdhcp04162.research.nokia.com ([172.21.41.62]) by
	esebh101.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 16:15:20 +0300
From: =?iso-8859-1?q?R=E9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: behave@ietf.org
Subject: Re: [MMUSIC] Re: [BEHAVE] Re: STUN/ICE: Username length inconstency
Date: Mon, 10 Sep 2007 16:15:16 +0300
User-Agent: KMail/1.9.6 (enterprise 0.20070831.706792)
References: <200708201437.55515.remi.denis-courmont@nokia.com>
	<46E164F2.207@cisco.com> <46E53E26.1090703@ericsson.com>
In-Reply-To: <46E53E26.1090703@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709101615.16398.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 10 Sep 2007 13:15:20.0337 (UTC)
	FILETIME=[A3529010:01C7F3AC]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 856eb5f76e7a34990d1d457d8e8e5b7f
Cc: Dan Wing <dwing@cisco.com>, mmusic@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Le Monday 10 September 2007 15:52:54 ext Magnus Westerlund, vous avez =E9cr=
it=A0:
> Jonathan Rosenberg skrev:
> > To be specific, let me propose the following:
> >
> > rfc3489bis: says the username MUST be less than 513 bytes. It also says
> > that the STUN message when sent over UDP MUST be less than the MTU.
> >
> > ICE: says that the username fragment MUST be less than 256 bytes.

That's fine with me. I haven't computed if that made ICE connectivity check=
s=20
less than 576 bytes, but at least it ensures they will be under any practic=
al=20
MTU limit on the real Internet.

=2D-=20
R=E9mi Denis-Courmont


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 10 09:19:30 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUjAo-0007qc-E7; Mon, 10 Sep 2007 09:19:30 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IUjAn-0007pe-UQ
	for behave-confirm+ok@megatron.ietf.org; Mon, 10 Sep 2007 09:19:29 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IUjAn-0007on-K4
	for behave@ietf.org; Mon, 10 Sep 2007 09:19:29 -0400
Received: from smtp.nokia.com ([131.228.20.171] helo=mgw-ext12.nokia.com)
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IUjAn-0005gH-4a
	for behave@ietf.org; Mon, 10 Sep 2007 09:19:29 -0400
Received: from esebh106.NOE.Nokia.com (esebh106.ntc.nokia.com [172.21.138.213])
	by mgw-ext12.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l8ADJ5to025057; Mon, 10 Sep 2007 16:19:26 +0300
Received: from esebh103.NOE.Nokia.com ([172.21.143.33]) by
	esebh106.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 16:19:09 +0300
Received: from esebh102.NOE.Nokia.com ([172.21.138.183]) by
	esebh103.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 16:19:09 +0300
Received: from esdhcp04162.research.nokia.com ([172.21.41.62]) by
	esebh102.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 10 Sep 2007 16:19:08 +0300
From: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: behave@ietf.org
Subject: Re: [BEHAVE] I-D Action:draft-ietf-behave-rfc3489bis-09.txt
Date: Mon, 10 Sep 2007 16:19:04 +0300
User-Agent: KMail/1.9.6 (enterprise 0.20070831.706792)
References: <E1IM3cH-0005ng-O5@stiedprstage1.ietf.org>
	<46E53F53.4030000@ericsson.com>
In-Reply-To: <46E53F53.4030000@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709101619.04747.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 10 Sep 2007 13:19:08.0387 (UTC)
	FILETIME=[2B403B30:01C7F3AD]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 7a6398bf8aaeabc7a7bb696b6b0a2aad
Cc: 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Le Monday 10 September 2007 15:57:55 ext Magnus Westerlund, vous avez =C3=
=A9crit=C2=A0:
> This version resolved a most of the comments raised during the WG last
> call. However one issue and editorial was not covered by this version.
> The plan is to request publication as soon as we have consensus on the
> single remaining issue and update is made available. So please check
> that this version resolves any WG last call comments you had.

Apart from s the USERNAME length inconsistency with ICE, -09 looks good to =
me.

=2D-=20
R=C3=A9mi Denis-Courmont


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 10 14:15:35 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUnnK-00042o-IW; Mon, 10 Sep 2007 14:15:34 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IUnmq-0002zG-BK
	for behave-confirm+ok@megatron.ietf.org; Mon, 10 Sep 2007 14:15:04 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IUnmo-0002yr-O6; Mon, 10 Sep 2007 14:15:02 -0400
Received: from ns1.neustar.com ([2001:503:c779:1a::9c9a:108a])
	by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IUnmo-0005Cg-9m; Mon, 10 Sep 2007 14:15:02 -0400
Received: from stiedprstage1.ietf.org (stiedprstage1.va.neustar.com
	[10.31.47.10]) by ns1.neustar.com (Postfix) with ESMTP id 1336126E82;
	Mon, 10 Sep 2007 18:15:02 +0000 (GMT)
Received: from ietf by stiedprstage1.ietf.org with local (Exim 4.43)
	id 1IUnmn-000619-Uc; Mon, 10 Sep 2007 14:15:01 -0400
Content-Type: Multipart/Mixed; Boundary="NextPart"
Mime-Version: 1.0
To: i-d-announce@ietf.org
From: Internet-Drafts@ietf.org
Message-Id: <E1IUnmn-000619-Uc@stiedprstage1.ietf.org>
Date: Mon, 10 Sep 2007 14:15:01 -0400
X-Spam-Score: -1.4 (-)
X-Scan-Signature: c3a18ef96977fc9bcc21a621cbf1174b
Cc: behave@ietf.org
Subject: [BEHAVE] I-D ACTION:draft-ietf-behave-multicast-10.txt 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts 
directories.
This draft is a work item of the Behavior Engineering for Hindrance Avoidance Working Group of the IETF.

	Title		: IP Multicast Requirements for a Network Address (and port) Translator (NAT)
	Author(s)	: T. Eckert, D. Wing
	Filename	: draft-ietf-behave-multicast-10.txt
	Pages		: 16
	Date		: 2007-9-10
	
This document specifies requirements for a Network Address (and port)
   Translator (NAT) that supports Any Source IP Multicast or Source-
   Specific IP Multicast.  An IP multicast-capable NAT device that
   adheres to the requirements of this document can optimize the
   operation of IP multicast applications that are generally unaware of
   IP multicast NAT devices.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-behave-multicast-10.txt

To remove yourself from the I-D Announcement list, send a message to 
i-d-announce-request@ietf.org with the word unsubscribe in the body of 
the message. 
You can also visit https://www1.ietf.org/mailman/listinfo/I-D-announce 
to change your subscription settings.

Internet-Drafts are also available by anonymous FTP. Login with the 
username "anonymous" and a password of your e-mail address. After 
logging in, type "cd internet-drafts" and then 
"get draft-ietf-behave-multicast-10.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html 
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt

Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-behave-multicast-10.txt".
	
NOTE:	The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.

Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body; access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID: <2007-9-10131006.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-behave-multicast-10.txt

--OtherAccess
Content-Type: Message/External-body; name="draft-ietf-behave-multicast-10.txt";
	site="ftp.ietf.org"; access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID: <2007-9-10131006.I-D@ietf.org>


--OtherAccess--

--NextPart
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

--NextPart--






From behave-bounces@ietf.org Tue Sep 11 05:46:52 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IV2KY-0007FQ-J1; Tue, 11 Sep 2007 05:46:50 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IV2KX-0007FK-Er
	for behave-confirm+ok@megatron.ietf.org; Tue, 11 Sep 2007 05:46:49 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IV2KX-0007FC-49
	for behave@ietf.org; Tue, 11 Sep 2007 05:46:49 -0400
Received: from mailgw3.ericsson.se ([193.180.251.60])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IV2KW-0006vv-CL
	for behave@ietf.org; Tue, 11 Sep 2007 05:46:48 -0400
Received: from mailgw3.ericsson.se (unknown [127.0.0.1])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	D11FE20780
	for <behave@ietf.org>; Tue, 11 Sep 2007 11:46:47 +0200 (CEST)
X-AuditID: c1b4fb3c-b0e80bb0000007e1-6c-46e664073eb6
Received: from esealmw126.eemea.ericsson.se (unknown [153.88.254.123])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	B690B2076E
	for <behave@ietf.org>; Tue, 11 Sep 2007 11:46:47 +0200 (CEST)
Received: from esealmw128.eemea.ericsson.se ([153.88.254.172]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 11 Sep 2007 11:46:47 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw128.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 11 Sep 2007 11:46:47 +0200
Message-ID: <46E66407.2060604@ericsson.com>
Date: Tue, 11 Sep 2007 11:46:47 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: behave@ietf.org
References: <E1IUnmn-000619-Uc@stiedprstage1.ietf.org>
In-Reply-To: <E1IUnmn-000619-Uc@stiedprstage1.ietf.org>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 11 Sep 2007 09:46:47.0354 (UTC)
	FILETIME=[AB6AA1A0:01C7F458]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: b7b9551d71acde901886cc48bfc088a6
Subject: [BEHAVE] Re: I-D ACTION:draft-ietf-behave-multicast-10.txt
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Hi,

I have now requested publication of this document as BCP.

/Magnus

Internet-Drafts@ietf.org skrev:
> A New Internet-Draft is available from the on-line Internet-Drafts 
> directories.
> This draft is a work item of the Behavior Engineering for Hindrance Avoidance Working Group of the IETF.
> 
> 	Title		: IP Multicast Requirements for a Network Address (and port) Translator (NAT)
> 	Author(s)	: T. Eckert, D. Wing
> 	Filename	: draft-ietf-behave-multicast-10.txt
> 	Pages		: 16
> 	Date		: 2007-9-10
> 	
> This document specifies requirements for a Network Address (and port)
>    Translator (NAT) that supports Any Source IP Multicast or Source-
>    Specific IP Multicast.  An IP multicast-capable NAT device that
>    adheres to the requirements of this document can optimize the
>    operation of IP multicast applications that are generally unaware of
>    IP multicast NAT devices.
> 
> A URL for this Internet-Draft is:
> http://www.ietf.org/internet-drafts/draft-ietf-behave-multicast-10.txt
> 
> To remove yourself from the I-D Announcement list, send a message to 
> i-d-announce-request@ietf.org with the word unsubscribe in the body of 
> the message. 
> You can also visit https://www1.ietf.org/mailman/listinfo/I-D-announce 
> to change your subscription settings.
> 
> Internet-Drafts are also available by anonymous FTP. Login with the 
> username "anonymous" and a password of your e-mail address. After 
> logging in, type "cd internet-drafts" and then 
> "get draft-ietf-behave-multicast-10.txt".
> 
> A list of Internet-Drafts directories can be found in
> http://www.ietf.org/shadow.html 
> or ftp://ftp.ietf.org/ietf/1shadow-sites.txt
> 
> Internet-Drafts can also be obtained by e-mail.
> 
> Send a message to:
> 	mailserv@ietf.org.
> In the body type:
> 	"FILE /internet-drafts/draft-ietf-behave-multicast-10.txt".
> 	
> NOTE:	The mail server at ietf.org can return the document in
> 	MIME-encoded form by using the "mpack" utility.  To use this
> 	feature, insert the command "ENCODING mime" before the "FILE"
> 	command.  To decode the response(s), you will need "munpack" or
> 	a MIME-compliant mail reader.  Different MIME-compliant mail readers
> 	exhibit different behavior, especially when dealing with
> 	"multipart" MIME messages (i.e. documents which have been split
> 	up into multiple messages), so check your local documentation on
> 	how to manipulate these messages.
> 
> Below is the data which will enable a MIME compliant mail reader
> implementation to automatically retrieve the ASCII version of the
> Internet-Draft.
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> I-D-Announce mailing list
> I-D-Announce@ietf.org
> https://www1.ietf.org/mailman/listinfo/i-d-announce


-- 

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 11 10:10:42 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IV6Rq-0000vR-Ev; Tue, 11 Sep 2007 10:10:38 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IV6Ro-0000qm-Kd
	for behave-confirm+ok@megatron.ietf.org; Tue, 11 Sep 2007 10:10:36 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IV6Rn-0000o1-MM; Tue, 11 Sep 2007 10:10:35 -0400
Received: from ns0.neustar.com ([156.154.16.158])
	by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IV6Rm-0003n4-Iq; Tue, 11 Sep 2007 10:10:35 -0400
Received: from stiedprstage1.ietf.org (stiedprstage1.va.neustar.com
	[10.31.47.10]) by ns0.neustar.com (Postfix) with ESMTP id 8790932880;
	Tue, 11 Sep 2007 14:10:34 +0000 (GMT)
Received: from ietf by stiedprstage1.ietf.org with local (Exim 4.43)
	id 1IV6Rm-0003rR-F5; Tue, 11 Sep 2007 10:10:34 -0400
X-test-idtracker: no
To: IETF-Announce <ietf-announce@ietf.org>
From: The IESG <iesg-secretary@ietf.org>
Message-Id: <E1IV6Rm-0003rR-F5@stiedprstage1.ietf.org>
Date: Tue, 11 Sep 2007 10:10:34 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 79899194edc4f33a41f49410777972f8
Cc: behave@ietf.org
Subject: [BEHAVE] Last Call: draft-ietf-behave-multicast (IP Multicast 
 Requirements for a Network Address (and port) Translator (NAT)) 
 to BCP 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: ietf@ietf.org
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

The IESG has received a request from the Behavior Engineering for 
Hindrance Avoidance WG (behave) to consider the following document:

- 'IP Multicast Requirements for a Network Address (and port) Translator
   (NAT) '
   <draft-ietf-behave-multicast-10.txt> as a BCP

The IESG plans to make a decision in the next few weeks, and solicits
final comments on this action.  Please send substantive comments to the
ietf@ietf.org mailing lists by 2007-09-25. Exceptionally, 
comments may be sent to iesg@ietf.org instead. In either case, please 
retain the beginning of the Subject line to allow automated sorting.

The file can be obtained via
http://www.ietf.org/internet-drafts/draft-ietf-behave-multicast-10.txt


IESG discussion can be tracked via
https://datatracker.ietf.org/public/pidtracker.cgi?command=view_id&dTag=13134&rfc_flag=0



_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Wed Sep 12 12:59:30 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVVYm-0004mS-Hx; Wed, 12 Sep 2007 12:59:28 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVVYl-0004hi-OK
	for behave-confirm+ok@megatron.ietf.org; Wed, 12 Sep 2007 12:59:27 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IVVYl-0004fT-9M
	for behave@ietf.org; Wed, 12 Sep 2007 12:59:27 -0400
Received: from sj-iport-6.cisco.com ([171.71.176.117])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IVVYk-0001mW-4S
	for behave@ietf.org; Wed, 12 Sep 2007 12:59:27 -0400
X-IronPort-AV: E=Sophos;i="4.20,245,1186383600"; d="scan'208";a="216849732"
Received: from sj-dkim-1.cisco.com ([171.71.179.21])
	by sj-iport-6.cisco.com with ESMTP; 12 Sep 2007 09:59:25 -0700
Received: from sj-core-1.cisco.com (sj-core-1.cisco.com [171.71.177.237])
	by sj-dkim-1.cisco.com (8.12.11/8.12.11) with ESMTP id l8CGxPxk007439
	for <behave@ietf.org>; Wed, 12 Sep 2007 09:59:25 -0700
Received: from xbh-sjc-231.amer.cisco.com (xbh-sjc-231.cisco.com
	[128.107.191.100])
	by sj-core-1.cisco.com (8.12.10/8.12.6) with ESMTP id l8CGxPEw027735
	for <behave@ietf.org>; Wed, 12 Sep 2007 16:59:25 GMT
Received: from xfe-sjc-211.amer.cisco.com ([171.70.151.174]) by
	xbh-sjc-231.amer.cisco.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 12 Sep 2007 09:59:25 -0700
Received: from [10.32.241.146] ([10.32.241.146]) by xfe-sjc-211.amer.cisco.com
	with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 12 Sep 2007 09:59:25 -0700
Message-ID: <46E81AFE.6040705@cisco.com>
Date: Wed, 12 Sep 2007 12:59:42 -0400
From: Jonathan Rosenberg <jdrosen@cisco.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Behave WG <behave@ietf.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 12 Sep 2007 16:59:25.0118 (UTC)
	FILETIME=[45DCD5E0:01C7F55E]
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=8381; t=1189616365;
	x=1190480365; c=relaxed/simple; s=sjdkim1004;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=jdrosen@cisco.com;
	z=From:=20Jonathan=20Rosenberg=20<jdrosen@cisco.com>
	|Subject:=20TURN=20-=20some=20proposed=20changes=20for=20improvement=20an
	d=20simplification |Sender:=20;
	bh=A4mVQOBEVVekPPjMSIygY31EARoX7gQ3OGFUGO2RZKI=;
	b=UJKxwxYI474GbeByp+2eQ5T8R+BEImI4BnW0gqRxaqkDa+VZ9yGzzghcFYFjbLkgj9WcB/2M
	s5RvdG0jnH/mIdHdvdnJpcAxCM745nDUf4Wx40dCeZ2yiBtXT8nn3qPHPVzEoKQTZQmLI1ZVpH
	DUciwBFLccf6AzkDjxRWG/CBM=;
Authentication-Results: sj-dkim-1; header.From=jdrosen@cisco.com; dkim=pass (
	sig from cisco.com/sjdkim1004 verified; ); 
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 0ff9c467ad7f19c2a6d058acd7faaec8
Subject: [BEHAVE] TURN - some proposed changes for improvement and
	simplification
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org


As Dan mentioned at the last IETF meeting, we've formed a small team to
actively work on TURN over the next month, in order to accelerate its
completion and produce a version ready for WGLC by the end of September.

As part of this, we took a step back and looked a bit at some of the
more troubling aspects of TURN. Three in particular seemed important:

* There are several different methods of framing and encapsulation of 
messages with a number of exceptions based on transport protocol, 
contents of the data, and the active destination.  The overall effect is 
confusing and rather complex.

* The active destination concept is inelegant, confusing, complex, and 
during forking results in undesirable overhead and at least one 
timing-related race condition.

* the use of multiple TCP connections from the TURN server towards
peers, when there is one TCP connection from the TURN server to the
client, has been a concern on the list recently.

We have a proposal on some non-trivial changes to TURN on these topics.

Firstly, we can eliminate the entire need for active destinations (and
the Send and Data primitives as well) by introducing the concept of a
'channel'. Each channel is a stream of application data sent to and from
a remote IP and port. In the case of UDP, we'd add a SINGLE BYTE to the
front of every packet to and from the TURN server, which is used to
signal the channel number. Channel 0 is reserved for TURN signaling to
and from the client and TURN server. The remaining 255 channels are for
remote destinations and carry pure data after the 1-byte channel number.
So when a client does a Send or a Connect, the server allocates a
remaining channel number and passes this channel number back to the
client. From that point on, when the TURN server receives data from the
peer, it sends it to the client and adds only the byte up front
containing the channel number. Similarly, when the client wants to send
data, it just adds this byte up front indicating the channel number, and
the TURN server sends the data to the corresponding remote peer. There
is no encapsulation of data as a TURN message ever. Thus, the Data
indication is removed. TURN currently requires the server to check for
the magic cookie in packets received from remote peers, and when it sees
it, encapsulate the resulting data in a Data indication. That
requirement would also be eliminated.

Consider a forking use case where A calls a URI and this forks to B and
C. Both send a 200 OK. A will end up running ICE simultaneously with
both B and C. All of the STUN checks and any resulting RTP would just be
sent over different channels, all with identical overhead (1 byte).
There would be no need to re-invite, as there would typically be in the
current TURN mechanism.

This also fixes the complexity that was raised on the list for
packetcable; the current mechanism complicates their QOS mechanisms
because the bandwidth for the stream is not deterministic. Here, it is
deterministic. Every packet always has an extra 1 byte per packet overhead.

The main reason we've avoided any kind of framing for UDP up until now
has been overhead. So what kind of overhead does this introduce? Well if
we assume something really nasty like g729 with 20ms framing, thats 20
bytes of payload plus 40 for RTP/UDP/IP gives us 60, which now goes to
61. Thats a 1.6 percent penalty which is pretty good, considering
especially the huge overhead you are already paying for RTP/UDP. Anyway
a more realistic number is g711 with 30ms framing, and there is .35%
overhead. This cost seems reasonable given the simplifications it brings.

Another possible concern with the 1-byte overhead is path MTU. If a UDP
packet was sent from a remote peer and it was exactly equal to the path
MTU from the remote peer to the TURN server, this will now fragment on
the TURN server to client link. Its not clear that this is a real
problem. But if you are aware of UDP applications that send data exactly
at the MTU for some reason, please speak up.

For TCP, the channel mechanism would work the same. The current framing
already has a 'type' field which we'd just turn into a channel number.
Thus both UDP and TCP would have framing, the difference is only that
TCP includes a length whereas UDP does not. Protocol operation is
otherwise more or less the same, and this would forge closer alignment
between UDP and TCP. As with UDP, this allows multiple TCP connections
from the TURN server to remote peers, each of which would have its own
channel number, allocated to the client when permissions are created for
a remote IP address, or when a connect request is sent. Connection
status indications would all come over the control channel, and each
such indication would contain, as an attribute, the channel number that
the status indication is associated with.

Now, that said, TCP has some additional complexities that UDP does not.
First and foremost is that you need to be very careful with how you
handle buffering. We want to avoid the case where data arrives quickly
on the connection from the client to the TURN server, but the TCP
connection from the TURN server to the remote peer is slow, and so data
is lost at the TURN server when buffers overflow. To handle this, we
will specify a buffering requirement and a possiblemechanism. There 
would be a maximum buffer size  per connection set in the kernel (for 
example). The TURN server application code reads  off of this buffer and 
writes to the other TCP connection. If it cannot write because the 
window is full, it doesn't read data from the incoming
connection. This will eventually cause the buffers in the kernel to
fill, and then packets are dropped (and not ACKed), causing backpressure
on the TCP connections. This effectively bridges the congestion control
between the connections, and ensures that, TURN server failures aside,
no packets get dropped in the TURN server. We believe that this kind of
mechanism is used in other similar boxes like HTTP proxies.

Now, lets consider the case of multiple TCP connections to a
multiplicity of remote peers. There have been suggestions on the list to
just disallow this. We cannot. It happens as an artifact of forking.
Consider client A that sends an offer. Amongst its candidates is a
passive TCP candidate learned from the TURN server. THis INVITE forks to
B and C, each of which return a single active candidate. Now, A will
need to open permissions on the TURN server for B and C's IP addresses
listed in the SDP answer, and both B and C will open a TCP connection
towards the TURN server in order for ICE checks to proceed. Note well,
that since ICE processing happens ideally prior to call answering, this
case happens NOT JUST when both B and C answer at the same time, but
whenever there is forking at all with ICE-tcp. So this is actually a
common case and not a corner case.

In this case of multiple TCP connections to remote peers, the congestion
control bridging described above gets complicated, and head-of-line
blocking gets introduced. Congestion on the connection from the TURN
server to one peer can slow down the connection to the client, and thus
limit the throughput to the other peer. This is only a problem when we
are transferring real data, not so much for things like STUN messages or
TLS handshakes. So the recommendation is going to be that, if a client
ends up in a case with multiple TCP connections that survive past the
connectivity checks, in that case the spec will recommend the client
allocate a new TCP candidate from the TURN server and restart ICE with
one of the peers, in order to eliminate the HOL blocking.

We're looking for feedback on this proposal (and questions and
clarifications of course) ASAP - within the next week. If this seems
like a good direction, we can work these changes in and address the
other comments and still deliver a revision on schedule.

Thanks,
Jonathan R.
-- 
Jonathan D. Rosenberg, Ph.D.                   600 Lanidex Plaza
Cisco Fellow                                   Parsippany, NJ 07054-2711
Cisco Systems
jdrosen@cisco.com                              FAX:   (973) 952-5050
http://www.jdrosen.net                         PHONE: (973) 952-5000
http://www.cisco.com


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Wed Sep 12 13:27:38 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVVzy-0004xS-Ig; Wed, 12 Sep 2007 13:27:34 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVVzx-0004xN-OA
	for behave-confirm+ok@megatron.ietf.org; Wed, 12 Sep 2007 13:27:33 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IVVzx-0004xF-ES
	for behave@ietf.org; Wed, 12 Sep 2007 13:27:33 -0400
Received: from ondar.cablelabs.com ([192.160.73.61])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IVVzv-0004w3-VR
	for behave@ietf.org; Wed, 12 Sep 2007 13:27:33 -0400
Received: from kyzyl.cablelabs.com (kyzyl.cablelabs.com [10.253.0.7])
	by ondar.cablelabs.com (8.13.8/8.13.8) with ESMTP id l8CHRUcj016853;
	Wed, 12 Sep 2007 11:27:30 -0600
Received: from srvxchg3.cablelabs.com (10.5.0.25)
	by kyzyl.cablelabs.com (F-Secure/fsigk_smtp/511/kyzyl.cablelabs.com);
	Wed, 12 Sep 2007 11:27:29 -0700 (MST)
X-Virus-Status: clean(F-Secure/fsigk_smtp/511/kyzyl.cablelabs.com)
X-MimeOLE: Produced By Microsoft Exchange V6.5
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [BEHAVE] TURN - some proposed changes for improvement
	andsimplification
Date: Wed, 12 Sep 2007 11:27:28 -0600
Message-ID: <9AAEDF491EF7CA48AB587781B8F5D7C65E0790@srvxchg3.cablelabs.com>
In-Reply-To: <46E81AFE.6040705@cisco.com>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: [BEHAVE] TURN - some proposed changes for improvement
	andsimplification
Thread-Index: Acf1Xky3pX4X9g9fSSmv88TgddD7twAAuFHg
References: <46E81AFE.6040705@cisco.com>
From: "Kevin Johns" <K.Johns@CableLabs.com>
To: "Jonathan Rosenberg" <jdrosen@cisco.com>, "Behave WG" <behave@ietf.org>
X-Approved: ondar
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 2bf730a014b318fd3efd65b39b48818c
Cc: 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Jonathan,

Thanks for the lengthy write-up on the proposal. It does naturally raise
a few questions for clarification...

How does the channel concept work if the TURN server receives data from
a peer prior to the client setting up a channel for that peer? In the
current model, that data would be forwarded to the client with a receive
indication which does not require any action by the client. This was
both nice and not-so-nice in that a client did not have to communicate
with the TURN server if a call forked and could wait for the 200 OK to
set the active destination. It did create some QoS complexity, but
minimized client communication with the TURN server.

The other question is around the details of how a channel is setup
between the client and TURN server and how this interacts with ICE. When
in the call setup/ICE process would the client setup a channel with the
TURN server? Would the client setup a channel for each candidate
received or only those which are validated as part of the ICE
connectivity checks?

Thanks
Kevin
=20

-----Original Message-----
From: Jonathan Rosenberg [mailto:jdrosen@cisco.com]=20
Sent: Wednesday, September 12, 2007 11:00 AM
To: Behave WG
Subject: [BEHAVE] TURN - some proposed changes for improvement
andsimplification


As Dan mentioned at the last IETF meeting, we've formed a small team to
actively work on TURN over the next month, in order to accelerate its
completion and produce a version ready for WGLC by the end of September.

As part of this, we took a step back and looked a bit at some of the
more troubling aspects of TURN. Three in particular seemed important:

* There are several different methods of framing and encapsulation of
messages with a number of exceptions based on transport protocol,
contents of the data, and the active destination.  The overall effect is
confusing and rather complex.

* The active destination concept is inelegant, confusing, complex, and
during forking results in undesirable overhead and at least one
timing-related race condition.

* the use of multiple TCP connections from the TURN server towards
peers, when there is one TCP connection from the TURN server to the
client, has been a concern on the list recently.

We have a proposal on some non-trivial changes to TURN on these topics.

Firstly, we can eliminate the entire need for active destinations (and
the Send and Data primitives as well) by introducing the concept of a
'channel'. Each channel is a stream of application data sent to and from
a remote IP and port. In the case of UDP, we'd add a SINGLE BYTE to the
front of every packet to and from the TURN server, which is used to
signal the channel number. Channel 0 is reserved for TURN signaling to
and from the client and TURN server. The remaining 255 channels are for
remote destinations and carry pure data after the 1-byte channel number.
So when a client does a Send or a Connect, the server allocates a
remaining channel number and passes this channel number back to the
client. From that point on, when the TURN server receives data from the
peer, it sends it to the client and adds only the byte up front
containing the channel number. Similarly, when the client wants to send
data, it just adds this byte up front indicating the channel number, and
the TURN server sends the data to the corresponding remote peer. There
is no encapsulation of data as a TURN message ever. Thus, the Data
indication is removed. TURN currently requires the server to check for
the magic cookie in packets received from remote peers, and when it sees
it, encapsulate the resulting data in a Data indication. That
requirement would also be eliminated.

Consider a forking use case where A calls a URI and this forks to B and
C. Both send a 200 OK. A will end up running ICE simultaneously with
both B and C. All of the STUN checks and any resulting RTP would just be
sent over different channels, all with identical overhead (1 byte).
There would be no need to re-invite, as there would typically be in the
current TURN mechanism.

This also fixes the complexity that was raised on the list for
packetcable; the current mechanism complicates their QOS mechanisms
because the bandwidth for the stream is not deterministic. Here, it is
deterministic. Every packet always has an extra 1 byte per packet
overhead.

The main reason we've avoided any kind of framing for UDP up until now
has been overhead. So what kind of overhead does this introduce? Well if
we assume something really nasty like g729 with 20ms framing, thats 20
bytes of payload plus 40 for RTP/UDP/IP gives us 60, which now goes to
61. Thats a 1.6 percent penalty which is pretty good, considering
especially the huge overhead you are already paying for RTP/UDP. Anyway
a more realistic number is g711 with 30ms framing, and there is .35%
overhead. This cost seems reasonable given the simplifications it
brings.

Another possible concern with the 1-byte overhead is path MTU. If a UDP
packet was sent from a remote peer and it was exactly equal to the path
MTU from the remote peer to the TURN server, this will now fragment on
the TURN server to client link. Its not clear that this is a real
problem. But if you are aware of UDP applications that send data exactly
at the MTU for some reason, please speak up.

For TCP, the channel mechanism would work the same. The current framing
already has a 'type' field which we'd just turn into a channel number.
Thus both UDP and TCP would have framing, the difference is only that
TCP includes a length whereas UDP does not. Protocol operation is
otherwise more or less the same, and this would forge closer alignment
between UDP and TCP. As with UDP, this allows multiple TCP connections
from the TURN server to remote peers, each of which would have its own
channel number, allocated to the client when permissions are created for
a remote IP address, or when a connect request is sent. Connection
status indications would all come over the control channel, and each
such indication would contain, as an attribute, the channel number that
the status indication is associated with.

Now, that said, TCP has some additional complexities that UDP does not.
First and foremost is that you need to be very careful with how you
handle buffering. We want to avoid the case where data arrives quickly
on the connection from the client to the TURN server, but the TCP
connection from the TURN server to the remote peer is slow, and so data
is lost at the TURN server when buffers overflow. To handle this, we
will specify a buffering requirement and a possiblemechanism. There
would be a maximum buffer size  per connection set in the kernel (for
example). The TURN server application code reads  off of this buffer and
writes to the other TCP connection. If it cannot write because the
window is full, it doesn't read data from the incoming connection. This
will eventually cause the buffers in the kernel to fill, and then
packets are dropped (and not ACKed), causing backpressure on the TCP
connections. This effectively bridges the congestion control between the
connections, and ensures that, TURN server failures aside, no packets
get dropped in the TURN server. We believe that this kind of mechanism
is used in other similar boxes like HTTP proxies.

Now, lets consider the case of multiple TCP connections to a
multiplicity of remote peers. There have been suggestions on the list to
just disallow this. We cannot. It happens as an artifact of forking.
Consider client A that sends an offer. Amongst its candidates is a
passive TCP candidate learned from the TURN server. THis INVITE forks to
B and C, each of which return a single active candidate. Now, A will
need to open permissions on the TURN server for B and C's IP addresses
listed in the SDP answer, and both B and C will open a TCP connection
towards the TURN server in order for ICE checks to proceed. Note well,
that since ICE processing happens ideally prior to call answering, this
case happens NOT JUST when both B and C answer at the same time, but
whenever there is forking at all with ICE-tcp. So this is actually a
common case and not a corner case.

In this case of multiple TCP connections to remote peers, the congestion
control bridging described above gets complicated, and head-of-line
blocking gets introduced. Congestion on the connection from the TURN
server to one peer can slow down the connection to the client, and thus
limit the throughput to the other peer. This is only a problem when we
are transferring real data, not so much for things like STUN messages or
TLS handshakes. So the recommendation is going to be that, if a client
ends up in a case with multiple TCP connections that survive past the
connectivity checks, in that case the spec will recommend the client
allocate a new TCP candidate from the TURN server and restart ICE with
one of the peers, in order to eliminate the HOL blocking.

We're looking for feedback on this proposal (and questions and
clarifications of course) ASAP - within the next week. If this seems
like a good direction, we can work these changes in and address the
other comments and still deliver a revision on schedule.

Thanks,
Jonathan R.
--=20
Jonathan D. Rosenberg, Ph.D.                   600 Lanidex Plaza
Cisco Fellow                                   Parsippany, NJ 07054-2711
Cisco Systems
jdrosen@cisco.com                              FAX:   (973) 952-5050
http://www.jdrosen.net                         PHONE: (973) 952-5000
http://www.cisco.com


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Wed Sep 12 13:30:35 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVW2s-0003Yw-Ks; Wed, 12 Sep 2007 13:30:34 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVW2r-0003Yo-83
	for behave-confirm+ok@megatron.ietf.org; Wed, 12 Sep 2007 13:30:33 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVW2q-0003Yg-Ub; Wed, 12 Sep 2007 13:30:32 -0400
Received: from sj-iport-4.cisco.com ([171.68.10.86])
	by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IVW2p-0004yM-M0; Wed, 12 Sep 2007 13:30:32 -0400
X-IronPort-AV: E=Sophos;i="4.20,245,1186383600"; 
   d="scan'208";a="9935789"
Received: from sj-dkim-2.cisco.com ([171.71.179.186])
	by sj-iport-4.cisco.com with ESMTP; 12 Sep 2007 10:30:27 -0700
Received: from sj-core-2.cisco.com (sj-core-2.cisco.com [171.71.177.254])
	by sj-dkim-2.cisco.com (8.12.11/8.12.11) with ESMTP id l8CHUQVY026379; 
	Wed, 12 Sep 2007 10:30:26 -0700
Received: from xbh-sjc-211.amer.cisco.com (xbh-sjc-211.cisco.com
	[171.70.151.144])
	by sj-core-2.cisco.com (8.12.10/8.12.6) with ESMTP id l8CHUQat021373;
	Wed, 12 Sep 2007 17:30:26 GMT
Received: from xfe-sjc-212.amer.cisco.com ([171.70.151.187]) by
	xbh-sjc-211.amer.cisco.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 12 Sep 2007 10:30:24 -0700
Received: from [10.32.241.146] ([10.32.241.146]) by xfe-sjc-212.amer.cisco.com
	with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 12 Sep 2007 10:30:24 -0700
Message-ID: <46E82240.5080108@cisco.com>
Date: Wed, 12 Sep 2007 13:30:40 -0400
From: Jonathan Rosenberg <jdrosen@cisco.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Magnus Westerlund <magnus.westerlund@ericsson.com>
Subject: Re: [MMUSIC] Re: [BEHAVE] Re: STUN/ICE: Username length inconstency
References: <200708201437.55515.remi.denis-courmont@nokia.com><46CA12C7.6020302@cisco.com>	<200708210926.45072.remi.denis-courmont@nokia.com>	<243901c7ef4d$05723700$c6f0200a@cisco.com>
	<46E164F2.207@cisco.com> <46E53E26.1090703@ericsson.com>
In-Reply-To: <46E53E26.1090703@ericsson.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 12 Sep 2007 17:30:24.0281 (UTC)
	FILETIME=[9A029890:01C7F562]
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=1690; t=1189618226;
	x=1190482226; c=relaxed/simple; s=sjdkim2002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=jdrosen@cisco.com;
	z=From:=20Jonathan=20Rosenberg=20<jdrosen@cisco.com>
	|Subject:=20Re=3A=20[MMUSIC]=20Re=3A=20[BEHAVE]=20Re=3A=20STUN/ICE=3A=20U
	sername=20length=20inconstency |Sender:=20;
	bh=1RBm9BIP/R4+zSLe/PlddYryDNtH8Vuko4xaoTdt/E0=;
	b=CFr7QPUCJLOGGBKV+6NDE/A4HqO+Y1kNKor8sKcg2DGmBh4Ezj0G2Ktktj6Mtb2VETy7aN/+
	Y5nT0oOidhgKADnQfkXBXuINKT/0LDRooaSQwkvG/Av1XZWadbLxuNU7;
Authentication-Results: sj-dkim-2; header.From=jdrosen@cisco.com; dkim=pass (
	sig from cisco.com/sjdkim2002 verified; ); 
X-Spam-Score: -4.0 (----)
X-Scan-Signature: 8b30eb7682a596edff707698f4a80f7d
Cc: behave@ietf.org, mmusic@ietf.org, Dan Wing <dwing@cisco.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

one quick response inline:

Magnus Westerlund wrote:
> Jonathan Rosenberg skrev:
>> To be specific, let me propose the following:
>>
>> rfc3489bis: says the username MUST be less than 513 bytes. It also says
>> that the STUN message when sent over UDP MUST be less than the MTU.
>>
>> ICE: says that the username fragment MUST be less than 256 bytes.
>>
>>
>> The desire for the ability to have larger usernames is not security
>> (i.e., this doesn't mean endpoints compute username fragments with 256
>> bytes of entropy), but that cases have been identified ala Dan's draft
>> of stuffing useful identifiers in there. So we want to allow that to be
>> possible in the future. Length limits always need to be carefully
>> managed, since they are nice in terms of implementation but you can
>> sometimes regret them in a really big way down the road (think SMS
>> message sizes, 640k memory limit, etc.)
>>
> 
> So do I interpret this correctly, in that STUN will have byte limit
> which is lower than in the previous draft version, but if one uses
> characters that has few bytes per character they will be allowed to be
> longer than 128? And now ICE and STUN will be aligned.

Yes. The limit is a total of 513 bytes, so this would be 513 characters 
if the characters are single byte (i.e., from the ASCII set).

Thanks,
Jonathan R.
-- 
Jonathan D. Rosenberg, Ph.D.                   600 Lanidex Plaza
Cisco Fellow                                   Parsippany, NJ 07054-2711
Cisco Systems
jdrosen@cisco.com                              FAX:   (973) 952-5050
http://www.jdrosen.net                         PHONE: (973) 952-5000
http://www.cisco.com


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Wed Sep 12 14:40:24 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVX8R-0005Ef-3d; Wed, 12 Sep 2007 14:40:23 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVX8P-0005EI-M5
	for behave-confirm+ok@megatron.ietf.org; Wed, 12 Sep 2007 14:40:21 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IVX8P-0005E9-CI
	for behave@ietf.org; Wed, 12 Sep 2007 14:40:21 -0400
Received: from sj-iport-3-in.cisco.com ([171.71.176.72]
	helo=sj-iport-3.cisco.com) by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IVX8O-0006hw-3L
	for behave@ietf.org; Wed, 12 Sep 2007 14:40:21 -0400
X-IronPort-AV: E=Sophos;i="4.20,245,1186383600"; d="scan'208";a="523214708"
Received: from sj-dkim-2.cisco.com ([171.71.179.186])
	by sj-iport-3.cisco.com with ESMTP; 12 Sep 2007 11:40:20 -0700
Received: from sj-core-2.cisco.com (sj-core-2.cisco.com [171.71.177.254])
	by sj-dkim-2.cisco.com (8.12.11/8.12.11) with ESMTP id l8CIeJj5009411; 
	Wed, 12 Sep 2007 11:40:19 -0700
Received: from xbh-sjc-211.amer.cisco.com (xbh-sjc-211.cisco.com
	[171.70.151.144])
	by sj-core-2.cisco.com (8.12.10/8.12.6) with ESMTP id l8CIeDb3013015;
	Wed, 12 Sep 2007 18:40:15 GMT
Received: from xfe-sjc-212.amer.cisco.com ([171.70.151.187]) by
	xbh-sjc-211.amer.cisco.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 12 Sep 2007 11:40:13 -0700
Received: from [10.32.241.146] ([10.32.241.146]) by xfe-sjc-212.amer.cisco.com
	with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 12 Sep 2007 11:40:13 -0700
Message-ID: <46E8329F.2050701@cisco.com>
Date: Wed, 12 Sep 2007 14:40:31 -0400
From: Jonathan Rosenberg <jdrosen@cisco.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Kevin Johns <K.Johns@CableLabs.com>
Subject: Re: [BEHAVE] TURN - some proposed changes for improvement
	andsimplification
References: <46E81AFE.6040705@cisco.com>
	<9AAEDF491EF7CA48AB587781B8F5D7C65E0790@srvxchg3.cablelabs.com>
In-Reply-To: <9AAEDF491EF7CA48AB587781B8F5D7C65E0790@srvxchg3.cablelabs.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 12 Sep 2007 18:40:13.0294 (UTC)
	FILETIME=[5ADB40E0:01C7F56C]
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=2287; t=1189622419;
	x=1190486419; c=relaxed/simple; s=sjdkim2002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=jdrosen@cisco.com;
	z=From:=20Jonathan=20Rosenberg=20<jdrosen@cisco.com>
	|Subject:=20Re=3A=20[BEHAVE]=20TURN=20-=20some=20proposed=20changes=20for
	=20improvement=20andsimplification |Sender:=20;
	bh=pCATt5i7kxL8N2qyXCCr9Z7O4RpSQBT/TJoK3rxrUpU=;
	b=k3nwNNpQ5FuUkz6M0qIyKCs8CsS8r7ttMdXN7WyyqVGp8pAYjeyeIr7s4q0tlkCpH9trIhP5
	0aIAPL7IZBlMUFJSVdikEq8jBL1dI5qeQRYY3tWQCwnVT06M1D/AkSiV;
Authentication-Results: sj-dkim-2; header.From=jdrosen@cisco.com; dkim=pass (
	sig from cisco.com/sjdkim2002 verified; ); 
X-Spam-Score: -4.0 (----)
X-Scan-Signature: f4c2cf0bccc868e4cc88dace71fb3f44
Cc: Behave WG <behave@ietf.org>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org



Kevin Johns wrote:
> Jonathan,
> 
> Thanks for the lengthy write-up on the proposal. It does naturally raise
> a few questions for clarification...
> 
> How does the channel concept work if the TURN server receives data from
> a peer prior to the client setting up a channel for that peer?

It has been the case for a long time that a TURN server won't just 
accept data from a random peer. You need to set up a permission by 
telling the TURN server the identity of the remote peer. So, when you do 
that, the TURN server will allocate the channel to you at that time.


> In the
> current model, that data would be forwarded to the client with a receive
> indication which does not require any action by the client. 

It had required the client to do a Send. It is the Send that will 
establish the channel.

> This was
> both nice and not-so-nice in that a client did not have to communicate
> with the TURN server if a call forked and could wait for the 200 OK to
> set the active destination. It did create some QoS complexity, but
> minimized client communication with the TURN server.
> 
> The other question is around the details of how a channel is setup
> between the client and TURN server and how this interacts with ICE. When
> in the call setup/ICE process would the client setup a channel with the
> TURN server?

In ICE processing, once the O/A exchanges happens an agent starts to 
send checks. It does so by issuing a Send request to the TURN server. It 
would continue to do that, but instead would issue a message to create a 
channel to the remote peer and then send data over the channel that has 
been allocated. Its possible to combine these into one transaction also.


  Would the client setup a channel for each candidate
> received or only those which are validated as part of the ICE
> connectivity checks?

For each remote candidate it will have a separate channel.

-Jonathan R.


-- 
Jonathan D. Rosenberg, Ph.D.                   600 Lanidex Plaza
Cisco Fellow                                   Parsippany, NJ 07054-2711
Cisco Systems
jdrosen@cisco.com                              FAX:   (973) 952-5050
http://www.jdrosen.net                         PHONE: (973) 952-5000
http://www.cisco.com


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Wed Sep 12 15:36:33 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVY0m-0003W6-Df; Wed, 12 Sep 2007 15:36:32 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVY0l-0003W0-EC
	for behave-confirm+ok@megatron.ietf.org; Wed, 12 Sep 2007 15:36:31 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IVY0l-0003Vl-4B
	for behave@ietf.org; Wed, 12 Sep 2007 15:36:31 -0400
Received: from ondar.cablelabs.com ([192.160.73.61])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IVY0j-000060-PW
	for behave@ietf.org; Wed, 12 Sep 2007 15:36:31 -0400
Received: from kyzyl.cablelabs.com (kyzyl.cablelabs.com [10.253.0.7])
	by ondar.cablelabs.com (8.13.8/8.13.8) with ESMTP id l8CJaSgu002341;
	Wed, 12 Sep 2007 13:36:28 -0600
Received: from srvxchg3.cablelabs.com (10.5.0.25)
	by kyzyl.cablelabs.com (F-Secure/fsigk_smtp/511/kyzyl.cablelabs.com);
	Wed, 12 Sep 2007 13:36:28 -0700 (MST)
X-Virus-Status: clean(F-Secure/fsigk_smtp/511/kyzyl.cablelabs.com)
X-MimeOLE: Produced By Microsoft Exchange V6.5
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [BEHAVE] TURN - some proposed changes for improvement
	andsimplification
Date: Wed, 12 Sep 2007 13:36:48 -0600
Message-ID: <9AAEDF491EF7CA48AB587781B8F5D7C65E07B8@srvxchg3.cablelabs.com>
In-Reply-To: <46E8329F.2050701@cisco.com>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: [BEHAVE] TURN - some proposed changes for improvement
	andsimplification
Thread-Index: Acf1bF7Jj9k2IHDxT8mH7ExXZEjIcQABf7xA
References: <46E81AFE.6040705@cisco.com>
	<9AAEDF491EF7CA48AB587781B8F5D7C65E0790@srvxchg3.cablelabs.com>
	<46E8329F.2050701@cisco.com>
From: "Kevin Johns" <K.Johns@CableLabs.com>
To: "Jonathan Rosenberg" <jdrosen@cisco.com>
X-Approved: ondar
X-Spam-Score: 0.0 (/)
X-Scan-Signature: b280b4db656c3ca28dd62e5e0b03daa8
Cc: Behave WG <behave@ietf.org>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

=20
Some more comments in line, if we can keep the setting up of a channel a
byproduct of sending a connectivity check, this would appear to be a
better approach. I certainly like having a more deterministic overhead.

Kevin
-----Original Message-----
From: Jonathan Rosenberg [mailto:jdrosen@cisco.com]=20
Sent: Wednesday, September 12, 2007 12:41 PM
To: Kevin Johns
Cc: Behave WG
Subject: Re: [BEHAVE] TURN - some proposed changes for improvement
andsimplification



Kevin Johns wrote:
> Jonathan,
>=20
> Thanks for the lengthy write-up on the proposal. It does naturally=20
> raise a few questions for clarification...
>=20
> How does the channel concept work if the TURN server receives data=20
> from a peer prior to the client setting up a channel for that peer?

It has been the case for a long time that a TURN server won't just
accept data from a random peer. You need to set up a permission by
telling the TURN server the identity of the remote peer. So, when you do
that, the TURN server will allocate the channel to you at that time.
<kcj> Right, but I think a permission is automatically created when the
client sends a send request which also contains the binding request for
the connectivity check.


> In the
> current model, that data would be forwarded to the client with a=20
> receive indication which does not require any action by the client.

It had required the client to do a Send. It is the Send that will
establish the channel.

> This was
> both nice and not-so-nice in that a client did not have to communicate

> with the TURN server if a call forked and could wait for the 200 OK to

> set the active destination. It did create some QoS complexity, but=20
> minimized client communication with the TURN server.
>=20
> The other question is around the details of how a channel is setup=20
> between the client and TURN server and how this interacts with ICE.=20
> When in the call setup/ICE process would the client setup a channel=20
> with the TURN server?

In ICE processing, once the O/A exchanges happens an agent starts to
send checks. It does so by issuing a Send request to the TURN server. It
would continue to do that, but instead would issue a message to create a
channel to the remote peer and then send data over the channel that has
been allocated. Its possible to combine these into one transaction also.
<kcj> Right, so the send request would add the remote endpoint to the
permissions list and forward the binding request at the same time. It
would be good if we could keep that as a single transaction.


  Would the client setup a channel for each candidate
> received or only those which are validated as part of the ICE=20
> connectivity checks?

For each remote candidate it will have a separate channel.
<kcj> Right and the channel would be setup as a byproduct of the binding
request (hopefully).

-Jonathan R.


--=20
Jonathan D. Rosenberg, Ph.D.                   600 Lanidex Plaza
Cisco Fellow                                   Parsippany, NJ 07054-2711
Cisco Systems
jdrosen@cisco.com                              FAX:   (973) 952-5050
http://www.jdrosen.net                         PHONE: (973) 952-5000
http://www.cisco.com


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Wed Sep 12 23:05:05 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVf0n-0000uq-UY; Wed, 12 Sep 2007 23:05:02 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVf0m-0000ub-On
	for behave-confirm+ok@megatron.ietf.org; Wed, 12 Sep 2007 23:05:00 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IVf0m-0000uR-Bb
	for behave@ietf.org; Wed, 12 Sep 2007 23:05:00 -0400
Received: from wa-out-1112.google.com ([209.85.146.178])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IVf0l-0008FS-0I
	for behave@ietf.org; Wed, 12 Sep 2007 23:05:00 -0400
Received: by wa-out-1112.google.com with SMTP id k40so578824wah
	for <behave@ietf.org>; Wed, 12 Sep 2007 20:04:58 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=beta;
	h=domainkey-signature:received:received:message-id:date:from:sender:to:subject:cc:mime-version:content-type:content-transfer-encoding:content-disposition:x-google-sender-auth;
	bh=Cu0Wj5ZdplOdApW41NysPiGPHQzs3QoXrfzTCDRr+Pk=;
	b=RL9ja2g3gKfnCZspfSH1Ya4hX6o3amO+eQIlYPE6ERi3CxJJo8kRLepL3aL+S1KRITyv1L4rXentbJdebkLkGO70iyt/dtHh9abDf3hk8oxV1OVju0mN8PL5foDbf58TQhEfsm3wl2xFTep/1kv/kzxzc2iRpuJT1nPRSTGBoFk=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta;
	h=received:message-id:date:from:sender:to:subject:cc:mime-version:content-type:content-transfer-encoding:content-disposition:x-google-sender-auth;
	b=lqSG9NqyzzfJsJ9l5B561wpdneFp9LML3iciqxPbbXh5py31nOGzZwUijfv/GRsMX/bTsUwWmWvTrKzVkDVxEm2vfayBVPEOtyHDfUuBAlK+UM/Pri7uuYA4WYMzGw63+t5nE3flsR03xECTLJ6TV44SJMG0Z9pjBUYA1goLvAU=
Received: by 10.114.178.1 with SMTP id a1mr189615waf.1189652697684;
	Wed, 12 Sep 2007 20:04:57 -0700 (PDT)
Received: by 10.114.145.17 with HTTP; Wed, 12 Sep 2007 20:04:57 -0700 (PDT)
Message-ID: <20d2bdfb0709122004o709211bbt9569b0f1d3ae1120@mail.gmail.com>
Date: Wed, 12 Sep 2007 23:04:57 -0400
From: "Bruce Lowekamp" <lowekamp@sipeerior.com>
To: "Behave WG" <behave@ietf.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
X-Google-Sender-Auth: fc544532d3d4a5ec
X-Spam-Score: 0.0 (/)
X-Scan-Signature: c83ccb5cc10e751496398f1233ca9c3a
Cc: 
Subject: [BEHAVE] behavior discovery category
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

In Chicago we agreed to bring the question of what category
(Experimental or Informational) was appropriate for behavior-discovery
to the list.  Derek and I spent some time looking at this and at
http://www.ietf.org/u/ietfchair/info-exp.html.  We believe that the
appropriate status is experimental.   We have used P2P supernode
selection as an example application, but believe that the case of
applications without a rendevous server is more general.

I'm including the text from a revision of the draft that addresses this issue.

Bruce


Abstract

   This specification defines an experimental usage of the Simple
   Traversal Underneath Network Address Translators (NAT) (STUN)
   Protocol that discovers the presence and current behaviour of NATs
   and firewalls between the STUN client and the STUN server.



1.  Applicability

   This experimental STUN usage does not allow an application behind a
   NAT to make an absolute determination of the NAT's characteristics.
   NAT devices do not behave consistently enough to predict future
   behaviour with any guarantee.  This STUN usage provides information
   about observable transient behavior; it only truly determines a NAT's
   behavior with regard to the STUN server used at the instant the test
   is run.  Applications requiring reliable reach between two particular
   endpoints must establish a communication channel through a NAT using
   another technique.  IETF has proposed standards including ICE
   [I-D.ietf-mmusic-ice] and OUTBOUND [I-D.ietf-sip-outbound] for
   establishing communication channels when a publically accessible
   rendevous service is available.

   This usage provides techniques known to be of use for diagnostic
   purposes.  These techniques are considered experimental when applied
   to real-time selection of parameters for protocols in situations
   where a publically accessible rendevous service is not available.
   One possible application might be optimizing supernode selection in
   P2P networks based on statistical experience with establishing
   connections and diagnosing NAT behavior with a variety of other
   peers.


2.  Introduction

   The Simple Traversal Underneath Network Address Translators (NAT)
   (STUN) [I-D.ietf-behave-rfc3489bis] provides a mechanism to discover
   the reflexive transport address toward the STUN server, using the
   Binding Request.  This specification defines the NAT Behavior
   Discovery STUN usage, which allows a STUN client to probe the current
   behaviour of the NAT/FW devices between the client and the STUN
   server.  This usage defines new STUN attributes for the Binding
   Request and Binding Response.

   Many NAT/FW devices do not behave consistently and will change their
   behaviour under load and over time.  Applications requiring high
   reliability must be prepared for the NAT's behaviour to become more
   restrictive.  Specifically, it has been found that under load NATs
   may transition to the most restrictive filtering and mapping
   behaviour and shorten the lifetime of new and existing bindings.  In
   short, applications can discover how bad things currently are, but
   not how bad things will get.

   Despite this limitation, instantaneous observations are often quite
   useful in troubleshooting network problems, and repeated tests over
   time, or in known load situations, may be used to characterize a
   NAT's behavior.  In particular, in the hands of a person
   knowledgeable about the needs of an application and the nodes an
   application needs to communicate with, it can be a powerful tool.
   For these diagnostics to be effectiev, the application must select a
   STUN server that is appropriately located with regards to its future
   communication partners.

2.1.  Example Use with P2P Overlays

   An application could use Behavior Discovery in a P2P protocol to
   determine if a particular endpoint is a reasonable supernode
   candidate.  This P2P network application is willing to select
   supernodes that might advertise unreliable IP addresses to avoid the
   cost of dedicated servers.  A supernode candidate requires that its
   NAT(s) offer(s) Address Independent Filtering.  It might periodically
   re-run tests and would remove itself as a supernode if its NAT/FW
   chain lost this characteristic.  These tests could be run with other
   supernode candidates acting as STUN servers as well as dedicated STUN
   servers.  As many P2P algorithms tolerate non-transitive connectivity
   between a portion of their peers, guaranteed pair-wise reliable reach
   might be sacrificed in order to distribute the P2P overlay's load
   across peers that can be directly contacted by the majority of users.

   Use of Behavior Discovery for such an application requires:

   o  Specification of protocols capable of offering reliable end-user
      performance using unreliable links between peers.

   o  The application is deployed behind NATs that provide Address
      Independent Filtering and that remain in this mode for an amount
      of time sufficient for the application to identify their behavior,
      distribute this information to the rest of the overlay, and
      provide useful work for the application.

   This draft is experimental as deployed applications implementing open
   protocols have yet to be deployed in such environments to demonstrate
   that these two requirements have been met.  However, apocryphal
   evidence suggests that househould and small business targeted NAT
   devices have stable behaviour, especially when there are few clients
   behind them.  And numerous P2P applications have been deployed that
   appear to have these properties, although their protocols have not
   yet been subjected to rigorous evaluation by standards bodies.


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 13 01:00:10 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVgo9-0007Xj-S4; Thu, 13 Sep 2007 01:00:06 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVgo8-0007XR-F9
	for behave-confirm+ok@megatron.ietf.org; Thu, 13 Sep 2007 01:00:04 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVgo8-0007XI-3r; Thu, 13 Sep 2007 01:00:04 -0400
Received: from ns4.neustar.com ([156.154.24.139])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43)
	id 1IVgo7-0000rY-NH; Thu, 13 Sep 2007 01:00:04 -0400
Received: from stiedprstage1.ietf.org (stiedprstage1.va.neustar.com
	[10.31.47.10]) by ns4.neustar.com (Postfix) with ESMTP id 8F8AA2AC49;
	Thu, 13 Sep 2007 05:00:03 +0000 (GMT)
Received: from ietf by stiedprstage1.ietf.org with local (Exim 4.43)
	id 1IVgo6-0002xL-0j; Thu, 13 Sep 2007 01:00:02 -0400
Content-Type: Multipart/Mixed; Boundary="NextPart"
Mime-Version: 1.0
To: i-d-announce@ietf.org
From: Internet-Drafts@ietf.org
Message-Id: <E1IVgo6-0002xL-0j@stiedprstage1.ietf.org>
Date: Thu, 13 Sep 2007 01:00:02 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: b5d20af10c334b36874c0264b10f59f1
Cc: behave@ietf.org
Subject: [BEHAVE] I-D Action:draft-ietf-behave-rfc3489bis-10.txt 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Behavior Engineering for Hindrance Avoidance Working Group of the IETF.


	Title           : Session Traversal Utilities for (NAT) (STUN)
	Author(s)       : J. Rosenberg, et al.
	Filename        : draft-ietf-behave-rfc3489bis-10.txt
	Pages           : 47
	Date            : 2007-09-13

Session Traversal Utilities for NAT (STUN) is a protocol that serves
as a tool for other protocols in dealing with NAT traversal.  It can
be used by an endpoint to determine the IP address and port allocated
to it by a NAT.  It can also be used to check connectivity between
two endpoints, and as a keep-alive protocol to maintain NAT bindings.
STUN works with many existing NATs, and does not require any special
behavior from them.

STUN is not a NAT traversal solution by itself.  Rather, it is a tool
to be used in the context of a NAT traversal solution.  This is an
important change from the previous version of this specification (RFC
3489), which presented STUN as a complete solution.

This document obsoletes RFC 3489.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-behave-rfc3489bis-10.txt

To remove yourself from the I-D Announcement list, send a message to
i-d-announce-request@ietf.org with the word unsubscribe in the body of 
the message.
You can also visit https://www1.ietf.org/mailman/listinfo/I-D-announce
to change your subscription settings.

Internet-Drafts are also available by anonymous FTP. Login with the 
username "anonymous" and a password of your e-mail address. After 
logging in, type "cd internet-drafts" and then
	"get draft-ietf-behave-rfc3489bis-10.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt

Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-behave-rfc3489bis-10.txt".

NOTE:   The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.

Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body; access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID: <2007-09-13005340.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-behave-rfc3489bis-10.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-behave-rfc3489bis-10.txt"; site="ftp.ietf.org";
	access-type="anon-ftp"; directory="internet-drafts"

Content-Type: text/plain
Content-ID: <2007-09-13005340.I-D\@ietf.org>


--OtherAccess--

--NextPart
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

--NextPart--





From behave-bounces@ietf.org Thu Sep 13 04:42:19 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVkHB-0006EK-Bb; Thu, 13 Sep 2007 04:42:17 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVkH9-0006DD-Bn
	for behave-confirm+ok@megatron.ietf.org; Thu, 13 Sep 2007 04:42:15 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IVkH5-00069t-7t
	for behave@ietf.org; Thu, 13 Sep 2007 04:42:11 -0400
Received: from smtp.nokia.com ([131.228.20.170] helo=mgw-ext11.nokia.com)
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IVkH4-00079q-6a
	for behave@ietf.org; Thu, 13 Sep 2007 04:42:11 -0400
Received: from esebh106.NOE.Nokia.com (esebh106.ntc.nokia.com [172.21.138.213])
	by mgw-ext11.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l8D8ffaF018694; Thu, 13 Sep 2007 11:42:06 +0300
Received: from esebh103.NOE.Nokia.com ([172.21.143.33]) by
	esebh106.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Thu, 13 Sep 2007 11:41:53 +0300
Received: from esebh101.NOE.Nokia.com ([172.21.138.177]) by
	esebh103.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Thu, 13 Sep 2007 11:41:51 +0300
Received: from esdhcp04184.research.nokia.com ([172.21.41.84]) by
	esebh101.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Thu, 13 Sep 2007 11:41:51 +0300
From: =?iso-8859-1?q?R=E9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: behave@ietf.org
Subject: Re: [BEHAVE] TURN - some proposed changes for improvement and
	simplification
Date: Thu, 13 Sep 2007 11:41:52 +0300
User-Agent: KMail/1.9.6 (enterprise 0.20070907.709405)
References: <46E81AFE.6040705@cisco.com>
In-Reply-To: <46E81AFE.6040705@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709131141.52765.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 13 Sep 2007 08:41:51.0193 (UTC)
	FILETIME=[EDF32490:01C7F5E1]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 52f7a77164458f8c7b36b66787c853da
Cc: 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Le Wednesday 12 September 2007 19:59:42 ext Jonathan Rosenberg, vous avez=20
=E9crit=A0:
(...)
> As Dan mentioned at the last IETF meeting, we've formed a small team to
> Firstly, we can eliminate the entire need for active destinations (and
> the Send and Data primitives as well) by introducing the concept of a
> 'channel'. Each channel is a stream of application data sent to and from
> a remote IP and port. In the case of UDP, we'd add a SINGLE BYTE to the
> front of every packet to and from the TURN server, which is used to
> signal the channel number. (...)

That looks ok.

This would break natural alignment of the inner payload. But the overhead i=
s=20
fixed size, so that could be easily compensated.

(...)
> Another possible concern with the 1-byte overhead is path MTU. If a UDP
> packet was sent from a remote peer and it was exactly equal to the path
> MTU from the remote peer to the TURN server, this will now fragment on
> the TURN server to client link. Its not clear that this is a real
> problem. But if you are aware of UDP applications that send data exactly
> at the MTU for some reason, please speak up.

Given how path MTU discovery is unreliable through NATs/firewalls, that wil=
l=20
hardly worsen the situation.

(...)
> In this case of multiple TCP connections to remote peers, the congestion
> control bridging described above gets complicated, and head-of-line
> blocking gets introduced. Congestion on the connection from the TURN
> server to one peer can slow down the connection to the client, and thus
> limit the throughput to the other peer. This is only a problem when we
> are transferring real data, not so much for things like STUN messages or
> TLS handshakes.

OK for connectivity checks, provided the output buffer of any TCP session i=
s=20
big enough to handle one connectivity check (I assume you don't send more=20
than one check in TCP case). And it most certainly is big enough.

It might actually be a problem for TLS handshakes? Though it was my=20
understanding ICE-TCP would do connectivity checks first. Would this not ma=
ke=20
TLS handshakes a non-issue?

> So the recommendation is going to be that, if a client=20
> ends up in a case with multiple TCP connections that survive past the
> connectivity checks, in that case the spec will recommend the client
> allocate a new TCP candidate from the TURN server and restart ICE with
> one of the peers, in order to eliminate the HOL blocking.

Precisely.

> We're looking for feedback on this proposal (and questions and
> clarifications of course) ASAP - within the next week. If this seems
> like a good direction, we can work these changes in and address the
> other comments and still deliver a revision on schedule.



=2D-=20
R=E9mi Denis-Courmont


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 13 15:15:01 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVu9R-0005RF-7B; Thu, 13 Sep 2007 15:14:57 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVu9Q-0005R8-48
	for behave-confirm+ok@megatron.ietf.org; Thu, 13 Sep 2007 15:14:56 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVu9P-0005Qp-GR; Thu, 13 Sep 2007 15:14:55 -0400
Received: from sj-iport-6.cisco.com ([171.71.176.117])
	by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IVu9N-0004O5-Uf; Thu, 13 Sep 2007 15:14:55 -0400
X-IronPort-AV: E=Sophos;i="4.20,250,1186383600"; d="scan'208";a="217654124"
Received: from sj-dkim-1.cisco.com ([171.71.179.21])
	by sj-iport-6.cisco.com with ESMTP; 13 Sep 2007 12:14:53 -0700
Received: from sj-core-1.cisco.com (sj-core-1.cisco.com [171.71.177.237])
	by sj-dkim-1.cisco.com (8.12.11/8.12.11) with ESMTP id l8DJEr0X030278; 
	Thu, 13 Sep 2007 12:14:53 -0700
Received: from dwingwxp01 ([10.32.240.198])
	by sj-core-1.cisco.com (8.12.10/8.12.6) with ESMTP id l8DJEpEw005549;
	Thu, 13 Sep 2007 19:14:51 GMT
From: "Dan Wing" <dwing@cisco.com>
To: "'Matt Mathis'" <mathis@psc.edu>
References: <Pine.LNX.4.58.0709131428280.6691@tesla.psc.edu>
Date: Thu, 13 Sep 2007 12:14:48 -0700
Message-ID: <1b1b01c7f63a$5b2388e0$c6f0200a@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
X-Mailer: Microsoft Office Outlook 11
X-Mimeole: Produced By Microsoft MimeOLE V6.00.2900.3138
In-Reply-To: <Pine.LNX.4.58.0709131428280.6691@tesla.psc.edu>
Thread-Index: Acf2NlUd1TxJkkJdRx2+4vGa9axioQAAGrGQ
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=11180; t=1189710893;
	x=1190574893; c=relaxed/simple; s=sjdkim1004;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=dwing@cisco.com;
	z=From:=20=22Dan=20Wing=22=20<dwing@cisco.com>
	|Subject:=20RE=3A=20Review=20of=20draft-ietf-behave-turn-04.txt
	|Sender:=20; bh=+vscJmgTfKJfVQmaWe8AXcimH+6n7pdF6RWi28pwEVs=;
	b=q/b6WR8LxsDJeLkpghEbCtS8caYdaevyKt9Z6x9UuIe/taLms000tK8mUZo/lXqnDuHIRoWx
	rR/NoJm4nHc6dF5D4sVVqJsRUSxk8qKZFjL4TKyE3WEt9b8J3RKAx1SDfm7bx3WgIMLJ9eowxj
	uwR18ZMwWAP2vJjMDRvDu2/GU=;
Authentication-Results: sj-dkim-1; header.From=dwing@cisco.com; dkim=pass (s
	ig from cisco.com/sjdkim1004 verified; ); 
X-Spam-Score: -4.0 (----)
X-Scan-Signature: 32029c790f79bd4a84a26bd2915c54b9
Cc: 'Rohan Mahy' <rohan@ekabal.com>, behave@ietf.org,
	'Transport Directorate' <tsv-dir@ietf.org>, huitema@microsoft.com,
	'Philip Matthews' <philip_matthews@magma.ca>
Subject: [BEHAVE] RE: Review of draft-ietf-behave-turn-04.txt
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Matt -- I don't recall seeing a response to your earlier email.  Please =
accept
my apologies for that.

There is a team of authors working on -05 (Rohan Mahy, Philip Matthews,
Jonathan Rosenberg, and me helping a bit).  As you can see from =
Jonathan's
post at =
http://www1.ietf.org/mail-archive/web/behave/current/msg02742.html,
the protocol may change if there is consensus in the working group for =
the
proposed changes.

See below for some specific responses.

> -----Original Message-----
> From: Matt Mathis [mailto:mathis@psc.edu]=20
> Sent: Thursday, September 13, 2007 11:46 AM
> To: Lars Eggert; Magnus Westerlund; Dan Wing
> Cc: jdrosen@cisco.com; rohan@ekabal.com;=20
> huitema@microsoft.com; Dan Wing; Transport Directorate;=20
> behave@ietf.org
> Subject: Review of draft-ietf-behave-turn-04.txt
>=20
> (Resend to include more participants and stakeholders.)
>=20
> I've reviewed this document as part of the transport area=20
> directorate's
> ongoing effort to review key IETF documents. These comments=20
> were written
> primarily for the transport area directors, but are copied to=20
> the document's
> authors for their information and to allow them to address=20
> any issues raised.
>=20
> Unfortunately I have not been a participating in the ongoing=20
> discussion about
> STUN and TURN, so some of my comments may reflect my lack of=20
> context, or
> otherwise repeat deeply reviewed features of these protocols.
>=20
> Please forward this to the WG list as appropriate (I am not=20
> subscribed).
>=20
> My comments are primarily focused on=20
> draft-ietf-behave-turn-04.txt, although
> some reflect architectural concerns about STUN itself.  These comments
> probably apply to additional documents.
>=20
> Summary: TCP over TURN needs some additional work.  It would=20
> improve the document not to intermix the TCP and UDP discussions.

We separated ICE into ICE-for-UDP (draft-ietf-mmusic-ice) and=20
ICE-for-TCP, draft-ietf-mmusic-ice-tcp.  This was done to accelerate
ICE for UDP, as most implementors were interested in ICE for UDP,
as well as to simplify things.

At this time, we hope to keep TCP in the TURN specification,=20
especially because TURN allows TCP-to-UDP interworking (see
section 4.1 of turn-04). =20

> First I would like to use "/" notation to indicate various TURN
> configurations: UDP/TURN/TCP would be UDP over TURN over TCP
> transport.  I will use */TCP when */TLS/TCP also applies.
>
> I have no problems with UDP/TURN/UDP - the semantics of all three
> layer seem to be well matched.
>
> UDP/TURN/TCP is likely to cause problems for real time application
> because losses on the TCP connection will cause the UDP to be
> delivered very late rather than lost.  It should be noted that this
> is a non-preferred configuration, and why.

Good point, thanks.

> I have multiple concerns about the framing and multiplexing layer
> between TURN and it's transport.  Although I have only recently
> become aware of this mechanism, and are far from being a expert, it
> really strikes me as being non-robust.  I do understand why it
> exists and why it was done the way it is done.
>
> Particularly problematic for me is the idea that TCP/TURN/TCP can be
> used without framing, but without prohibiting further STUN messages.
> First it is a violation of TCP semantics to manage or expose segment
> boundaries.  TCP is not obligated to make segment boundaries match
> write() buffer edges and is even permitted to change segment
> boundaries on retransmission or for example as part of RFC 4821 MTU
> discovery.  Also read() buffers are not required to match segment
> boundaries.  This means that embedded TURN messages can not be
> assumed to be aligned with buffer edges and have to be discovered by
> scanning all client data.  This by itself is not a total show
> stopper, however, if I wanted to attack STUN, I would create data
> objects that resembled STUN messages. They could be placed anywhere:
> comments in web pages, etc., and although the authentication
> machinery would probably cause them to be discarded, the messages
> could first cause arbitrary data to be discarded, including possibly
> legitimate following control messages.  At this point is I do not
> believe that there is a robust or secure way to process any
> additional STUN messages, once a connection has carried any
> non-framed TCP/TURN/TCP data, and doing so should be forbidden.
> (The situation is much better for UDP/TURN/UDP, since the UDP syntax
> in not ambiguous with the STUN syntax).

TURN's TCP framing is described in sectoin 5 of turn-04, and
I do not believe suffers from the problems you describe above.

> Although I understand why STUN permits lots of flexibility in the
> framing/multiplexing layer, doing so creates the opportunity for
> unanticipated robustness problems that are the consequence of
> interactions between different STUN services specified in different
> documents.  There should be a unified analysis of threats against
> the weakly specified framing and multiplexing layer (please excuse
> me if I have overlooked one).  The simplest fix is to always require
> explicit framing for STUN/TCP, except perhaps a few specific events
> such as one transition to TCP/TURN/TCP, which would forbid all
> further STUN messages.

It would be helpful if you looked at the new 'channel' proposal
from Jonathan,
http://www1.ietf.org/mail-archive/web/behave/current/msg02742.html,=20
and see if it suffers a similar problem or if the use of 'channels' =
would be
superior.


> The document does not describe how TURN would be used or deployed.

During IESG review, Jonathan added an "Operational Considerations"
section to ICE (it's in draft-ietf-mmusic-ice-18).  Perhaps something
akin to that would be useful for TURN.

> I will make the assumption that TURN is not designed to only be used
> by TURN specific applications.  Ordinary Internet application would
> be adapted to use TURN by partially replacing run-time libraries
> and/or system functions with TURN enabled versions of the same
> functions.  This places additional constraints on TURN.  Not only
> must there be a clear mapping between TURN messages and actions on
> the connection to the external client, there also has to be a clear
> mapping between events at the application API in the STUN client and
> corresponding TURN messages.  This implicitly requires that there is
> a reasonable mapping between TCP API semantics and TURN messages at
> the STUN client.
>
> There are a number of problems in this area.  For starters no TCP
> application ever knows what bandwidth it is going to use, so "should
> include a BANDWIDTH attribute" is pointless.=20

Some TCP applications do; for example, if you're running RTP over
TCP and you know the codec you're using, you do know the bandwidth
you're going to consume.

> More serious, there
> are a number of places where it is specified that TCP payload data
> is silently discarded.  Under no circumstances does TCP silently
> discard data, it always reports errors through the API, usually
> synchronously at that.  Many application depend on this to shutdown
> failed connections.  It is guaranteed that there are applications
> that will hang or loop forever if data is silently discarded.

I concur that if a TURN server is doing TCP-to-TCP, it shouldn't=20
be allowed to discard data.  However, if it's doing TCP-to-UDP,
I don't see the harm in it discarding data; such discarding can
even be helpful for realtime applications (packets that arrive
late are worthless).

> I have a feeling that I overlooked other API glitches.
>=20
> It makes the document extra obtuse to have UDP and TCP discussions
> interleaved, when the underlying assumptions to the protocols=20
> are totally different.  It would help to separate the descriptions=20
> of TCP/TURN/* from UDP/TURN/*. Although the TURN messages look=20
> similar for TCP and UDP, they have substantially different=20
> semantics.
>=20
> Finally, the question that triggered this review: a point=20
> raise by R=E9mi Denis-Courmont and Rohan Mahy on about Aug 23rd. =20
> The connection multiplexing mechanism does not implement per=20
> stream flow control and is subject to head-of-queue blocking=20
> if one of the streams is making insufficient progress
> for some reason.=20

Yep.

> Put anther way, any application or context that causes "TCP
> persist" or "zero window probing" will cause deadlock like=20
> TURN protocol freezes.  It is also possible to construct=20
> examples that can exhibit true deadlocks.

This problem only occurs when we have SIP forking, and when
SIP forking occurs we're only likely to need ICE connectivity
checks and TLS (or DTLS) handshakes. =20

In -05, we will provide reasons why the duration of this
activity needs to be limited, and describe how (with SIP)
re-Invites can be performed to regain 1-to-1 TCP connections
so that HOL blocking is avoided.

> Early versions of SSH port forwarding also had this this problem.
> To fix it each stream has it's own buffer at the demultiplexer, and
> there is an explicit flow control signaling mechanism to inform the
> multiplexer how much buffer space is available for each channel.
> The multiplexer is not permitted to send data for any channel that
> would overflow that channels buffer at the demultiplexer.  Search
> RFC4254 for "window".

Thank you very much for that pointer -- Rohan, Jonathan, Philip
and I were discussing that very problem Monday evening.

> BTW, user experience with the older non-flow-control SSH=20
> was not good.  Head-of-queue blocking is a bad thing and=20
> must be avoided.
>=20
> To further complicate the matter, an optimal implementation=20
> requires that each of the buffers at the demultiplexer be=20
> the same size as the delay bandwidth product of the=20
> underlying TCP connection itself.  This has been implemented in
> HPN-SSH http://www.psc.edu/networking/projects/hpn-ssh/ ,=20
> although not described in a standards document.
>=20
> The security section for TURN (and STUN) are focused on=20
> attacks against the protocols themselves, and do not adequately=20
> address ways in which these protocols might be used to attack=20
> other protocols or resources.  In particular a TURN/STUN server=20
> that does weak Authentication, Authorization or Accounting
> (logging) can be used to implement untraceable anonymous=20
> attacks on other resources.  These configurations must be=20
> forbidden.

Good point.  We will add some text to the Operational=20
Considerations section or, failing that, to the Security=20
Considerations section.
=20
> Although UDP/TURN might be ready to progress, TCP/TURN needs=20
> to be reworked.
> It might make sense to divide (or restructure) the document.

We will consider it, especially that we split ICE-TCP long
ago.

> Please explicitly include me on any follow up discussion.

Thanks for your thorough comments!

-d


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 13 15:15:32 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVu9x-0006j5-Ka; Thu, 13 Sep 2007 15:15:29 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVthc-00043T-Ct
	for behave-confirm+ok@megatron.ietf.org; Thu, 13 Sep 2007 14:46:12 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVthc-00043D-2b; Thu, 13 Sep 2007 14:46:12 -0400
Received: from mailer1.psc.edu ([2001:5e8:1:3a::64])
	by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IVtha-0003ur-PA; Thu, 13 Sep 2007 14:46:12 -0400
Received: from tesla.psc.edu (tesla.psc.edu [128.182.58.233])
	by mailer1.psc.edu (8.14.1/8.13.3) with ESMTP id l8DIjlYH004094
	(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO);
	Thu, 13 Sep 2007 14:45:47 -0400 (EDT)
Received: from localhost.psc.edu (localhost.psc.edu [127.0.0.1])
	by tesla.psc.edu (8.13.1/8.13.1) with ESMTP id l8DIjk4M007928;
	Thu, 13 Sep 2007 14:45:46 -0400
Date: Thu, 13 Sep 2007 14:45:46 -0400 (EDT)
From: Matt Mathis <mathis@psc.edu>
To: Lars Eggert <lars.eggert@nokia.com>,
	Magnus Westerlund <magnus.westerlund@ericsson.com>,
	Dan Wing <dwing@cisco.com>
Message-ID: <Pine.LNX.4.58.0709131428280.6691@tesla.psc.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=X-UNKNOWN
Content-Transfer-Encoding: QUOTED-PRINTABLE
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 995b2e24d23b953c94bac5288c432399
X-Mailman-Approved-At: Thu, 13 Sep 2007 15:15:28 -0400
Cc: rohan@ekabal.com, behave@ietf.org, Dan Wing <dwing@cisco.com>,
	Transport Directorate <tsv-dir@ietf.org>, huitema@microsoft.com
Subject: [BEHAVE] Review of draft-ietf-behave-turn-04.txt
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

(Resend to include more participants and stakeholders.)

I've reviewed this document as part of the transport area directorate's
ongoing effort to review key IETF documents. These comments were written
primarily for the transport area directors, but are copied to the document'=
s
authors for their information and to allow them to address any issues raise=
d.

Unfortunately I have not been a participating in the ongoing discussion abo=
ut
STUN and TURN, so some of my comments may reflect my lack of context, or
otherwise repeat deeply reviewed features of these protocols.

Please forward this to the WG list as appropriate (I am not subscribed).

My comments are primarily focused on draft-ietf-behave-turn-04.txt, althoug=
h
some reflect architectural concerns about STUN itself.  These comments
probably apply to additional documents.

Summary: TCP over TURN needs some additional work.  It would improve the
document not to intermix the TCP and UDP discussions.

First I would like to use "/" notation to indicate various TURN
configurations: UDP/TURN/TCP would be UDP over TURN over TCP transport.  I
will use */TCP when */TLS/TCP also applies.

I have no problems with UDP/TURN/UDP - the semantics of all three layer see=
m
to be well matched.

UDP/TURN/TCP is likely to cause problems for real time application because
losses on the TCP connection will cause the UDP to be delivered very late
rather than lost.  It should be noted that this is a non-preferred
configuration, and why.

I have multiple concerns about the framing and multiplexing layer between T=
URN
and it's transport.  Although I have only recently become aware of this
mechanism, and are far from being a expert, it really strikes me as being
non-robust.  I do understand why it exists and why it was done the way it i=
s
done.

Particularly problematic for me is the idea that TCP/TURN/TCP can be used
without framing, but without prohibiting further STUN messages.  First it i=
s a
violation of TCP semantics to manage or expose segment boundaries.  TCP is =
not
obligated to make segment boundaries match write() buffer edges and is even
permitted to change segment boundaries on retransmission or for example as
part of RFC 4821 MTU discovery.  Also read() buffers are not required to ma=
tch
segment boundaries.  This means that embedded TURN messages can not be assu=
med
to be aligned with buffer edges and have to be discovered by scanning all
client data.  This by itself is not a total show stopper, however, if I wan=
ted
to attack STUN, I would create data objects that resembled STUN messages.
They could be placed anywhere: comments in web pages, etc., and although th=
e
authentication machinery would probably cause them to be discarded, the
messages could first cause arbitrary data to be discarded, including possib=
ly
legitimate following control messages.  At this point is I do not believe t=
hat
there is a robust or secure way to process any additional STUN messages, on=
ce
a connection has carried any non-framed TCP/TURN/TCP data, and doing so sho=
uld
be forbidden.  (The situation is much better for UDP/TURN/UDP, since the UD=
P
syntax in not ambiguous with the STUN syntax).

Although I understand why STUN permits lots of flexibility in the
framing/multiplexing layer, doing so creates the opportunity for unanticipa=
ted
robustness problems that are the consequence of interactions between differ=
ent
STUN services specified in different documents.  There should be a unified
analysis of threats against the weakly specified framing and multiplexing
layer (please excuse me if I have overlooked one).  The simplest fix is to
always require explicit framing for STUN/TCP, except perhaps a few specific
events such as one transition to TCP/TURN/TCP, which would forbid all furth=
er
STUN messages.

The document does not describe how TURN would be used or deployed.  I will
make the assumption that TURN is not designed to only be used by TURN speci=
fic
applications.  Ordinary Internet application would be adapted to use TURN b=
y
partially replacing run-time libraries and/or system functions with TURN
enabled versions of the same functions.  This places additional constraints=
 on
TURN.  Not only must there be a clear mapping between TURN messages and
actions on the connection to the external client, there also has to be a cl=
ear
mapping between events at the application API in the STUN client and
corresponding TURN messages.  This implicitly requires that there is a
reasonable mapping between TCP API semantics and TURN messages at the STUN
client.

There are a number of problems in this area.  For starters no TCP applicati=
on
ever knows what bandwidth it is going to use, so "should include a BANDWIDT=
H
attribute" is pointless.  More serious, there are a number of places where =
it
is specified that TCP payload data is silently discarded.  Under no
circumstances does TCP silently discard data, it always reports errors thro=
ugh
the API, usually synchronously at that.  Many application depend on this to
shutdown failed connections.  It is guaranteed that there are applications
that will hang or loop forever if data is silently discarded.

I have a feeling that I overlooked other API glitches.

It makes the document extra obtuse to have UDP and TCP discussions
interleaved, when the underlying assumptions to the protocols are totally
different.  It would help to separate the descriptions of TCP/TURN/* from
UDP/TURN/*. Although the TURN messages look similar for TCP and UDP, they h=
ave
substantially different semantics.

Finally, the question that triggered this review: a point raise by R=E9mi
Denis-Courmont and Rohan Mahy on about Aug 23rd.  The connection multiplexi=
ng
mechanism does not implement per stream flow control and is subject to
head-of-queue blocking if one of the streams is making insufficient progres=
s
for some reason.  Put anther way, any application or context that causes "T=
CP
persist" or "zero window probing" will cause deadlock like TURN protocol
freezes.  It is also possible to construct examples that can exhibit true
deadlocks.

Early versions of SSH port forwarding also had this this problem.  To fix i=
t
each stream has it's own buffer at the demultiplexer, and there is an expli=
cit
flow control signaling mechanism to inform the multiplexer how much buffer
space is available for each channel.  The multiplexer is not permitted to s=
end
data for any channel that would overflow that channels buffer at the
demultiplexer.  Search RFC4254 for "window".  BTW, user experience with the
older non-flow-control SSH was not good.  Head-of-queue blocking is a bad
thing and must be avoided.

To further complicate the matter, an optimal implementation requires that e=
ach
of the buffers at the demultiplexer be the same size as the delay bandwidth
product of the underlying TCP connection itself.  This has been implemented=
 in
HPN-SSH http://www.psc.edu/networking/projects/hpn-ssh/ , although not
described in a standards document.

The security section for TURN (and STUN) are focused on attacks against the
protocols themselves, and do not adequately address ways in which these
protocols might be used to attack other protocols or resources.  In particu=
lar
a TURN/STUN server that does weak Authentication, Authorization or Accounti=
ng
(logging) can be used to implement untraceable anonymous attacks on other
resources.  These configurations must be forbidden.

Although UDP/TURN might be ready to progress, TCP/TURN needs to be reworked=
=2E
It might make sense to divide (or restructure) the document.

Please explicitly include me on any follow up discussion.
Thanks,
--MM--


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 13 17:07:23 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVvuC-0007wk-B8; Thu, 13 Sep 2007 17:07:20 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IVvuB-0007wS-FY
	for behave-confirm+ok@megatron.ietf.org; Thu, 13 Sep 2007 17:07:19 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IVvuB-0007wI-4e; Thu, 13 Sep 2007 17:07:19 -0400
Received: from mailer2.psc.edu ([128.182.66.106])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43)
	id 1IVvuA-0004NR-MI; Thu, 13 Sep 2007 17:07:19 -0400
Received: from tesla.psc.edu (tesla.psc.edu [128.182.58.233])
	by mailer2.psc.edu (8.14.1/8.13.3) with ESMTP id l8DL76rT014359
	(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO);
	Thu, 13 Sep 2007 17:07:07 -0400 (EDT)
Received: from localhost.psc.edu (localhost.psc.edu [127.0.0.1])
	by tesla.psc.edu (8.13.1/8.13.1) with ESMTP id l8DL75kj013083;
	Thu, 13 Sep 2007 17:07:05 -0400
Date: Thu, 13 Sep 2007 17:07:05 -0400 (EDT)
From: Matt Mathis <mathis@psc.edu>
To: Dan Wing <dwing@cisco.com>
In-Reply-To: <1b1b01c7f63a$5b2388e0$c6f0200a@cisco.com>
Message-ID: <Pine.LNX.4.58.0709131559080.6691@tesla.psc.edu>
References: <Pine.LNX.4.58.0709131428280.6691@tesla.psc.edu>
	<1b1b01c7f63a$5b2388e0$c6f0200a@cisco.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Spam-Score: 0.0 (/)
X-Scan-Signature: e1e48a527f609d1be2bc8d8a70eb76cb
Cc: 'Rohan Mahy' <rohan@ekabal.com>, behave@ietf.org,
	'Transport Directorate' <tsv-dir@ietf.org>, huitema@microsoft.com,
	'Philip Matthews' <philip_matthews@magma.ca>
Subject: [BEHAVE] RED FLAG: Review of draft-ietf-behave-turn-04.txt
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org


On Thu, 13 Sep 2007, Dan Wing wrote:
> > Put anther way, any application or context that causes "TCP
> > persist" or "zero window probing" will cause deadlock like
> > TURN protocol freezes.  It is also possible to construct
> > examples that can exhibit true deadlocks.
>
> This problem only occurs when we have SIP forking, and when
> SIP forking occurs we're only likely to need ICE connectivity
> checks and TLS (or DTLS) handshakes.

This comment implies that TURN might only be used with SIP - This is not a
good assumption at all.

If STUN and/or TURN run as stand alone services, they will come to carry
arbitrary other protocols.  SIP might tolerate head-of-queue blocking under
normal conditions, but that is not an acceptable assumption to make for a
protocol that might be appropriate for solving a wide range of problems.

TURN MUST be robust and free from deadlocks, no matter what client protocols
are using it.  It is ok to explicitly implement only a subset of the functions
(e.g. no general listen() facility), but this has to be stated up front. "It
will do weird things with unexpected client protocols" will be a show stopper
further down the standards track.

As for testing future versions of TURN for head-of-queue blocking, think about
pathologies when one channel has a sender with lots of data to send and a
receiver that dawdles.  Does this interfere with other channels at all?  Does
it interfere with the return data on the same or other channels?  If any of
these are true, you can be sure that there will be pathological behavior
(deadlocks?) under some unexpected conditions.

Thanks,
--MM--
-------------------------------------------
Matt Mathis      http://www.psc.edu/~mathis
Work:412.268.3319    Home/Cell:412.654.7529
-------------------------------------------
Evil is defined by mortals who think they know
"The Truth" and use force to apply it to others.



_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 17 05:55:10 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXDJp-00074G-Sy; Mon, 17 Sep 2007 05:55:05 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXDJo-00074A-Mf
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 05:55:04 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXDJo-000741-BZ
	for behave@ietf.org; Mon, 17 Sep 2007 05:55:04 -0400
Received: from mailgw4.ericsson.se ([193.180.251.62])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXDJn-0000Sh-Dn
	for behave@ietf.org; Mon, 17 Sep 2007 05:55:04 -0400
Received: from mailgw4.ericsson.se (unknown [127.0.0.1])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	935D220087; Mon, 17 Sep 2007 11:55:02 +0200 (CEST)
X-AuditID: c1b4fb3e-ae831bb0000007e1-bc-46ee4ef6600f
Received: from esealmw129.eemea.ericsson.se (unknown [153.88.254.124])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	7657F20083; Mon, 17 Sep 2007 11:55:02 +0200 (CEST)
Received: from esealmw127.eemea.ericsson.se ([153.88.254.171]) by
	esealmw129.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 17 Sep 2007 11:55:01 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw127.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 17 Sep 2007 11:55:02 +0200
Message-ID: <46EE4EF5.9080804@ericsson.com>
Date: Mon, 17 Sep 2007 11:55:01 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: draft-ietf-behave-rfc3489bis@tools.ietf.org, Behave WG <behave@ietf.org>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 17 Sep 2007 09:55:03.0012 (UTC)
	FILETIME=[D1548A40:01C7F910]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 5d7a7e767f20255fce80fa0b77fb2433
Cc: 
Subject: [BEHAVE] AD comments on draft-ietf-behave-rfc3489bis-10
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Hi,

I have performed a combined AD-review / shepherding review and have some
comments. I think some of these will require an updated version before
proceeding. I hope this can be accomplished quickly to avoid to much delay.


1. Section 7.2.1:
It is not obvious why the number 7900 ms is the end of listening for
responses. How did one arrive on that number?

2. Section 12.2:
"   A STUN server can detect when a given Binding Request message was
   sent from an RFC 3489 [RFC3489] client by the absence of the correct
   value in the Magic Cookie field.  When the server detects an RFC 3489
   client, it SHOULD copy the value seen in the Magic Cookie field in
   the Binding Request to the Magic Cookie field in the Binding Response
   message, and insert a MAPPED-ADDRESS attribute instead of an XOR-
   MAPPED-ADDRESS attribute."

I think this text should have some warning about what functionality
issues regarding MUXing STUN with other protocols may have due to this
backwards compatibility issue. If I understand things correctly, having
3489 compatibility means that demuxer only can use the two first bits
and two last of the first word to separate STUN from normal data packets?

3. Section 14.4:

"   The key used as input to HMAC is the password."

Section 10.2:

   "For STUN servers used in conjunction with SIP servers, it is
   desirable to use the same credentials for authentication to the SIP
   server and STUN server.  Typically, SIP systems utilizing SIP's
   digest authentication mechanism do not actually store the password in
   the database.  Rather, they store a value called H(A1), which is
   computed as:

                H(A1) = MD5(username ":" realm ":" password)

   If a system wishes to utilize this credential, the STUN password
   would be computed by taking the user-entered username and password,
   and using H(A1) as the STUN password.  It is RECOMMENDED that clients
   utilize this construction for the STUN password."

Isn't these two in contradiction. In addition aren't they a potential
for interoperability issues between client and server. This as the
actual key used in the HMAC must be the same to allow correct integrity
verification. So, how is it inteded to work? Is it something the usage
needs to define?

4. Section 15.1.2:

"  A rogue client may use a STUN server as a reflector, sending it
   requests with a falsified source IP address and port.  In such a
   case, the response would be delivered to that source IP and port.
   There is no amplification with this attack, and it is mitigated by
   ingress source address filtering."

I think there is missing a clarification on what type of amplication
that is not possible with STUN. To my understanding there can't be any
packet rate amplification in this attack. However some percent of packet
size amplification seems possible due to NONCE and other error
indication added by the server. Maybe a clarification that there is no
amplification in the number of packets is warrented.

5. Section 15.2:

"   Most of the attacks in this section revolve around an attacker
   modifying the reflexive address learned by a STUN client through a
   Binding Request/Binding Response transaction.  Since the usage of the
   reflexive address is a function of the usage, the applicability and
   remediation of these attacks is usage-specific.  In common
   situations, modification of the reflexive address by an on-path
   attacker is easy to do.  Consider, for example, the common situation
   where STUN is run directly over UDP.  In this case, an on-path
   attacker can modify the source IP address of the Binding Request
   before it arrives at the STUN server.  The STUN server will then
   return this IP address in the XOR-MAPPED-ADDRESS attribute to the
   client.  Protecting against this attack by using a message-integrity
   check is impossible, since a message-integrity value cannot cover the
   source IP address, since the intervening NAT must be able to modify
   this value.  Instead, one solution to preventing the attacks listed
   below is for the client to verify the reflexive address learned, as
   is done in ICE [I-D.ietf-mmusic-ice].  Other usages may use other
   means to prevent these attacks."

I think there are a serious thinking error that affects most of the
attack variants in this description. If on on-path attacker changes the
source IP-address of the packet the packet will not be routed back to
the client issuing the request. Thus they are looking like lost packets.
To me the only attack being really viable is if the attacker changes the
source address to itself and then forwards the packet to the client
(target of the attack). This is of course assuming that the attacker is
not sitting very closely to the server and can see all/most traffic to
and from the server?

Can we be a bit clearer on when the attacks are possible. Thus I think
that 15.2.1 and 15.2.2 are more difficult to achieve (assuming integrity
checks between client and server) than 15.2.3 and 15.2.4.

6. Section 15.3:

"   This specification uses SHA-1 for computation of the message
   integrity. "

Shouldn't there be a HMAC before SHA-1 here?


Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 17 10:09:37 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXHI7-00069j-NO; Mon, 17 Sep 2007 10:09:35 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXHI6-00069d-Mr
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 10:09:34 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXHI6-00069V-Ca
	for behave@ietf.org; Mon, 17 Sep 2007 10:09:34 -0400
Received: from web33310.mail.mud.yahoo.com ([68.142.206.125])
	by ietf-mx.ietf.org with smtp (Exim 4.43) id 1IXHI5-0004qX-2L
	for behave@ietf.org; Mon, 17 Sep 2007 10:09:34 -0400
Received: (qmail 45670 invoked by uid 60001); 17 Sep 2007 14:09:32 -0000
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID;
	b=PRjQFvXHqgaWuVI6UKqqcrcyl+9xi5rxCLMD0W4SeDke5/0H0+AvI5dlrN/+4qvKVmCzakSX9WXWNI8BBqYP3l5IlR0XKWUdxHwJzhU4XZogQCs74YGgeG8OvzGikmK2LEjfYbNid4m4CzoLpRGaJqEesiFBYYwXMH4olri6j6Q=;
X-YMail-OSG: n04cimcVM1n1S.IinCZIBeLtDKnpmtTn2MeeVw2fR.x5Stl.DXos6m6V0SoqfepnwBNbkQ--
Received: from [69.236.94.7] by web33310.mail.mud.yahoo.com via HTTP;
	Mon, 17 Sep 2007 07:09:32 PDT
Date: Mon, 17 Sep 2007 07:09:32 -0700 (PDT)
From: Pyda Srisuresh <srisuresh@yahoo.com>
To: Magnus Westerlund <magnus.westerlund@ericsson.com>,
	Bryan Ford <baford@mit.edu>, Senthil Sivakumar <ssenthil@cisco.com>,
	Saikat Guha <saikat@cs.cornell.edu>, Behave WG <behave@ietf.org>
In-Reply-To: <46DD5FAB.5070607@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <266984.45543.qm@web33310.mail.mud.yahoo.com>
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 3d7f2f6612d734db849efa86ea692407
Cc: 
Subject: [BEHAVE] Re: AD comments on draft-ietf-behave-icmp-04
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Magnus,

Thank you for the review. Apologies for the delay in responding. Please see my
responses inline below.

regards,
suresh

--- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:

> Hi,
> 
> I have now done my AD review. Here are my comments.
> 
> 1. Section 1, third paragraph:
> 
> "Traditional NAT inherently mandates a certain level
>    of firewall-like functionality. However, firewall functionality in
>    general or any other middlebox functionality is out of the
>    scope of this document."
> 
> I assume that the firewall-like behavior is the filtering behavior. I
> think there might be some benefit to express it that way and to refer to
> the terms in UDP NAT RFC.
> 

[suresh] OK. Would the following rewording work for you?

  "Traditional NAT inherently mandates firewall-like filtering behavior 
   [BEH-UDP]. However, firewall functionality in general or any other 
   middlebox functionality is out of the scope of this document."

> 2. Section 2, last sentence:
> 
> "A NAT MAY drop
>    or appropriately handle a post-1122 ICMP messages. "
> 
> I think a normative statement of this type do not belong in the
> terminology section. Please find a better place to locate it.
> 
[suresh] OK. I will move this to the first paragraph of section 8 (Summary of
Requirements).

> 3. Section 3.2, REQ-2:
> 
>   "Setting the ICMP NAT session timeout to a very large duration (say,
>    240 seconds) could potentially tie up precious NAT resources such as
>    query mappings and NAT Sessions for the whole duration. On the other
>    hand, setting the timeout very low can result in premature freeing
>    of NAT resources and applications failing to complete gracefully.
>    The ICMP Query session timeout needs to be a balance between the two
>    extremes. 60 seconds timeout is a balance between the two extremes.
>    ICMP query session timer MUST not expire in less than 60 seconds. We
>    RECOMMEND however that the administrator(s) be allowed to configure
>    the timer.
> 
>    REQ-2: An ICMP Query session timer MUST NOT expire in less than 60
>    seconds.
>    a) It is RECOMMENDED that the ICMP Query session timer be made
>    configurable."
> 
> Once more we use NATs as a motivation to redefine what our constants
> are. I can't understand the motivation behind making the requirement to
> anything other than 2xMSL, i.e. 240 s. I fear that people will start use
> lower values for MSL, due to the fact that they know that if there are a
> NAT there is anyway no possibility for the packet to live. Yes, it is a
> long time, but MSL exist for being a real outer boundary for how long
> packets can live.
> 
[suresh] I didnt understand the comment - "I fear that people will start use
lower values for MSL, due to the fact that they know that if there are a NAT
there is anyway no possibility for the packet to live."  Why is there no
possibility for the packet to live when there is a NAT?

There was much discussion on the ICMP timeouts on the list.  RFC 4787 (Req-5a)
also provided some precedence on this. The WG agreed on 60sec timeout. In
practice, users of query based applications (ex: ICMP ping and ICMP traceroute)
are likely to be impatient if the response packets took longer than 60 secs.
240 secs., woudl be longer than even the udp TIMEOUT (120secs).

> If a NAT get too many querys aren't there better ways to deal with that?
> After all we are talking about requests comming from the internal side
> of the NAT. Are there ways of providing error response to query style ICMP?

[suresh] Most NAT sessions are initiated from the internal side, anyways,
right. In that sense, ICMP query based sessions are no different from other NAT
sessions. NATs could respond with  ICMP destination unreachable message (REQ-8)
when out of resources.

> 
> 4. Section 4.1,
> "This
>    requirement is meant primarily for future applications. Current
>    applications may not be checking for mismatched checksum."
> 
> I don't find these two sentences providing much benefit. I think it is
> well motivated that payload validation checking is supported for current
> and any future applications.
> 

[suresh] I could remove the two sentences. Woudl that work for you?

> 5. Section 4.2, first paragraph:
> "Section 4.3 of RFC 3022 describes the fields of an ICMP error
>    message that a NAT device translates. In this section, we describe
>    the requirements a NAT device MUST conform to while performing the
>    translations. Requirements identified in this section are necessary
>    for the current applications to work correctly."
> 
> I interpret this as an normative reference. This would be a normative
> reference from a BCP to a informative document, which I like to avoid.
> Can we please rework this, for example by including how the translation
> is done in this document instead.
> 
[suresh] Implicitly, both [BEH-UDP] and [BEH-TCP] also defer to RFC 3022 for
the NAT translation. In that sense, this docuemnt is no different. Would it
work for you if I simply removed the first sentence.  

> 6. Section 4.2.2, page 12, second paragraph:
> 
>    In addition, if the ICMP Error payload contains ICMP extensions
>    ([ICMP-EXT]), the NATs are encouraged to support ICMP extension
>    objects.
> 
> I think this sentence is strangely formulated. Please reword it.
> 
[suresh] would the following rewording work for you? 

   ICMP Error payload may contain ICMP extension objects ([ICMP-EXT]). 
   NATs are  encouraged to support ICMP extension objects.

> 
> Please comment if there are any unclarities or disagrement. But I do
> expect a revision.
> 
[suresh] If you are Ok with my responses, I could get the next rev out shortly.
Thanks.

regards,
suresh

> Cheers
> 
> Magnus Westerlund
> 
> IETF Transport Area Director & TSVWG Chair
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM/M
> ----------------------------------------------------------------------
> Ericsson AB                | Phone +46 8 4048287
> Torshamsgatan 23           | Fax   +46 8 7575550
> S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
> ----------------------------------------------------------------------
> 





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 17 11:10:06 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXIEb-00047V-JA; Mon, 17 Sep 2007 11:10:01 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXIEa-00047H-80
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 11:10:00 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXIEZ-000479-Sl
	for behave@ietf.org; Mon, 17 Sep 2007 11:09:59 -0400
Received: from mailgw3.ericsson.se ([193.180.251.60])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXIEY-0000SZ-SI
	for behave@ietf.org; Mon, 17 Sep 2007 11:09:59 -0400
Received: from mailgw3.ericsson.se (unknown [127.0.0.1])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	D0B0720767; Mon, 17 Sep 2007 17:09:57 +0200 (CEST)
X-AuditID: c1b4fb3c-ae67bbb0000007e1-2e-46ee98c5bfc0
Received: from esealmw126.eemea.ericsson.se (unknown [153.88.254.123])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	ABCDC205CB; Mon, 17 Sep 2007 17:09:57 +0200 (CEST)
Received: from esealmw128.eemea.ericsson.se ([153.88.254.172]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 17 Sep 2007 17:09:57 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw128.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Mon, 17 Sep 2007 17:09:56 +0200
Message-ID: <46EE98C5.4080403@ericsson.com>
Date: Mon, 17 Sep 2007 17:09:57 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Pyda Srisuresh <srisuresh@yahoo.com>
References: <266984.45543.qm@web33310.mail.mud.yahoo.com>
In-Reply-To: <266984.45543.qm@web33310.mail.mud.yahoo.com>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 17 Sep 2007 15:09:56.0595 (UTC)
	FILETIME=[CEC88430:01C7F93C]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: -1.0 (-)
X-Scan-Signature: 2a76bcd37b1c8a21336eb0a1ea6bbf48
Cc: Bryan Ford <baford@mit.edu>, Behave WG <behave@ietf.org>
Subject: [BEHAVE] Re: AD comments on draft-ietf-behave-icmp-04
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

See inline

Pyda Srisuresh skrev:
> Magnus,
> 
> Thank you for the review. Apologies for the delay in responding. Please see my
> responses inline below.
> 
> regards,
> suresh
> 
> --- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:
> 
>> Hi,
>>
>> I have now done my AD review. Here are my comments.
>>
>> 1. Section 1, third paragraph:
>>
>> "Traditional NAT inherently mandates a certain level
>>    of firewall-like functionality. However, firewall functionality in
>>    general or any other middlebox functionality is out of the
>>    scope of this document."
>>
>> I assume that the firewall-like behavior is the filtering behavior. I
>> think there might be some benefit to express it that way and to refer to
>> the terms in UDP NAT RFC.
>>
> 
> [suresh] OK. Would the following rewording work for you?
> 
>   "Traditional NAT inherently mandates firewall-like filtering behavior 
>    [BEH-UDP]. However, firewall functionality in general or any other 
>    middlebox functionality is out of the scope of this document."

Yes.

> 
>> 2. Section 2, last sentence:
>>
>> "A NAT MAY drop
>>    or appropriately handle a post-1122 ICMP messages. "
>>
>> I think a normative statement of this type do not belong in the
>> terminology section. Please find a better place to locate it.
>>
> [suresh] OK. I will move this to the first paragraph of section 8 (Summary of
> Requirements).

Will that not be strange without having a requirement that only is
listed in the summary?

> 
>> 3. Section 3.2, REQ-2:
>>
>>   "Setting the ICMP NAT session timeout to a very large duration (say,
>>    240 seconds) could potentially tie up precious NAT resources such as
>>    query mappings and NAT Sessions for the whole duration. On the other
>>    hand, setting the timeout very low can result in premature freeing
>>    of NAT resources and applications failing to complete gracefully.
>>    The ICMP Query session timeout needs to be a balance between the two
>>    extremes. 60 seconds timeout is a balance between the two extremes.
>>    ICMP query session timer MUST not expire in less than 60 seconds. We
>>    RECOMMEND however that the administrator(s) be allowed to configure
>>    the timer.
>>
>>    REQ-2: An ICMP Query session timer MUST NOT expire in less than 60
>>    seconds.
>>    a) It is RECOMMENDED that the ICMP Query session timer be made
>>    configurable."
>>
>> Once more we use NATs as a motivation to redefine what our constants
>> are. I can't understand the motivation behind making the requirement to
>> anything other than 2xMSL, i.e. 240 s. I fear that people will start use
>> lower values for MSL, due to the fact that they know that if there are a
>> NAT there is anyway no possibility for the packet to live. Yes, it is a
>> long time, but MSL exist for being a real outer boundary for how long
>> packets can live.
>>
> [suresh] I didnt understand the comment - "I fear that people will start use
> lower values for MSL, due to the fact that they know that if there are a NAT
> there is anyway no possibility for the packet to live."  Why is there no
> possibility for the packet to live when there is a NAT?

Well, the packet may live, but the response will never reach the
requestor if the context needed to translate it has been timed out. That
is why I fear that one might start optimizing these timers for something
else.

> 
> There was much discussion on the ICMP timeouts on the list.  RFC 4787 (Req-5a)
> also provided some precedence on this. The WG agreed on 60sec timeout. In
> practice, users of query based applications (ex: ICMP ping and ICMP traceroute)
> are likely to be impatient if the response packets took longer than 60 secs.
> 240 secs., woudl be longer than even the udp TIMEOUT (120secs).

Well, the WG can consider my displeasure with this, but I do accept the
WG consensus on it.

> 
>> If a NAT get too many querys aren't there better ways to deal with that?
>> After all we are talking about requests comming from the internal side
>> of the NAT. Are there ways of providing error response to query style ICMP?
> 
> [suresh] Most NAT sessions are initiated from the internal side, anyways,
> right. In that sense, ICMP query based sessions are no different from other NAT
> sessions. NATs could respond with  ICMP destination unreachable message (REQ-8)
> when out of resources.

Okay.


> 
>> 4. Section 4.1,
>> "This
>>    requirement is meant primarily for future applications. Current
>>    applications may not be checking for mismatched checksum."
>>
>> I don't find these two sentences providing much benefit. I think it is
>> well motivated that payload validation checking is supported for current
>> and any future applications.
>>
> 
> [suresh] I could remove the two sentences. Woudl that work for you?

Yes, that seems to be a okay solution.

> 
>> 5. Section 4.2, first paragraph:
>> "Section 4.3 of RFC 3022 describes the fields of an ICMP error
>>    message that a NAT device translates. In this section, we describe
>>    the requirements a NAT device MUST conform to while performing the
>>    translations. Requirements identified in this section are necessary
>>    for the current applications to work correctly."
>>
>> I interpret this as an normative reference. This would be a normative
>> reference from a BCP to a informative document, which I like to avoid.
>> Can we please rework this, for example by including how the translation
>> is done in this document instead.
>>
> [suresh] Implicitly, both [BEH-UDP] and [BEH-TCP] also defer to RFC 3022 for
> the NAT translation. In that sense, this docuemnt is no different. Would it
> work for you if I simply removed the first sentence.  

Well, I don't see how removing the sentence will make things better. It
seems that we likely need to apply RFC 3967 exception on this reference.
So please upgrade it to a real refence in the normative part.

> 
>> 6. Section 4.2.2, page 12, second paragraph:
>>
>>    In addition, if the ICMP Error payload contains ICMP extensions
>>    ([ICMP-EXT]), the NATs are encouraged to support ICMP extension
>>    objects.
>>
>> I think this sentence is strangely formulated. Please reword it.
>>
> [suresh] would the following rewording work for you? 
> 
>    ICMP Error payload may contain ICMP extension objects ([ICMP-EXT]). 
>    NATs are  encouraged to support ICMP extension objects.


Yes.

Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 17 14:29:18 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXLLP-0002wX-8l; Mon, 17 Sep 2007 14:29:15 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXLLN-0002wR-Ve
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 14:29:13 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXLLN-0002wF-LI
	for behave@ietf.org; Mon, 17 Sep 2007 14:29:13 -0400
Received: from ranger.systems.pipex.net ([62.241.162.32])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXLLN-00077Y-9K
	for behave@ietf.org; Mon, 17 Sep 2007 14:29:13 -0400
Received: from [61.94.220.90] (unknown [61.94.220.90])
	by ranger.systems.pipex.net (Postfix) with ESMTP id D6291E00066E;
	Mon, 17 Sep 2007 19:29:07 +0100 (BST)
Message-ID: <46EEC6E5.9000406@pjsip.org>
Date: Mon, 17 Sep 2007 19:26:45 +0100
From: Benny Prijono <bennylp@pjsip.org>
User-Agent: Thunderbird 1.5.0.10 (Windows/20070221)
MIME-Version: 1.0
To: Jonathan Rosenberg <jdrosen@cisco.com>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>	<200708081203.19729.remi.denis-courmont@nokia.com>	<46BA4268.2060204@db.org>	<200708090918.13691.remi.denis-courmont@nokia.com>	<46BB4B0C.6030204@db.org>
	<46C1ACF3.7000202@cisco.com>
In-Reply-To: <46C1ACF3.7000202@cisco.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
X-Spam-Score: -4.0 (----)
X-Scan-Signature: b19722fc8d3865b147c75ae2495625f2
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Jonathan Rosenberg wrote:
> An important point of clarification here.
> 
> Though FINGERPRINT is currently the only attribute beyond the MI, there 
> could potentially be others in the future so you can't depend on it 
> being the only one. So inserting a dummy FINGERPRINT won't help.

Jonathan,

sorry for joining late on the discussion, but your comment concerns 
me. Allowing any other attribute, except FINGERPRINT, beyond M-I, 
not only violates rfc3489bis-10 but could also break existing 
implementation as well.

The part where it violates rfc3489bis-10 is in Section 14.5:

"When present, the FINGERPRINT attribute MUST be the last attribute 
in the message, and thus will appear after MESSAGE-INTEGRITY."

And the reason why it may break existing implementation is because, 
in order to avoid linear searching for M-I throughout all 
attributes, currently we could probe M-I at position 
(20+hdr.length-24), if the M-I is the last attribute, or at position 
(20+hdr.length-24-8), if the M-I appears as the attribute before 
FINGERPRINT.

Allowing some other attributes to exist between M-I and FINGERPRINT 
will break above algorithm, and it will mean linear search for the 
M-I. So I propose we disallow this.

  -benny




_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 17 15:50:35 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXMbS-0007T0-A5; Mon, 17 Sep 2007 15:49:54 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXMbQ-0007Sp-Jv
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 15:49:52 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXMbQ-0007Sh-A7
	for behave@ietf.org; Mon, 17 Sep 2007 15:49:52 -0400
Received: from sj-iport-2-in.cisco.com ([171.71.176.71]
	helo=sj-iport-2.cisco.com) by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IXMbO-0007m8-UF
	for behave@ietf.org; Mon, 17 Sep 2007 15:49:52 -0400
X-IronPort-AV: E=Sophos;i="4.20,265,1186383600"; d="scan'208";a="399598032"
Received: from sj-dkim-1.cisco.com ([171.71.179.21])
	by sj-iport-2.cisco.com with ESMTP; 17 Sep 2007 12:49:45 -0700
Received: from sj-core-2.cisco.com (sj-core-2.cisco.com [171.71.177.254])
	by sj-dkim-1.cisco.com (8.12.11/8.12.11) with ESMTP id l8HJnjol018414; 
	Mon, 17 Sep 2007 12:49:45 -0700
Received: from dwingwxp01 ([10.32.240.198])
	by sj-core-2.cisco.com (8.12.10/8.12.6) with ESMTP id l8HJni03025985;
	Mon, 17 Sep 2007 19:49:44 GMT
From: "Dan Wing" <dwing@cisco.com>
To: "'Magnus Westerlund'" <magnus.westerlund@ericsson.com>,
	<draft-ietf-behave-rfc3489bis@tools.ietf.org>,
	"'Behave WG'" <behave@ietf.org>
References: <46EE4EF5.9080804@ericsson.com>
Subject: RE: [BEHAVE] AD comments on draft-ietf-behave-rfc3489bis-10
Date: Mon, 17 Sep 2007 12:49:44 -0700
Message-ID: <03b701c7f963$e5914590$c6f0200a@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Office Outlook 11
Thread-Index: Acf5ENt0xODu0krmQ06YK0qRSKjJQAAUfb+w
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138
In-Reply-To: <46EE4EF5.9080804@ericsson.com>
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=7029; t=1190058585;
	x=1190922585; c=relaxed/simple; s=sjdkim1004;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=dwing@cisco.com;
	z=From:=20=22Dan=20Wing=22=20<dwing@cisco.com>
	|Subject:=20RE=3A=20[BEHAVE]=20AD=20comments=20on=20draft-ietf-behave-rfc
	3489bis-10 |Sender:=20;
	bh=cC3h+OZpKBuMyHDvGv7tCpGyAba95mVQszzNPVxagqA=;
	b=rGb/W80EQdm4pl5HD89hxFlK9Efv0sKns2UWLiv9Lk+pSnXbpsEguCFh5S1bI10JTnYIq/S2
	MqlZksYMY9q6r3Xdr44aMw75/ChZCaRFBvCp28u9eYmj6mYW1Z0zw8Suy7D0fLesZ/TLYS+PGP
	pMkwyHtDHS0tg45yOTTsZ2Bns=;
Authentication-Results: sj-dkim-1; header.From=dwing@cisco.com; dkim=pass (s
	ig from cisco.com/sjdkim1004 verified; ); 
X-Spam-Score: -4.0 (----)
X-Scan-Signature: ee80a2074afbfe28d15369f4e74e579d
Cc: 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

 

> -----Original Message-----
> From: Magnus Westerlund [mailto:magnus.westerlund@ericsson.com] 
> Sent: Monday, September 17, 2007 2:55 AM
> To: draft-ietf-behave-rfc3489bis@tools.ietf.org; Behave WG
> Subject: [BEHAVE] AD comments on draft-ietf-behave-rfc3489bis-10
> 
> Hi,
> 
> I have performed a combined AD-review / shepherding review 
> and have some
> comments. I think some of these will require an updated version before
> proceeding. I hope this can be accomplished quickly to avoid 
> to much delay.
> 
> 
> 1. Section 7.2.1:
> It is not obvious why the number 7900 ms is the end of listening for
> responses. How did one arrive on that number?
> 
> 2. Section 12.2:
> "   A STUN server can detect when a given Binding Request message was
>    sent from an RFC 3489 [RFC3489] client by the absence of 
> the correct
>    value in the Magic Cookie field.  When the server detects 
> an RFC 3489
>    client, it SHOULD copy the value seen in the Magic Cookie field in
>    the Binding Request to the Magic Cookie field in the 
> Binding Response
>    message, and insert a MAPPED-ADDRESS attribute instead of an XOR-
>    MAPPED-ADDRESS attribute."
> 
> I think this text should have some warning about what functionality
> issues regarding MUXing STUN with other protocols may have due to this
> backwards compatibility issue. If I understand things 
> correctly, having
> 3489 compatibility means that demuxer only can use the two first bits
> and two last of the first word to separate STUN from normal 
> data packets?

(Writing as an individual:)

RFC3489 doesn't do multiplexing -- multiplexing of STUN with other
protocols was introduced with RFC3489bis (not RFC3489).

Should we make a note of how this is not a problem, though?  Perhaps
something along the lines of:

  "Note:  Backwards compatibility with RFC3489 does not affect 
   multiplexing of STUN with other protocols (e.g., [ICE], 
   [SIP-OUTBOUND]), because such multiplexing was not available
   in RFC3489."

-d


> 3. Section 14.4:
> 
> "   The key used as input to HMAC is the password."
> 
> Section 10.2:
> 
>    "For STUN servers used in conjunction with SIP servers, it is
>    desirable to use the same credentials for authentication to the SIP
>    server and STUN server.  Typically, SIP systems utilizing SIP's
>    digest authentication mechanism do not actually store the 
> password in
>    the database.  Rather, they store a value called H(A1), which is
>    computed as:
> 
>                 H(A1) = MD5(username ":" realm ":" password)
> 
>    If a system wishes to utilize this credential, the STUN password
>    would be computed by taking the user-entered username and password,
>    and using H(A1) as the STUN password.  It is RECOMMENDED 
> that clients
>    utilize this construction for the STUN password."
> 
> Isn't these two in contradiction. In addition aren't they a potential
> for interoperability issues between client and server. This as the
> actual key used in the HMAC must be the same to allow correct 
> integrity
> verification. So, how is it inteded to work? Is it something the usage
> needs to define?
> 
> 4. Section 15.1.2:
> 
> "  A rogue client may use a STUN server as a reflector, sending it
>    requests with a falsified source IP address and port.  In such a
>    case, the response would be delivered to that source IP and port.
>    There is no amplification with this attack, and it is mitigated by
>    ingress source address filtering."
> 
> I think there is missing a clarification on what type of amplication
> that is not possible with STUN. To my understanding there can't be any
> packet rate amplification in this attack. However some 
> percent of packet
> size amplification seems possible due to NONCE and other error
> indication added by the server. Maybe a clarification that there is no
> amplification in the number of packets is warrented.
> 
> 5. Section 15.2:
> 
> "   Most of the attacks in this section revolve around an attacker
>    modifying the reflexive address learned by a STUN client through a
>    Binding Request/Binding Response transaction.  Since the 
> usage of the
>    reflexive address is a function of the usage, the applicability and
>    remediation of these attacks is usage-specific.  In common
>    situations, modification of the reflexive address by an on-path
>    attacker is easy to do.  Consider, for example, the common 
> situation
>    where STUN is run directly over UDP.  In this case, an on-path
>    attacker can modify the source IP address of the Binding Request
>    before it arrives at the STUN server.  The STUN server will then
>    return this IP address in the XOR-MAPPED-ADDRESS attribute to the
>    client.  Protecting against this attack by using a 
> message-integrity
>    check is impossible, since a message-integrity value 
> cannot cover the
>    source IP address, since the intervening NAT must be able to modify
>    this value.  Instead, one solution to preventing the attacks listed
>    below is for the client to verify the reflexive address learned, as
>    is done in ICE [I-D.ietf-mmusic-ice].  Other usages may use other
>    means to prevent these attacks."
> 
> I think there are a serious thinking error that affects most of the
> attack variants in this description. If on on-path attacker 
> changes the
> source IP-address of the packet the packet will not be routed back to
> the client issuing the request. Thus they are looking like 
> lost packets.
> To me the only attack being really viable is if the attacker 
> changes the
> source address to itself and then forwards the packet to the client
> (target of the attack). This is of course assuming that the 
> attacker is
> not sitting very closely to the server and can see all/most traffic to
> and from the server?
> 
> Can we be a bit clearer on when the attacks are possible. Thus I think
> that 15.2.1 and 15.2.2 are more difficult to achieve 
> (assuming integrity
> checks between client and server) than 15.2.3 and 15.2.4.
> 
> 6. Section 15.3:
> 
> "   This specification uses SHA-1 for computation of the message
>    integrity. "
> 
> Shouldn't there be a HMAC before SHA-1 here?
> 
> 
> Cheers
> 
> Magnus Westerlund
> 
> IETF Transport Area Director & TSVWG Chair
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM/M
> ----------------------------------------------------------------------
> Ericsson AB                | Phone +46 8 4048287
> Torshamsgatan 23           | Fax   +46 8 7575550
> S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
> ----------------------------------------------------------------------
> 
> 
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www1.ietf.org/mailman/listinfo/behave


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 17 16:45:37 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXNSc-0002Jg-Gs; Mon, 17 Sep 2007 16:44:50 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXNSb-0002HN-DY
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 16:44:49 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXNSb-0002HF-3q
	for behave@ietf.org; Mon, 17 Sep 2007 16:44:49 -0400
Received: from server.implementers.org ([69.55.225.91])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IXNSU-0000xb-TU
	for behave@ietf.org; Mon, 17 Sep 2007 16:44:49 -0400
Received: by server.implementers.org (Postfix, from userid 1001)
	id 44255E1CD6D4; Mon, 17 Sep 2007 13:44:27 -0700 (PDT)
Received: from [192.168.1.3] (c-24-4-253-5.hsd1.ca.comcast.net [24.4.253.5])
	by server.implementers.org (Postfix) with ESMTP id C5F58E1CD6D0;
	Mon, 17 Sep 2007 13:44:24 -0700 (PDT)
Message-ID: <46EEE728.9070505@acm.org>
Date: Mon, 17 Sep 2007 13:44:24 -0700
From: Marc Petit-Huguenin <petithug@acm.org>
User-Agent: Mozilla-Thunderbird 2.0.0.4 (X11/20070828)
MIME-Version: 1.0
To: Benny Prijono <bennylp@pjsip.org>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>	<200708081203.19729.remi.denis-courmont@nokia.com>	<46BA4268.2060204@db.org>	<200708090918.13691.remi.denis-courmont@nokia.com>	<46BB4B0C.6030204@db.org>	<46C1ACF3.7000202@cisco.com>
	<46EEC6E5.9000406@pjsip.org>
In-Reply-To: <46EEC6E5.9000406@pjsip.org>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 7baded97d9887f7a0c7e8a33c2e3ea1b
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Benny Prijono wrote:
> Jonathan Rosenberg wrote:
>> An important point of clarification here.
>>
>> Though FINGERPRINT is currently the only attribute beyond the MI,
>> there could potentially be others in the future so you can't depend on
>> it being the only one. So inserting a dummy FINGERPRINT won't help.
> 
> Jonathan,
> 
> sorry for joining late on the discussion, but your comment concerns me.
> Allowing any other attribute, except FINGERPRINT, beyond M-I, not only
> violates rfc3489bis-10 but could also break existing implementation as
> well.
> 
> The part where it violates rfc3489bis-10 is in Section 14.5:
> 
> "When present, the FINGERPRINT attribute MUST be the last attribute in
> the message, and thus will appear after MESSAGE-INTEGRITY."
> 
> And the reason why it may break existing implementation is because, in
> order to avoid linear searching for M-I throughout all attributes,
> currently we could probe M-I at position (20+hdr.length-24), if the M-I
> is the last attribute, or at position (20+hdr.length-24-8), if the M-I
> appears as the attribute before FINGERPRINT.
> 
> Allowing some other attributes to exist between M-I and FINGERPRINT will
> break above algorithm, and it will mean linear search for the M-I. So I
> propose we disallow this.
> 

I think that changing the MUST in rfc3489bis-10 to a SHOULD is a better
solution.  Who knows if another attribute after the FINGERPRINT
attribute will not be needed in the future.

-- 
Marc Petit-Huguenin           [                                 ]
Home: marc@petit-huguenin.org [RFC1855-compliant space for rent ]
Work: marc@8x8.com            [                                 ]
[                                                               ]


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 17 19:40:59 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXQC7-00077C-QO; Mon, 17 Sep 2007 19:39:59 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXQC6-00076y-8i
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 19:39:58 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXQC5-00076p-Uw
	for behave@ietf.org; Mon, 17 Sep 2007 19:39:57 -0400
Received: from sj-iport-3-in.cisco.com ([171.71.176.72]
	helo=sj-iport-3.cisco.com) by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IXQC4-0005Vj-Ik
	for behave@ietf.org; Mon, 17 Sep 2007 19:39:57 -0400
X-IronPort-AV: E=Sophos;i="4.20,266,1186383600"; d="scan'208";a="524875227"
Received: from sj-dkim-2.cisco.com ([171.71.179.186])
	by sj-iport-3.cisco.com with ESMTP; 17 Sep 2007 16:39:52 -0700
Received: from sj-core-5.cisco.com (sj-core-5.cisco.com [171.71.177.238])
	by sj-dkim-2.cisco.com (8.12.11/8.12.11) with ESMTP id l8HNdufT021124; 
	Mon, 17 Sep 2007 16:39:56 -0700
Received: from dwingwxp01 ([10.32.240.198])
	by sj-core-5.cisco.com (8.12.10/8.12.6) with ESMTP id l8HNdkfB022845;
	Mon, 17 Sep 2007 23:39:47 GMT
From: "Dan Wing" <dwing@cisco.com>
To: "'Behave WG'" <behave@ietf.org>
References: <20d2bdfb0709122004o709211bbt9569b0f1d3ae1120@mail.gmail.com>
Subject: RE: [BEHAVE] behavior discovery category
Date: Mon, 17 Sep 2007 16:39:46 -0700
Message-ID: <04f401c7f984$0d6bdb00$c6f0200a@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Office Outlook 11
Thread-Index: Acf1suity6k6b2eEQIe6uRu9e+WW6AD0C+sA
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138
In-Reply-To: <20d2bdfb0709122004o709211bbt9569b0f1d3ae1120@mail.gmail.com>
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=6963; t=1190072396;
	x=1190936396; c=relaxed/simple; s=sjdkim2002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=dwing@cisco.com;
	z=From:=20=22Dan=20Wing=22=20<dwing@cisco.com>
	|Subject:=20RE=3A=20[BEHAVE]=20behavior=20discovery=20category
	|Sender:=20; bh=5oi1EanMMfnfnOca3HdCBzDB/xsZF4JvrAaLv2VHaXI=;
	b=DZ+AxnJy+I2ZqGDs1Y04RObrw+fk0lnDizj+KnCuJgHsYIJ8Qm9OGXdB2UnSRorMdT71IWhV
	k75uP18RqG8NBsrRx+JcUqXujC62dZt+ug16Yw6siSYWLUYvXF2/4b/R;
Authentication-Results: sj-dkim-2; header.From=dwing@cisco.com; dkim=pass (s
	ig from cisco.com/sjdkim2002 verified; ); 
X-Spam-Score: -4.0 (----)
X-Scan-Signature: 3d7f2f6612d734db849efa86ea692407
Cc: 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Folks, please look over Bruce's email below.  We need some discussion on
Informational versus Experimental for
draft-ietf-behave-nat-behavior-discovery.  The choice will control how other
specifications can reference the document.  As Bruce mentions below, see
http://www.ietf.org/u/ietfchair/info-exp.html for guidelines on Informational
versus Experimental.

Thanks,
-d

> -----Original Message-----
> From: Bruce Lowekamp [mailto:lowekamp@sipeerior.com] 
> Sent: Wednesday, September 12, 2007 8:05 PM
> To: Behave WG
> Subject: [BEHAVE] behavior discovery category
> 
> In Chicago we agreed to bring the question of what category
> (Experimental or Informational) was appropriate for behavior-discovery
> to the list.  Derek and I spent some time looking at this and at
> http://www.ietf.org/u/ietfchair/info-exp.html.  We believe that the
> appropriate status is experimental.   We have used P2P supernode
> selection as an example application, but believe that the case of
> applications without a rendevous server is more general.
> 
> I'm including the text from a revision of the draft that 
> addresses this issue.
> 
> Bruce
> 
> 
> Abstract
> 
>    This specification defines an experimental usage of the Simple
>    Traversal Underneath Network Address Translators (NAT) (STUN)
>    Protocol that discovers the presence and current behaviour of NATs
>    and firewalls between the STUN client and the STUN server.
> 
> 
> 
> 1.  Applicability
> 
>    This experimental STUN usage does not allow an application behind a
>    NAT to make an absolute determination of the NAT's characteristics.
>    NAT devices do not behave consistently enough to predict future
>    behaviour with any guarantee.  This STUN usage provides information
>    about observable transient behavior; it only truly 
> determines a NAT's
>    behavior with regard to the STUN server used at the 
> instant the test
>    is run.  Applications requiring reliable reach between two 
> particular
>    endpoints must establish a communication channel through a 
> NAT using
>    another technique.  IETF has proposed standards including ICE
>    [I-D.ietf-mmusic-ice] and OUTBOUND [I-D.ietf-sip-outbound] for
>    establishing communication channels when a publically accessible
>    rendevous service is available.
> 
>    This usage provides techniques known to be of use for diagnostic
>    purposes.  These techniques are considered experimental 
> when applied
>    to real-time selection of parameters for protocols in situations
>    where a publically accessible rendevous service is not available.
>    One possible application might be optimizing supernode selection in
>    P2P networks based on statistical experience with establishing
>    connections and diagnosing NAT behavior with a variety of other
>    peers.
> 
> 
> 2.  Introduction
> 
>    The Simple Traversal Underneath Network Address Translators (NAT)
>    (STUN) [I-D.ietf-behave-rfc3489bis] provides a mechanism 
> to discover
>    the reflexive transport address toward the STUN server, using the
>    Binding Request.  This specification defines the NAT Behavior
>    Discovery STUN usage, which allows a STUN client to probe 
> the current
>    behaviour of the NAT/FW devices between the client and the STUN
>    server.  This usage defines new STUN attributes for the Binding
>    Request and Binding Response.
> 
>    Many NAT/FW devices do not behave consistently and will 
> change their
>    behaviour under load and over time.  Applications requiring high
>    reliability must be prepared for the NAT's behaviour to become more
>    restrictive.  Specifically, it has been found that under load NATs
>    may transition to the most restrictive filtering and mapping
>    behaviour and shorten the lifetime of new and existing 
> bindings.  In
>    short, applications can discover how bad things currently are, but
>    not how bad things will get.
> 
>    Despite this limitation, instantaneous observations are often quite
>    useful in troubleshooting network problems, and repeated tests over
>    time, or in known load situations, may be used to characterize a
>    NAT's behavior.  In particular, in the hands of a person
>    knowledgeable about the needs of an application and the nodes an
>    application needs to communicate with, it can be a powerful tool.
>    For these diagnostics to be effectiev, the application 
> must select a
>    STUN server that is appropriately located with regards to 
> its future
>    communication partners.
> 
> 2.1.  Example Use with P2P Overlays
> 
>    An application could use Behavior Discovery in a P2P protocol to
>    determine if a particular endpoint is a reasonable supernode
>    candidate.  This P2P network application is willing to select
>    supernodes that might advertise unreliable IP addresses to 
> avoid the
>    cost of dedicated servers.  A supernode candidate requires that its
>    NAT(s) offer(s) Address Independent Filtering.  It might 
> periodically
>    re-run tests and would remove itself as a supernode if its NAT/FW
>    chain lost this characteristic.  These tests could be run 
> with other
>    supernode candidates acting as STUN servers as well as 
> dedicated STUN
>    servers.  As many P2P algorithms tolerate non-transitive 
> connectivity
>    between a portion of their peers, guaranteed pair-wise 
> reliable reach
>    might be sacrificed in order to distribute the P2P overlay's load
>    across peers that can be directly contacted by the 
> majority of users.
> 
>    Use of Behavior Discovery for such an application requires:
> 
>    o  Specification of protocols capable of offering reliable end-user
>       performance using unreliable links between peers.
> 
>    o  The application is deployed behind NATs that provide Address
>       Independent Filtering and that remain in this mode for an amount
>       of time sufficient for the application to identify 
> their behavior,
>       distribute this information to the rest of the overlay, and
>       provide useful work for the application.
> 
>    This draft is experimental as deployed applications 
> implementing open
>    protocols have yet to be deployed in such environments to 
> demonstrate
>    that these two requirements have been met.  However, apocryphal
>    evidence suggests that househould and small business targeted NAT
>    devices have stable behaviour, especially when there are 
> few clients
>    behind them.  And numerous P2P applications have been deployed that
>    appear to have these properties, although their protocols have not
>    yet been subjected to rigorous evaluation by standards bodies.
> 
> 
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www1.ietf.org/mailman/listinfo/behave


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Mon Sep 17 20:14:54 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXQit-0002c9-6j; Mon, 17 Sep 2007 20:13:51 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXQis-0002c3-2N
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 20:13:50 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXQir-0002bv-OE
	for behave@ietf.org; Mon, 17 Sep 2007 20:13:49 -0400
Received: from server.implementers.org ([69.55.225.91])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXQir-0001ay-CR
	for behave@ietf.org; Mon, 17 Sep 2007 20:13:49 -0400
Received: by server.implementers.org (Postfix, from userid 1001)
	id 8AB00E1CD6D4; Mon, 17 Sep 2007 17:13:48 -0700 (PDT)
Received: from [192.168.1.3] (c-24-4-253-5.hsd1.ca.comcast.net [24.4.253.5])
	by server.implementers.org (Postfix) with ESMTP id 15A75E1CD6D0;
	Mon, 17 Sep 2007 17:13:46 -0700 (PDT)
Message-ID: <46EF1839.70605@acm.org>
Date: Mon, 17 Sep 2007 17:13:45 -0700
From: Marc Petit-Huguenin <petithug@acm.org>
User-Agent: Mozilla-Thunderbird 2.0.0.4 (X11/20070828)
MIME-Version: 1.0
To: Brian Korver <briank@briank.com>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>	<200708081203.19729.remi.denis-courmont@nokia.com>	<46BA4268.2060204@db.org>	<200708090918.13691.remi.denis-courmont@nokia.com>	<46BB4B0C.6030204@db.org>	<46C1ACF3.7000202@cisco.com>
	<46EEC6E5.9000406@pjsip.org> <46EEE728.9070505@acm.org>
	<7329187B-6236-4E43-A305-8E2FB437F7E3@briank.com>
In-Reply-To: <7329187B-6236-4E43-A305-8E2FB437F7E3@briank.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 798b2e660f1819ae38035ac1d8d5e3ab
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Brian Korver wrote:
> On Sep 17, 2007, at 1:44 PM, Marc Petit-Huguenin wrote:
>> I think that changing the MUST in rfc3489bis-10 to a SHOULD is a better
>> solution.  Who knows if another attribute after the FINGERPRINT
>> attribute will not be needed in the future.
> 
> Making this a SHOULD means we can't count on it being last, that we must
> always do a sequential search for it.  I vote for sticking with MUST and
> revisiting the issue if/when a good reason presents itself.  

Well, the problem is that it will not be possible to change this
decision in a future RFC, because it will be incompatible with the
current version.

> The
> proposed change seems like premature future-proofing.

-- 
Marc Petit-Huguenin           [                                 ]
Home: marc@petit-huguenin.org [RFC1855-compliant space for rent ]
Work: marc@8x8.com            [                                 ]
[                                                               ]


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 18 04:24:26 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXYMl-0007hP-V7; Tue, 18 Sep 2007 04:23:31 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXYMk-0007g3-EQ
	for behave-confirm+ok@megatron.ietf.org; Tue, 18 Sep 2007 04:23:30 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXYMf-0007aJ-NI
	for behave@ietf.org; Tue, 18 Sep 2007 04:23:25 -0400
Received: from mailgw3.ericsson.se ([193.180.251.60])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXYMc-00042b-W5
	for behave@ietf.org; Tue, 18 Sep 2007 04:23:23 -0400
Received: from mailgw3.ericsson.se (unknown [127.0.0.1])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	4E6BD20752; Tue, 18 Sep 2007 10:23:22 +0200 (CEST)
X-AuditID: c1b4fb3c-b067fbb0000007e1-33-46ef8afacd27
Received: from esealmw129.eemea.ericsson.se (unknown [153.88.254.124])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	32B4120572; Tue, 18 Sep 2007 10:23:22 +0200 (CEST)
Received: from esealmw129.eemea.ericsson.se ([153.88.254.177]) by
	esealmw129.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 10:23:21 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw129.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 10:23:21 +0200
Message-ID: <46EF8AF9.1080608@ericsson.com>
Date: Tue, 18 Sep 2007 10:23:21 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Dan Wing <dwing@cisco.com>
Subject: Re: [BEHAVE] AD comments on draft-ietf-behave-rfc3489bis-10
References: <46EE4EF5.9080804@ericsson.com>
	<03b701c7f963$e5914590$c6f0200a@cisco.com>
In-Reply-To: <03b701c7f963$e5914590$c6f0200a@cisco.com>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 18 Sep 2007 08:23:21.0729 (UTC)
	FILETIME=[2CB91F10:01C7F9CD]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 31247fb3be228bb596db9127becad0bc
Cc: 'Behave WG' <behave@ietf.org>, draft-ietf-behave-rfc3489bis@tools.ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Dan Wing skrev:
>  
> 
>> -----Original Message-----
>> From: Magnus Westerlund [mailto:magnus.westerlund@ericsson.com] 
>> Sent: Monday, September 17, 2007 2:55 AM
>> To: draft-ietf-behave-rfc3489bis@tools.ietf.org; Behave WG
>> Subject: [BEHAVE] AD comments on draft-ietf-behave-rfc3489bis-10
>>
>> Hi,
>>
>> I have performed a combined AD-review / shepherding review 
>> and have some
>> comments. I think some of these will require an updated version before
>> proceeding. I hope this can be accomplished quickly to avoid 
>> to much delay.
>>
>>
>> 1. Section 7.2.1:
>> It is not obvious why the number 7900 ms is the end of listening for
>> responses. How did one arrive on that number?
>>
>> 2. Section 12.2:
>> "   A STUN server can detect when a given Binding Request message was
>>    sent from an RFC 3489 [RFC3489] client by the absence of 
>> the correct
>>    value in the Magic Cookie field.  When the server detects 
>> an RFC 3489
>>    client, it SHOULD copy the value seen in the Magic Cookie field in
>>    the Binding Request to the Magic Cookie field in the 
>> Binding Response
>>    message, and insert a MAPPED-ADDRESS attribute instead of an XOR-
>>    MAPPED-ADDRESS attribute."
>>
>> I think this text should have some warning about what functionality
>> issues regarding MUXing STUN with other protocols may have due to this
>> backwards compatibility issue. If I understand things 
>> correctly, having
>> 3489 compatibility means that demuxer only can use the two first bits
>> and two last of the first word to separate STUN from normal 
>> data packets?
> 
> (Writing as an individual:)
> 
> RFC3489 doesn't do multiplexing -- multiplexing of STUN with other
> protocols was introduced with RFC3489bis (not RFC3489).
> 
> Should we make a note of how this is not a problem, though?  Perhaps
> something along the lines of:
> 
>   "Note:  Backwards compatibility with RFC3489 does not affect 
>    multiplexing of STUN with other protocols (e.g., [ICE], 
>    [SIP-OUTBOUND]), because such multiplexing was not available
>    in RFC3489."
> 
> -d

Well, is that really obvious so that not anyone which have an 3489
implementation will try to apply it to a multiplexed case. I think
having a warning in this section which explains both why 3489 should not
be used, and therefore it really is not a problem.

"The RFC 3489 version of STUN lacks both the Magic Cookie and the
FINGERPRINT attribute that allows for a very high probablility of
correctly identifying STUN messages when multiplexed with other
protocols. Therefore, STUN implementations that are backwards compatible
with RFC 3489 SHOULD NOT be used in cases where STUN will be multiplexed
with another protocol. However, that should not be an issues as such
multiplexing was not available in RFC 3489."


Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 18 06:00:15 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXZqk-00005P-2q; Tue, 18 Sep 2007 05:58:34 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXZqj-00005K-4K
	for behave-confirm+ok@megatron.ietf.org; Tue, 18 Sep 2007 05:58:33 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXZqi-00005C-Qq
	for behave@ietf.org; Tue, 18 Sep 2007 05:58:32 -0400
Received: from smtp.nokia.com ([131.228.20.170] helo=mgw-ext11.nokia.com)
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IXZqc-0002NR-Fk
	for behave@ietf.org; Tue, 18 Sep 2007 05:58:32 -0400
Received: from esebh106.NOE.Nokia.com (esebh106.ntc.nokia.com [172.21.138.213])
	by mgw-ext11.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l8I9vs8G025774; Tue, 18 Sep 2007 12:58:03 +0300
Received: from esebh104.NOE.Nokia.com ([172.21.143.34]) by
	esebh106.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 12:57:59 +0300
Received: from esebh102.NOE.Nokia.com ([172.21.138.183]) by
	esebh104.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 12:57:59 +0300
Received: from esdhcp04176.research.nokia.com ([172.21.41.76]) by
	esebh102.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 12:57:58 +0300
From: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: behave@ietf.org
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
Date: Tue, 18 Sep 2007 12:57:58 +0300
User-Agent: KMail/1.9.6 (enterprise 0.20070907.709405)
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>
	<46C1ACF3.7000202@cisco.com> <46EEC6E5.9000406@pjsip.org>
In-Reply-To: <46EEC6E5.9000406@pjsip.org>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709181257.58901.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 18 Sep 2007 09:57:58.0462 (UTC)
	FILETIME=[6451D5E0:01C7F9DA]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: a7d6aff76b15f3f56fcb94490e1052e4
Cc: 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Le Monday 17 September 2007 21:26:45 ext Benny Prijono, vous avez =C3=A9cri=
t=C2=A0:
> sorry for joining late on the discussion, but your comment concerns
> me. Allowing any other attribute, except FINGERPRINT, beyond M-I,
> not only violates rfc3489bis-10 but could also break existing
> implementation as well.

If there is a known attribute after FINGERPRINT, someone is trying to screw=
=20
you and you MUST ignore the attribute.

If there is an unknown optional attribute after FINGERPRINT, you MUST ignor=
e=20
it anyway, regardless of where FINGERPRINT is.

If there is an unknown mandatory attribute after FINGERPRINT, you MUST drop=
=20
the entire message regardless of where FINGERPRINT is.

As such I do not see much of an issue, except the likeliness that someone w=
ill=20
implement the spec wrong.

> And the reason why it may break existing implementation is because,
> in order to avoid linear searching for M-I throughout all
> attributes, currently we could probe M-I at position
> (20+hdr.length-24), if the M-I is the last attribute, or at position
> (20+hdr.length-24-8), if the M-I appears as the attribute before
> FINGERPRINT.

You MUST do at least one linear pass over the entire STUN message, because =
you=20
MUST look for unknown mandatory attributes. Hence, that is a non-issue.

> Allowing some other attributes to exist between M-I and FINGERPRINT
> will break above algorithm, and it will mean linear search for the
> M-I. So I propose we disallow this.

IMHO, that algorithm is already broken.

=2D-=20
R=C3=A9mi Denis-Courmont


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 18 07:44:49 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXbUX-0006Wo-VO; Tue, 18 Sep 2007 07:43:45 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXbUW-0006Wj-Lc
	for behave-confirm+ok@megatron.ietf.org; Tue, 18 Sep 2007 07:43:44 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXbUW-0006Pn-Ar
	for behave@ietf.org; Tue, 18 Sep 2007 07:43:44 -0400
Received: from blaster.systems.pipex.net ([62.241.163.7])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXbUH-0000iB-Lz
	for behave@ietf.org; Tue, 18 Sep 2007 07:43:30 -0400
Received: from [61.94.220.135] (unknown [61.94.220.135])
	by blaster.systems.pipex.net (Postfix) with ESMTP id 82FECE000705;
	Tue, 18 Sep 2007 12:43:21 +0100 (BST)
Message-ID: <46EFB87C.5010307@pjsip.org>
Date: Tue, 18 Sep 2007 12:37:32 +0100
From: Benny Prijono <bennylp@pjsip.org>
User-Agent: Thunderbird 1.5.0.10 (Windows/20070221)
MIME-Version: 1.0
To: =?UTF-8?B?UsOpbWkgRGVuaXMtQ291cm1vbnQ=?= <remi.denis-courmont@nokia.com>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>
	<46C1ACF3.7000202@cisco.com> <46EEC6E5.9000406@pjsip.org>
	<200709181257.58901.remi.denis-courmont@nokia.com>
In-Reply-To: <200709181257.58901.remi.denis-courmont@nokia.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: quoted-printable
X-Spam-Score: 0.0 (/)
X-Scan-Signature: f60d0f7806b0c40781eee6b9cd0b2135
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

R=C3=A9mi Denis-Courmont wrote:
> Le Monday 17 September 2007 21:26:45 ext Benny Prijono, vous avez =C3=A9=
crit :
>> sorry for joining late on the discussion, but your comment concerns
>> me. Allowing any other attribute, except FINGERPRINT, beyond M-I,
>> not only violates rfc3489bis-10 but could also break existing
>> implementation as well.
>=20
> If there is a known attribute after FINGERPRINT, someone is trying to s=
crew=20
> you and you MUST ignore the attribute.
>=20
> If there is an unknown optional attribute after FINGERPRINT, you MUST i=
gnore=20
> it anyway, regardless of where FINGERPRINT is.
>=20
> If there is an unknown mandatory attribute after FINGERPRINT, you MUST =
drop=20
> the entire message regardless of where FINGERPRINT is.
>=20
> As such I do not see much of an issue, except the likeliness that someo=
ne will=20
> implement the spec wrong.

The issue that I was asking was for attributes between M-I and
FINGERPRINT, not the ones after FINGERPRINT.

Interestingly though, the processing for attributes beyond
FINGERPRINT like above is not mentioned anywhere in the spec. What I
found is:

Section 14.5
   When present, the FINGERPRINT attribute MUST be the last attribute
   in the message, and thus will appear after MESSAGE-INTEGRITY.

So I think we should drop the entire message if attributes are found
beyond FINGERPRINT, regardless whether its comprehension is required
or not.


>> And the reason why it may break existing implementation is because,
>> in order to avoid linear searching for M-I throughout all
>> attributes, currently we could probe M-I at position
>> (20+hdr.length-24), if the M-I is the last attribute, or at position
>> (20+hdr.length-24-8), if the M-I appears as the attribute before
>> FINGERPRINT.
>=20
> You MUST do at least one linear pass over the entire STUN message, beca=
use you=20
> MUST look for unknown mandatory attributes. Hence, that is a non-issue.

True that linear searching will have to be done in parsing, and
checking for invalid attributes is done in parsing stage. But not
everybody does everything in a single function. Some will have
authentication in separate layer after parsing is done.

>> Allowing some other attributes to exist between M-I and FINGERPRINT
>> will break above algorithm, and it will mean linear search for the
>> M-I. So I propose we disallow this.
>=20
> IMHO, that algorithm is already broken.

Not if we only allow FINGERPRINT after M-I. And certainly that was=20
valid in bis-06, which I based my implementation on. :-(

  -benny




_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 18 08:15:31 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXbyi-0001qI-UR; Tue, 18 Sep 2007 08:14:56 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXbyh-0001oO-Iq
	for behave-confirm+ok@megatron.ietf.org; Tue, 18 Sep 2007 08:14:55 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXbyh-0001oG-7t
	for behave@ietf.org; Tue, 18 Sep 2007 08:14:55 -0400
Received: from smtp.nokia.com ([131.228.20.172] helo=mgw-ext13.nokia.com)
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXbyg-0001Mp-Fv
	for behave@ietf.org; Tue, 18 Sep 2007 08:14:55 -0400
Received: from esebh105.NOE.Nokia.com (esebh105.ntc.nokia.com [172.21.138.211])
	by mgw-ext13.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l8ICE1ds020998; Tue, 18 Sep 2007 15:14:48 +0300
Received: from esebh104.NOE.Nokia.com ([172.21.143.34]) by
	esebh105.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 15:14:30 +0300
Received: from esebh101.NOE.Nokia.com ([172.21.138.177]) by
	esebh104.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 15:10:58 +0300
Received: from esdhcp04176.research.nokia.com ([172.21.41.76]) by
	esebh101.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 15:10:58 +0300
From: =?utf-8?q?R=C3=A9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: "ext Benny Prijono" <bennylp@pjsip.org>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
Date: Tue, 18 Sep 2007 15:10:58 +0300
User-Agent: KMail/1.9.6 (enterprise 0.20070907.709405)
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>
	<200709181257.58901.remi.denis-courmont@nokia.com>
	<46EFB87C.5010307@pjsip.org>
In-Reply-To: <46EFB87C.5010307@pjsip.org>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200709181510.58421.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 18 Sep 2007 12:10:58.0062 (UTC)
	FILETIME=[F8882AE0:01C7F9EC]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 4b800b1eab964a31702fa68f1ff0e955
Cc: behave@ietf.org, Dan Wing <dwing@cisco.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Le Tuesday 18 September 2007 14:37:32 ext Benny Prijono, vous avez =C3=A9cr=
it=C2=A0:
> > If there is a known attribute after FINGERPRINT, someone is trying to
> > screw you and you MUST ignore the attribute.
> >
> > If there is an unknown optional attribute after FINGERPRINT, you MUST
> > ignore it anyway, regardless of where FINGERPRINT is.
> >
> > If there is an unknown mandatory attribute after FINGERPRINT, you MUST
> > drop the entire message regardless of where FINGERPRINT is.
> >
> > As such I do not see much of an issue, except the likeliness that someo=
ne
> > will implement the spec wrong.
>
> The issue that I was asking was for attributes between M-I and
> FINGERPRINT, not the ones after FINGERPRINT.

My mistake. Simply s/FINGERPRINT/MESSAGE-INTEGRITY/g in the previous=20
sentences, and you get my point.

> Interestingly though, the processing for attributes beyond
> FINGERPRINT like above is not mentioned anywhere in the spec. What I
> found is:
>
> Section 14.5
>    When present, the FINGERPRINT attribute MUST be the last attribute
>    in the message, and thus will appear after MESSAGE-INTEGRITY.
>
> So I think we should drop the entire message if attributes are found
> beyond FINGERPRINT, regardless whether its comprehension is required
> or not.

Right. I meant MESSAGE-INTEGRITY.

> >> And the reason why it may break existing implementation is because,
> >> in order to avoid linear searching for M-I throughout all
> >> attributes, currently we could probe M-I at position
> >> (20+hdr.length-24), if the M-I is the last attribute, or at position
> >> (20+hdr.length-24-8), if the M-I appears as the attribute before
> >> FINGERPRINT.
> >
> > You MUST do at least one linear pass over the entire STUN message,
> > because you MUST look for unknown mandatory attributes. Hence, that is a
> > non-issue.
>
> True that linear searching will have to be done in parsing, and
> checking for invalid attributes is done in parsing stage. But not
> everybody does everything in a single function. Some will have
> authentication in separate layer after parsing is done.

You can always flag attributes that are after M-I during the parsing phase.=
=20
Then it only adds O(1) time per attribute (and O(1) memory).

> >> Allowing some other attributes to exist between M-I and FINGERPRINT
> >> will break above algorithm, and it will mean linear search for the
> >> M-I. So I propose we disallow this.
> >
> > IMHO, that algorithm is already broken.
>
> Not if we only allow FINGERPRINT after M-I. And certainly that was
> valid in bis-06, which I based my implementation on. :-(

STUN NAT control needs to be able to add stuff after M-I already. This chan=
ge=20
has been introduced very much on-purpose, as far as I understand.

=2D-=20
R=C3=A9mi Denis-Courmont


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 18 09:48:47 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXdQn-0001gd-Ce; Tue, 18 Sep 2007 09:48:01 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXdQm-0001gY-NH
	for behave-confirm+ok@megatron.ietf.org; Tue, 18 Sep 2007 09:48:00 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXdQm-0001fS-Aq
	for behave@ietf.org; Tue, 18 Sep 2007 09:48:00 -0400
Received: from web33301.mail.mud.yahoo.com ([68.142.206.116])
	by chiedprmail1.ietf.org with smtp (Exim 4.43) id 1IXdQl-0003Pi-6J
	for behave@ietf.org; Tue, 18 Sep 2007 09:48:00 -0400
Received: (qmail 64732 invoked by uid 60001); 18 Sep 2007 13:47:58 -0000
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID;
	b=nZXD+EHGW+InUSOVl/UMtknLlqlRks5Djg/MbdvuxLRk21A6efkfItevdOBOZDZfu9IMyhsqL+q2Mb40MvgQV/7lC/yiMsMBq+jJ46TEkWMw/jCQ90Qw+48VrE4G0KSLPIzGevQ0NPmg990CKPucyP1FKrh+btFV9Ypl+8wJRS8=;
X-YMail-OSG: RIrwTYAVM1lAmbBfrhTjYx_eTzFZSd6fSgszrvXjkIY_.VxOe9VDuf0T4u1ZZbaDaqhnOw--
Received: from [69.236.94.7] by web33301.mail.mud.yahoo.com via HTTP;
	Tue, 18 Sep 2007 06:47:58 PDT
Date: Tue, 18 Sep 2007 06:47:58 -0700 (PDT)
From: Pyda Srisuresh <srisuresh@yahoo.com>
To: Magnus Westerlund <magnus.westerlund@ericsson.com>,
	Bryan Ford <baford@mit.edu>, Dan Kegel <dank@kegel.com>,
	Behave WG <behave@ietf.org>
In-Reply-To: <46DD3F15.2030808@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <639293.64700.qm@web33301.mail.mud.yahoo.com>
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 3f3e54d3c03ed638c06aa9fa6861237e
Cc: 
Subject: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Magnus,

Thank you for the review. Apologize for my delay in responding. Please see my
responses inline.

regards,
suresh
--- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:

> Hi,
> 
> I have done my AD review of this document and have the below comments.
> If you don't agree with them, please discuss them with me.
> 
> 1. Section 1, page 3, first paragraph:
> "These issues are
>    likely to persist even into the IPv6 world, where a NAT may be used
>    as an IPv4 compatibility mechanism [NAT-PT]."
> 
> I think you should rewrite this considering that NAT-PT was deprecated.
> 
[suresh] Would the following work for you? I believe, there might be NAT-PT
deployments, even as [NAT-PT] is deprecated.

     These issues are 
     likely to persist even into the IPv6 world and in deployments 
     using [NAT-PT].

> 2. Section 1, page 3, second paragraph:
> 
> "The
>    anonymity and inaccessibility of the internal hosts behind a NAT
>    device is not a problem for applications such as web browsers, which
>    only need to initiate outgoing connections. This inaccessibility is
>    sometimes perceived as a privacy benefit."
> 
> Is really "privacy" the main benefit. Isn't it the filtering behavior of
> the NAT that blocks incoming requests and thus avoid getting the traffic
> onto your local network that people like, i.e. the "firewalling" aspect
> of NATs.
> 
[suresh] The internal hosts have private IP addresses, which outsiders have no
knowledge of. This anonymity of IP addressing is considered as privacy by many
folks. In addition, a host behind NAT is not accessible to outsiders, unless a
session is initiated from inside. This inaccessibility is also considered as
privacy by many folks. 

You might say, the filtering behavior of NATs provides inaccessibility to
internal hosts; and mapping behavior of NATs provides anonymity to intenral
hosts.

> 3. Section 1, page 3, second to last paragraph:
> 
> "  NAT traversal strategies that involve explicit signaling between
>    applications and  NAT devices, namely [NAT-PMP], [NSIS-NSLP],
>    [SOCKS], [RSIP], [MIDCOM], and [UPNP] are out of the scope of this
>    document. [UNSAF] is in scope."
> 
> I agree that these are not in scope, but I think there should be some
> mention that these techiques if available are a complement to the
> techniques mentioned, allowing a client behind a NAT behave as if it was
> having a public IP address, or at least an NAT external address.
> 
[suresh] No problem. I could add a sentence to that effect.

> 4. Section 2, first paragraph:
> "As
>    for the notation for ports usage, all clients use ports in the
>    range of 1-1000 and servers use ports in the range of
>    20000-21000. NAT devices use ports 30000 and above for endpoint
>    mapping."
> 
> I think there are an error in this description. To me it appears that
> clients are using addresses in the range 1025-2000 in the examples.
> 
[suresh] Good catch. I will reword the first sentence to say that clients use
ports in the raneg of 1-2000. Thanks.

> 5. Section 2.7:
> 
>   "A P2P application is an application that uses the same endpoint to
>    initiate outgoing sessions to peering hosts as well as accept
>    incoming sessions from peering hosts."
> 
> Considering the definition of "endpoint" in this document I don't think
> this definition is characteristic for peer to peer applications. An peer
> to peer application clearly can use different address and ports to
> communicate with other peers.
> 
[suresh] Magnus - Would it work for you if the definition is expanded to say
that a P2P application may use multiple endpoints for P2P communication?  The
primary characteristic of a P2P applictaion is that the same endpoint is used
for initiating an outgoing connection as well to accept incoming connections.
All examples and deployment instances described in the document assume this
definition.

An application that uses one endpoint to receive incoming connections and a
different endpoint for outgoing connections is not a P2P application.

> 6. Section 2.8:
> "  A NAT-friendly P2P application registers with a publicly
>    addressable rendezvous server, used for registration and peer
>    discovery purposes. Pursuant to registering with rendezvous
>    server, the application uses its private endpoint, public
>    endpoint, or a combination thereof to establish peering sessions."
> 
> I don't think this is a strict requierment for a NAT-friendly P2P
> application. It is rather an effect of being one of the simplest way of
> accomplishing the rendezvous between two peers.
> 
[suresh] Would the following rewording work for you? Thanks.

     One common way P2P applications establish peering sessions and 
     remain NAT friendly is by using a publicly addressable rendezvous
     server for registration and  peer discovery purposes.

> 7. Section 2.9. I think one should be explicit that an EIM-NAT can have
> any filtering behavior.
> 
[suresh] Not a problem. That is the intent. I will state that explicitly.

> 8. Section 3.1, last paragraph:
> 
>   "The obvious disadvantages of
>    relaying are that it consumes the server's processing power and
>    network bandwidth, and communication latency between the peering
>    clients is likely to be increased even if the server is
>    well-connected."
> 
> I think "well-connected" is the wrong word here. I think it is rather
> topology wise correctly located.
> 
[suresh] By Well-connected, I was refering to having large bandwidth (say, in a
data center with GigE connectivity). This is required, as relay servers can be
in the path of many diverse, simultaneous sessions. The relay server ought to
have aggregate bandwidth to support the numerous parallel connections. I am not
refering to any topology-wise correctness.


> 9. Section 3.1, last paragraph:
> 
> "The TURN protocol [TURN] defines a method of
>    implementing relaying in a relatively secure fashion."
> 
> I think there should be some expansion on what type of relay service
> that TURN provides. For certain relay services a more application aware
> relay may be the better choice rather then packet payload relaying.
> 
[suresh] I am referign to application agnostic, session specific packet relay
with TURN. I will state that explicitly.

> 10. Section 3.2, page 10:
> 
> "In the first case, traffic directed to IP address
>    10.0.0.1 will simply be dropped by the network because 10.0.0.1 is
>    not a publicly routable IP address."
> 
> I think, this sentence needs to be expanded to cover the issue that the
> traffic might hit another host and any application listening on this port.
> 

[suresh] In the specific example, the connection is initiated by a public
client. SO, there is no issue of private addresses beign in the same domain,
right?

> 11. Section 3.3 or 3.3.1:
> 
> Somewhere here there need to be a description on why this fails with non
> EIM-NATs.
> 

[suresh] OK, I will add description to that effect. When one of the hosts is
behind a NAT that is not EIM-NAT, the peering host cannot predictably know the
mapped endpoint to which to initiate connection. Further, the application on
the host behind non EIM-NAT would be unable to reuse an already established
endpoint mapping for communication with different external destinations and the
hole-punching technique would fail.


> 12. Section 3.3.1:
> 
> "The application does
>    not need to attempt to detect the kind of NAT device it is behind,
>    as in [STUN], since the procedure above will establish peer-to-peer
>    communication channels equally well if either or both clients do not
>    happen to be behind a NAT device."
> 
> Well, in most usages of STUN you don't need to detect the bevior of the
> NAT. I assume that you refer to the model of  using STUN to find the
> external address and port and then advertise it. Not knowing the bevior
> of the NAT only results in one knowing if it works or not. I think the
> wording is a bit missfortunate, as it seems to say that STUN is really
> bad, when it is only a specific usage that is problematic.
> 

[suresh] I dont believe, the wording says that STUN is bad. If you like, I
could remove reference to [STUN] here.


> 13. Section 3.3.2, Figure 4:
> 
> I think the example is strange. What I can see if A and B sends its
> request starting with Registry, then try1 and then try2, try1 should
> succed. Also these messages are not well connected to the text. I think
> you should swap the order and then reference them a bit better in the text.
> 

[suresh] You are right. Must have been a cut and paste error. I will swap the
try order in the diagram and reword accordingly. Thanks.

> 14. Section 3.3.2, last paragraph:
>   "[ICE] protocol employs this technique effectively, in that
>    multiple candidate endpoints (both private and public) are
>    communicated between peering end hosts during offer/response
>    exchange."
> 
> I think you should use either offer/answer or request/response, not mix it.
> 
[suresh] OK.

> 15. section 5.1, second paragraph:
> 
> "When the NAT device
>    enroute is not EIM-NAT, the application is unable to reuse an
>    already established endpoint mapping for communication with
>    different external destinations and the technique would fail."
> 
> I think "enroute" is the wrong word. It is the outermost NAT that isn't
> an EIM-NAT for it to matter.
> 
[suresh] As discussed with Remi on another thread, I believe, "enroute" is
proper usage here.

Thanks again, for your detailed comments. Once you are Ok with my responses, i
will go ahead and generate a new rev of the draft. 

regards,
suresh

> Cheers
> 
> Magnus Westerlund
> 
> IETF Transport Area Director & TSVWG Chair
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM/M
> ----------------------------------------------------------------------
> Ericsson AB                | Phone +46 8 4048287
> Torshamsgatan 23           | Fax   +46 8 7575550
> S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
> ----------------------------------------------------------------------
> 





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 18 10:06:56 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXdi6-0007pB-Ra; Tue, 18 Sep 2007 10:05:54 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXdi6-0007ow-3c
	for behave-confirm+ok@megatron.ietf.org; Tue, 18 Sep 2007 10:05:54 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXdi5-0007og-Q0
	for behave@ietf.org; Tue, 18 Sep 2007 10:05:53 -0400
Received: from web33315.mail.mud.yahoo.com ([68.142.206.130])
	by ietf-mx.ietf.org with smtp (Exim 4.43) id 1IXdhy-0007gA-Do
	for behave@ietf.org; Tue, 18 Sep 2007 10:05:53 -0400
Received: (qmail 86561 invoked by uid 60001); 18 Sep 2007 14:05:34 -0000
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID;
	b=2u6n19IVPflNhYelOnKGohofdTptHLEbWPZT75zb/tuwumUOCL39Fo87tgP9+oE/2S/9FX9znGTvLUCxK76oER50VeQ87xdgGSrAUypm5mcw8Hn7ZUKuA5CUf6WvJo3l6uaJPgpTvz+JVRCzcBi0zv7duiSZAnMta58+L7js+1E=;
X-YMail-OSG: fk_hItgVM1nMro8EM1yjYuVjtPtymtNmaReEGHD4Xs3FsExeXzuShH_xrvXaW6GfJIQvJw--
Received: from [69.236.94.7] by web33315.mail.mud.yahoo.com via HTTP;
	Tue, 18 Sep 2007 07:05:33 PDT
Date: Tue, 18 Sep 2007 07:05:33 -0700 (PDT)
From: Pyda Srisuresh <srisuresh@yahoo.com>
To: Magnus Westerlund <magnus.westerlund@ericsson.com>
In-Reply-To: <46EE98C5.4080403@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <9493.86537.qm@web33315.mail.mud.yahoo.com>
X-Scan-Signature: 3971661e40967acfc35f708dd5f33760
Cc: Bryan Ford <baford@mit.edu>, Behave WG <behave@ietf.org>
Subject: [BEHAVE] Re: AD comments on draft-ietf-behave-icmp-04
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Magnus,

Thank you for your responses. Please see my comments inline. I am only keeping
items needing additional clarity. Thanks.

regards,
suresh

--- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:

<.... stuff deletec ...>

> >> 2. Section 2, last sentence:
> >>
> >> "A NAT MAY drop
> >>    or appropriately handle a post-1122 ICMP messages. "
> >>
> >> I think a normative statement of this type do not belong in the
> >> terminology section. Please find a better place to locate it.
> >>
> > [suresh] OK. I will move this to the first paragraph of section 8 (Summary
> of
> > Requirements).
> 
> Will that not be strange without having a requirement that only is
> listed in the summary?
> 
[suresh] I will preface by saying that the document focuses on the requirements
concerning ICMP queries and ICMP error messages. SO, i think, it is Ok to
follow this up and say that "a NAT MAY drop or appropriately handle a post-1122
ICMP messages". All this would be in the top paragraph of the summary section.

> > 
> >> 3. Section 3.2, REQ-2:
> >>
> >>   "Setting the ICMP NAT session timeout to a very large duration (say,
> >>    240 seconds) could potentially tie up precious NAT resources such as
> >>    query mappings and NAT Sessions for the whole duration. On the other
> >>    hand, setting the timeout very low can result in premature freeing
> >>    of NAT resources and applications failing to complete gracefully.
> >>    The ICMP Query session timeout needs to be a balance between the two
> >>    extremes. 60 seconds timeout is a balance between the two extremes.
> >>    ICMP query session timer MUST not expire in less than 60 seconds. We
> >>    RECOMMEND however that the administrator(s) be allowed to configure
> >>    the timer.
> >>
> >>    REQ-2: An ICMP Query session timer MUST NOT expire in less than 60
> >>    seconds.
> >>    a) It is RECOMMENDED that the ICMP Query session timer be made
> >>    configurable."
> >>
> >> Once more we use NATs as a motivation to redefine what our constants
> >> are. I can't understand the motivation behind making the requirement to
> >> anything other than 2xMSL, i.e. 240 s. I fear that people will start use
> >> lower values for MSL, due to the fact that they know that if there are a
> >> NAT there is anyway no possibility for the packet to live. Yes, it is a
> >> long time, but MSL exist for being a real outer boundary for how long
> >> packets can live.
> >>
> > [suresh] I didnt understand the comment - "I fear that people will start
> use
> > lower values for MSL, due to the fact that they know that if there are a
> NAT
> > there is anyway no possibility for the packet to live."  Why is there no
> > possibility for the packet to live when there is a NAT?
> 
> Well, the packet may live, but the response will never reach the
> requestor if the context needed to translate it has been timed out. That
> is why I fear that one might start optimizing these timers for something
> else.
> 
[suresh] Ah, now I understand what you meant.

> > 
> > There was much discussion on the ICMP timeouts on the list.  RFC 4787
> (Req-5a)
> > also provided some precedence on this. The WG agreed on 60sec timeout. In
> > practice, users of query based applications (ex: ICMP ping and ICMP
> traceroute)
> > are likely to be impatient if the response packets took longer than 60
> secs.
> > 240 secs., woudl be longer than even the udp TIMEOUT (120secs).
> 
> Well, the WG can consider my displeasure with this, but I do accept the
> WG consensus on it.
> 
[suresh] Got it. Thanks.

<... stuff deleted>


> >> 5. Section 4.2, first paragraph:
> >> "Section 4.3 of RFC 3022 describes the fields of an ICMP error
> >>    message that a NAT device translates. In this section, we describe
> >>    the requirements a NAT device MUST conform to while performing the
> >>    translations. Requirements identified in this section are necessary
> >>    for the current applications to work correctly."
> >>
> >> I interpret this as an normative reference. This would be a normative
> >> reference from a BCP to a informative document, which I like to avoid.
> >> Can we please rework this, for example by including how the translation
> >> is done in this document instead.
> >>
> > [suresh] Implicitly, both [BEH-UDP] and [BEH-TCP] also defer to RFC 3022
> for
> > the NAT translation. In that sense, this docuemnt is no different. Would it
> > work for you if I simply removed the first sentence.  
> 
> Well, I don't see how removing the sentence will make things better. It
> seems that we likely need to apply RFC 3967 exception on this reference.
> So please upgrade it to a real refence in the normative part.
> 

[suresh] You are saying that I should move [NAT-TRAD] to the normative section,
right. 

Thanks.

regards,
suresh

> > 
> >> 6. Section 4.2.2, page 12, second paragraph:
> >>
> >>    In addition, if the ICMP Error payload contains ICMP extensions
> >>    ([ICMP-EXT]), the NATs are encouraged to support ICMP extension
> >>    objects.
> >>
> >> I think this sentence is strangely formulated. Please reword it.
> >>
> > [suresh] would the following rewording work for you? 
> > 
> >    ICMP Error payload may contain ICMP extension objects ([ICMP-EXT]). 
> >    NATs are  encouraged to support ICMP extension objects.
> 
> 
> Yes.
> 
> Cheers
> 
> Magnus Westerlund
> 
> IETF Transport Area Director & TSVWG Chair
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM/M
> ----------------------------------------------------------------------
> Ericsson AB                | Phone +46 8 4048287
> Torshamsgatan 23           | Fax   +46 8 7575550
> S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
> ----------------------------------------------------------------------
> 





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 18 12:05:59 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXfa2-0000vA-E7; Tue, 18 Sep 2007 12:05:42 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXfa1-0000v4-Hq
	for behave-confirm+ok@megatron.ietf.org; Tue, 18 Sep 2007 12:05:41 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXfa0-0000td-4C
	for behave@ietf.org; Tue, 18 Sep 2007 12:05:41 -0400
Received: from astro.systems.pipex.net ([62.241.163.6])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IXfZs-00030z-Cz
	for behave@ietf.org; Tue, 18 Sep 2007 12:05:40 -0400
Received: from [61.94.220.132] (unknown [61.94.220.132])
	by astro.systems.pipex.net (Postfix) with ESMTP id 8A2E6E0007A1;
	Tue, 18 Sep 2007 17:04:47 +0100 (BST)
Message-ID: <46EFCB0E.5000306@pjsip.org>
Date: Tue, 18 Sep 2007 13:56:46 +0100
From: Benny Prijono <bennylp@pjsip.org>
User-Agent: Thunderbird 1.5.0.10 (Windows/20070221)
MIME-Version: 1.0
To: =?UTF-8?B?UsOpbWkgRGVuaXMtQ291cm1vbnQ=?= <remi.denis-courmont@nokia.com>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>
	<200709181257.58901.remi.denis-courmont@nokia.com>
	<46EFB87C.5010307@pjsip.org>
	<200709181510.58421.remi.denis-courmont@nokia.com>
In-Reply-To: <200709181510.58421.remi.denis-courmont@nokia.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: quoted-printable
X-Spam-Score: 0.4 (/)
X-Scan-Signature: f4c2cf0bccc868e4cc88dace71fb3f44
Cc: behave@ietf.org, Dan Wing <dwing@cisco.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

R=C3=A9mi Denis-Courmont wrote:
> Le Tuesday 18 September 2007 14:37:32 ext Benny Prijono, vous avez =C3=A9=
crit :
>>> If there is a known attribute after FINGERPRINT, someone is trying to
>>> screw you and you MUST ignore the attribute.
>>>
>>> If there is an unknown optional attribute after FINGERPRINT, you MUST
>>> ignore it anyway, regardless of where FINGERPRINT is.
>>>
>>> If there is an unknown mandatory attribute after FINGERPRINT, you MUS=
T
>>> drop the entire message regardless of where FINGERPRINT is.
>>>
>>> As such I do not see much of an issue, except the likeliness that som=
eone
>>> will implement the spec wrong.
>> The issue that I was asking was for attributes between M-I and
>> FINGERPRINT, not the ones after FINGERPRINT.
>=20
> My mistake. Simply s/FINGERPRINT/MESSAGE-INTEGRITY/g in the previous=20
> sentences, and you get my point.

In that case, the attributes must be ignored instead, as per 14.4:

  "With the exception of the FINGERPRINT attribute, which appears=20
after MESSAGE-INTEGRITY, agents MUST ignore all other attributes=20
that follow MESSAGE-INTEGRITY."

>> True that linear searching will have to be done in parsing, and
>> checking for invalid attributes is done in parsing stage. But not
>> everybody does everything in a single function. Some will have
>> authentication in separate layer after parsing is done.
>=20
> You can always flag attributes that are after M-I during the parsing ph=
ase.=20
> Then it only adds O(1) time per attribute (and O(1) memory).

True, software can always be changed (unfortunately!).

>> Not if we only allow FINGERPRINT after M-I. And certainly that was
>> valid in bis-06, which I based my implementation on. :-(
>=20
> STUN NAT control needs to be able to add stuff after M-I already. This =
change=20
> has been introduced very much on-purpose, as far as I understand.

Okay, I guess I'll have to accept that.

I was just updating my software from 06 to 10, and it seems that I=20
missed reading the part where non-FINGERPRINT attribute is now=20
allowed past M-I, thus I was quite surprised to hear Jonathan's comment.

Back to coding now..

thanks,
  -benny




_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 18 13:27:55 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXgqj-0000TR-0d; Tue, 18 Sep 2007 13:27:01 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXgqd-0000Ne-Vm
	for behave-confirm+ok@megatron.ietf.org; Tue, 18 Sep 2007 13:26:55 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXgqd-0000NA-4j
	for behave@ietf.org; Tue, 18 Sep 2007 13:26:55 -0400
Received: from server.implementers.org ([69.55.225.91])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IXgqW-0004r1-Ob
	for behave@ietf.org; Tue, 18 Sep 2007 13:26:55 -0400
Received: by server.implementers.org (Postfix, from userid 1001)
	id 0367CE1CD6D4; Tue, 18 Sep 2007 10:26:13 -0700 (PDT)
Received: from [192.168.1.3] (c-24-4-253-5.hsd1.ca.comcast.net [24.4.253.5])
	by server.implementers.org (Postfix) with ESMTP id A30E0E1CD6D0;
	Tue, 18 Sep 2007 10:26:10 -0700 (PDT)
Message-ID: <46F00A32.8090708@acm.org>
Date: Tue, 18 Sep 2007 10:26:10 -0700
From: Marc Petit-Huguenin <petithug@acm.org>
User-Agent: Mozilla-Thunderbird 2.0.0.4 (X11/20070828)
MIME-Version: 1.0
To: Jonathan Rosenberg <jdrosen@cisco.com>
Subject: Re: [BEHAVE] TURN - some proposed changes for improvement
	and	simplification
References: <46E81AFE.6040705@cisco.com>
In-Reply-To: <46E81AFE.6040705@cisco.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 2e8fc473f5174be667965460bd5288ba
Cc: Behave WG <behave@ietf.org>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Hi Jonathan,

I like very much the simplification but I have some comments.

As I understand it, permissions contains only the IP address of the
peer.  So if a channel is created for a specific IP address and port,
when data will be coming from the peer from a different port they will
be accepted by the permission but there is no mechanism in the channel
to send the real port to the TURN client.

Also I think that perhaps the Send and Connect methods can be merged
into one request, like Create Channel (and add a Destroy Channel for
symmetry).


Jonathan Rosenberg wrote:
> 
> As Dan mentioned at the last IETF meeting, we've formed a small team to
> actively work on TURN over the next month, in order to accelerate its
> completion and produce a version ready for WGLC by the end of September.
> 
> As part of this, we took a step back and looked a bit at some of the
> more troubling aspects of TURN. Three in particular seemed important:
> 
> * There are several different methods of framing and encapsulation of
> messages with a number of exceptions based on transport protocol,
> contents of the data, and the active destination.  The overall effect is
> confusing and rather complex.
> 
> * The active destination concept is inelegant, confusing, complex, and
> during forking results in undesirable overhead and at least one
> timing-related race condition.
> 
> * the use of multiple TCP connections from the TURN server towards
> peers, when there is one TCP connection from the TURN server to the
> client, has been a concern on the list recently.
> 
> We have a proposal on some non-trivial changes to TURN on these topics.
> 
> Firstly, we can eliminate the entire need for active destinations (and
> the Send and Data primitives as well) by introducing the concept of a
> 'channel'. Each channel is a stream of application data sent to and from
> a remote IP and port. In the case of UDP, we'd add a SINGLE BYTE to the
> front of every packet to and from the TURN server, which is used to
> signal the channel number. Channel 0 is reserved for TURN signaling to
> and from the client and TURN server. The remaining 255 channels are for
> remote destinations and carry pure data after the 1-byte channel number.
> So when a client does a Send or a Connect, the server allocates a
> remaining channel number and passes this channel number back to the
> client. From that point on, when the TURN server receives data from the
> peer, it sends it to the client and adds only the byte up front
> containing the channel number. Similarly, when the client wants to send
> data, it just adds this byte up front indicating the channel number, and
> the TURN server sends the data to the corresponding remote peer. There
> is no encapsulation of data as a TURN message ever. Thus, the Data
> indication is removed. TURN currently requires the server to check for
> the magic cookie in packets received from remote peers, and when it sees
> it, encapsulate the resulting data in a Data indication. That
> requirement would also be eliminated.
> 
> Consider a forking use case where A calls a URI and this forks to B and
> C. Both send a 200 OK. A will end up running ICE simultaneously with
> both B and C. All of the STUN checks and any resulting RTP would just be
> sent over different channels, all with identical overhead (1 byte).
> There would be no need to re-invite, as there would typically be in the
> current TURN mechanism.
> 
> This also fixes the complexity that was raised on the list for
> packetcable; the current mechanism complicates their QOS mechanisms
> because the bandwidth for the stream is not deterministic. Here, it is
> deterministic. Every packet always has an extra 1 byte per packet overhead.
> 
> The main reason we've avoided any kind of framing for UDP up until now
> has been overhead. So what kind of overhead does this introduce? Well if
> we assume something really nasty like g729 with 20ms framing, thats 20
> bytes of payload plus 40 for RTP/UDP/IP gives us 60, which now goes to
> 61. Thats a 1.6 percent penalty which is pretty good, considering
> especially the huge overhead you are already paying for RTP/UDP. Anyway
> a more realistic number is g711 with 30ms framing, and there is .35%
> overhead. This cost seems reasonable given the simplifications it brings.
> 
> Another possible concern with the 1-byte overhead is path MTU. If a UDP
> packet was sent from a remote peer and it was exactly equal to the path
> MTU from the remote peer to the TURN server, this will now fragment on
> the TURN server to client link. Its not clear that this is a real
> problem. But if you are aware of UDP applications that send data exactly
> at the MTU for some reason, please speak up.
> 
> For TCP, the channel mechanism would work the same. The current framing
> already has a 'type' field which we'd just turn into a channel number.
> Thus both UDP and TCP would have framing, the difference is only that
> TCP includes a length whereas UDP does not. Protocol operation is
> otherwise more or less the same, and this would forge closer alignment
> between UDP and TCP. As with UDP, this allows multiple TCP connections
> from the TURN server to remote peers, each of which would have its own
> channel number, allocated to the client when permissions are created for
> a remote IP address, or when a connect request is sent. Connection
> status indications would all come over the control channel, and each
> such indication would contain, as an attribute, the channel number that
> the status indication is associated with.
> 
> Now, that said, TCP has some additional complexities that UDP does not.
> First and foremost is that you need to be very careful with how you
> handle buffering. We want to avoid the case where data arrives quickly
> on the connection from the client to the TURN server, but the TCP
> connection from the TURN server to the remote peer is slow, and so data
> is lost at the TURN server when buffers overflow. To handle this, we
> will specify a buffering requirement and a possiblemechanism. There
> would be a maximum buffer size  per connection set in the kernel (for
> example). The TURN server application code reads  off of this buffer and
> writes to the other TCP connection. If it cannot write because the
> window is full, it doesn't read data from the incoming
> connection. This will eventually cause the buffers in the kernel to
> fill, and then packets are dropped (and not ACKed), causing backpressure
> on the TCP connections. This effectively bridges the congestion control
> between the connections, and ensures that, TURN server failures aside,
> no packets get dropped in the TURN server. We believe that this kind of
> mechanism is used in other similar boxes like HTTP proxies.
> 
> Now, lets consider the case of multiple TCP connections to a
> multiplicity of remote peers. There have been suggestions on the list to
> just disallow this. We cannot. It happens as an artifact of forking.
> Consider client A that sends an offer. Amongst its candidates is a
> passive TCP candidate learned from the TURN server. THis INVITE forks to
> B and C, each of which return a single active candidate. Now, A will
> need to open permissions on the TURN server for B and C's IP addresses
> listed in the SDP answer, and both B and C will open a TCP connection
> towards the TURN server in order for ICE checks to proceed. Note well,
> that since ICE processing happens ideally prior to call answering, this
> case happens NOT JUST when both B and C answer at the same time, but
> whenever there is forking at all with ICE-tcp. So this is actually a
> common case and not a corner case.
> 
> In this case of multiple TCP connections to remote peers, the congestion
> control bridging described above gets complicated, and head-of-line
> blocking gets introduced. Congestion on the connection from the TURN
> server to one peer can slow down the connection to the client, and thus
> limit the throughput to the other peer. This is only a problem when we
> are transferring real data, not so much for things like STUN messages or
> TLS handshakes. So the recommendation is going to be that, if a client
> ends up in a case with multiple TCP connections that survive past the
> connectivity checks, in that case the spec will recommend the client
> allocate a new TCP candidate from the TURN server and restart ICE with
> one of the peers, in order to eliminate the HOL blocking.
> 
> We're looking for feedback on this proposal (and questions and
> clarifications of course) ASAP - within the next week. If this seems
> like a good direction, we can work these changes in and address the
> other comments and still deliver a revision on schedule.
> 
> Thanks,
> Jonathan R.


-- 
Marc Petit-Huguenin           [                                 ]
Home: marc@petit-huguenin.org [RFC1855-compliant space for rent ]
Work: marc@8x8.com            [                                 ]
[                                                               ]


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Tue Sep 18 13:39:57 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXh2i-0003xB-4y; Tue, 18 Sep 2007 13:39:24 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXSSF-00048r-4W
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 22:04:47 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXSSE-00047N-PD
	for behave@ietf.org; Mon, 17 Sep 2007 22:04:46 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXQCO-0007ZC-Ep
	for behave@ietf.org; Mon, 17 Sep 2007 19:40:16 -0400
Received: from shell.rawbw.com ([198.144.192.42])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXQCN-0000Cz-TL
	for behave@ietf.org; Mon, 17 Sep 2007 19:40:16 -0400
Received: from [192.168.0.13] (IDENT:stunnel@localhost [127.0.0.1])
	(authenticated bits=0)
	by shell.rawbw.com (8.13.6/8.13.6) with ESMTP id l8HNe8DN038823;
	Mon, 17 Sep 2007 16:40:09 -0700 (PDT)
In-Reply-To: <46EEE728.9070505@acm.org>
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>	<200708081203.19729.remi.denis-courmont@nokia.com>	<46BA4268.2060204@db.org>	<200708090918.13691.remi.denis-courmont@nokia.com>	<46BB4B0C.6030204@db.org>	<46C1ACF3.7000202@cisco.com>
	<46EEC6E5.9000406@pjsip.org> <46EEE728.9070505@acm.org>
Mime-Version: 1.0 (Apple Message framework v752.3)
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <7329187B-6236-4E43-A305-8E2FB437F7E3@briank.com>
Content-Transfer-Encoding: 7bit
From: Brian Korver <briank@briank.com>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
Date: Mon, 17 Sep 2007 16:38:37 -0700
To: Marc Petit-Huguenin <petithug@acm.org>
X-Mailer: Apple Mail (2.752.3)
X-Spam-Score: 0.0 (/)
X-Scan-Signature: cf4fa59384e76e63313391b70cd0dd25
X-TMDA-Confirmed: Mon, 17 Sep 2007 22:04:46 -0400
X-Mailman-Approved-At: Tue, 18 Sep 2007 13:39:22 -0400
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

On Sep 17, 2007, at 1:44 PM, Marc Petit-Huguenin wrote:
> I think that changing the MUST in rfc3489bis-10 to a SHOULD is a  
> better
> solution.  Who knows if another attribute after the FINGERPRINT
> attribute will not be needed in the future.

Making this a SHOULD means we can't count on it being last, that we  
must always do a sequential search for it.  I vote for sticking with  
MUST and revisiting the issue if/when a good reason presents itself.   
The proposed change seems like premature future-proofing.

-brian
briank@briank.com






_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

From behave-bounces@ietf.org Tue Sep 18 13:39:57 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXh2i-000418-TP; Tue, 18 Sep 2007 13:39:24 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXSTn-0004e9-N1
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 22:06:23 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXSTn-0004e0-D5
	for behave@ietf.org; Mon, 17 Sep 2007 22:06:23 -0400
Received: from shell.rawbw.com ([198.144.192.42])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IXSTh-0000tg-5B
	for behave@ietf.org; Mon, 17 Sep 2007 22:06:23 -0400
Received: from [192.168.0.13] (IDENT:stunnel@localhost [127.0.0.1])
	(authenticated bits=0)
	by shell.rawbw.com (8.13.6/8.13.6) with ESMTP id l8I25xJ9098556;
	Mon, 17 Sep 2007 19:05:59 -0700 (PDT)
In-Reply-To: <46EF1839.70605@acm.org>
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>	<200708081203.19729.remi.denis-courmont@nokia.com>	<46BA4268.2060204@db.org>	<200708090918.13691.remi.denis-courmont@nokia.com>	<46BB4B0C.6030204@db.org>	<46C1ACF3.7000202@cisco.com>
	<46EEC6E5.9000406@pjsip.org> <46EEE728.9070505@acm.org>
	<7329187B-6236-4E43-A305-8E2FB437F7E3@briank.com>
	<46EF1839.70605@acm.org>
Mime-Version: 1.0 (Apple Message framework v752.3)
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <22CC5667-341F-42E6-8D5D-343A70388F82@briank.com>
Content-Transfer-Encoding: 7bit
From: Brian Korver <briank@briank.com>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
Date: Mon, 17 Sep 2007 19:04:27 -0700
To: Marc Petit-Huguenin <petithug@acm.org>
X-Mailer: Apple Mail (2.752.3)
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 9182cfff02fae4f1b6e9349e01d62f32
X-Mailman-Approved-At: Tue, 18 Sep 2007 13:39:22 -0400
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

On Sep 17, 2007, at 5:13 PM, Marc Petit-Huguenin wrote:
> Brian Korver wrote:
>> On Sep 17, 2007, at 1:44 PM, Marc Petit-Huguenin wrote:
>>> I think that changing the MUST in rfc3489bis-10 to a SHOULD is a  
>>> better
>>> solution.  Who knows if another attribute after the FINGERPRINT
>>> attribute will not be needed in the future.
>>
>> Making this a SHOULD means we can't count on it being last, that  
>> we must
>> always do a sequential search for it.  I vote for sticking with  
>> MUST and
>> revisiting the issue if/when a good reason presents itself.
>
> Well, the problem is that it will not be possible to change this
> decision in a future RFC, because it will be incompatible with the
> current version.

Sure, but you could say that of any MUST.

-brian
briank@briank.com





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave





From behave-bounces@ietf.org Tue Sep 18 13:39:57 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXh2i-0003xB-4y; Tue, 18 Sep 2007 13:39:24 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXSSF-00048r-4W
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 22:04:47 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXSSE-00047N-PD
	for behave@ietf.org; Mon, 17 Sep 2007 22:04:46 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXQCO-0007ZC-Ep
	for behave@ietf.org; Mon, 17 Sep 2007 19:40:16 -0400
Received: from shell.rawbw.com ([198.144.192.42])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXQCN-0000Cz-TL
	for behave@ietf.org; Mon, 17 Sep 2007 19:40:16 -0400
Received: from [192.168.0.13] (IDENT:stunnel@localhost [127.0.0.1])
	(authenticated bits=0)
	by shell.rawbw.com (8.13.6/8.13.6) with ESMTP id l8HNe8DN038823;
	Mon, 17 Sep 2007 16:40:09 -0700 (PDT)
In-Reply-To: <46EEE728.9070505@acm.org>
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>	<200708081203.19729.remi.denis-courmont@nokia.com>	<46BA4268.2060204@db.org>	<200708090918.13691.remi.denis-courmont@nokia.com>	<46BB4B0C.6030204@db.org>	<46C1ACF3.7000202@cisco.com>
	<46EEC6E5.9000406@pjsip.org> <46EEE728.9070505@acm.org>
Mime-Version: 1.0 (Apple Message framework v752.3)
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <7329187B-6236-4E43-A305-8E2FB437F7E3@briank.com>
Content-Transfer-Encoding: 7bit
From: Brian Korver <briank@briank.com>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
Date: Mon, 17 Sep 2007 16:38:37 -0700
To: Marc Petit-Huguenin <petithug@acm.org>
X-Mailer: Apple Mail (2.752.3)
X-Spam-Score: 0.0 (/)
X-Scan-Signature: cf4fa59384e76e63313391b70cd0dd25
X-TMDA-Confirmed: Mon, 17 Sep 2007 22:04:46 -0400
X-Mailman-Approved-At: Tue, 18 Sep 2007 13:39:22 -0400
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

On Sep 17, 2007, at 1:44 PM, Marc Petit-Huguenin wrote:
> I think that changing the MUST in rfc3489bis-10 to a SHOULD is a  
> better
> solution.  Who knows if another attribute after the FINGERPRINT
> attribute will not be needed in the future.

Making this a SHOULD means we can't count on it being last, that we  
must always do a sequential search for it.  I vote for sticking with  
MUST and revisiting the issue if/when a good reason presents itself.   
The proposed change seems like premature future-proofing.

-brian
briank@briank.com






_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

From behave-bounces@ietf.org Tue Sep 18 13:39:57 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXh2i-000418-TP; Tue, 18 Sep 2007 13:39:24 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXSTn-0004e9-N1
	for behave-confirm+ok@megatron.ietf.org; Mon, 17 Sep 2007 22:06:23 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXSTn-0004e0-D5
	for behave@ietf.org; Mon, 17 Sep 2007 22:06:23 -0400
Received: from shell.rawbw.com ([198.144.192.42])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IXSTh-0000tg-5B
	for behave@ietf.org; Mon, 17 Sep 2007 22:06:23 -0400
Received: from [192.168.0.13] (IDENT:stunnel@localhost [127.0.0.1])
	(authenticated bits=0)
	by shell.rawbw.com (8.13.6/8.13.6) with ESMTP id l8I25xJ9098556;
	Mon, 17 Sep 2007 19:05:59 -0700 (PDT)
In-Reply-To: <46EF1839.70605@acm.org>
References: <72e3f01c7d91e$e82fc589$7b8f15ac@NOE.Nokia.com>	<200708081203.19729.remi.denis-courmont@nokia.com>	<46BA4268.2060204@db.org>	<200708090918.13691.remi.denis-courmont@nokia.com>	<46BB4B0C.6030204@db.org>	<46C1ACF3.7000202@cisco.com>
	<46EEC6E5.9000406@pjsip.org> <46EEE728.9070505@acm.org>
	<7329187B-6236-4E43-A305-8E2FB437F7E3@briank.com>
	<46EF1839.70605@acm.org>
Mime-Version: 1.0 (Apple Message framework v752.3)
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <22CC5667-341F-42E6-8D5D-343A70388F82@briank.com>
Content-Transfer-Encoding: 7bit
From: Brian Korver <briank@briank.com>
Subject: Re: [BEHAVE] STUN fingerprint and integrity tests
Date: Mon, 17 Sep 2007 19:04:27 -0700
To: Marc Petit-Huguenin <petithug@acm.org>
X-Mailer: Apple Mail (2.752.3)
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 9182cfff02fae4f1b6e9349e01d62f32
X-Mailman-Approved-At: Tue, 18 Sep 2007 13:39:22 -0400
Cc: behave@ietf.org
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

On Sep 17, 2007, at 5:13 PM, Marc Petit-Huguenin wrote:
> Brian Korver wrote:
>> On Sep 17, 2007, at 1:44 PM, Marc Petit-Huguenin wrote:
>>> I think that changing the MUST in rfc3489bis-10 to a SHOULD is a  
>>> better
>>> solution.  Who knows if another attribute after the FINGERPRINT
>>> attribute will not be needed in the future.
>>
>> Making this a SHOULD means we can't count on it being last, that  
>> we must
>> always do a sequential search for it.  I vote for sticking with  
>> MUST and
>> revisiting the issue if/when a good reason presents itself.
>
> Well, the problem is that it will not be possible to change this
> decision in a future RFC, because it will be incompatible with the
> current version.

Sure, but you could say that of any MUST.

-brian
briank@briank.com





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave





From behave-bounces@ietf.org Tue Sep 18 16:53:22 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXk43-0003No-S2; Tue, 18 Sep 2007 16:52:59 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXk42-0003LZ-AT
	for behave-confirm+ok@megatron.ietf.org; Tue, 18 Sep 2007 16:52:58 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXk41-0003Fu-NI
	for behave@ietf.org; Tue, 18 Sep 2007 16:52:57 -0400
Received: from sj-iport-6.cisco.com ([171.71.176.117])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXk40-00080R-Cd
	for behave@ietf.org; Tue, 18 Sep 2007 16:52:57 -0400
X-IronPort-AV: E=Sophos;i="4.20,270,1186383600"; d="scan'208";a="220526822"
Received: from sj-dkim-2.cisco.com ([171.71.179.186])
	by sj-iport-6.cisco.com with ESMTP; 18 Sep 2007 13:52:55 -0700
Received: from sj-core-3.cisco.com (sj-core-3.cisco.com [171.68.223.137])
	by sj-dkim-2.cisco.com (8.12.11/8.12.11) with ESMTP id l8IKqtl8010976; 
	Tue, 18 Sep 2007 13:52:55 -0700
Received: from xbh-sjc-221.amer.cisco.com (xbh-sjc-221.cisco.com
	[128.107.191.63])
	by sj-core-3.cisco.com (8.12.10/8.12.6) with ESMTP id l8IKqdRO020847;
	Tue, 18 Sep 2007 20:52:55 GMT
Received: from xfe-sjc-211.amer.cisco.com ([171.70.151.174]) by
	xbh-sjc-221.amer.cisco.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 13:52:47 -0700
Received: from [10.32.241.153] ([10.32.241.153]) by xfe-sjc-211.amer.cisco.com
	with Microsoft SMTPSVC(6.0.3790.1830); 
	Tue, 18 Sep 2007 13:52:46 -0700
Message-ID: <46F03AB2.7010406@cisco.com>
Date: Tue, 18 Sep 2007 16:53:06 -0400
From: Jonathan Rosenberg <jdrosen@cisco.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Marc Petit-Huguenin <petithug@acm.org>
Subject: Re: [BEHAVE] TURN - some proposed changes for improvement
	and	simplification
References: <46E81AFE.6040705@cisco.com> <46F00A32.8090708@acm.org>
In-Reply-To: <46F00A32.8090708@acm.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 18 Sep 2007 20:52:46.0713 (UTC)
	FILETIME=[DDF12E90:01C7FA35]
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=9532; t=1190148775;
	x=1191012775; c=relaxed/simple; s=sjdkim2002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=jdrosen@cisco.com;
	z=From:=20Jonathan=20Rosenberg=20<jdrosen@cisco.com>
	|Subject:=20Re=3A=20[BEHAVE]=20TURN=20-=20some=20proposed=20changes=20for
	=20improvement=20and=09simplification |Sender:=20;
	bh=P6SV8DCghP5/qkgizI21lINJBvmlEi1uRv66Nk724TI=;
	b=2Nb3bc0eMaiWKZjqd8987ClWHVTP6JV1Z/oOVWekB6rranw9VzHRoOrLCeukV3jigjMSY1SL
	9gQvtrJbkHBM/D5ecIJ0845YoSPwy0tgPRWyVPXMVoWXW98Pt4jhP/fg;
Authentication-Results: sj-dkim-2; header.From=jdrosen@cisco.com; dkim=pass (
	sig from cisco.com/sjdkim2002 verified; ); 
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 21be852dc93f0971708678c18d38c096
Cc: Behave WG <behave@ietf.org>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Yeah - we noticed this also. We've got a few ideas on how to deal with 
this. Stay tuned.

-Jonathan R.

Marc Petit-Huguenin wrote:
> Hi Jonathan,
> 
> I like very much the simplification but I have some comments.
> 
> As I understand it, permissions contains only the IP address of the
> peer.  So if a channel is created for a specific IP address and port,
> when data will be coming from the peer from a different port they will
> be accepted by the permission but there is no mechanism in the channel
> to send the real port to the TURN client.
> 
> Also I think that perhaps the Send and Connect methods can be merged
> into one request, like Create Channel (and add a Destroy Channel for
> symmetry).
> 
> 
> Jonathan Rosenberg wrote:
>> As Dan mentioned at the last IETF meeting, we've formed a small team to
>> actively work on TURN over the next month, in order to accelerate its
>> completion and produce a version ready for WGLC by the end of September.
>>
>> As part of this, we took a step back and looked a bit at some of the
>> more troubling aspects of TURN. Three in particular seemed important:
>>
>> * There are several different methods of framing and encapsulation of
>> messages with a number of exceptions based on transport protocol,
>> contents of the data, and the active destination.  The overall effect is
>> confusing and rather complex.
>>
>> * The active destination concept is inelegant, confusing, complex, and
>> during forking results in undesirable overhead and at least one
>> timing-related race condition.
>>
>> * the use of multiple TCP connections from the TURN server towards
>> peers, when there is one TCP connection from the TURN server to the
>> client, has been a concern on the list recently.
>>
>> We have a proposal on some non-trivial changes to TURN on these topics.
>>
>> Firstly, we can eliminate the entire need for active destinations (and
>> the Send and Data primitives as well) by introducing the concept of a
>> 'channel'. Each channel is a stream of application data sent to and from
>> a remote IP and port. In the case of UDP, we'd add a SINGLE BYTE to the
>> front of every packet to and from the TURN server, which is used to
>> signal the channel number. Channel 0 is reserved for TURN signaling to
>> and from the client and TURN server. The remaining 255 channels are for
>> remote destinations and carry pure data after the 1-byte channel number.
>> So when a client does a Send or a Connect, the server allocates a
>> remaining channel number and passes this channel number back to the
>> client. From that point on, when the TURN server receives data from the
>> peer, it sends it to the client and adds only the byte up front
>> containing the channel number. Similarly, when the client wants to send
>> data, it just adds this byte up front indicating the channel number, and
>> the TURN server sends the data to the corresponding remote peer. There
>> is no encapsulation of data as a TURN message ever. Thus, the Data
>> indication is removed. TURN currently requires the server to check for
>> the magic cookie in packets received from remote peers, and when it sees
>> it, encapsulate the resulting data in a Data indication. That
>> requirement would also be eliminated.
>>
>> Consider a forking use case where A calls a URI and this forks to B and
>> C. Both send a 200 OK. A will end up running ICE simultaneously with
>> both B and C. All of the STUN checks and any resulting RTP would just be
>> sent over different channels, all with identical overhead (1 byte).
>> There would be no need to re-invite, as there would typically be in the
>> current TURN mechanism.
>>
>> This also fixes the complexity that was raised on the list for
>> packetcable; the current mechanism complicates their QOS mechanisms
>> because the bandwidth for the stream is not deterministic. Here, it is
>> deterministic. Every packet always has an extra 1 byte per packet overhead.
>>
>> The main reason we've avoided any kind of framing for UDP up until now
>> has been overhead. So what kind of overhead does this introduce? Well if
>> we assume something really nasty like g729 with 20ms framing, thats 20
>> bytes of payload plus 40 for RTP/UDP/IP gives us 60, which now goes to
>> 61. Thats a 1.6 percent penalty which is pretty good, considering
>> especially the huge overhead you are already paying for RTP/UDP. Anyway
>> a more realistic number is g711 with 30ms framing, and there is .35%
>> overhead. This cost seems reasonable given the simplifications it brings.
>>
>> Another possible concern with the 1-byte overhead is path MTU. If a UDP
>> packet was sent from a remote peer and it was exactly equal to the path
>> MTU from the remote peer to the TURN server, this will now fragment on
>> the TURN server to client link. Its not clear that this is a real
>> problem. But if you are aware of UDP applications that send data exactly
>> at the MTU for some reason, please speak up.
>>
>> For TCP, the channel mechanism would work the same. The current framing
>> already has a 'type' field which we'd just turn into a channel number.
>> Thus both UDP and TCP would have framing, the difference is only that
>> TCP includes a length whereas UDP does not. Protocol operation is
>> otherwise more or less the same, and this would forge closer alignment
>> between UDP and TCP. As with UDP, this allows multiple TCP connections
>> from the TURN server to remote peers, each of which would have its own
>> channel number, allocated to the client when permissions are created for
>> a remote IP address, or when a connect request is sent. Connection
>> status indications would all come over the control channel, and each
>> such indication would contain, as an attribute, the channel number that
>> the status indication is associated with.
>>
>> Now, that said, TCP has some additional complexities that UDP does not.
>> First and foremost is that you need to be very careful with how you
>> handle buffering. We want to avoid the case where data arrives quickly
>> on the connection from the client to the TURN server, but the TCP
>> connection from the TURN server to the remote peer is slow, and so data
>> is lost at the TURN server when buffers overflow. To handle this, we
>> will specify a buffering requirement and a possiblemechanism. There
>> would be a maximum buffer size  per connection set in the kernel (for
>> example). The TURN server application code reads  off of this buffer and
>> writes to the other TCP connection. If it cannot write because the
>> window is full, it doesn't read data from the incoming
>> connection. This will eventually cause the buffers in the kernel to
>> fill, and then packets are dropped (and not ACKed), causing backpressure
>> on the TCP connections. This effectively bridges the congestion control
>> between the connections, and ensures that, TURN server failures aside,
>> no packets get dropped in the TURN server. We believe that this kind of
>> mechanism is used in other similar boxes like HTTP proxies.
>>
>> Now, lets consider the case of multiple TCP connections to a
>> multiplicity of remote peers. There have been suggestions on the list to
>> just disallow this. We cannot. It happens as an artifact of forking.
>> Consider client A that sends an offer. Amongst its candidates is a
>> passive TCP candidate learned from the TURN server. THis INVITE forks to
>> B and C, each of which return a single active candidate. Now, A will
>> need to open permissions on the TURN server for B and C's IP addresses
>> listed in the SDP answer, and both B and C will open a TCP connection
>> towards the TURN server in order for ICE checks to proceed. Note well,
>> that since ICE processing happens ideally prior to call answering, this
>> case happens NOT JUST when both B and C answer at the same time, but
>> whenever there is forking at all with ICE-tcp. So this is actually a
>> common case and not a corner case.
>>
>> In this case of multiple TCP connections to remote peers, the congestion
>> control bridging described above gets complicated, and head-of-line
>> blocking gets introduced. Congestion on the connection from the TURN
>> server to one peer can slow down the connection to the client, and thus
>> limit the throughput to the other peer. This is only a problem when we
>> are transferring real data, not so much for things like STUN messages or
>> TLS handshakes. So the recommendation is going to be that, if a client
>> ends up in a case with multiple TCP connections that survive past the
>> connectivity checks, in that case the spec will recommend the client
>> allocate a new TCP candidate from the TURN server and restart ICE with
>> one of the peers, in order to eliminate the HOL blocking.
>>
>> We're looking for feedback on this proposal (and questions and
>> clarifications of course) ASAP - within the next week. If this seems
>> like a good direction, we can work these changes in and address the
>> other comments and still deliver a revision on schedule.
>>
>> Thanks,
>> Jonathan R.
> 
> 

-- 
Jonathan D. Rosenberg, Ph.D.                   600 Lanidex Plaza
Cisco Fellow                                   Parsippany, NJ 07054-2711
Cisco Systems
jdrosen@cisco.com                              FAX:   (973) 952-5050
http://www.jdrosen.net                         PHONE: (973) 952-5000
http://www.cisco.com


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Wed Sep 19 04:32:29 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXuxX-0005Aa-F8; Wed, 19 Sep 2007 04:30:59 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXuxW-0005AR-2n
	for behave-confirm+ok@megatron.ietf.org; Wed, 19 Sep 2007 04:30:58 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXuxV-00053K-KD
	for behave@ietf.org; Wed, 19 Sep 2007 04:30:57 -0400
Received: from mailgw3.ericsson.se ([193.180.251.60])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IXuxN-0002uf-06
	for behave@ietf.org; Wed, 19 Sep 2007 04:30:55 -0400
Received: from mailgw3.ericsson.se (unknown [127.0.0.1])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	4BA4F20612; Wed, 19 Sep 2007 10:30:26 +0200 (CEST)
X-AuditID: c1b4fb3c-b067fbb0000007e1-2e-46f0de226bcf
Received: from esealmw126.eemea.ericsson.se (unknown [153.88.254.123])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	2C615207C9; Wed, 19 Sep 2007 10:30:26 +0200 (CEST)
Received: from esealmw126.eemea.ericsson.se ([153.88.254.174]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 19 Sep 2007 10:30:25 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 19 Sep 2007 10:30:25 +0200
Message-ID: <46F0DE21.70101@ericsson.com>
Date: Wed, 19 Sep 2007 10:30:25 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Pyda Srisuresh <srisuresh@yahoo.com>
References: <639293.64700.qm@web33301.mail.mud.yahoo.com>
In-Reply-To: <639293.64700.qm@web33301.mail.mud.yahoo.com>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 19 Sep 2007 08:30:25.0825 (UTC)
	FILETIME=[53EAC910:01C7FA97]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 9e5c23589e6cce06555030c0194c9e2b
Cc: Bryan Ford <baford@mit.edu>, Behave WG <behave@ietf.org>,
	Dan Kegel <dank@kegel.com>
Subject: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Hi,

See inline.

Pyda Srisuresh skrev:
> Magnus,
> 
> Thank you for the review. Apologize for my delay in responding. Please see my
> responses inline.
> 
> regards,
> suresh
> --- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:
> 
>> Hi,
>>
>> I have done my AD review of this document and have the below comments.
>> If you don't agree with them, please discuss them with me.
>>
>> 1. Section 1, page 3, first paragraph:
>> "These issues are
>>    likely to persist even into the IPv6 world, where a NAT may be used
>>    as an IPv4 compatibility mechanism [NAT-PT]."
>>
>> I think you should rewrite this considering that NAT-PT was deprecated.
>>
> [suresh] Would the following work for you? I believe, there might be NAT-PT
> deployments, even as [NAT-PT] is deprecated.
> 
>      These issues are 
>      likely to persist even into the IPv6 world and in deployments 
>      using [NAT-PT].
> 

Yes.

>> 2. Section 1, page 3, second paragraph:
>>
>> "The
>>    anonymity and inaccessibility of the internal hosts behind a NAT
>>    device is not a problem for applications such as web browsers, which
>>    only need to initiate outgoing connections. This inaccessibility is
>>    sometimes perceived as a privacy benefit."
>>
>> Is really "privacy" the main benefit. Isn't it the filtering behavior of
>> the NAT that blocks incoming requests and thus avoid getting the traffic
>> onto your local network that people like, i.e. the "firewalling" aspect
>> of NATs.
>>
> [suresh] The internal hosts have private IP addresses, which outsiders have no
> knowledge of. This anonymity of IP addressing is considered as privacy by many
> folks. In addition, a host behind NAT is not accessible to outsiders, unless a
> session is initiated from inside. This inaccessibility is also considered as
> privacy by many folks. 
> 
> You might say, the filtering behavior of NATs provides inaccessibility to
> internal hosts; and mapping behavior of NATs provides anonymity to intenral
> hosts.

I agree with NATs filtering behavior making hosts inaccessible in most
cases. But I don't they think they provide "privacy" in the meaning most
 people think about it. If I deploy a NAT in my home, it is still
possible to learn that what ever traffic was sent come from my home, but
maybe not which of the 5 computers in my home that sent it. But that I
don't call privacy. There are many aspects of privacy, maybe a more
precise description of what it really provides.

> 
>> 3. Section 1, page 3, second to last paragraph:
>>
>> "  NAT traversal strategies that involve explicit signaling between
>>    applications and  NAT devices, namely [NAT-PMP], [NSIS-NSLP],
>>    [SOCKS], [RSIP], [MIDCOM], and [UPNP] are out of the scope of this
>>    document. [UNSAF] is in scope."
>>
>> I agree that these are not in scope, but I think there should be some
>> mention that these techiques if available are a complement to the
>> techniques mentioned, allowing a client behind a NAT behave as if it was
>> having a public IP address, or at least an NAT external address.
>>
> [suresh] No problem. I could add a sentence to that effect.

Good

> 
>> 4. Section 2, first paragraph:
>> "As
>>    for the notation for ports usage, all clients use ports in the
>>    range of 1-1000 and servers use ports in the range of
>>    20000-21000. NAT devices use ports 30000 and above for endpoint
>>    mapping."
>>
>> I think there are an error in this description. To me it appears that
>> clients are using addresses in the range 1025-2000 in the examples.
>>
> [suresh] Good catch. I will reword the first sentence to say that clients use
> ports in the raneg of 1-2000. Thanks.

ok.

> 
>> 5. Section 2.7:
>>
>>   "A P2P application is an application that uses the same endpoint to
>>    initiate outgoing sessions to peering hosts as well as accept
>>    incoming sessions from peering hosts."
>>
>> Considering the definition of "endpoint" in this document I don't think
>> this definition is characteristic for peer to peer applications. An peer
>> to peer application clearly can use different address and ports to
>> communicate with other peers.
>>
> [suresh] Magnus - Would it work for you if the definition is expanded to say
> that a P2P application may use multiple endpoints for P2P communication?  The
> primary characteristic of a P2P applictaion is that the same endpoint is used
> for initiating an outgoing connection as well to accept incoming connections.
> All examples and deployment instances described in the document assume this
> definition.

I think that will resolve the main conflict in the definitions.

> 
> An application that uses one endpoint to receive incoming connections and a
> different endpoint for outgoing connections is not a P2P application.

I don't think I agree with that characterisation. Isn't that a result of
the need for making NAT traversal work. My interpretation is more along
the following definition:

"Peer-to-peer is a communications model in which each party has the same
capabilities and either party can initiate a communication session."

> 
>> 6. Section 2.8:
>> "  A NAT-friendly P2P application registers with a publicly
>>    addressable rendezvous server, used for registration and peer
>>    discovery purposes. Pursuant to registering with rendezvous
>>    server, the application uses its private endpoint, public
>>    endpoint, or a combination thereof to establish peering sessions."
>>
>> I don't think this is a strict requierment for a NAT-friendly P2P
>> application. It is rather an effect of being one of the simplest way of
>> accomplishing the rendezvous between two peers.
>>
> [suresh] Would the following rewording work for you? Thanks.
> 
>      One common way P2P applications establish peering sessions and 
>      remain NAT friendly is by using a publicly addressable rendezvous
>      server for registration and  peer discovery purposes.
> 

Yes.

>> 7. Section 2.9. I think one should be explicit that an EIM-NAT can have
>> any filtering behavior.
>>
> [suresh] Not a problem. That is the intent. I will state that explicitly.
> 
>> 8. Section 3.1, last paragraph:
>>
>>   "The obvious disadvantages of
>>    relaying are that it consumes the server's processing power and
>>    network bandwidth, and communication latency between the peering
>>    clients is likely to be increased even if the server is
>>    well-connected."
>>
>> I think "well-connected" is the wrong word here. I think it is rather
>> topology wise correctly located.
>>
> [suresh] By Well-connected, I was refering to having large bandwidth (say, in a
> data center with GigE connectivity). This is required, as relay servers can be
> in the path of many diverse, simultaneous sessions. The relay server ought to
> have aggregate bandwidth to support the numerous parallel connections. I am not
> refering to any topology-wise correctness.
> 

Well, for "communication latency" the topologhy do matter. And clearly
it needs enough bandwidth to handle the number of clients it accept to
serve. However, that it necessirely to have very high bandwidth is in my
view wrong. One clearly can build a relaying system that builds on
whatever node that exist with a real public IP address. As long as one
keep the relaying bandwidth within what exist there should be no extra
delay due to lower bandwidth. The path latency will still be the major
issue.

> 
>> 9. Section 3.1, last paragraph:
>>
>> "The TURN protocol [TURN] defines a method of
>>    implementing relaying in a relatively secure fashion."
>>
>> I think there should be some expansion on what type of relay service
>> that TURN provides. For certain relay services a more application aware
>> relay may be the better choice rather then packet payload relaying.
>>
> [suresh] I am referign to application agnostic, session specific packet relay
> with TURN. I will state that explicitly.

Okay.

> 
>> 10. Section 3.2, page 10:
>>
>> "In the first case, traffic directed to IP address
>>    10.0.0.1 will simply be dropped by the network because 10.0.0.1 is
>>    not a publicly routable IP address."
>>
>> I think, this sentence needs to be expanded to cover the issue that the
>> traffic might hit another host and any application listening on this port.
>>
> 
> [suresh] In the specific example, the connection is initiated by a public
> client. SO, there is no issue of private addresses beign in the same domain,
> right?

Yes, my error.

> 
>> 11. Section 3.3 or 3.3.1:
>>
>> Somewhere here there need to be a description on why this fails with non
>> EIM-NATs.
>>
> 
> [suresh] OK, I will add description to that effect. When one of the hosts is
> behind a NAT that is not EIM-NAT, the peering host cannot predictably know the
> mapped endpoint to which to initiate connection. Further, the application on
> the host behind non EIM-NAT would be unable to reuse an already established
> endpoint mapping for communication with different external destinations and the
> hole-punching technique would fail.
> 

Good.

> 
>> 12. Section 3.3.1:
>>
>> "The application does
>>    not need to attempt to detect the kind of NAT device it is behind,
>>    as in [STUN], since the procedure above will establish peer-to-peer
>>    communication channels equally well if either or both clients do not
>>    happen to be behind a NAT device."
>>
>> Well, in most usages of STUN you don't need to detect the bevior of the
>> NAT. I assume that you refer to the model of  using STUN to find the
>> external address and port and then advertise it. Not knowing the bevior
>> of the NAT only results in one knowing if it works or not. I think the
>> wording is a bit missfortunate, as it seems to say that STUN is really
>> bad, when it is only a specific usage that is problematic.
>>
> 
> [suresh] I dont believe, the wording says that STUN is bad. If you like, I
> could remove reference to [STUN] here.

I think removing the reference may be the best way of resolving this.

> 
> 
>> 13. Section 3.3.2, Figure 4:
>>
>> I think the example is strange. What I can see if A and B sends its
>> request starting with Registry, then try1 and then try2, try1 should
>> succed. Also these messages are not well connected to the text. I think
>> you should swap the order and then reference them a bit better in the text.
>>
> 
> [suresh] You are right. Must have been a cut and paste error. I will swap the
> try order in the diagram and reword accordingly. Thanks.

Ok.

> 
>> 14. Section 3.3.2, last paragraph:
>>   "[ICE] protocol employs this technique effectively, in that
>>    multiple candidate endpoints (both private and public) are
>>    communicated between peering end hosts during offer/response
>>    exchange."
>>
>> I think you should use either offer/answer or request/response, not mix it.
>>
> [suresh] OK.
> 
>> 15. section 5.1, second paragraph:
>>
>> "When the NAT device
>>    enroute is not EIM-NAT, the application is unable to reuse an
>>    already established endpoint mapping for communication with
>>    different external destinations and the technique would fail."
>>
>> I think "enroute" is the wrong word. It is the outermost NAT that isn't
>> an EIM-NAT for it to matter.
>>
> [suresh] As discussed with Remi on another thread, I believe, "enroute" is
> proper usage here.
> 
> 

I don't know which thread you are refering to. Well, enroute will not be
missunderstood. But isn't "on path" what is commonly used in this case.
But, this is really nit-picking. So lets ignore this.

Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------



_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Wed Sep 19 04:33:22 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IXuzl-0001G6-3Z; Wed, 19 Sep 2007 04:33:17 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IXuzj-00014k-76
	for behave-confirm+ok@megatron.ietf.org; Wed, 19 Sep 2007 04:33:15 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IXuzi-0000yX-KE
	for behave@ietf.org; Wed, 19 Sep 2007 04:33:14 -0400
Received: from mailgw4.ericsson.se ([193.180.251.62])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IXuzh-0000fE-Vr
	for behave@ietf.org; Wed, 19 Sep 2007 04:33:14 -0400
Received: from mailgw4.ericsson.se (unknown [127.0.0.1])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	A1A5C211AD; Wed, 19 Sep 2007 10:33:12 +0200 (CEST)
X-AuditID: c1b4fb3e-b1837bb0000007e1-9f-46f0dec8f98e
Received: from esealmw127.eemea.ericsson.se (unknown [153.88.254.122])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	7ACA221142; Wed, 19 Sep 2007 10:33:12 +0200 (CEST)
Received: from esealmw127.eemea.ericsson.se ([153.88.254.175]) by
	esealmw127.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 19 Sep 2007 10:33:12 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw127.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Wed, 19 Sep 2007 10:33:12 +0200
Message-ID: <46F0DEC8.80204@ericsson.com>
Date: Wed, 19 Sep 2007 10:33:12 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Pyda Srisuresh <srisuresh@yahoo.com>
References: <9493.86537.qm@web33315.mail.mud.yahoo.com>
In-Reply-To: <9493.86537.qm@web33315.mail.mud.yahoo.com>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 19 Sep 2007 08:33:12.0401 (UTC)
	FILETIME=[B7344410:01C7FA97]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: b4a0a5f5992e2a4954405484e7717d8c
Cc: Bryan Ford <baford@mit.edu>, Behave WG <behave@ietf.org>
Subject: [BEHAVE] Re: AD comments on draft-ietf-behave-icmp-04
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Pyda Srisuresh skrev:

>>>> 2. Section 2, last sentence:
>>>>
>>>> "A NAT MAY drop
>>>>    or appropriately handle a post-1122 ICMP messages. "
>>>>
>>>> I think a normative statement of this type do not belong in the
>>>> terminology section. Please find a better place to locate it.
>>>>
>>> [suresh] OK. I will move this to the first paragraph of section 8 (Summary
>> of
>>> Requirements).
>> Will that not be strange without having a requirement that only is
>> listed in the summary?
>>
> [suresh] I will preface by saying that the document focuses on the requirements
> concerning ICMP queries and ICMP error messages. SO, i think, it is Ok to
> follow this up and say that "a NAT MAY drop or appropriately handle a post-1122
> ICMP messages". All this would be in the top paragraph of the summary section.
> 

Okay, lets see how it looks.



> 
>>>> 5. Section 4.2, first paragraph:
>>>> "Section 4.3 of RFC 3022 describes the fields of an ICMP error
>>>>    message that a NAT device translates. In this section, we describe
>>>>    the requirements a NAT device MUST conform to while performing the
>>>>    translations. Requirements identified in this section are necessary
>>>>    for the current applications to work correctly."
>>>>
>>>> I interpret this as an normative reference. This would be a normative
>>>> reference from a BCP to a informative document, which I like to avoid.
>>>> Can we please rework this, for example by including how the translation
>>>> is done in this document instead.
>>>>
>>> [suresh] Implicitly, both [BEH-UDP] and [BEH-TCP] also defer to RFC 3022
>> for
>>> the NAT translation. In that sense, this docuemnt is no different. Would it
>>> work for you if I simply removed the first sentence.  
>> Well, I don't see how removing the sentence will make things better. It
>> seems that we likely need to apply RFC 3967 exception on this reference.
>> So please upgrade it to a real refence in the normative part.
>>
> 
> [suresh] You are saying that I should move [NAT-TRAD] to the normative section,
> right. 
> 

Yes.

Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 20 01:16:38 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYENn-0004VK-Ft; Thu, 20 Sep 2007 01:15:23 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYENm-0004Tq-Jo
	for behave-confirm+ok@megatron.ietf.org; Thu, 20 Sep 2007 01:15:22 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYENl-0004QO-9k
	for behave@ietf.org; Thu, 20 Sep 2007 01:15:22 -0400
Received: from web33312.mail.mud.yahoo.com ([68.142.206.127])
	by ietf-mx.ietf.org with smtp (Exim 4.43) id 1IYENe-0001YD-UP
	for behave@ietf.org; Thu, 20 Sep 2007 01:15:21 -0400
Received: (qmail 69344 invoked by uid 60001); 20 Sep 2007 05:14:59 -0000
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID;
	b=RclUqKR6/BA71r8tQ2+rUJnVBw3uZOYCKTvjZSPmm3br937aJPKc6qZoCOhl9zfs1QctP9qx5jkTrDBEbZ8g78blv7g47NzFHxjmbwK08kIq3w1wLvcyZQnRzJDc0F84ev7KO2dOjZw1ib+9oqq/6RWjTdST8t78Gl3J2xJ5lV0=;
X-YMail-OSG: NjTHWs8VM1kG5AGNuO7T0aUMieM41O3OIk4CvpqhspDO3HEd5z19NyTO6CC8hvh..9nsbA--
Received: from [69.236.94.7] by web33312.mail.mud.yahoo.com via HTTP;
	Wed, 19 Sep 2007 22:14:59 PDT
Date: Wed, 19 Sep 2007 22:14:59 -0700 (PDT)
From: Pyda Srisuresh <srisuresh@yahoo.com>
To: Magnus Westerlund <magnus.westerlund@ericsson.com>
In-Reply-To: <46F0DE21.70101@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <250174.68788.qm@web33312.mail.mud.yahoo.com>
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 7a0494a0224ca59418dd8f92694c1fdb
Cc: Bryan Ford <baford@mit.edu>, Behave WG <behave@ietf.org>,
	Dan Kegel <dank@kegel.com>
Subject: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Magnus,

Please see my responses inline.

regards,
suresh

--- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:

<... stuff deleted>

> 
> >> 2. Section 1, page 3, second paragraph:
> >>
> >> "The
> >>    anonymity and inaccessibility of the internal hosts behind a NAT
> >>    device is not a problem for applications such as web browsers, which
> >>    only need to initiate outgoing connections. This inaccessibility is
> >>    sometimes perceived as a privacy benefit."
> >>
> >> Is really "privacy" the main benefit. Isn't it the filtering behavior of
> >> the NAT that blocks incoming requests and thus avoid getting the traffic
> >> onto your local network that people like, i.e. the "firewalling" aspect
> >> of NATs.
> >>
> > [suresh] The internal hosts have private IP addresses, which outsiders have
> no
> > knowledge of. This anonymity of IP addressing is considered as privacy by
> many
> > folks. In addition, a host behind NAT is not accessible to outsiders,
> unless a
> > session is initiated from inside. This inaccessibility is also considered
> as
> > privacy by many folks. 
> > 
> > You might say, the filtering behavior of NATs provides inaccessibility to
> > internal hosts; and mapping behavior of NATs provides anonymity to intenral
> > hosts.
> 
> I agree with NATs filtering behavior making hosts inaccessible in most
> cases. But I don't they think they provide "privacy" in the meaning most
>  people think about it. If I deploy a NAT in my home, it is still
> possible to learn that what ever traffic was sent come from my home, but
> maybe not which of the 5 computers in my home that sent it. But that I
> don't call privacy. There are many aspects of privacy, maybe a more
> precise description of what it really provides.
> 

[suresh] As I said earlier, the privacy I am referign to is principally the
combination of private Ip address (anonymity) and inaccessibility from external
hosts. The mapping behavior of NATs provides anonymity and the filtering
behavior of NATs provides inaccessibility to internal hosts. Would the
following rewording work for you?

  "The
   anonymity (private IP addresses) and inaccessibility of the internal 
   hosts behind a NAT device is not a problem for applications such as
   web browsers, which only need to initiate outgoing connections. This
   anonymity and inaccessibility is sometimes perceived as a privacy
   benefit."

<... stuff deleted>

> 
> > 
> >> 5. Section 2.7:
> >>
> >>   "A P2P application is an application that uses the same endpoint to
> >>    initiate outgoing sessions to peering hosts as well as accept
> >>    incoming sessions from peering hosts."
> >>
> >> Considering the definition of "endpoint" in this document I don't think
> >> this definition is characteristic for peer to peer applications. An peer
> >> to peer application clearly can use different address and ports to
> >> communicate with other peers.
> >>
> > [suresh] Magnus - Would it work for you if the definition is expanded to
> say
> > that a P2P application may use multiple endpoints for P2P communication? 
> The
> > primary characteristic of a P2P applictaion is that the same endpoint is
> used
> > for initiating an outgoing connection as well to accept incoming
> connections.
> > All examples and deployment instances described in the document assume this
> > definition.
> 
> I think that will resolve the main conflict in the definitions.
> 
[suresh] Will do. 

> > 
> > An application that uses one endpoint to receive incoming connections and a
> > different endpoint for outgoing connections is not a P2P application.
> 
> I don't think I agree with that characterisation. Isn't that a result of
> the need for making NAT traversal work. My interpretation is more along
> the following definition:
> 
> "Peer-to-peer is a communications model in which each party has the same
> capabilities and either party can initiate a communication session."
> 

[suresh] I see what you mean. However, the definition I provided is appropriate
for the document, I believe.

<... stuff deleted>


> >> 8. Section 3.1, last paragraph:
> >>
> >>   "The obvious disadvantages of
> >>    relaying are that it consumes the server's processing power and
> >>    network bandwidth, and communication latency between the peering
> >>    clients is likely to be increased even if the server is
> >>    well-connected."
> >>
> >> I think "well-connected" is the wrong word here. I think it is rather
> >> topology wise correctly located.
> >>
> > [suresh] By Well-connected, I was refering to having large bandwidth (say,
> in a
> > data center with GigE connectivity). This is required, as relay servers can
> be
> > in the path of many diverse, simultaneous sessions. The relay server ought
> to
> > have aggregate bandwidth to support the numerous parallel connections. I am
> not
> > refering to any topology-wise correctness.
> > 
> 
> Well, for "communication latency" the topologhy do matter. And clearly
> it needs enough bandwidth to handle the number of clients it accept to
> serve. However, that it necessirely to have very high bandwidth is in my
> view wrong. One clearly can build a relaying system that builds on
> whatever node that exist with a real public IP address. As long as one
> keep the relaying bandwidth within what exist there should be no extra
> delay due to lower bandwidth. The path latency will still be the major
> issue.
> 
[suresh] Would it it work for if I reworded this as follows? 

  "The obvious disadvantages of
   relaying are that it consumes the server's processing power and
   network bandwidth, and communication latency between the peering
   clients is likely to be increased even if the server has sufficient
   I/O bandwidth and is located correctly topology wise."

<.... stuff deleted>

> > 
> >> 15. section 5.1, second paragraph:
> >>
> >> "When the NAT device
> >>    enroute is not EIM-NAT, the application is unable to reuse an
> >>    already established endpoint mapping for communication with
> >>    different external destinations and the technique would fail."
> >>
> >> I think "enroute" is the wrong word. It is the outermost NAT that isn't
> >> an EIM-NAT for it to matter.
> >>
> > [suresh] As discussed with Remi on another thread, I believe, "enroute" is
> > proper usage here.
> > 
> > 
> 
> I don't know which thread you are refering to. Well, enroute will not be
> missunderstood. But isn't "on path" what is commonly used in this case.
> But, this is really nit-picking. So lets ignore this.
> 
[suresh] I was refering to the e-mail exchange between you and Remi on this
item on 09/04/07.

Thank you for your comments.

regards,
suresh

> Cheers
> 
> Magnus Westerlund
> 
> IETF Transport Area Director & TSVWG Chair
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM/M
> ----------------------------------------------------------------------
> Ericsson AB                | Phone +46 8 4048287
> Torshamsgatan 23           | Fax   +46 8 7575550
> S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
> ----------------------------------------------------------------------
> 
> 





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 20 01:17:37 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYEPx-00076G-AD; Thu, 20 Sep 2007 01:17:37 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYEPw-000766-82
	for behave-confirm+ok@megatron.ietf.org; Thu, 20 Sep 2007 01:17:36 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYEPv-00075v-Cc
	for behave@ietf.org; Thu, 20 Sep 2007 01:17:35 -0400
Received: from web33310.mail.mud.yahoo.com ([68.142.206.125])
	by chiedprmail1.ietf.org with smtp (Exim 4.43) id 1IYEPu-000240-RT
	for behave@ietf.org; Thu, 20 Sep 2007 01:17:35 -0400
Received: (qmail 69001 invoked by uid 60001); 20 Sep 2007 05:17:34 -0000
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID;
	b=G7cYIe778lRKrRHj8eUY8GIMWi1b1phx4FvG3KjWEAPy4ejDuyrMNgMIkp78M41vnZyTa1/YfdbfDT6BtY/PKv6pg22g3/CN1psZ+jwi4uJLgNNZSd6mjV+Oba3TZcuEW7y2L9z+mqewDTnviNi0I9mcZusLSafjqHWZSA+I+Jc=;
X-YMail-OSG: iQLBEucVM1k9MdMDN3bw7MY9vMLfCw7OlZIl1neT1R_laifsVV._8F1qaX7uUV8X0cF31w--
Received: from [69.236.94.7] by web33310.mail.mud.yahoo.com via HTTP;
	Wed, 19 Sep 2007 22:17:34 PDT
Date: Wed, 19 Sep 2007 22:17:34 -0700 (PDT)
From: Pyda Srisuresh <srisuresh@yahoo.com>
To: Magnus Westerlund <magnus.westerlund@ericsson.com>
In-Reply-To: <46F0DEC8.80204@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <275670.68825.qm@web33310.mail.mud.yahoo.com>
X-Spam-Score: 0.0 (/)
X-Scan-Signature: c3a18ef96977fc9bcc21a621cbf1174b
Cc: Bryan Ford <baford@mit.edu>, Behave WG <behave@ietf.org>
Subject: [BEHAVE] Re: AD comments on draft-ietf-behave-icmp-04
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Magnus,

Thanks for all your comments and responses. I will fold in your feedback in the
next rev of the draft.

regards,
suresh

--- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:

> Pyda Srisuresh skrev:
> 
> >>>> 2. Section 2, last sentence:
> >>>>
> >>>> "A NAT MAY drop
> >>>>    or appropriately handle a post-1122 ICMP messages. "
> >>>>
> >>>> I think a normative statement of this type do not belong in the
> >>>> terminology section. Please find a better place to locate it.
> >>>>
> >>> [suresh] OK. I will move this to the first paragraph of section 8
> (Summary
> >> of
> >>> Requirements).
> >> Will that not be strange without having a requirement that only is
> >> listed in the summary?
> >>
> > [suresh] I will preface by saying that the document focuses on the
> requirements
> > concerning ICMP queries and ICMP error messages. SO, i think, it is Ok to
> > follow this up and say that "a NAT MAY drop or appropriately handle a
> post-1122
> > ICMP messages". All this would be in the top paragraph of the summary
> section.
> > 
> 
> Okay, lets see how it looks.
> 
> 
> 
> > 
> >>>> 5. Section 4.2, first paragraph:
> >>>> "Section 4.3 of RFC 3022 describes the fields of an ICMP error
> >>>>    message that a NAT device translates. In this section, we describe
> >>>>    the requirements a NAT device MUST conform to while performing the
> >>>>    translations. Requirements identified in this section are necessary
> >>>>    for the current applications to work correctly."
> >>>>
> >>>> I interpret this as an normative reference. This would be a normative
> >>>> reference from a BCP to a informative document, which I like to avoid.
> >>>> Can we please rework this, for example by including how the translation
> >>>> is done in this document instead.
> >>>>
> >>> [suresh] Implicitly, both [BEH-UDP] and [BEH-TCP] also defer to RFC 3022
> >> for
> >>> the NAT translation. In that sense, this docuemnt is no different. Would
> it
> >>> work for you if I simply removed the first sentence.  
> >> Well, I don't see how removing the sentence will make things better. It
> >> seems that we likely need to apply RFC 3967 exception on this reference.
> >> So please upgrade it to a real refence in the normative part.
> >>
> > 
> > [suresh] You are saying that I should move [NAT-TRAD] to the normative
> section,
> > right. 
> > 
> 
> Yes.
> 
> Cheers
> 
> Magnus Westerlund
> 
> IETF Transport Area Director & TSVWG Chair
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM/M
> ----------------------------------------------------------------------
> Ericsson AB                | Phone +46 8 4048287
> Torshamsgatan 23           | Fax   +46 8 7575550
> S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
> ----------------------------------------------------------------------
> 





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 20 04:41:06 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYHaD-0008P1-F1; Thu, 20 Sep 2007 04:40:25 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYHaB-0008Ol-BG
	for behave-confirm+ok@megatron.ietf.org; Thu, 20 Sep 2007 04:40:23 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYHaA-0008Od-SU
	for behave@ietf.org; Thu, 20 Sep 2007 04:40:22 -0400
Received: from mailgw3.ericsson.se ([193.180.251.60])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IYHa4-0006XY-DY
	for behave@ietf.org; Thu, 20 Sep 2007 04:40:22 -0400
Received: from mailgw3.ericsson.se (unknown [127.0.0.1])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	E1C35204F9; Thu, 20 Sep 2007 10:40:00 +0200 (CEST)
X-AuditID: c1b4fb3c-ae67bbb0000007e1-61-46f231e0ea6e
Received: from esealmw129.eemea.ericsson.se (unknown [153.88.254.124])
	by mailgw3.ericsson.se (Symantec Mail Security) with ESMTP id
	BD302204D8; Thu, 20 Sep 2007 10:40:00 +0200 (CEST)
Received: from esealmw129.eemea.ericsson.se ([153.88.254.177]) by
	esealmw129.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Thu, 20 Sep 2007 10:40:00 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw129.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Thu, 20 Sep 2007 10:40:00 +0200
Message-ID: <46F231E0.5070701@ericsson.com>
Date: Thu, 20 Sep 2007 10:40:00 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Pyda Srisuresh <srisuresh@yahoo.com>
References: <250174.68788.qm@web33312.mail.mud.yahoo.com>
In-Reply-To: <250174.68788.qm@web33312.mail.mud.yahoo.com>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 20 Sep 2007 08:40:00.0149 (UTC)
	FILETIME=[D4A76C50:01C7FB61]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: -1.0 (-)
X-Scan-Signature: d890c9ddd0b0a61e8c597ad30c1c2176
Cc: Bryan Ford <baford@mit.edu>, Behave WG <behave@ietf.org>,
	Dan Kegel <dank@kegel.com>
Subject: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

See inline

Pyda Srisuresh skrev:
> Magnus,
> 
> Please see my responses inline.
> 
> regards,
> suresh
> 
> --- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:
> 
> <... stuff deleted>
> 
>>>> 2. Section 1, page 3, second paragraph:
>>>>
>>>> "The
>>>>    anonymity and inaccessibility of the internal hosts behind a NAT
>>>>    device is not a problem for applications such as web browsers, which
>>>>    only need to initiate outgoing connections. This inaccessibility is
>>>>    sometimes perceived as a privacy benefit."
>>>>
>>>> Is really "privacy" the main benefit. Isn't it the filtering behavior of
>>>> the NAT that blocks incoming requests and thus avoid getting the traffic
>>>> onto your local network that people like, i.e. the "firewalling" aspect
>>>> of NATs.
>>>>
>>> [suresh] The internal hosts have private IP addresses, which outsiders have
>> no
>>> knowledge of. This anonymity of IP addressing is considered as privacy by
>> many
>>> folks. In addition, a host behind NAT is not accessible to outsiders,
>> unless a
>>> session is initiated from inside. This inaccessibility is also considered
>> as
>>> privacy by many folks. 
>>>
>>> You might say, the filtering behavior of NATs provides inaccessibility to
>>> internal hosts; and mapping behavior of NATs provides anonymity to intenral
>>> hosts.
>> I agree with NATs filtering behavior making hosts inaccessible in most
>> cases. But I don't they think they provide "privacy" in the meaning most
>>  people think about it. If I deploy a NAT in my home, it is still
>> possible to learn that what ever traffic was sent come from my home, but
>> maybe not which of the 5 computers in my home that sent it. But that I
>> don't call privacy. There are many aspects of privacy, maybe a more
>> precise description of what it really provides.
>>
> 
> [suresh] As I said earlier, the privacy I am referign to is principally the
> combination of private Ip address (anonymity) and inaccessibility from external
> hosts. The mapping behavior of NATs provides anonymity and the filtering
> behavior of NATs provides inaccessibility to internal hosts. Would the
> following rewording work for you?
> 
>   "The
>    anonymity (private IP addresses) and inaccessibility of the internal 
>    hosts behind a NAT device is not a problem for applications such as
>    web browsers, which only need to initiate outgoing connections. This
>    anonymity and inaccessibility is sometimes perceived as a privacy
>    benefit."
> 

I have some concerns about using "anonymity" in this context, because it
is mostly an illusion. In many deployements there are very few hosts
NATed behind a specific IP address (at least in sweden), thus NATs
provide with very little privacy or anonymity.

And this should be evident from how targeted commericals one do gets
even when websurfing behind a NAT. So I don't want to build on this. But
if you include the word "illusion of" prior to "anonymity" then I am fine.


> 
>>> An application that uses one endpoint to receive incoming connections and a
>>> different endpoint for outgoing connections is not a P2P application.
>> I don't think I agree with that characterisation. Isn't that a result of
>> the need for making NAT traversal work. My interpretation is more along
>> the following definition:
>>
>> "Peer-to-peer is a communications model in which each party has the same
>> capabilities and either party can initiate a communication session."
>>
> 
> [suresh] I see what you mean. However, the definition I provided is appropriate
> for the document, I believe.
> 

Yes, that is okay, the basic issue was that conflict in the usage in
this case. So if you resolve that, then it is okay.

> <... stuff deleted>
> 
> 
>>>> 8. Section 3.1, last paragraph:
>>>>
>>>>   "The obvious disadvantages of
>>>>    relaying are that it consumes the server's processing power and
>>>>    network bandwidth, and communication latency between the peering
>>>>    clients is likely to be increased even if the server is
>>>>    well-connected."
>>>>
>>>> I think "well-connected" is the wrong word here. I think it is rather
>>>> topology wise correctly located.
>>>>
>>> [suresh] By Well-connected, I was refering to having large bandwidth (say,
>> in a
>>> data center with GigE connectivity). This is required, as relay servers can
>> be
>>> in the path of many diverse, simultaneous sessions. The relay server ought
>> to
>>> have aggregate bandwidth to support the numerous parallel connections. I am
>> not
>>> refering to any topology-wise correctness.
>>>
>> Well, for "communication latency" the topologhy do matter. And clearly
>> it needs enough bandwidth to handle the number of clients it accept to
>> serve. However, that it necessirely to have very high bandwidth is in my
>> view wrong. One clearly can build a relaying system that builds on
>> whatever node that exist with a real public IP address. As long as one
>> keep the relaying bandwidth within what exist there should be no extra
>> delay due to lower bandwidth. The path latency will still be the major
>> issue.
>>
> [suresh] Would it it work for if I reworded this as follows? 
> 
>   "The obvious disadvantages of
>    relaying are that it consumes the server's processing power and
>    network bandwidth, and communication latency between the peering
>    clients is likely to be increased even if the server has sufficient
>    I/O bandwidth and is located correctly topology wise."

Yes, sound good.

> 
> <.... stuff deleted>
> 
>>>> 15. section 5.1, second paragraph:
>>>>
>>>> "When the NAT device
>>>>    enroute is not EIM-NAT, the application is unable to reuse an
>>>>    already established endpoint mapping for communication with
>>>>    different external destinations and the technique would fail."
>>>>
>>>> I think "enroute" is the wrong word. It is the outermost NAT that isn't
>>>> an EIM-NAT for it to matter.
>>>>
>>> [suresh] As discussed with Remi on another thread, I believe, "enroute" is
>>> proper usage here.
>>>
>>>
>> I don't know which thread you are refering to. Well, enroute will not be
>> missunderstood. But isn't "on path" what is commonly used in this case.
>> But, this is really nit-picking. So lets ignore this.
>>
> [suresh] I was refering to the e-mail exchange between you and Remi on this
> item on 09/04/07.
> 

Okay, I think that the focus of that discussion was my misstake in that
all NATs must be EIM-NATs, rather then if enroute or on path is the best
word. But both do work and I was being to nit picking.

Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 20 10:02:39 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYMbc-0004w1-Ig; Thu, 20 Sep 2007 10:02:12 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYMbb-0004vc-3K
	for behave-confirm+ok@megatron.ietf.org; Thu, 20 Sep 2007 10:02:11 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYMba-0004tl-H3
	for behave@ietf.org; Thu, 20 Sep 2007 10:02:10 -0400
Received: from web33301.mail.mud.yahoo.com ([68.142.206.116])
	by chiedprmail1.ietf.org with smtp (Exim 4.43) id 1IYMbZ-0007UA-E9
	for behave@ietf.org; Thu, 20 Sep 2007 10:02:10 -0400
Received: (qmail 53685 invoked by uid 60001); 20 Sep 2007 14:02:08 -0000
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID;
	b=3w62tchH/bX1xIuvV/l/tMsdsVwKd1H5lUy03v3GWDdKPpgpbhZgvT7h/1zO046ESZioWLoihTfjbu+sfCDCaCFt0JBD/HHuFddUhZ9o0RC8GyFEW70OOM6wHzicp7iFJm9VK9CzHxeJKMp+shNCz47p/wQrY3vYjc5VdfMYRHU=;
X-YMail-OSG: DBXPB3sVM1n_HA4S9uFlm9r2pFx71BK5gnc3oiZoI0uzqXFvm.9u7dbXDvRz1RBFGpvTbg--
Received: from [69.236.94.7] by web33301.mail.mud.yahoo.com via HTTP;
	Thu, 20 Sep 2007 07:02:08 PDT
Date: Thu, 20 Sep 2007 07:02:08 -0700 (PDT)
From: Pyda Srisuresh <srisuresh@yahoo.com>
To: Magnus Westerlund <magnus.westerlund@ericsson.com>
In-Reply-To: <46F231E0.5070701@ericsson.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <464267.53402.qm@web33301.mail.mud.yahoo.com>
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 97c820c82c68af374c4e382a80dc5017
Cc: Bryan Ford <baford@mit.edu>, Behave WG <behave@ietf.org>,
	Dan Kegel <dank@kegel.com>
Subject: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Magnus,

Thanks. I will bring out the next rev with your feedback folded in. 

regards,
suresh

--- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:

> See inline
> 
> Pyda Srisuresh skrev:
> > Magnus,
> > 
> > Please see my responses inline.
> > 
> > regards,
> > suresh
> > 
> > --- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:
> > 
> > <... stuff deleted>
> > 
> >>>> 2. Section 1, page 3, second paragraph:
> >>>>
> >>>> "The
> >>>>    anonymity and inaccessibility of the internal hosts behind a NAT
> >>>>    device is not a problem for applications such as web browsers, which
> >>>>    only need to initiate outgoing connections. This inaccessibility is
> >>>>    sometimes perceived as a privacy benefit."
> >>>>
> >>>> Is really "privacy" the main benefit. Isn't it the filtering behavior of
> >>>> the NAT that blocks incoming requests and thus avoid getting the traffic
> >>>> onto your local network that people like, i.e. the "firewalling" aspect
> >>>> of NATs.
> >>>>
> >>> [suresh] The internal hosts have private IP addresses, which outsiders
> have
> >> no
> >>> knowledge of. This anonymity of IP addressing is considered as privacy by
> >> many
> >>> folks. In addition, a host behind NAT is not accessible to outsiders,
> >> unless a
> >>> session is initiated from inside. This inaccessibility is also considered
> >> as
> >>> privacy by many folks. 
> >>>
> >>> You might say, the filtering behavior of NATs provides inaccessibility to
> >>> internal hosts; and mapping behavior of NATs provides anonymity to
> intenral
> >>> hosts.
> >> I agree with NATs filtering behavior making hosts inaccessible in most
> >> cases. But I don't they think they provide "privacy" in the meaning most
> >>  people think about it. If I deploy a NAT in my home, it is still
> >> possible to learn that what ever traffic was sent come from my home, but
> >> maybe not which of the 5 computers in my home that sent it. But that I
> >> don't call privacy. There are many aspects of privacy, maybe a more
> >> precise description of what it really provides.
> >>
> > 
> > [suresh] As I said earlier, the privacy I am referign to is principally the
> > combination of private Ip address (anonymity) and inaccessibility from
> external
> > hosts. The mapping behavior of NATs provides anonymity and the filtering
> > behavior of NATs provides inaccessibility to internal hosts. Would the
> > following rewording work for you?
> > 
> >   "The
> >    anonymity (private IP addresses) and inaccessibility of the internal 
> >    hosts behind a NAT device is not a problem for applications such as
> >    web browsers, which only need to initiate outgoing connections. This
> >    anonymity and inaccessibility is sometimes perceived as a privacy
> >    benefit."
> > 
> 
> I have some concerns about using "anonymity" in this context, because it
> is mostly an illusion. In many deployements there are very few hosts
> NATed behind a specific IP address (at least in sweden), thus NATs
> provide with very little privacy or anonymity.
> 
> And this should be evident from how targeted commericals one do gets
> even when websurfing behind a NAT. So I don't want to build on this. But
> if you include the word "illusion of" prior to "anonymity" then I am fine.
> 
> 
> > 
> >>> An application that uses one endpoint to receive incoming connections and
> a
> >>> different endpoint for outgoing connections is not a P2P application.
> >> I don't think I agree with that characterisation. Isn't that a result of
> >> the need for making NAT traversal work. My interpretation is more along
> >> the following definition:
> >>
> >> "Peer-to-peer is a communications model in which each party has the same
> >> capabilities and either party can initiate a communication session."
> >>
> > 
> > [suresh] I see what you mean. However, the definition I provided is
> appropriate
> > for the document, I believe.
> > 
> 
> Yes, that is okay, the basic issue was that conflict in the usage in
> this case. So if you resolve that, then it is okay.
> 
> > <... stuff deleted>
> > 
> > 
> >>>> 8. Section 3.1, last paragraph:
> >>>>
> >>>>   "The obvious disadvantages of
> >>>>    relaying are that it consumes the server's processing power and
> >>>>    network bandwidth, and communication latency between the peering
> >>>>    clients is likely to be increased even if the server is
> >>>>    well-connected."
> >>>>
> >>>> I think "well-connected" is the wrong word here. I think it is rather
> >>>> topology wise correctly located.
> >>>>
> >>> [suresh] By Well-connected, I was refering to having large bandwidth
> (say,
> >> in a
> >>> data center with GigE connectivity). This is required, as relay servers
> can
> >> be
> >>> in the path of many diverse, simultaneous sessions. The relay server
> ought
> >> to
> >>> have aggregate bandwidth to support the numerous parallel connections. I
> am
> >> not
> >>> refering to any topology-wise correctness.
> >>>
> >> Well, for "communication latency" the topologhy do matter. And clearly
> >> it needs enough bandwidth to handle the number of clients it accept to
> >> serve. However, that it necessirely to have very high bandwidth is in my
> >> view wrong. One clearly can build a relaying system that builds on
> >> whatever node that exist with a real public IP address. As long as one
> >> keep the relaying bandwidth within what exist there should be no extra
> >> delay due to lower bandwidth. The path latency will still be the major
> >> issue.
> >>
> > [suresh] Would it it work for if I reworded this as follows? 
> > 
> >   "The obvious disadvantages of
> >    relaying are that it consumes the server's processing power and
> >    network bandwidth, and communication latency between the peering
> >    clients is likely to be increased even if the server has sufficient
> >    I/O bandwidth and is located correctly topology wise."
> 
> Yes, sound good.
> 
> > 
> > <.... stuff deleted>
> > 
> >>>> 15. section 5.1, second paragraph:
> >>>>
> >>>> "When the NAT device
> >>>>    enroute is not EIM-NAT, the application is unable to reuse an
> >>>>    already established endpoint mapping for communication with
> >>>>    different external destinations and the technique would fail."
> >>>>
> >>>> I think "enroute" is the wrong word. It is the outermost NAT that isn't
> >>>> an EIM-NAT for it to matter.
> >>>>
> >>> [suresh] As discussed with Remi on another thread, I believe, "enroute"
> is
> >>> proper usage here.
> >>>
> >>>
> >> I don't know which thread you are refering to. Well, enroute will not be
> >> missunderstood. But isn't "on path" what is commonly used in this case.
> >> But, this is really nit-picking. So lets ignore this.
> >>
> > [suresh] I was refering to the e-mail exchange between you and Remi on this
> > item on 09/04/07.
> > 
> 
> Okay, I think that the focus of that discussion was my misstake in that
> all NATs must be EIM-NATs, rather then if enroute or on path is the best
> word. But both do work and I was being to nit picking.
> 
> Cheers
> 
> Magnus Westerlund
> 
> IETF Transport Area Director & TSVWG Chair
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM/M
> ----------------------------------------------------------------------
> Ericsson AB                | Phone +46 8 4048287
> Torshamsgatan 23           | Fax   +46 8 7575550
> S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
> ----------------------------------------------------------------------
> 





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 20 14:47:24 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYR3F-0002pX-UL; Thu, 20 Sep 2007 14:47:01 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYR3E-0002nn-Qg
	for behave-confirm+ok@megatron.ietf.org; Thu, 20 Sep 2007 14:47:00 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYR3E-0002in-Eq
	for behave@ietf.org; Thu, 20 Sep 2007 14:47:00 -0400
Received: from mail-out4.apple.com ([17.254.13.23])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IYR2y-0000oI-LY
	for behave@ietf.org; Thu, 20 Sep 2007 14:46:44 -0400
Received: from relay13.apple.com (relay13.apple.com [17.128.113.29])
	by mail-out4.apple.com (Postfix) with ESMTP id AC3F1124C322
	for <behave@ietf.org>; Thu, 20 Sep 2007 11:46:43 -0700 (PDT)
Received: from relay13.apple.com (unknown [127.0.0.1])
	by relay13.apple.com (Symantec Mail Security) with ESMTP id 8CB6128085
	for <behave@ietf.org>; Thu, 20 Sep 2007 11:46:42 -0700 (PDT)
X-AuditID: 1180711d-a4361bb000006cd8-15-46f2c0122667
Received: from [17.206.23.199] (il0602a-dhcp71.apple.com [17.206.23.199])
	by relay13.apple.com (Apple SCV relay) with ESMTP id 769402805B
	for <behave@ietf.org>; Thu, 20 Sep 2007 11:46:42 -0700 (PDT)
Mime-Version: 1.0 (Apple Message framework v752.3)
In-Reply-To: <639293.64700.qm@web33301.mail.mud.yahoo.com>
References: <639293.64700.qm@web33301.mail.mud.yahoo.com>
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <107930F7-D59E-4297-9E2E-DF4C2E821292@apple.com>
Content-Transfer-Encoding: 7bit
From: james woodyatt <jhw@apple.com>
Subject: Re: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
Date: Thu, 20 Sep 2007 11:46:39 -0700
To: Behave WG <behave@ietf.org>
X-Mailer: Apple Mail (2.752.3)
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: b19722fc8d3865b147c75ae2495625f2
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

On Sep 18, 2007, at 06:47, Pyda Srisuresh wrote:
> --- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:
>>
>> 1. Section 1, page 3, first paragraph:
>> "These issues are
>>    likely to persist even into the IPv6 world, where a NAT may be  
>> used
>>    as an IPv4 compatibility mechanism [NAT-PT]."
>>
>> I think you should rewrite this considering that NAT-PT was  
>> deprecated.
>
> [suresh] Would the following work for you? I believe, there might  
> be NAT-PT
> deployments, even as [NAT-PT] is deprecated.
>
>      These issues are
>      likely to persist even into the IPv6 world and in deployments
>      using [NAT-PT].

It might also help to recognize that the IPv6 world will be plagued  
with stateful filtering firewalls with all the same basic filtering  
behaviors of NAT but without the address translation.  It's the  
filtering behavior more than anything else that interferes with P2P  
applications, and most of the considerations in this document apply  
for IPv6 as well as for IPv4/NAT.


--
james woodyatt <jhw@apple.com>
member of technical staff, communications engineering




_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 20 18:25:05 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYURh-0001hH-CN; Thu, 20 Sep 2007 18:24:29 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYURg-0001dd-Hx
	for behave-confirm+ok@megatron.ietf.org; Thu, 20 Sep 2007 18:24:28 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYURg-0001No-6L
	for behave@ietf.org; Thu, 20 Sep 2007 18:24:28 -0400
Received: from mail-out4.apple.com ([17.254.13.23])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IYURI-0006W8-Uh
	for behave@ietf.org; Thu, 20 Sep 2007 18:24:05 -0400
Received: from relay11.apple.com (relay11.apple.com [17.128.113.48])
	by mail-out4.apple.com (Postfix) with ESMTP id 537D91251683
	for <behave@ietf.org>; Thu, 20 Sep 2007 15:24:04 -0700 (PDT)
Received: from relay11.apple.com (unknown [127.0.0.1])
	by relay11.apple.com (Symantec Mail Security) with ESMTP id 3928728096
	for <behave@ietf.org>; Thu, 20 Sep 2007 15:24:04 -0700 (PDT)
X-AuditID: 11807130-a4bc4bb000004daf-0e-46f2f3042431
Received: from [17.206.23.199] (il0602a-dhcp71.apple.com [17.206.23.199])
	by relay11.apple.com (Apple SCV relay) with ESMTP id 2956128087
	for <behave@ietf.org>; Thu, 20 Sep 2007 15:24:04 -0700 (PDT)
Mime-Version: 1.0 (Apple Message framework v752.3)
In-Reply-To: <21801.1190319068@marajade.sandelman.ca>
References: <639293.64700.qm@web33301.mail.mud.yahoo.com>
	<107930F7-D59E-4297-9E2E-DF4C2E821292@apple.com>
	<21801.1190319068@marajade.sandelman.ca>
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <FE4D831F-F658-40E2-997E-1EEF4E38604D@apple.com>
Content-Transfer-Encoding: 7bit
From: james woodyatt <jhw@apple.com>
Subject: Re: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
Date: Thu, 20 Sep 2007 15:24:01 -0700
To: Behave WG <behave@ietf.org>
X-Mailer: Apple Mail (2.752.3)
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: -4.0 (----)
X-Scan-Signature: e5ba305d0e64821bf3d8bc5d3bb07228
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

On Sep 20, 2007, at 13:11, Michael Richardson wrote:
>
> Yes, but if you don't need any filtering, with IPv6, there is no  
> reason for your "router" to do anything other than routing.  That  
> eliminates *almost* all the situations where people are being  
> filtered in ways that they didn't ask for.

No, it does not.  Please see draft-ietf-v6ops-cpe-simple-security and  
the related discussion on V6OPS about the recommended deployment of  
stateful IPv6 packet filters by default in most consumer-grade  
routers.  These filters will be turned on in the factory default  
configuration, and in most cases will require uncommon networking  
expertise even to turn them off, much less customize them.  For all  
practical purposes, there will not be *ANY* situations where people  
are using IPv6 networks without being subjected to stateful filtering.

> Almost, because it doesn't get rid of wireless-cafe/hotel type  
> networks, where you are being filtered for payment reasons.
>
> However, in theory, "PANA" and "802.1x" are supposed to solve this  
> right?

Once again, the system works in theory, but how does it hold up in  
practice?


--
james woodyatt <jhw@apple.com>
member of technical staff, communications engineering




_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 20 18:58:58 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYUxj-0001p5-Dm; Thu, 20 Sep 2007 18:57:35 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYUxh-0001p0-Vf
	for behave-confirm+ok@megatron.ietf.org; Thu, 20 Sep 2007 18:57:33 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYUxh-0001os-Ls
	for behave@ietf.org; Thu, 20 Sep 2007 18:57:33 -0400
Received: from sj-iport-6.cisco.com ([171.71.176.117])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IYUxg-0004bA-8m
	for behave@ietf.org; Thu, 20 Sep 2007 18:57:33 -0400
X-IronPort-AV: E=Sophos;i="4.20,280,1186383600"; d="scan'208";a="222151047"
Received: from sj-dkim-4.cisco.com ([171.71.179.196])
	by sj-iport-6.cisco.com with ESMTP; 20 Sep 2007 15:57:31 -0700
Received: from sj-core-5.cisco.com (sj-core-5.cisco.com [171.71.177.238])
	by sj-dkim-4.cisco.com (8.12.11/8.12.11) with ESMTP id l8KMvVwT028825; 
	Thu, 20 Sep 2007 15:57:31 -0700
Received: from dwingwxp01 ([10.32.240.198])
	by sj-core-5.cisco.com (8.12.10/8.12.6) with ESMTP id l8KMvMDr027790;
	Thu, 20 Sep 2007 22:57:22 GMT
From: "Dan Wing" <dwing@cisco.com>
To: "'Pyda Srisuresh'" <srisuresh@yahoo.com>
References: <46F0DE21.70101@ericsson.com>
	<250174.68788.qm@web33312.mail.mud.yahoo.com>
Subject: RE: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
Date: Thu, 20 Sep 2007 15:57:22 -0700
Message-ID: <029d01c7fbd9$a02b8470$c6f0200a@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Office Outlook 11
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138
In-Reply-To: <250174.68788.qm@web33312.mail.mud.yahoo.com>
Thread-Index: Acf7RfOBASrJ8JGERVmyGw5nlKT9WwAk0fjw
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=8615; t=1190329051;
	x=1191193051; c=relaxed/simple; s=sjdkim4002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=dwing@cisco.com;
	z=From:=20=22Dan=20Wing=22=20<dwing@cisco.com>
	|Subject:=20RE=3A=20[BEHAVE]=20Re=3A=20Ad=20comments=20on=20draft-ietf-be
	have-p2p-state-03 |Sender:=20;
	bh=GS6fnWOYCOAWjdwNjzwB1yKx34U0JAG5Zfr/XdjE4FQ=;
	b=TPelkpaF6QFhtUiQZvk/SeMtYpt0AKfg3z0aqa0jaHQIUqtQQvA7wwW4fAICpZrtNI8u0CDv
	QARn7CbETaq7p+TYx3Dfm42hV1UG5IEBH6rxf88IqzUujBaggM9pxNnB;
Authentication-Results: sj-dkim-4; header.From=dwing@cisco.com; dkim=pass (s
	ig from cisco.com/sjdkim4002 verified; ); 
X-Spam-Score: -4.0 (----)
X-Scan-Signature: 155726d2f5fe5eb5c40a9f079fd9e841
Cc: 'Bryan Ford' <baford@mit.edu>, behave@ietf.org,
	'Dan Kegel' <dank@kegel.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

As an individual, I suggest that there may be useful text in section 2 of
RFC3041 that could be incorporated into -behave-p2p-state around 'privacy'.

-d

 

> -----Original Message-----
> From: Pyda Srisuresh [mailto:srisuresh@yahoo.com] 
> Sent: Wednesday, September 19, 2007 10:15 PM
> To: Magnus Westerlund
> Cc: Bryan Ford; Behave WG; Dan Kegel
> Subject: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
> 
> Magnus,
> 
> Please see my responses inline.
> 
> regards,
> suresh
> 
> --- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:
> 
> <... stuff deleted>
> 
> > 
> > >> 2. Section 1, page 3, second paragraph:
> > >>
> > >> "The
> > >>    anonymity and inaccessibility of the internal hosts 
> behind a NAT
> > >>    device is not a problem for applications such as web 
> browsers, which
> > >>    only need to initiate outgoing connections. This 
> inaccessibility is
> > >>    sometimes perceived as a privacy benefit."
> > >>
> > >> Is really "privacy" the main benefit. Isn't it the 
> filtering behavior of
> > >> the NAT that blocks incoming requests and thus avoid 
> getting the traffic
> > >> onto your local network that people like, i.e. the 
> "firewalling" aspect
> > >> of NATs.
> > >>
> > > [suresh] The internal hosts have private IP addresses, 
> which outsiders have
> > no
> > > knowledge of. This anonymity of IP addressing is 
> considered as privacy by
> > many
> > > folks. In addition, a host behind NAT is not accessible 
> to outsiders,
> > unless a
> > > session is initiated from inside. This inaccessibility is 
> also considered
> > as
> > > privacy by many folks. 
> > > 
> > > You might say, the filtering behavior of NATs provides 
> inaccessibility to
> > > internal hosts; and mapping behavior of NATs provides 
> anonymity to intenral
> > > hosts.
> > 
> > I agree with NATs filtering behavior making hosts 
> inaccessible in most
> > cases. But I don't they think they provide "privacy" in the 
> meaning most
> >  people think about it. If I deploy a NAT in my home, it is still
> > possible to learn that what ever traffic was sent come from 
> my home, but
> > maybe not which of the 5 computers in my home that sent it. 
> But that I
> > don't call privacy. There are many aspects of privacy, maybe a more
> > precise description of what it really provides.
> > 
> 
> [suresh] As I said earlier, the privacy I am referign to is 
> principally the
> combination of private Ip address (anonymity) and 
> inaccessibility from external
> hosts. The mapping behavior of NATs provides anonymity and 
> the filtering
> behavior of NATs provides inaccessibility to internal hosts. Would the
> following rewording work for you?
> 
>   "The
>    anonymity (private IP addresses) and inaccessibility of 
> the internal 
>    hosts behind a NAT device is not a problem for applications such as
>    web browsers, which only need to initiate outgoing 
> connections. This
>    anonymity and inaccessibility is sometimes perceived as a privacy
>    benefit."
> 
> <... stuff deleted>
> 
> > 
> > > 
> > >> 5. Section 2.7:
> > >>
> > >>   "A P2P application is an application that uses the 
> same endpoint to
> > >>    initiate outgoing sessions to peering hosts as well as accept
> > >>    incoming sessions from peering hosts."
> > >>
> > >> Considering the definition of "endpoint" in this 
> document I don't think
> > >> this definition is characteristic for peer to peer 
> applications. An peer
> > >> to peer application clearly can use different address 
> and ports to
> > >> communicate with other peers.
> > >>
> > > [suresh] Magnus - Would it work for you if the definition 
> is expanded to
> > say
> > > that a P2P application may use multiple endpoints for P2P 
> communication? 
> > The
> > > primary characteristic of a P2P applictaion is that the 
> same endpoint is
> > used
> > > for initiating an outgoing connection as well to accept incoming
> > connections.
> > > All examples and deployment instances described in the 
> document assume this
> > > definition.
> > 
> > I think that will resolve the main conflict in the definitions.
> > 
> [suresh] Will do. 
> 
> > > 
> > > An application that uses one endpoint to receive incoming 
> connections and a
> > > different endpoint for outgoing connections is not a P2P 
> application.
> > 
> > I don't think I agree with that characterisation. Isn't 
> that a result of
> > the need for making NAT traversal work. My interpretation 
> is more along
> > the following definition:
> > 
> > "Peer-to-peer is a communications model in which each party 
> has the same
> > capabilities and either party can initiate a communication session."
> > 
> 
> [suresh] I see what you mean. However, the definition I 
> provided is appropriate
> for the document, I believe.
> 
> <... stuff deleted>
> 
> 
> > >> 8. Section 3.1, last paragraph:
> > >>
> > >>   "The obvious disadvantages of
> > >>    relaying are that it consumes the server's processing 
> power and
> > >>    network bandwidth, and communication latency between 
> the peering
> > >>    clients is likely to be increased even if the server is
> > >>    well-connected."
> > >>
> > >> I think "well-connected" is the wrong word here. I think 
> it is rather
> > >> topology wise correctly located.
> > >>
> > > [suresh] By Well-connected, I was refering to having 
> large bandwidth (say,
> > in a
> > > data center with GigE connectivity). This is required, as 
> relay servers can
> > be
> > > in the path of many diverse, simultaneous sessions. The 
> relay server ought
> > to
> > > have aggregate bandwidth to support the numerous parallel 
> connections. I am
> > not
> > > refering to any topology-wise correctness.
> > > 
> > 
> > Well, for "communication latency" the topologhy do matter. 
> And clearly
> > it needs enough bandwidth to handle the number of clients 
> it accept to
> > serve. However, that it necessirely to have very high 
> bandwidth is in my
> > view wrong. One clearly can build a relaying system that builds on
> > whatever node that exist with a real public IP address. As 
> long as one
> > keep the relaying bandwidth within what exist there should 
> be no extra
> > delay due to lower bandwidth. The path latency will still 
> be the major
> > issue.
> > 
> [suresh] Would it it work for if I reworded this as follows? 
> 
>   "The obvious disadvantages of
>    relaying are that it consumes the server's processing power and
>    network bandwidth, and communication latency between the peering
>    clients is likely to be increased even if the server has sufficient
>    I/O bandwidth and is located correctly topology wise."
> 
> <.... stuff deleted>
> 
> > > 
> > >> 15. section 5.1, second paragraph:
> > >>
> > >> "When the NAT device
> > >>    enroute is not EIM-NAT, the application is unable to reuse an
> > >>    already established endpoint mapping for communication with
> > >>    different external destinations and the technique would fail."
> > >>
> > >> I think "enroute" is the wrong word. It is the outermost 
> NAT that isn't
> > >> an EIM-NAT for it to matter.
> > >>
> > > [suresh] As discussed with Remi on another thread, I 
> believe, "enroute" is
> > > proper usage here.
> > > 
> > > 
> > 
> > I don't know which thread you are refering to. Well, 
> enroute will not be
> > missunderstood. But isn't "on path" what is commonly used 
> in this case.
> > But, this is really nit-picking. So lets ignore this.
> > 
> [suresh] I was refering to the e-mail exchange between you 
> and Remi on this
> item on 09/04/07.
> 
> Thank you for your comments.
> 
> regards,
> suresh
> 
> > Cheers
> > 
> > Magnus Westerlund
> > 
> > IETF Transport Area Director & TSVWG Chair
> > 
> ----------------------------------------------------------------------
> > Multimedia Technologies, Ericsson Research EAB/TVM/M
> > 
> ----------------------------------------------------------------------
> > Ericsson AB                | Phone +46 8 4048287
> > Torshamsgatan 23           | Fax   +46 8 7575550
> > S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
> > 
> ----------------------------------------------------------------------
> > 
> > 
> 
> 
> 
> 
> 
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www1.ietf.org/mailman/listinfo/behave


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 20 21:49:48 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYXd5-0006ml-7p; Thu, 20 Sep 2007 21:48:27 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYXd2-0006md-SB
	for behave-confirm+ok@megatron.ietf.org; Thu, 20 Sep 2007 21:48:24 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYXd2-0006mU-G7
	for behave@ietf.org; Thu, 20 Sep 2007 21:48:24 -0400
Received: from web33315.mail.mud.yahoo.com ([68.142.206.130])
	by chiedprmail1.ietf.org with smtp (Exim 4.43) id 1IYXd1-00036I-U9
	for behave@ietf.org; Thu, 20 Sep 2007 21:48:24 -0400
Received: (qmail 40378 invoked by uid 60001); 21 Sep 2007 01:48:23 -0000
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID;
	b=qpMbqNa+tUBux7SdftYSVongDbVY0Uwn9jpvwHqlZInaUf+GddDIXYdw55mnY+BGat+hGBVZQc0SxdISD0kFusxEfu9R2uefVlckON8q9Nk33bpzENjnfyqgZoCXZH60ZrouXqzUt2lOqKl6CswGKpPSZnlHrtOtFtASh3q2bek=;
X-YMail-OSG: WzVmlXUVM1ltjanWGIGAf9LRsW.COtFMLmyjZ.pMmLGqJ0UkRJd7RIJ3xJ6WnejczA--
Received: from [209.172.74.45] by web33315.mail.mud.yahoo.com via HTTP;
	Thu, 20 Sep 2007 18:48:23 PDT
Date: Thu, 20 Sep 2007 18:48:23 -0700 (PDT)
From: Pyda Srisuresh <srisuresh@yahoo.com>
Subject: Re: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
To: james woodyatt <jhw@apple.com>, Behave WG <behave@ietf.org>
In-Reply-To: <107930F7-D59E-4297-9E2E-DF4C2E821292@apple.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <378522.39281.qm@web33315.mail.mud.yahoo.com>
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 8b431ad66d60be2d47c7bfeb879db82c
Cc: 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org


--- james woodyatt <jhw@apple.com> wrote:

> On Sep 18, 2007, at 06:47, Pyda Srisuresh wrote:
> > --- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:
> >>
> >> 1. Section 1, page 3, first paragraph:
> >> "These issues are
> >>    likely to persist even into the IPv6 world, where a NAT may be  
> >> used
> >>    as an IPv4 compatibility mechanism [NAT-PT]."
> >>
> >> I think you should rewrite this considering that NAT-PT was  
> >> deprecated.
> >
> > [suresh] Would the following work for you? I believe, there might  
> > be NAT-PT
> > deployments, even as [NAT-PT] is deprecated.
> >
> >      These issues are
> >      likely to persist even into the IPv6 world and in deployments
> >      using [NAT-PT].
> 
> It might also help to recognize that the IPv6 world will be plagued  
> with stateful filtering firewalls with all the same basic filtering  
> behaviors of NAT but without the address translation.  It's the  
> filtering behavior more than anything else that interferes with P2P  
> applications, and most of the considerations in this document apply  
> for IPv6 as well as for IPv4/NAT.
> 
[suresh] Agreed. I will include this text in the next rev of the draft. Thanks.

regards,
suresh

> 
> --
> james woodyatt <jhw@apple.com>
> member of technical staff, communications engineering
> 
> 
> 
> 
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www1.ietf.org/mailman/listinfo/behave
> 





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 20 21:52:48 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYXh3-0006lM-J4; Thu, 20 Sep 2007 21:52:33 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYXh1-0006iv-IL
	for behave-confirm+ok@megatron.ietf.org; Thu, 20 Sep 2007 21:52:31 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYXh1-0006im-8m
	for behave@ietf.org; Thu, 20 Sep 2007 21:52:31 -0400
Received: from web33313.mail.mud.yahoo.com ([68.142.206.128])
	by ietf-mx.ietf.org with smtp (Exim 4.43) id 1IYXgu-0008Qz-Uu
	for behave@ietf.org; Thu, 20 Sep 2007 21:52:31 -0400
Received: (qmail 54770 invoked by uid 60001); 21 Sep 2007 01:52:09 -0000
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID;
	b=vGzPhNyMYIMe41Dfrv8+EpCAXYAJ0mxw9+HA2p0DYwKRiN27sqRASzB0zJqUTqGeOr4+IU6BABShhtJc9Ov7uJspcyoNMsWBKsMa9jkQYa9yzq7uRYPQ3Ir2z3NuUv3w+crk1MrUFDeUhXS2dQZelK0zAQbna9y1aTggWDLYNy4=;
X-YMail-OSG: 6kdztwMVM1m8SoiyNPc52zwKs.v3EVqUYOURDJgwxo3uZE6c4g5VpHCbQECtp0ZRS2QSYth.5yr0YQyqrv2UVb3HkDHpS4HPC6B3WErpSHcBqyI-
Received: from [209.172.74.45] by web33313.mail.mud.yahoo.com via HTTP;
	Thu, 20 Sep 2007 18:52:09 PDT
Date: Thu, 20 Sep 2007 18:52:09 -0700 (PDT)
From: Pyda Srisuresh <srisuresh@yahoo.com>
Subject: RE: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
To: Dan Wing <dwing@cisco.com>
In-Reply-To: <029d01c7fbd9$a02b8470$c6f0200a@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Message-ID: <483411.54608.qm@web33313.mail.mud.yahoo.com>
X-Spam-Score: 0.0 (/)
X-Scan-Signature: f60fbf3dbcaca652b6d10036f0630412
Cc: 'Bryan Ford' <baford@mit.edu>, behave@ietf.org,
	'Dan Kegel' <dank@kegel.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

I will take a look. Thanks.

regards,
suresh

--- Dan Wing <dwing@cisco.com> wrote:

> As an individual, I suggest that there may be useful text in section 2 of
> RFC3041 that could be incorporated into -behave-p2p-state around 'privacy'.
> 
> -d
> 
>  
> 
> > -----Original Message-----
> > From: Pyda Srisuresh [mailto:srisuresh@yahoo.com] 
> > Sent: Wednesday, September 19, 2007 10:15 PM
> > To: Magnus Westerlund
> > Cc: Bryan Ford; Behave WG; Dan Kegel
> > Subject: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
> > 
> > Magnus,
> > 
> > Please see my responses inline.
> > 
> > regards,
> > suresh
> > 
> > --- Magnus Westerlund <magnus.westerlund@ericsson.com> wrote:
> > 
> > <... stuff deleted>
> > 
> > > 
> > > >> 2. Section 1, page 3, second paragraph:
> > > >>
> > > >> "The
> > > >>    anonymity and inaccessibility of the internal hosts 
> > behind a NAT
> > > >>    device is not a problem for applications such as web 
> > browsers, which
> > > >>    only need to initiate outgoing connections. This 
> > inaccessibility is
> > > >>    sometimes perceived as a privacy benefit."
> > > >>
> > > >> Is really "privacy" the main benefit. Isn't it the 
> > filtering behavior of
> > > >> the NAT that blocks incoming requests and thus avoid 
> > getting the traffic
> > > >> onto your local network that people like, i.e. the 
> > "firewalling" aspect
> > > >> of NATs.
> > > >>
> > > > [suresh] The internal hosts have private IP addresses, 
> > which outsiders have
> > > no
> > > > knowledge of. This anonymity of IP addressing is 
> > considered as privacy by
> > > many
> > > > folks. In addition, a host behind NAT is not accessible 
> > to outsiders,
> > > unless a
> > > > session is initiated from inside. This inaccessibility is 
> > also considered
> > > as
> > > > privacy by many folks. 
> > > > 
> > > > You might say, the filtering behavior of NATs provides 
> > inaccessibility to
> > > > internal hosts; and mapping behavior of NATs provides 
> > anonymity to intenral
> > > > hosts.
> > > 
> > > I agree with NATs filtering behavior making hosts 
> > inaccessible in most
> > > cases. But I don't they think they provide "privacy" in the 
> > meaning most
> > >  people think about it. If I deploy a NAT in my home, it is still
> > > possible to learn that what ever traffic was sent come from 
> > my home, but
> > > maybe not which of the 5 computers in my home that sent it. 
> > But that I
> > > don't call privacy. There are many aspects of privacy, maybe a more
> > > precise description of what it really provides.
> > > 
> > 
> > [suresh] As I said earlier, the privacy I am referign to is 
> > principally the
> > combination of private Ip address (anonymity) and 
> > inaccessibility from external
> > hosts. The mapping behavior of NATs provides anonymity and 
> > the filtering
> > behavior of NATs provides inaccessibility to internal hosts. Would the
> > following rewording work for you?
> > 
> >   "The
> >    anonymity (private IP addresses) and inaccessibility of 
> > the internal 
> >    hosts behind a NAT device is not a problem for applications such as
> >    web browsers, which only need to initiate outgoing 
> > connections. This
> >    anonymity and inaccessibility is sometimes perceived as a privacy
> >    benefit."
> > 
> > <... stuff deleted>
> > 
> > > 
> > > > 
> > > >> 5. Section 2.7:
> > > >>
> > > >>   "A P2P application is an application that uses the 
> > same endpoint to
> > > >>    initiate outgoing sessions to peering hosts as well as accept
> > > >>    incoming sessions from peering hosts."
> > > >>
> > > >> Considering the definition of "endpoint" in this 
> > document I don't think
> > > >> this definition is characteristic for peer to peer 
> > applications. An peer
> > > >> to peer application clearly can use different address 
> > and ports to
> > > >> communicate with other peers.
> > > >>
> > > > [suresh] Magnus - Would it work for you if the definition 
> > is expanded to
> > > say
> > > > that a P2P application may use multiple endpoints for P2P 
> > communication? 
> > > The
> > > > primary characteristic of a P2P applictaion is that the 
> > same endpoint is
> > > used
> > > > for initiating an outgoing connection as well to accept incoming
> > > connections.
> > > > All examples and deployment instances described in the 
> > document assume this
> > > > definition.
> > > 
> > > I think that will resolve the main conflict in the definitions.
> > > 
> > [suresh] Will do. 
> > 
> > > > 
> > > > An application that uses one endpoint to receive incoming 
> > connections and a
> > > > different endpoint for outgoing connections is not a P2P 
> > application.
> > > 
> > > I don't think I agree with that characterisation. Isn't 
> > that a result of
> > > the need for making NAT traversal work. My interpretation 
> > is more along
> > > the following definition:
> > > 
> > > "Peer-to-peer is a communications model in which each party 
> > has the same
> > > capabilities and either party can initiate a communication session."
> > > 
> > 
> > [suresh] I see what you mean. However, the definition I 
> > provided is appropriate
> > for the document, I believe.
> > 
> > <... stuff deleted>
> > 
> > 
> > > >> 8. Section 3.1, last paragraph:
> > > >>
> > > >>   "The obvious disadvantages of
> > > >>    relaying are that it consumes the server's processing 
> > power and
> > > >>    network bandwidth, and communication latency between 
> > the peering
> > > >>    clients is likely to be increased even if the server is
> > > >>    well-connected."
> > > >>
> > > >> I think "well-connected" is the wrong word here. I think 
> > it is rather
> > > >> topology wise correctly located.
> > > >>
> > > > [suresh] By Well-connected, I was refering to having 
> > large bandwidth (say,
> > > in a
> > > > data center with GigE connectivity). This is required, as 
> > relay servers can
> > > be
> > > > in the path of many diverse, simultaneous sessions. The 
> > relay server ought
> > > to
> > > > have aggregate bandwidth to support the numerous parallel 
> > connections. I am
> > > not
> > > > refering to any topology-wise correctness.
> > > > 
> > > 
> > > Well, for "communication latency" the topologhy do matter. 
> > And clearly
> > > it needs enough bandwidth to handle the number of clients 
> > it accept to
> > > serve. However, that it necessirely to have very high 
> > bandwidth is in my
> > > view wrong. One clearly can build a relaying system that builds on
> > > whatever node that exist with a real public IP address. As 
> > long as one
> > > keep the relaying bandwidth within what exist there should 
> > be no extra
> > > delay due to lower bandwidth. The path latency will still 
> > be the major
> > > issue.
> > > 
> > [suresh] Would it it work for if I reworded this as follows? 
> > 
> >   "The obvious disadvantages of
> >    relaying are that it consumes the server's processing power and
> >    network bandwidth, and communication latency between the peering
> >    clients is likely to be increased even if the server has sufficient
> >    I/O bandwidth and is located correctly topology wise."
> > 
> > <.... stuff deleted>
> > 
> > > > 
> > > >> 15. section 5.1, second paragraph:
> > > >>
> > > >> "When the NAT device
> > > >>    enroute is not EIM-NAT, the application is unable to reuse an
> > > >>    already established endpoint mapping for communication with
> > > >>    different external destinations and the technique would fail."
> > > >>
> > > >> I think "enroute" is the wrong word. It is the outermost 
> > NAT that isn't
> > > >> an EIM-NAT for it to matter.
> > > >>
> > > > [suresh] As discussed with Remi on another thread, I 
> > believe, "enroute" is
> > > > proper usage here.
> > > > 
> > > > 
> > > 
> > > I don't know which thread you are refering to. Well, 
> > enroute will not be
> > > missunderstood. But isn't "on path" what is commonly used 
> > in this case.
> > > But, this is really nit-picking. So lets ignore this.
> > > 
> > [suresh] I was refering to the e-mail exchange between you 
> > and Remi on this
> > item on 09/04/07.
> > 
> > Thank you for your comments.
> > 
> > regards,
> > suresh
> > 
> > > Cheers
> > > 
> > > Magnus Westerlund
> > > 
> > > IETF Transport Area Director & TSVWG Chair
> > > 
> > ----------------------------------------------------------------------
> > > Multimedia Technologies, Ericsson Research EAB/TVM/M
> > > 
> > ----------------------------------------------------------------------
> > > Ericsson AB                | Phone +46 8 4048287
> > > Torshamsgatan 23           | Fax   +46 8 7575550
> > > S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
> > > 
> > ----------------------------------------------------------------------
> > > 
> > > 
> > 
> > 
> > 
> > 
> > 
> > _______________________________________________
> > Behave mailing list
> > Behave@ietf.org
> > https://www1.ietf.org/mailman/listinfo/behave
> 





_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Fri Sep 21 05:01:38 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYeME-0001gb-2R; Fri, 21 Sep 2007 04:59:30 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYeMC-0001ey-O0
	for behave-confirm+ok@megatron.ietf.org; Fri, 21 Sep 2007 04:59:28 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYeMC-0001ep-CL
	for behave@ietf.org; Fri, 21 Sep 2007 04:59:28 -0400
Received: from mailgw4.ericsson.se ([193.180.251.62])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IYeMB-00051a-QV
	for behave@ietf.org; Fri, 21 Sep 2007 04:59:28 -0400
Received: from mailgw4.ericsson.se (unknown [127.0.0.1])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	778C42186D; Fri, 21 Sep 2007 10:59:08 +0200 (CEST)
X-AuditID: c1b4fb3e-b0835bb0000007e1-ee-46f387dc0c5e
Received: from esealmw126.eemea.ericsson.se (unknown [153.88.254.123])
	by mailgw4.ericsson.se (Symantec Mail Security) with ESMTP id
	596B5217FA; Fri, 21 Sep 2007 10:59:08 +0200 (CEST)
Received: from esealmw126.eemea.ericsson.se ([153.88.254.174]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 21 Sep 2007 10:59:08 +0200
Received: from [147.214.30.247] ([147.214.30.247]) by
	esealmw126.eemea.ericsson.se with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 21 Sep 2007 10:59:07 +0200
Message-ID: <46F387D8.3060003@ericsson.com>
Date: Fri, 21 Sep 2007 10:59:04 +0200
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: Dan Wing <dwing@cisco.com>
Subject: Re: [BEHAVE] Re: Ad comments on draft-ietf-behave-p2p-state-03
References: <46F0DE21.70101@ericsson.com>
	<250174.68788.qm@web33312.mail.mud.yahoo.com>
	<029d01c7fbd9$a02b8470$c6f0200a@cisco.com>
In-Reply-To: <029d01c7fbd9$a02b8470$c6f0200a@cisco.com>
X-Enigmail-Version: 0.95.3
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-OriginalArrivalTime: 21 Sep 2007 08:59:07.0915 (UTC)
	FILETIME=[AB305DB0:01C7FC2D]
X-Brightmail-Tracker: AAAAAA==
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 7d33c50f3756db14428398e2bdedd581
Cc: behave@ietf.org, 'Bryan Ford' <baford@mit.edu>,
	'Dan Kegel' <dank@kegel.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Dan Wing skrev:
> As an individual, I suggest that there may be useful text in section 2 of
> RFC3041 that could be incorporated into -behave-p2p-state around 'privacy'.
> 

Yes, I think the information in Section 2.2 is very relevant to this
discussion and shows that SOHO nats really do not provide any "privacy"
feature, even if the general public may considered it providing something.

Cheers

Magnus Westerlund

IETF Transport Area Director & TSVWG Chair
----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM/M
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Fri Sep 21 09:15:57 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IYiLR-0002mq-Iv; Fri, 21 Sep 2007 09:14:57 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IYiLQ-0002jd-4j
	for behave-confirm+ok@megatron.ietf.org; Fri, 21 Sep 2007 09:14:56 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IYiLP-0002jU-R4
	for behave@ietf.org; Fri, 21 Sep 2007 09:14:55 -0400
Received: from smtp.nokia.com ([131.228.20.171] helo=mgw-ext12.nokia.com)
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IYiLJ-0007RC-EW
	for behave@ietf.org; Fri, 21 Sep 2007 09:14:55 -0400
Received: from esebh108.NOE.Nokia.com (esebh108.ntc.nokia.com [172.21.143.145])
	by mgw-ext12.nokia.com (Switch-3.2.5/Switch-3.2.5) with ESMTP id
	l8LDEdVq012629 for <behave@ietf.org>; Fri, 21 Sep 2007 16:14:42 +0300
Received: from esebh103.NOE.Nokia.com ([172.21.143.33]) by
	esebh108.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 21 Sep 2007 16:14:31 +0300
Received: from esebh102.NOE.Nokia.com ([172.21.138.183]) by
	esebh103.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 21 Sep 2007 16:14:30 +0300
Received: from esdhcp04327.research.nokia.com ([172.21.43.27]) by
	esebh102.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 21 Sep 2007 16:14:30 +0300
From: =?iso-8859-1?q?R=E9mi_Denis-Courmont?= <remi.denis-courmont@nokia.com>
Organization: Nokia TP-SP-SWD
To: behave@ietf.org
Date: Fri, 21 Sep 2007 16:14:32 +0300
User-Agent: KMail/1.9.6 (enterprise 0.20070907.709405)
MIME-Version: 1.0
Content-Type: Multipart/Mixed;
  boundary="Boundary-00=_4O88G6eQy0LyFiw"
Message-Id: <200709211614.32826.remi.denis-courmont@nokia.com>
X-OriginalArrivalTime: 21 Sep 2007 13:14:30.0314 (UTC)
	FILETIME=[580CE8A0:01C7FC51]
X-Nokia-AV: Clean
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 6e922792024732fb1bb6f346e63517e4
Subject: [BEHAVE] Fwd: New Version Notification for
	draft-denis-behave-rfc3489bis-test-vectors-02
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

--Boundary-00=_4O88G6eQy0LyFiw
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

=46YI.

I have added the embedded xor'd address (changed the vectors to use=20
documentation IP address space). Also added an IPv6 example.

Comments welcome.

=2D-=20
R=E9mi Denis-Courmont

--Boundary-00=_4O88G6eQy0LyFiw
Content-Type: message/rfc822;
  name="forwarded message"
Content-Transfer-Encoding: 7bit
Content-Description: "ext IETF I-D Submission Tool" <developers@ietf.org>: New
	Version Notification for
	draft-denis-behave-rfc3489bis-test-vectors-02
Content-Disposition: inline

Received: from esebh103.NOE.Nokia.com ([172.21.143.33]) by
	esebe108.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 21 Sep 2007 16:09:49 +0300
Received: from esebh106.NOE.Nokia.com ([172.21.138.213]) by
	esebh103.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 21 Sep 2007 16:09:48 +0300
Received: from mgw-mx06.nokia.com ([192.100.122.233]) by
	esebh106.NOE.Nokia.com with Microsoft SMTPSVC(6.0.3790.1830); 
	Fri, 21 Sep 2007 16:09:48 +0300
Received: from ns0.neustar.com (ns0.neustar.com [156.154.16.158])
	by mgw-mx06.nokia.com (Switch-3.2.6/Switch-3.2.6) with ESMTP id
	l8LD9jqQ015239
	(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=FAIL)
	for <remi.denis-courmont@nokia.com>; Fri, 21 Sep 2007 16:09:47 +0300
Received: from ietf.org (stiedprweb1.va.neustar.com [10.91.34.42])
	by ns0.neustar.com (Postfix) with ESMTP id 951F7328CD
	for <remi.denis-courmont@nokia.com>;
	Fri, 21 Sep 2007 13:09:34 +0000 (GMT)
Received: from mirror by ietf.org with local (Exim 4.43) id 1IYiGE-00039T-Gl
	for remi.denis-courmont@nokia.com; Fri, 21 Sep 2007 09:09:34 -0400
Content-Type: text/plain;
  charset="utf-8"
Mime-Version: 1.0
To: remi.denis-courmont@nokia.com
Cc: 
From: "ext IETF I-D Submission Tool" <developers@ietf.org>
Subject: New Version Notification for 
	draft-denis-behave-rfc3489bis-test-vectors-02 
Message-Id: <E1IYiGE-00039T-Gl@ietf.org>
Date: Fri, 21 Sep 2007 09:09:34 -0400
X-Nokia-AV: Clean
X-pstn-spam: W
X-Spam-Score: 0.00%
Return-Path: mirror@ietf.org
X-OriginalArrivalTime: 21 Sep 2007 13:09:48.0638 (UTC)
	FILETIME=[B0288FE0:01C7FC50]
X-UID: 11636
X-Length: 3447


A new version of I-D, draft-denis-behave-rfc3489bis-test-vectors-02.txt has been successfuly submitted by Remi Denis-Courmont and posted to the IETF repository.

Filename:	 draft-denis-behave-rfc3489bis-test-vectors
Revision:	 02
Title:		 Test vectors for RFC3489bis
Creation_date:	 2007-09-21
WG ID:		 Independent Submission
Number_of_pages: 8

Abstract:
This document includes test vectors for the MESSAGE-INTEGRITY and
FINGERPRINT attributes of the STUN protocol.Table of Contents

1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.  Test vectors  . . . . . . . . . . . . . . . . . . . . . . . . . 3
  2.1.  Sample request  . . . . . . . . . . . . . . . . . . . . . . 3
  2.2.  Sample IPv4 response  . . . . . . . . . . . . . . . . . . . 4
  2.3.  Sample IPv6 response  . . . . . . . . . . . . . . . . . . . 5
3.  Security Considerations . . . . . . . . . . . . . . . . . . . . 5
4.  IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 5
5.  Acknowledgements  . . . . . . . . . . . . . . . . . . . . . . . 5
6.  Normative References  . . . . . . . . . . . . . . . . . . . . . 6
Appendix A.  Source code for test vectors . . . . . . . . . . . . . 61.  Introduction

The Session Traversal Utilities for NAT
(STUN)[I-D.ietf-behave-rfc3489bis] protocol defines two different
hashes that may be included in messages exchanged by peers
implementing that protocol:

FINGERPRINT attribute:  a 32-bits Circular Redundancy Check.
MESSAGE-INTEGRITY attribute:  a HMAC-SHA1 authentication code.

This document documents sample properly-formatted STUN messages
including these hashes, for the sake of testing implementations of
the STUN protocol.
                                                                                  


The IETF Secretariat.



--Boundary-00=_4O88G6eQy0LyFiw
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

--Boundary-00=_4O88G6eQy0LyFiw--





From behave-bounces@ietf.org Sun Sep 23 16:31:26 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IZY5t-0005LY-KM; Sun, 23 Sep 2007 16:30:21 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IZY5s-0005KW-AU
	for behave-confirm+ok@megatron.ietf.org; Sun, 23 Sep 2007 16:30:20 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IZY5r-0005K4-Ps
	for behave@ietf.org; Sun, 23 Sep 2007 16:30:20 -0400
Received: from sj-iport-5.cisco.com ([171.68.10.87])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IZY5q-00017a-Gk
	for behave@ietf.org; Sun, 23 Sep 2007 16:30:19 -0400
X-IronPort-AV: E=Sophos;i="4.20,289,1186383600"; d="scan'208";a="177285558"
Received: from sj-dkim-4.cisco.com ([171.71.179.196])
	by sj-iport-5.cisco.com with ESMTP; 23 Sep 2007 13:30:17 -0700
Received: from sj-core-1.cisco.com (sj-core-1.cisco.com [171.71.177.237])
	by sj-dkim-4.cisco.com (8.12.11/8.12.11) with ESMTP id l8NKUHfQ012636; 
	Sun, 23 Sep 2007 13:30:17 -0700
Received: from [192.168.4.177] (sjc-fluffy-vpn1.cisco.com [10.25.236.82])
	by sj-core-1.cisco.com (8.12.10/8.12.6) with SMTP id l8NKUGio016185;
	Sun, 23 Sep 2007 20:30:17 GMT
In-Reply-To: <04f401c7f984$0d6bdb00$c6f0200a@cisco.com>
References: <20d2bdfb0709122004o709211bbt9569b0f1d3ae1120@mail.gmail.com>
	<04f401c7f984$0d6bdb00$c6f0200a@cisco.com>
Impp: xmpp:cullenfluffyjennings@jabber.org
Mime-Version: 1.0 (Apple Message framework v752.3)
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <72D21363-52D8-4F3B-B02C-1F2730547F27@cisco.com>
Content-Transfer-Encoding: 7bit
From: Cullen Jennings <fluffy@cisco.com>
Subject: Re: [BEHAVE] behavior discovery category
Date: Sun, 23 Sep 2007 13:29:53 -0700
To: Dan Wing <dwing@cisco.com>
X-Mailer: Apple Mail (2.752.3)
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=8366; t=1190579417;
	x=1191443417; c=relaxed/simple; s=sjdkim4002;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=fluffy@cisco.com;
	z=From:=20Cullen=20Jennings=20<fluffy@cisco.com>
	|Subject:=20Re=3A=20[BEHAVE]=20behavior=20discovery=20category
	|Sender:=20; bh=wohgd4TuNV6IT9o/6sXCEJP4s1xEzOW6sWwvcWLoMPI=;
	b=I0WsZILgvsNs7xLdBMbaeTJuCXgbqWQ8XAQ+5qtSljJ7Z7qnUWqSPJCWGAJLiXPsQ1TuUJbZ
	gg5KgWWFpyn0Bat9AtwGfZk2V+QQ64M+ZhgvWTCnH2wTqjhZDaoOctp6;
Authentication-Results: sj-dkim-4; header.From=fluffy@cisco.com; dkim=pass (
	sig from cisco.com/sjdkim4002 verified; ); 
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 311e798ce51dbeacf5cdfcc8e9fda21b
Cc: 'Behave WG' <behave@ietf.org>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org


One of the key reasons that we needed to do a bis for RFC 3489  is  
because we have run the experiment to find out if nodes can reliably  
detect what type of NAT they are behind using a STUN like techniques.  
We know the answer - they can not and this is not a rare case but  
happens surprisingly often. This resulted in the change away from the  
advice in 3489 to the advice in bis (not to mention the move from  
unilateral NAT traversal techniques to bilateral techniques). There  
are test results document in draft-jennings-behave-test-results.  
These results are for specific NATs but we know from other deployment  
experience, and from other test results such as Saikat's and Ford's,  
that we wee similar issues in deployments.

I agree it would be nice if a client could detect what type of NAT it  
was behind using STUN, however, I believe the only experiment being  
proposed here we already clearly know the answer to.

Cullen <with my individual hat on>


On Sep 17, 2007, at 4:39 PM, Dan Wing wrote:

> Folks, please look over Bruce's email below.  We need some  
> discussion on
> Informational versus Experimental for
> draft-ietf-behave-nat-behavior-discovery.  The choice will control  
> how other
> specifications can reference the document.  As Bruce mentions  
> below, see
> http://www.ietf.org/u/ietfchair/info-exp.html for guidelines on  
> Informational
> versus Experimental.
>
> Thanks,
> -d
>
>> -----Original Message-----
>> From: Bruce Lowekamp [mailto:lowekamp@sipeerior.com]
>> Sent: Wednesday, September 12, 2007 8:05 PM
>> To: Behave WG
>> Subject: [BEHAVE] behavior discovery category
>>
>> In Chicago we agreed to bring the question of what category
>> (Experimental or Informational) was appropriate for behavior- 
>> discovery
>> to the list.  Derek and I spent some time looking at this and at
>> http://www.ietf.org/u/ietfchair/info-exp.html.  We believe that the
>> appropriate status is experimental.   We have used P2P supernode
>> selection as an example application, but believe that the case of
>> applications without a rendevous server is more general.
>>
>> I'm including the text from a revision of the draft that
>> addresses this issue.
>>
>> Bruce
>>
>>
>> Abstract
>>
>>    This specification defines an experimental usage of the Simple
>>    Traversal Underneath Network Address Translators (NAT) (STUN)
>>    Protocol that discovers the presence and current behaviour of NATs
>>    and firewalls between the STUN client and the STUN server.
>>
>>
>>
>> 1.  Applicability
>>
>>    This experimental STUN usage does not allow an application  
>> behind a
>>    NAT to make an absolute determination of the NAT's  
>> characteristics.
>>    NAT devices do not behave consistently enough to predict future
>>    behaviour with any guarantee.  This STUN usage provides  
>> information
>>    about observable transient behavior; it only truly
>> determines a NAT's
>>    behavior with regard to the STUN server used at the
>> instant the test
>>    is run.  Applications requiring reliable reach between two
>> particular
>>    endpoints must establish a communication channel through a
>> NAT using
>>    another technique.  IETF has proposed standards including ICE
>>    [I-D.ietf-mmusic-ice] and OUTBOUND [I-D.ietf-sip-outbound] for
>>    establishing communication channels when a publically accessible
>>    rendevous service is available.
>>
>>    This usage provides techniques known to be of use for diagnostic
>>    purposes.  These techniques are considered experimental
>> when applied
>>    to real-time selection of parameters for protocols in situations
>>    where a publically accessible rendevous service is not available.
>>    One possible application might be optimizing supernode  
>> selection in
>>    P2P networks based on statistical experience with establishing
>>    connections and diagnosing NAT behavior with a variety of other
>>    peers.
>>
>>
>> 2.  Introduction
>>
>>    The Simple Traversal Underneath Network Address Translators (NAT)
>>    (STUN) [I-D.ietf-behave-rfc3489bis] provides a mechanism
>> to discover
>>    the reflexive transport address toward the STUN server, using the
>>    Binding Request.  This specification defines the NAT Behavior
>>    Discovery STUN usage, which allows a STUN client to probe
>> the current
>>    behaviour of the NAT/FW devices between the client and the STUN
>>    server.  This usage defines new STUN attributes for the Binding
>>    Request and Binding Response.
>>
>>    Many NAT/FW devices do not behave consistently and will
>> change their
>>    behaviour under load and over time.  Applications requiring high
>>    reliability must be prepared for the NAT's behaviour to become  
>> more
>>    restrictive.  Specifically, it has been found that under load NATs
>>    may transition to the most restrictive filtering and mapping
>>    behaviour and shorten the lifetime of new and existing
>> bindings.  In
>>    short, applications can discover how bad things currently are, but
>>    not how bad things will get.
>>
>>    Despite this limitation, instantaneous observations are often  
>> quite
>>    useful in troubleshooting network problems, and repeated tests  
>> over
>>    time, or in known load situations, may be used to characterize a
>>    NAT's behavior.  In particular, in the hands of a person
>>    knowledgeable about the needs of an application and the nodes an
>>    application needs to communicate with, it can be a powerful tool.
>>    For these diagnostics to be effectiev, the application
>> must select a
>>    STUN server that is appropriately located with regards to
>> its future
>>    communication partners.
>>
>> 2.1.  Example Use with P2P Overlays
>>
>>    An application could use Behavior Discovery in a P2P protocol to
>>    determine if a particular endpoint is a reasonable supernode
>>    candidate.  This P2P network application is willing to select
>>    supernodes that might advertise unreliable IP addresses to
>> avoid the
>>    cost of dedicated servers.  A supernode candidate requires that  
>> its
>>    NAT(s) offer(s) Address Independent Filtering.  It might
>> periodically
>>    re-run tests and would remove itself as a supernode if its NAT/FW
>>    chain lost this characteristic.  These tests could be run
>> with other
>>    supernode candidates acting as STUN servers as well as
>> dedicated STUN
>>    servers.  As many P2P algorithms tolerate non-transitive
>> connectivity
>>    between a portion of their peers, guaranteed pair-wise
>> reliable reach
>>    might be sacrificed in order to distribute the P2P overlay's load
>>    across peers that can be directly contacted by the
>> majority of users.
>>
>>    Use of Behavior Discovery for such an application requires:
>>
>>    o  Specification of protocols capable of offering reliable end- 
>> user
>>       performance using unreliable links between peers.
>>
>>    o  The application is deployed behind NATs that provide Address
>>       Independent Filtering and that remain in this mode for an  
>> amount
>>       of time sufficient for the application to identify
>> their behavior,
>>       distribute this information to the rest of the overlay, and
>>       provide useful work for the application.
>>
>>    This draft is experimental as deployed applications
>> implementing open
>>    protocols have yet to be deployed in such environments to
>> demonstrate
>>    that these two requirements have been met.  However, apocryphal
>>    evidence suggests that househould and small business targeted NAT
>>    devices have stable behaviour, especially when there are
>> few clients
>>    behind them.  And numerous P2P applications have been deployed  
>> that
>>    appear to have these properties, although their protocols have not
>>    yet been subjected to rigorous evaluation by standards bodies.
>>
>>
>> _______________________________________________
>> Behave mailing list
>> Behave@ietf.org
>> https://www1.ietf.org/mailman/listinfo/behave
>
>
> _______________________________________________
> Behave mailing list
> Behave@ietf.org
> https://www1.ietf.org/mailman/listinfo/behave


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Wed Sep 26 12:21:07 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IaZcN-0007Dh-6T; Wed, 26 Sep 2007 12:20:07 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IaZcL-00072b-FH
	for behave-confirm+ok@megatron.ietf.org; Wed, 26 Sep 2007 12:20:05 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IaZcI-0006vQ-Kf; Wed, 26 Sep 2007 12:20:03 -0400
Received: from ns4.neustar.com ([156.154.24.139])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43)
	id 1IaZcI-0004MW-97; Wed, 26 Sep 2007 12:20:02 -0400
Received: from stiedprstage1.ietf.org (stiedprstage1.va.neustar.com
	[10.31.47.10]) by ns4.neustar.com (Postfix) with ESMTP id 34AA82AC65;
	Wed, 26 Sep 2007 16:20:02 +0000 (GMT)
Received: from ietf by stiedprstage1.ietf.org with local (Exim 4.43)
	id 1IaZcH-000857-Vg; Wed, 26 Sep 2007 12:20:01 -0400
Content-Type: Multipart/Mixed; Boundary="NextPart"
Mime-Version: 1.0
To: i-d-announce@ietf.org
From: Internet-Drafts@ietf.org
Message-Id: <E1IaZcH-000857-Vg@stiedprstage1.ietf.org>
Date: Wed, 26 Sep 2007 12:20:01 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 3002fc2e661cd7f114cb6bae92fe88f1
Cc: behave@ietf.org
Subject: [BEHAVE] I-D Action:draft-ietf-behave-nat-icmp-05.txt 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Behavior Engineering for Hindrance Avoidance Working Group of the IETF.


	Title           : NAT Behavioral Requirements for ICMP protocol
	Author(s)       : P. Srisuresh, et al.
	Filename        : draft-ietf-behave-nat-icmp-05.txt
	Pages           : 24
	Date            : 2007-09-26

This document specifies the behavioral properties required of the 
Network Address Translator (NAT) devices in conjunction with the
ICMP protocol. The objective of this memo is to make NAT devices
more predictable and compatible with diverse application protocols
that traverse the devices. Companion documents provide behavioral
recommendations specific to TCP, UDP and other protocols.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-behave-nat-icmp-05.txt

To remove yourself from the I-D Announcement list, send a message to
i-d-announce-request@ietf.org with the word unsubscribe in the body of 
the message.
You can also visit https://www1.ietf.org/mailman/listinfo/I-D-announce
to change your subscription settings.

Internet-Drafts are also available by anonymous FTP. Login with the 
username "anonymous" and a password of your e-mail address. After 
logging in, type "cd internet-drafts" and then
	"get draft-ietf-behave-nat-icmp-05.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt

Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-behave-nat-icmp-05.txt".

NOTE:   The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up iFrom behave-bounces@ietf.org Wed Sep 26 12:21:07 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IaZcN-0007Dh-6T; Wed, 26 Sep 2007 12:20:07 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IaZcL-00072b-FH
	for behave-confirm+ok@megatron.ietf.org; Wed, 26 Sep 2007 12:20:05 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IaZcI-0006vQ-Kf; Wed, 26 Sep 2007 12:20:03 -0400
Received: from ns4.neustar.com ([156.154.24.139])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43)
	id 1IaZcI-0004MW-97; Wed, 26 Sep 2007 12:20:02 -0400
Received: from stiedprstage1.ietf.org (stiedprstage1.va.neustar.com
	[10.31.47.10]) by ns4.neustar.com (Postfix) with ESMTP id 34AA82AC65;
	Wed, 26 Sep 2007 16:20:02 +0000 (GMT)
Received: from ietf by stiedprstage1.ietf.org with local (Exim 4.43)
	id 1IaZcH-000857-Vg; Wed, 26 Sep 2007 12:20:01 -0400
Content-Type: Multipart/Mixed; Boundary="NextPart"
Mime-Version: 1.0
To: i-d-announce@ietf.org
From: Internet-Drafts@ietf.org
Message-Id: <E1IaZcH-000857-Vg@stiedprstage1.ietf.org>
Date: Wed, 26 Sep 2007 12:20:01 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 3002fc2e661cd7f114cb6bae92fe88f1
Cc: behave@ietf.org
Subject: [BEHAVE] I-D Action:draft-ietf-behave-nat-icmp-05.txt 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Behavior Engineering for Hindrance Avoidance Working Group of the IETF.


	Title           : NAT Behavioral Requirements for ICMP protocol
	Author(s)       : P. Srisuresh, et al.
	Filename        : draft-ietf-behave-nat-icmp-05.txt
	Pages           : 24
	Date            : 2007-09-26

This document specifies the behavioral properties required of the 
Network Address Translator (NAT) devices in conjunction with the
ICMP protocol. The objective of this memo is to make NAT devices
more predictable and compatible with diverse application protocols
that traverse the devices. Companion documents provide behavioral
recommendations specific to TCP, UDP and other protocols.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-behave-nat-icmp-05.txt

To remove yourself from the I-D Announcement list, send a message to
i-d-announce-request@ietf.org with the word unsubscribe in the body of 
the message.
You can also visit https://www1.ietf.org/mailman/listinfo/I-D-announce
to change your subscription settings.

Internet-Drafts are also available by anonymous FTP. Login with the 
username "anonymous" and a password of your e-mail address. After 
logging in, type "cd internet-drafts" and then
	"get draft-ietf-behave-nat-icmp-05.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt

Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-behave-nat-icmp-05.txt".

NOTE:   The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.

Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body; access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID: <2007-09-26121300.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-behave-nat-icmp-05.txt

--OtherAccess
Content-Type: Message/External-body; name="draft-ietf-behave-nat-icmp-05.txt";
	site="ftp.ietf.org"; access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID: <2007-09-26121300.I-D\@ietf.org>


--OtherAccess--

--NextPart
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

--NextPart--



From behave-bounces@ietf.org Wed Sep 26 12:21:07 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IaZcP-0007Mk-T1; Wed, 26 Sep 2007 12:20:09 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IaZcL-00073F-HI
	for behave-confirm+ok@megatron.ietf.org; Wed, 26 Sep 2007 12:20:05 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IaZcJ-0006vT-5e; Wed, 26 Sep 2007 12:20:04 -0400
Received: from ns0.neustar.com ([156.154.16.158])
	by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IaZcI-0007A7-1s; Wed, 26 Sep 2007 12:20:03 -0400
Received: from stiedprstage1.ietf.org (stiedprstage1.va.neustar.com
	[10.31.47.10]) by ns0.neustar.com (Postfix) with ESMTP id 065903288F;
	Wed, 26 Sep 2007 16:20:02 +0000 (GMT)
Received: from ietf by stiedprstage1.ietf.org with local (Exim 4.43)
	id 1IaZcH-000852-Tv; Wed, 26 Sep 2007 12:20:01 -0400
Content-Type: Multipart/Mixed; Boundary="NextPart"
Mime-Version: 1.0
To: i-d-announce@ietf.org
From: Internet-Drafts@ietf.org
Message-Id: <E1IaZcH-000852-Tv@stiedprstage1.ietf.org>
Date: Wed, 26 Sep 2007 12:20:01 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 3002fc2e661cd7f114cb6bae92fe88f1
Cc: behave@ietf.org
Subject: [BEHAVE] I-D Action:draft-ietf-behave-p2p-state-04.txt 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Behavior Engineering for Hindrance Avoidance Working Group of the IETF.


	Title           : State of Peer-to-Peer(P2P) Communication Across Network Address Translators(NATs)
	Author(s)       : P. Srisuresh, et al.
	Filename        : draft-ietf-behave-p2p-state-04.txt
	Pages           : 32
	Date            : 2007-09-26

This memo documents the various methods known to be in use by
applications to establish direct communication in the presence
of Network Address Translators (NATs) at the current time. This
memo covers NAT traversal approaches used by both TCP and UDP
based applications. This memo is not an endorsement of the methods 
described, but merely an attempt to capture them in a document.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-behave-p2p-state-04.txt

To remove yourself from the I-D Announcement list, send a message to
i-d-announce-request@ietf.org with the wornto multiple messages), so check your local documentation on
	how to manipulate these messages.

Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body; access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID: <2007-09-26121300.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-behave-nat-icmp-05.txt

--OtherAccess
Content-Type: Message/External-body; name="draft-ietf-behave-nat-icmp-05.txt";
	site="ftp.ietf.org"; access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID: <2007-09-26121300.I-D\@ietf.org>


--OtherAccess--

--NextPart
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

--NextPart--



From behave-bounces@ietf.org Wed Sep 26 12:21:07 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IaZcP-0007Mk-T1; Wed, 26 Sep 2007 12:20:09 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IaZcL-00073F-HI
	for behave-confirm+ok@megatron.ietf.org; Wed, 26 Sep 2007 12:20:05 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IaZcJ-0006vT-5e; Wed, 26 Sep 2007 12:20:04 -0400
Received: from ns0.neustar.com ([156.154.16.158])
	by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IaZcI-0007A7-1s; Wed, 26 Sep 2007 12:20:03 -0400
Received: from stiedprstage1.ietf.org (stiedprstage1.va.neustar.com
	[10.31.47.10]) by ns0.neustar.com (Postfix) with ESMTP id 065903288F;
	Wed, 26 Sep 2007 16:20:02 +0000 (GMT)
Received: from ietf by stiedprstage1.ietf.org with local (Exim 4.43)
	id 1IaZcH-000852-Tv; Wed, 26 Sep 2007 12:20:01 -0400
Content-Type: Multipart/Mixed; Boundary="NextPart"
Mime-Version: 1.0
To: i-d-announce@ietf.org
From: Internet-Drafts@ietf.org
Message-Id: <E1IaZcH-000852-Tv@stiedprstage1.ietf.org>
Date: Wed, 26 Sep 2007 12:20:01 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 3002fc2e661cd7f114cb6bae92fe88f1
Cc: behave@ietf.org
Subject: [BEHAVE] I-D Action:draft-ietf-behave-p2p-state-04.txt 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Behavior Engineering for Hindrance Avoidance Working Group of the IETF.


	Title           : State of Peer-to-Peer(P2P) Communication Across Network Address Translators(NATs)
	Author(s)       : P. Srisuresh, et al.
	Filename        : draft-ietf-behave-p2p-state-04.txt
	Pages           : 32
	Date            : 2007-09-26

This memo documents the various methods known to be in use by
applications to establish direct communication in the presence
of Network Address Translators (NATs) at the current time. This
memo covers NAT traversal approaches used by both TCP and UDP
based applications. This memo is not an endorsement of the methods 
described, but merely an attempt to capture them in a document.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-behave-p2p-state-04.txt

To remove yourself from the I-D Announcement list, send a message to
i-d-announce-request@ietf.org with the word unsubscribe in the body of 
the message.
You can also visit https://www1.ietf.org/mailman/listinfo/I-D-announce
to change your subscription settings.

Internet-Drafts are also available by anonymous FTP. Login with the 
username "anonymous" and a password of your e-mail address. After 
logging in, type "cd internet-drafts" and then
	"get draft-ietf-behave-p2p-state-04.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt

Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-behave-p2p-state-04.txt".

NOTE:   The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.

Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body; access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID: <2007-09-26121204.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-behave-p2p-state-04.txt

--OtherAccess
Content-Type: Message/External-body; name="draft-ietf-behave-p2p-state-04.txt";
	site="ftp.ietf.org"; access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID: <2007-09-26121204.I-D\@ietf.org>


--OtherAccess--

--NextPart
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

--NextPart--







d unsubscribe in the body of 
the message.
You can also visit https://www1.ietf.org/mailman/listinfo/I-D-announce
to change your subscription settings.

Internet-Drafts are also available by anonymous FTP. Login with the 
username "anonymous" and a password of your e-mail address. After 
logging in, type "cd internet-drafts" and then
	"get draft-ietf-behave-p2p-state-04.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt

Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-behave-p2p-state-04.txt".

NOTE:   The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.

Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body; access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID: <2007-09-26121204.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-behave-p2p-state-04.txt

--OtherAccess
Content-Type: Message/External-body; name="draft-ietf-behave-p2p-state-04.txt";
	site="ftp.ietf.org"; access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID: <2007-09-26121204.I-D\@ietf.org>


--OtherAccess--

--NextPart
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave

--NextPart--







From behave-bounces@ietf.org Wed Sep 26 16:15:02 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IadGz-0008VS-Kp; Wed, 26 Sep 2007 16:14:17 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IadGx-0008EV-DL
	for behave-confirm+ok@megatron.ietf.org; Wed, 26 Sep 2007 16:14:15 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IadGx-0008Do-3I
	for behave@ietf.org; Wed, 26 Sep 2007 16:14:15 -0400
Received: from nz-out-0506.google.com ([64.233.162.228])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IadGs-0007K9-0I
	for behave@ietf.org; Wed, 26 Sep 2007 16:14:15 -0400
Received: by nz-out-0506.google.com with SMTP id n1so1509688nzf
	for <behave@ietf.org>; Wed, 26 Sep 2007 13:13:47 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=beta;
	h=domainkey-signature:received:received:message-id:date:from:sender:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references:x-google-sender-auth;
	bh=N1Y5qfbWpedmmEO+6TzToSFLQcMXlt/3lr4RzYWtzuA=;
	b=lozTV+HL/CIvvYXJT7oa/FjCiCC4Bwhyl9A4Y0+waaYs9vo5/OGuqQnLg6fXn0O5PNIhk8MFwfyD3sdnKpycFBaKVLB7e9ZdlN+JMESRok51u+7qwK4cRfr7dfZ8vyrf+tkRodywjoxRu2bcs5oQMDoBeKlia84oFXoEwiIhRHA=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta;
	h=received:message-id:date:from:sender:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references:x-google-sender-auth;
	b=nRPniyFMt13HBkdvdx1rs1zt82EgcFya6ubDkILPoeowD6LPzseuIZXp5tz1o80Lf6hiE0B2sE9es8wvyGln/P0LL1wxm4mFao44IO3XKVmS13XHgkj46Rd9Vql/3Q9nfkW+Nr0P+BVeQbaJCiTTYwdaJ+hRYOrmtRnHAvmr0NY=
Received: by 10.115.107.1 with SMTP id j1mr1557507wam.1190837626885;
	Wed, 26 Sep 2007 13:13:46 -0700 (PDT)
Received: by 10.142.52.2 with HTTP; Wed, 26 Sep 2007 13:13:46 -0700 (PDT)
Message-ID: <20d2bdfb0709261313p50f4dfd9m649af68f0a259e2d@mail.gmail.com>
Date: Wed, 26 Sep 2007 16:13:46 -0400
From: "Bruce Lowekamp" <lowekamp@sipeerior.com>
To: "Cullen Jennings" <fluffy@cisco.com>
Subject: Re: [BEHAVE] behavior discovery category
In-Reply-To: <72D21363-52D8-4F3B-B02C-1F2730547F27@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
References: <20d2bdfb0709122004o709211bbt9569b0f1d3ae1120@mail.gmail.com>
	<04f401c7f984$0d6bdb00$c6f0200a@cisco.com>
	<72D21363-52D8-4F3B-B02C-1F2730547F27@cisco.com>
X-Google-Sender-Auth: 2285ac63089911b1
X-Spam-Score: 0.0 (/)
X-Scan-Signature: e8a67952aa972b528dd04570d58ad8fe
Cc: Behave WG <behave@ietf.org>, Dan Wing <dwing@cisco.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

Cullen,

I think the argument of "this didn't solve the problem before" is a
little bit off target.  The original focus of 3489 was static
type-discovery, with the goal of obtaining information that could be
used for permanent configuration of a device.  We know that doesn't
work.

The focus of the behavior-discovery draft is explicitly on dynamic
behavior observation.    The experimental question is whether the
basic techniques that originally came from 3489 are useful in contexts
other than the original.  The current draft proposes two specific
uses.

- the first is just diagnostics for administrators. I don't think
there should be any dispute over whether this is useful.  I would also
argue that this is actually a PS type of use by itself.
- the second is the P2P example.  In this case, behavior discovery is
proposed to be used for initial guidance on how a new potential
supernode might perform.  The peer (and rest of the overlay) is
assumed to support ICE for connections regardless, but many overlays
would not want a node that requires ICE (and god-forbid relaying) to
act as a supernode.  So if behavior discovery reports an overlay
restrictive NAT behavior, the node won't try to become a supernode.
Otherwise, it might try and then if it discovered through usage that
there is too much overhead in contacting it, would withdraw as a
supernode.

The experimental question in the second case is whether such a test is
useful.  If becoming a supernode when you shouldn't and then
withdrawing is expensive or leads to unreliable or poorly performing
operation, then even if the behavior discovery check is only "correct"
75% of the time, its relative cheapness may make it very useful for
the application.

This is very different than the original use of 3489, because it's an
explicit statement that it may not be right 100% of the time, but is
still very useful.  In my opinion, making this available for new
protocols as an experimental option would be very useful.

Bruce



On 9/23/07, Cullen Jennings <fluffy@cisco.com> wrote:
>
> One of the key reasons that we needed to do a bis for RFC 3489  is
> because we have run the experiment to find out if nodes can reliably
> detect what type of NAT they are behind using a STUN like techniques.
> We know the answer - they can not and this is not a rare case but
> happens surprisingly often. This resulted in the change away from the
> advice in 3489 to the advice in bis (not to mention the move from
> unilateral NAT traversal techniques to bilateral techniques). There
> are test results document in draft-jennings-behave-test-results.
> These results are for specific NATs but we know from other deployment
> experience, and from other test results such as Saikat's and Ford's,
> that we wee similar issues in deployments.
>
> I agree it would be nice if a client could detect what type of NAT it
> was behind using STUN, however, I believe the only experiment being
> proposed here we already clearly know the answer to.
>
> Cullen <with my individual hat on>


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 27 09:34:57 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IatVn-0001CL-U9; Thu, 27 Sep 2007 09:34:39 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IatVn-0001CF-44
	for behave-confirm+ok@megatron.ietf.org; Thu, 27 Sep 2007 09:34:39 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IatVm-0001C7-HN
	for behave@ietf.org; Thu, 27 Sep 2007 09:34:38 -0400
Received: from rtp-iport-1.cisco.com ([64.102.122.148])
	by ietf-mx.ietf.org with esmtp (Exim 4.43) id 1IatVg-0006G6-DC
	for behave@ietf.org; Thu, 27 Sep 2007 09:34:38 -0400
X-IronPort-AV: E=Sophos;i="4.21,203,1188792000"; d="scan'208";a="72276321"
Received: from rtp-dkim-1.cisco.com ([64.102.121.158])
	by rtp-iport-1.cisco.com with ESMTP; 27 Sep 2007 09:34:17 -0400
Received: from rtp-core-2.cisco.com (rtp-core-2.cisco.com [64.102.124.13])
	by rtp-dkim-1.cisco.com (8.12.11/8.12.11) with ESMTP id l8RDYHIW010188; 
	Thu, 27 Sep 2007 09:34:17 -0400
Received: from [172.17.114.244] (rtp-vpn3-700.cisco.com [10.82.218.191])
	by rtp-core-2.cisco.com (8.12.10/8.12.6) with ESMTP id l8RDYGVD019359; 
	Thu, 27 Sep 2007 13:34:16 GMT
In-Reply-To: <20d2bdfb0709261313p50f4dfd9m649af68f0a259e2d@mail.gmail.com>
References: <20d2bdfb0709122004o709211bbt9569b0f1d3ae1120@mail.gmail.com>
	<04f401c7f984$0d6bdb00$c6f0200a@cisco.com>
	<72D21363-52D8-4F3B-B02C-1F2730547F27@cisco.com>
	<20d2bdfb0709261313p50f4dfd9m649af68f0a259e2d@mail.gmail.com>
Mime-Version: 1.0 (Apple Message framework v752.3)
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <BA51DE3B-7F82-4BC2-9D07-0D2FD2051C55@cisco.com>
Content-Transfer-Encoding: 7bit
From: Cullen Jennings <fluffy@cisco.com>
Subject: Re: [BEHAVE] behavior discovery category
Date: Thu, 27 Sep 2007 09:34:09 -0400
To: Bruce Lowekamp <lowekamp@sipeerior.com>
X-Mailer: Apple Mail (2.752.3)
DKIM-Signature: v=0.5; a=rsa-sha256; q=dns/txt; l=2581; t=1190900057;
	x=1191764057; c=relaxed/simple; s=rtpdkim1001;
	h=Content-Type:From:Subject:Content-Transfer-Encoding:MIME-Version;
	d=cisco.com; i=fluffy@cisco.com;
	z=From:=20Cullen=20Jennings=20<fluffy@cisco.com>
	|Subject:=20Re=3A=20[BEHAVE]=20behavior=20discovery=20category
	|Sender:=20 |To:=20Bruce=20Lowekamp=20<lowekamp@sipeerior.com>;
	bh=YAK9CEWBzrbt9PUU43O2ZV51ypA2/Ye+9uUDuRUvLeY=;
	b=Iy3+E69UGTpYH1PWGLuUagf0Kh9qRmbfzPIYn5pLbQbkNFBuEu4bT7+qjRSv1E3g4DzEc9uD
	0KWRR/gjrsFbenMy4v44Gu5E/rRSojiMZJD2Txp/c3uNf++f8x740eRo;
Authentication-Results: rtp-dkim-1; header.From=fluffy@cisco.com; dkim=pass (
	sig from cisco.com/rtpdkim1001 verified; ); 
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 4d87d2aa806f79fed918a62e834505ca
Cc: Behave WG <behave@ietf.org>, "Dan Wing \(dwing\)" <dwing@cisco.com>
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org


I think we have all converged on experimental but I'm going to reply  
to bit of stuff bellow ... only because I think it might impact some  
of thinking on what goes in this draft and on how it is used in p2p.

On Sep 26, 2007, at 4:13 PM, Bruce Lowekamp wrote:

> - the second is the P2P example.  In this case, behavior discovery is
> proposed to be used for initial guidance on how a new potential
> supernode might perform.  The peer (and rest of the overlay) is
> assumed to support ICE for connections regardless, but many overlays
> would not want a node that requires ICE (and god-forbid relaying) to
> act as a supernode.

small nit here - uh - I suspect that it is going to require ICE even  
if node has public IP. But I get what you are saying about wanting a  
way to detect if a relay is needed - I just would not phrase this as  
requires ICE

>   So if behavior discovery reports an overlay
> restrictive NAT behavior, the node won't try to become a supernode.
> Otherwise, it might try and then if it discovered through usage that
> there is too much overhead in contacting it, would withdraw as a
> supernode.

>
> The experimental question in the second case is whether such a test is
> useful.  If becoming a supernode when you shouldn't and then
> withdrawing is expensive or leads to unreliable or poorly performing
> operation, then even if the behavior discovery check is only "correct"
> 75% of the time, its relative cheapness may make it very useful for
> the application.

One thing we do know about testing NATs like this is that to run any  
sort of test that produces useful results, you need the client behind  
the NAT to have at least two and preferably more IP addresses. I am  
assuming that this nat test document is going to make that very  
clear. I suspect this will not be easy for a P2P node to do.

The second thing is that I think we are solving the goal the wrong  
way here. In the supernode case you are thinking of, what we want to  
know is that we won't need relays (or some other characteristic we  
deem desirable). The way to do this is let the node form a few  
connections and keep track of how many of them needed relays. If the  
number is good (like 0) then let it become a supernode and if the  
number is bad don't. I think this type of design will work out far  
better than any sort of STUN testing.

Culen <with my individual hat on>

PS - I hate all of the many ways IETF people are using the word  
"supernode" but fully understand what you mean by it above :-)


_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 27 10:21:47 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IauEy-0000MH-Oy; Thu, 27 Sep 2007 10:21:20 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IauEx-0000Jd-De
	for behave-confirm+ok@megatron.ietf.org; Thu, 27 Sep 2007 10:21:19 -0400
Received: from [10.91.34.44] (helo=ietf-mx.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IauEx-0000JV-4B; Thu, 27 Sep 2007 10:21:19 -0400
Received: from ns0.neustar.com ([156.154.16.158])
	by ietf-mx.ietf.org with esmtp (Exim 4.43)
	id 1IauEw-0007oc-0b; Thu, 27 Sep 2007 10:21:19 -0400
Received: from stiedprstage1.ietf.org (stiedprstage1.va.neustar.com
	[10.31.47.10]) by ns0.neustar.com (Postfix) with ESMTP id F21E13287E;
	Thu, 27 Sep 2007 14:20:47 +0000 (GMT)
Received: from ietf by stiedprstage1.ietf.org with local (Exim 4.43)
	id 1IauER-0006qv-T3; Thu, 27 Sep 2007 10:20:47 -0400
X-test-idtracker: no
To: IETF-Announce <ietf-announce@ietf.org>
From: The IESG <iesg-secretary@ietf.org>
Message-Id: <E1IauER-0006qv-T3@stiedprstage1.ietf.org>
Date: Thu, 27 Sep 2007 10:20:47 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 7655788c23eb79e336f5f8ba8bce7906
Cc: behave@ietf.org
Subject: [BEHAVE] Last Call: draft-ietf-behave-nat-icmp (NAT Behavioral 
 Requirements for ICMP protocol) to BCP 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: ietf@ietf.org
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

The IESG has received a request from the Behavior Engineering for 
Hindrance Avoidance WG (behave) to consider the following document:

- 'NAT Behavioral Requirements for ICMP protocol '
   <draft-ietf-behave-nat-icmp-05.txt> as a BCP

The IESG plans to make a decision in the next few weeks, and solicits
final comments on this action.  Please send substantive comments to the
ietf@ietf.org mailing lists by 2007-10-11. Exceptionally, 
comments may be sent to iesg@ietf.org instead. In either case, please 
retain the beginning of the Subject line to allow automated sorting.

This document contains a downward normative reference to the
informational RFC 3022 "Traditional IP Network Address Translator
(Traditional NAT)". 

The file can be obtained via
http://www.ietf.org/internet-drafts/draft-ietf-behave-nat-icmp-05.txt


IESG discussion can be tracked via
https://datatracker.ietf.org/public/pidtracker.cgi?command=view_id&dTag=14702&rfc_flag=0



_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 27 10:22:11 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IauFj-00012H-BP; Thu, 27 Sep 2007 10:22:07 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IauFf-0000zO-7q
	for behave-confirm+ok@megatron.ietf.org; Thu, 27 Sep 2007 10:22:03 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1IauFe-0000y7-13; Thu, 27 Sep 2007 10:22:02 -0400
Received: from ns3.neustar.com ([156.154.24.138])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43)
	id 1IauFd-0002MI-PZ; Thu, 27 Sep 2007 10:22:01 -0400
Received: from stiedprstage1.ietf.org (stiedprstage1.va.neustar.com
	[10.31.47.10]) by ns3.neustar.com (Postfix) with ESMTP id 75CAB175D3;
	Thu, 27 Sep 2007 14:21:26 +0000 (GMT)
Received: from ietf by stiedprstage1.ietf.org with local (Exim 4.43)
	id 1IauF4-0006sb-0T; Thu, 27 Sep 2007 10:21:26 -0400
X-test-idtracker: no
To: IETF-Announce <ietf-announce@ietf.org>
From: The IESG <iesg-secretary@ietf.org>
Message-Id: <E1IauF4-0006sb-0T@stiedprstage1.ietf.org>
Date: Thu, 27 Sep 2007 10:21:26 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 79899194edc4f33a41f49410777972f8
Cc: behave@ietf.org
Subject: [BEHAVE] Last Call: draft-ietf-behave-p2p-state (State of 
 Peer-to-Peer(P2P) Communication Across Network Address 
 Translators(NATs)) to Informational RFC 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: ietf@ietf.org
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org

The IESG has received a request from the Behavior Engineering for 
Hindrance Avoidance WG (behave) to consider the following document:

- 'State of Peer-to-Peer(P2P) Communication Across Network Address 
   Translators(NATs) '
   <draft-ietf-behave-p2p-state-04.txt> as an Informational RFC

The IESG plans to make a decision in the next few weeks, and solicits
final comments on this action.  Please send substantive comments to the
ietf@ietf.org mailing lists by 2007-10-11. Exceptionally, 
comments may be sent to iesg@ietf.org instead. In either case, please 
retain the beginning of the Subject line to allow automated sorting.

The file can be obtained via
http://www.ietf.org/internet-drafts/draft-ietf-behave-p2p-state-04.txt


IESG discussion can be tracked via
https://datatracker.ietf.org/public/pidtracker.cgi?command=view_id&dTag=15405&rfc_flag=0



_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 27 12:15:29 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1Iaw0i-0007Uq-5s; Thu, 27 Sep 2007 12:14:44 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IaZW6-00073x-Ki
	for behave-confirm+ok@megatron.ietf.org; Wed, 26 Sep 2007 12:13:38 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IaZW6-00073p-A5
	for behave@ietf.org; Wed, 26 Sep 2007 12:13:38 -0400
Received: from ns3.neustar.com ([156.154.24.138])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IaZW6-00046W-2u
	for behave@ietf.org; Wed, 26 Sep 2007 12:13:38 -0400
Received: from ietf.org (stiedprweb1.va.neustar.com [10.91.34.42])
	by ns3.neustar.com (Postfix) with ESMTP id A9BE5175C8;
	Wed, 26 Sep 2007 16:13:02 +0000 (GMT)
Received: from mirror by ietf.org with local (Exim 4.43)
	id 1IaZVW-00040w-6J; Wed, 26 Sep 2007 12:13:02 -0400
Content-Type: text/plain; charset="utf-8"
Mime-Version: 1.0
To: srisuresh@yahoo.com
From: IETF I-D Submission Tool <developers@ietf.org>
Message-Id: <E1IaZVW-00040w-6J@ietf.org>
Date: Wed, 26 Sep 2007 12:13:02 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 7655788c23eb79e336f5f8ba8bce7906
X-Mailman-Approved-At: Thu, 27 Sep 2007 12:14:42 -0400
Cc: baford@mit.edu, behave@ietf.org
Subject: [BEHAVE] New Version Notification for draft-ietf-behave-nat-icmp-05
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org


A new version of I-D, draft-ietf-behave-nat-icmp-05.txt has been successfuly submitted by Pyda Srisuresh and posted to the IETF repository.

Filename:	 draft-ietf-behave-nat-icmp
Revision:	 05
Title:		 NAT Behavioral Requirements for ICMP protocol
Creation_date:	 2007-09-26
WG ID:		 behave
Number_of_pages: 24

Abstract:
This document specifies the behavioral properties required of the 
Network Address Translator (NAT) devices in conjunction with the
ICMP protocol. The objective of this memo is to make NAT devices
more predictable and compatible with diverse application protocols
that traverse the devices. Companion documents provide behavioral
recommendations specific to TCP, UDP and other protocols.
                                                                                  


The IETF Secretariat.




_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



From behave-bounces@ietf.org Thu Sep 27 12:15:29 2007
Return-path: <behave-bounces@ietf.org>
Received: from [127.0.0.1] (helo=stiedprmman1.va.neustar.com)
	by megatron.ietf.org with esmtp (Exim 4.43)
	id 1Iaw0h-0007Ug-RK; Thu, 27 Sep 2007 12:14:44 -0400
Received: from behave by megatron.ietf.org with local (Exim 4.43)
	id 1IaZV7-0004yS-Qr
	for behave-confirm+ok@megatron.ietf.org; Wed, 26 Sep 2007 12:12:37 -0400
Received: from [10.90.34.44] (helo=chiedprmail1.ietf.org)
	by megatron.ietf.org with esmtp (Exim 4.43) id 1IaZV7-0004yK-Gi
	for behave@ietf.org; Wed, 26 Sep 2007 12:12:37 -0400
Received: from ns4.neustar.com ([156.154.24.139])
	by chiedprmail1.ietf.org with esmtp (Exim 4.43) id 1IaZV7-0003zq-9N
	for behave@ietf.org; Wed, 26 Sep 2007 12:12:37 -0400
Received: from ietf.org (stiedprweb1.va.neustar.com [10.91.34.42])
	by ns4.neustar.com (Postfix) with ESMTP id 3BD962AC63;
	Wed, 26 Sep 2007 16:12:07 +0000 (GMT)
Received: from mirror by ietf.org with local (Exim 4.43)
	id 1IaZUd-00040h-0Q; Wed, 26 Sep 2007 12:12:07 -0400
Content-Type: text/plain; charset="utf-8"
Mime-Version: 1.0
To: srisuresh@yahoo.com
From: IETF I-D Submission Tool <developers@ietf.org>
Message-Id: <E1IaZUd-00040h-0Q@ietf.org>
Date: Wed, 26 Sep 2007 12:12:07 -0400
X-Spam-Score: 0.0 (/)
X-Scan-Signature: 7655788c23eb79e336f5f8ba8bce7906
X-Mailman-Approved-At: Thu, 27 Sep 2007 12:14:42 -0400
Cc: baford@mit.edu, dank06@kegel.com, behave@ietf.org
Subject: [BEHAVE] New Version Notification for
	draft-ietf-behave-p2p-state-04 
X-BeenThere: behave@ietf.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: mailing list of BEHAVE IETF WG <behave.ietf.org>
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www1.ietf.org/pipermail/behave>
List-Post: <mailto:behave@ietf.org>
List-Help: <mailto:behave-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/behave>,
	<mailto:behave-request@ietf.org?subject=subscribe>
Errors-To: behave-bounces@ietf.org


A new version of I-D, draft-ietf-behave-p2p-state-04.txt has been successfuly submitted by Pyda Srisuresh and posted to the IETF repository.

Filename:	 draft-ietf-behave-p2p-state
Revision:	 04
Title:		 State of Peer-to-Peer(P2P) Communication Across Network Address Translators(NATs)
Creation_date:	 2007-09-26
WG ID:		 behave
Number_of_pages: 32

Abstract:
This memo documents the various methods known to be in use by
applications to establish direct communication in the presence
of Network Address Translators (NATs) at the current time. This
memo covers NAT traversal approaches used by both TCP and UDP
based applications. This memo is not an endorsement of the methods 
described, but merely an attempt to capture them in a document.
                                                                                  


The IETF Secretariat.




_______________________________________________
Behave mailing list
Behave@ietf.org
https://www1.ietf.org/mailman/listinfo/behave



