Re: [MMUSIC] Question about RTSP proxy

["Y.J. Gu" <guyingjie at huawei.com>]

Hi Jamie, see in line please.

Regards

Yingjie Gu

 

> -----Original Message-----
> From: Jamie Gordon [mailto:jgordon at real.com] 
> Sent: Thursday, August 13, 2009 2:37 AM
> To: Y.J. Gu
> Cc: mmusic at ietf.org
> Subject: Re: [MMUSIC] Question about RTSP proxy
> 
> On 8/10/2009 11:04 PM, Y.J. Gu wrote:
> > Hi all,
> > In RTSP 2.0, it is permitted to transmit several client requests in 
> > one Transport Connection.
> > I think this may introduce new problems when RTSP proxy is 
> intervened, 
> > and existing RTSP can not deal with the scenario correctly.
> > The following is an example. To display the figure 
> correctly, please 
> > maximize this email window.
> 
> This has not changed from RTSP 1.0, and even some existing 
> clients will do similar, for instance when streaming multiple 
> separate contents which it is syncing via smil. The RTSP 
> sequence number applies to the RTSP session on which it is 
> carried. A proxy must handle any associated sequence number 
> mapping between client connections and server connections.
> 
> So for your example:
> 
> >  
> >    The interactive messages are listed as following.
> >  
> >    Session issued by Client A:
> >    C->P: SETUP rtsp://live.example.com/concert/audio RTSP/2.0
> >          CSeq: 111
> >          Transport: RTP/AVP;multicast
> >  
> >    P->S: SETUP rtsp://live.example.com/concert/audio RTSP/2.0
> >          CSeq: 333
> >          Transport: RTP/AVP;multicast
> >    Session issued by Client A:
> >    C->P: SETUP rtsp://live.example.com/anotherconcert/audio RTSP/2.0
> >          CSeq: 222
> >          Transport: RTP/AVP;multicast
> >    P->S: SETUP rtsp://live.example.com/anotherconcert/audio RTSP/2.0
> >          CSeq: 444
> >          Transport: RTP/AVP;multicast
> >  
> > 
> >    Proxy will receive the following responses from server:
> >    S->P: RTSP/2.0 200 OK
> >          CSeq: 333
> >          Server: PhonyServer/1.0
> >          Date: Thu, 23 Jan 2009 15:35:06 GMT
> >          Transport: RTP/AVP;multicast;
> >               dest_addr="224.2.0.1:3456"/"224.2.0.1:3457";ttl=16
> >          Session: 12345678
> >    S->P: RTSP/2.0 200 OK
> >          CSeq: 444
> >          Server: PhonyServer/1.0
> >          Date: Thu, 23 Jan 2009 15:35:06 GMT
> >          Transport: RTP/AVP;multicast;
> >               dest_addr="224.2.0.2:1234"/"224.2.0.2:1235";ttl=16
> >          Session: 23456789
> >  
> 
> This is not correct. If the proxy is using the same RTSP 
> proxy<->server connection for multiple client<->proxy 
> connections, it must resequence and internally track the 
> associations. It can't just blindly pass through the sequence 
> numbers from it's client connections, those are clearly not 
> correct. So if Client A and Client B send
> 
> CA -> P:
> SETUP rtsp://live.example.com/concert/audio RTSP/2.0
> CSeq: 111
> Transport: RTP/AVP;multicast
> 
> CB->P:
> SETUP rtsp://live.example.com/anotherconcert/audio RTSP/2.0
> CSeq: 222
> Transport: RTP/AVP;multicast
> 
> The proxy must then send to the server
> 
> P->S:
> SETUP rtsp://live.example.com/concert/audio RTSP/2.0
> CSeq: 1
> Transport: RTP/AVP;multicast
> 
> SETUP rtsp://live.example.com/anotherconcert/audio RTSP/2.0
> CSeq: 2
> Transport: RTP/AVP;multicast

Welll, I do resequence the client CSeq to proxy CSeq, 
For client A, CSeq is from 111 to 333
For client B, CSeq is from 222 to 444
And yes of course the proxy will store the mapping as you said. 

> and store the mappings [Client A CSeq 111 -> Server CSeq 1] 
> and [Client B CSeq 222 -> Server CSeq 2]
> 
> and when it receives the responses it certainly knows the 
> correct response for each client. Similarly if the proxy is 
> not in pure pass through mode and is actually handling some 
> requests and/or inserting some requests, it would need to 
> adjust sequence numbers appropriately between server and client.

RTSP surely can distinguish the response by CSeq. I gave other problem
statement in my following email, please check them.

> Thanks,
> Jamie
> 
> > In this case, RTSP proxy can not distinguish the responses, then it 
> > may respond client with wrong message and induce client to join a 
> > wrong mutlicast group.
> >  
> > Do anybody think this is an possible trap/deficiency, and 
> RTSP should 
> > try to resolve?
> > Thanks.
> >  
> > 
> > Regards
> > 
> > Yingjie Gu
> > 
> >  
> > 
> > 
> > 
> ----------------------------------------------------------------------
> > --
> > 
> > _______________________________________________
> > mmusic mailing list
> > mmusic at ietf.org
> > https://www.ietf.org/mailman/listinfo/mmusic