Re: [MMUSIC] WGLC: draft-ietf-mmusic-rfc2326bis-22

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

Hi Magnus,
Thank you.
It'll be helpful to hint at some solutions.


Regards

Yingjie Gu

 

> -----Original Message-----
> From: Magnus Westerlund [mailto:magnus.westerlund at ericsson.com] 
> Sent: Monday, November 02, 2009 12:06 AM
> To: Y.J. Gu
> Cc: 'Joerg Ott'; mmusic at ietf.org; 'Jean-Francois Mule'; 
> 'Martin Stiemerling'
> Subject: Re: [MMUSIC] WGLC: draft-ietf-mmusic-rfc2326bis-22
> 
> Hi,
> 
> I agree about expanding the proxy chapter. How the proxy 
> really keep tracks of stuff is implementation specific as 
> long as what is sent out in both directions are following the 
> spec and consistent. But we can clearly hint at some solutions.
> 
> /Magnus
> 
> Y.J. Gu skrev:
> > Hi all,
> > According to the discussion in the mailing list, there exist some 
> > confusions on proxy and CSeq. I conclude the confusions as 
> below. Pls 
> > feel free to correct my understanding.
> > 
> > 1.  How Cseq series should be organized on proxy when the proxy 
> > aggregates several RTSP sessions. Magnus has answered this question 
> > and I suggest to make the text in Section 16.19 clearer.
> > 
> > 2.  Is Cseq a good way to matching responses to requests if the 
> > requests are pipelined? I have gave an implementation and explains 
> > that Cseq could work but relies on the TCP connection in some case. 
> > Pls refer to my post in 2009-9-10. I think, as a protocol, 
> RTSP should 
> > not bind tightly to any transport protocol.
> > 
> > 
> > So I suggest that we expand on the proxies charter. How 
> about adding a 
> > subsection 17.2 to discuss Proxy aggregating multiple RTSP sessions 
> > and give another optional solution other than Cseq. Cseq is 
> useful in 
> > some cases, but might there be an easier and clearer one 
> that can also 
> > achieve the goal? My gut feeling is that a local 
> identifier, generated 
> > by proxy and included without change in both request and 
> response, may 
> > help proxy match requests&responses. Proxies assign a local 
> identifier to each new session.
> > Cseq can be used as well for any intention rather than 
> matching R/r. 
> > 
> 
> -- 
> 
> Magnus Westerlund
> 
> IETF Transport Area Director
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM
> ----------------------------------------------------------------------
> Ericsson AB                | Phone  +46 10 7148287
> Färögatan 6                | Mobile +46 73 0949079
> SE-164 80 Stockholm, Sweden| mailto: magnus.westerlund at ericsson.com
> --------------------------------------------------------------
> --------