Re: [MMUSIC] Some issues with draft-ietf-mmusic-sctp-sdp-02

Paul Kyzivat <pkyzivat@alum.mit.edu> Fri, 11 January 2013 18:05 UTC

Return-Path: <pkyzivat@alum.mit.edu>
X-Original-To: mmusic@ietfa.amsl.com
Delivered-To: mmusic@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 000D721F89AA for <mmusic@ietfa.amsl.com>; Fri, 11 Jan 2013 10:05:27 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 0.186
X-Spam-Level:
X-Spam-Status: No, score=0.186 tagged_above=-999 required=5 tests=[AWL=0.023, BAYES_00=-2.599, FH_RELAY_NODNS=1.451, HELO_MISMATCH_NET=0.611, J_CHICKENPOX_111=0.6, RDNS_NONE=0.1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id diEUrkDp4xSz for <mmusic@ietfa.amsl.com>; Fri, 11 Jan 2013 10:05:27 -0800 (PST)
Received: from qmta09.westchester.pa.mail.comcast.net (qmta09.westchester.pa.mail.comcast.net [IPv6:2001:558:fe14:43:76:96:62:96]) by ietfa.amsl.com (Postfix) with ESMTP id 7EB2121F892A for <mmusic@ietf.org>; Fri, 11 Jan 2013 10:05:26 -0800 (PST)
Received: from omta09.westchester.pa.mail.comcast.net ([76.96.62.20]) by qmta09.westchester.pa.mail.comcast.net with comcast id mdQQ1k0020SCNGk59i5Rdk; Fri, 11 Jan 2013 18:05:25 +0000
Received: from Paul-Kyzivats-MacBook-Pro.local ([50.138.229.164]) by omta09.westchester.pa.mail.comcast.net with comcast id mi5R1k00R3ZTu2S3Vi5Rhg; Fri, 11 Jan 2013 18:05:25 +0000
Message-ID: <50F05464.8040202@alum.mit.edu>
Date: Fri, 11 Jan 2013 13:05:24 -0500
From: Paul Kyzivat <pkyzivat@alum.mit.edu>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:17.0) Gecko/20130107 Thunderbird/17.0.2
MIME-Version: 1.0
To: "Salvatore Loreto (JO/LMF)" <salvatore.loreto@ericsson.com>, Gonzalo Camarillo <Gonzalo.Camarillo@ericsson.com>
References: <50BD0BA8.3010901@alum.mit.edu>
In-Reply-To: <50BD0BA8.3010901@alum.mit.edu>
Content-Type: text/plain; charset="ISO-8859-1"; format="flowed"
Content-Transfer-Encoding: 7bit
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=comcast.net; s=q20121106; t=1357927525; bh=PVJBtnCYGqcEGQ7YsXY7FUgKYgKigUt9t5adP9z2jnw=; h=Received:Received:Message-ID:Date:From:MIME-Version:To:Subject: Content-Type; b=hrA5i5bCc+l1TbQCJ6vGpHidPGvPawkBuevZ8/duwIO6SId4DBk/TpK3UDFKKooSN eQYtt2DLl3PLJLK/xLYmZ2rPVr9UROhOnRWyNg3wzfIhSjqofw/zZG7UDCRx/AjzCA rk59i7mRgZpMLNztiWIjmivXnLacWM7LCEAyY9Rb6EKX8rGFbw3w7UvZ41iKnE2/v7 m/lPvnPoVyXZoHKnr62w+Z4+jwkdbebhgGZXuhYLkF+5O201gFzcW6tehDtew/hBP9 XCX1XeVEkav0CpexYnjL9/OFytwqM3331U+tUHkgqOy2sm9uwgRvAazTIJjkHFFWC5 e38xzqPbbV9gQ==
Cc: IETF MMUSIC WG <mmusic@ietf.org>
Subject: Re: [MMUSIC] Some issues with draft-ietf-mmusic-sctp-sdp-02
X-BeenThere: mmusic@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Multiparty Multimedia Session Control Working Group <mmusic.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mmusic>, <mailto:mmusic-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mmusic>
List-Post: <mailto:mmusic@ietf.org>
List-Help: <mailto:mmusic-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mmusic>, <mailto:mmusic-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 11 Jan 2013 18:05:32 -0000

Replying to myself, since nobody else did.

There has been more discussion recently (subject: SCTP question: Where 
does it multiplex?) on the same issue I raise below.

More inline.

On 12/3/12 3:29 PM, Paul Kyzivat wrote:
> While commenting on some related discussions I had occasion to look at
> this draft again and I found a few issues:
>
> First, some very specific ones::
>
> Section 4.1 says:
>
>     If the <port> sub-field is 'SCTP' or ...
>
> This is a typo. s/port/proto/
>
> But the major issue is that this draft doesn't explain the semantics of
> the <port> field. The only things I can find that touch on this are:
>
> - section 10.1 refers to "SCTP port 54111" and then shows an m-line
>    with: m=application 54111 SCTP 0 1 2
>    But this is just an example, not normative text.
>
> - section 4.1 says:
>
>     [OPEN ISSUE 3] do we need also an attribute to specify the SCTP port
>     number for the SCTP over DTLS scenario? that can be useful in the
>     case where multiple association are running on top of the same DTLS
>     session.
>
>    This implies that the SCTP port number is *not* what is in the port
>    field in this case. Presumably because the UDP port number is in
>    that field.
>
> ISTM that there needs to be a clear discussion of all the required port
> numbers for each case, and how they are specified. When there is more
> than one (namely in the SCTP/DTLS and DTLS/SCTP cases), *one* of them
> can be in the m-line and the other one must be somewhere else. And if
> the one that is not in the m-line is optional, then I guess there must
> be a default.

For *native* SCTP (directly over IPv* - <proto>=SCTP) it seems obvious 
that the port in the m-line should be the SCTP port.

For <proto> of SCTP/DTLS (RFC6083) there is again only the SCTP port to 
consider, so it should presumably go in the m-line. (My earlier comment 
about this one was wrong.)

For <proto> of DTLS/SCTP (draft-tuexen-sctp-udp-encaps) we need both the 
SCTP port and the UDP port, so we must pick which one goes in the 
m-line. What arguments are there for which should go in the m-line?

- Consistency with the other SCTP protos argues for the SCTP port.
- The port in the m-line is mandatory. AFAIK there is no default
   SCTP port. But there *is* a default UDP port for DTLS/SCTP
   (9899, defined in draft-tuexen-sctp-udp-encaps). This is also
   an argument for the SCTP port in the m-line.
- are there any arguments in favor of the UDP port in the m-line?
   (I haven't thought of any.)

Absent some arguments on the other side, I propose that this document 
specify that the <port> field of the m-line denote the SCTP port for all 
of the SCTP <proto>s, and specify an attribute to carry the UDP port, 
with a default of 9899.

	Thanks,
	Paul