Re: [rtcweb] [MMUSIC] Is bundle just a port override?

["Mo Zanaty (mzanaty)" <[e-mail]>]

Hi Dale,

In your draft (and Richard's), I understand your motivation to answer with port 0 (or 9 or IP 0), to plug the nostrils of intermediaries that sniff ports so they don't smell anything fishy. I won't comment on whether they will react better to nostril-plugging than fishy smells, because I don't know, but that is orthogonal to what I'm describing here.

Your bundle alternatives (and Richard's) also work fine with a simple attribute without complex grouping semantics.

Offer to mux audio and video on the same port:
m=audio 10000 RTP/AVP 0
m=video 10002 RTP/AVP 31
a=port 10000
i=I really want 10000, but lied on the m-line for fear of confusing you. Confused yet?
 
Answer supports the port override attribute and agrees to mux audio and video:
m=audio 40000 RTP/AVP 0
m=video 0 RTP/AVP 31         <-- Dale's variant to answer with port 0
a=port 40000
...so audio and video ports are 10000<->40000.
 
Answer doesn't support the port override attribute, or doesn't want to mux:
m=audio 40000 RTP/AVP 0
m=video 40002 RTP/AVP 31
...so audio ports are 10000<->40000 and video ports are 10002<->40002.

Cheers,
Mo


-----Original Message-----
From: Dale R. Worley [mailto:[e-mail]] 
Sent: Tuesday, March 19, 2013 3:37 PM
To: Mo Zanaty (mzanaty)
Cc: [e-mail]; [e-mail]; [e-mail]
Subject: Re: [rtcweb] [MMUSIC] Is bundle just a port override?

> From: "Mo Zanaty (mzanaty)" <[e-mail]>
> 
> Those drafts use the SDP grouping framework
> (a=group:BUNDLE/TOGETHER), which I think is unnecessary complexity
> and ambiguity for something which can be solved in a much simpler
> and clearer way with a single attribute: a=port <port>. (Or, if we
> want to warn humans about RTP muxing, then name it a=rtp-mux:<port>,
> but machines won't care about the name, and their RTP
> implementations already know how to deal with muxing if they use
> this attribute, so stronger warnings are unnecessary.)
> 
> Regarding the concern about non-use of the discarded RTP ports
> confusing SBCs, see this part of my original message [with
> clarifications], which is what the current bundle draft also
> suggests:
> 
> > Either end can re-offer without lies about ports on the m-line after
> > confirming the peer is not confused by muxing, if they want to avoid
> > confusing intermediaries [that may monitor the unused ports].

It's messier than that, unfortunately.

One requirement is that after negotiation is completed, the negotiated
transport associations (as seen by a legacy intermediary) must match
the actual media flows.

Another requirement is that two media descriptions shouldn't use the
same port (because a lot of legacy SBCs can't handle that).  This
applies to offer/answer updates as well as original offers/answers.

So what we want is a way for a media description to say, "I'm
reporting port zero so SBCs think this MD is rejected, but I'm
actually using port 1234 (which is officially reported in another
MD)."

There are also situations where the we want to signal perferences as
to which media descriptions share ports with which other media
descriptions.

Dale