I already did this for myself and I wrote it in a previous post: http://lists.live555.com/pipermail/live-devel/2010-February/011827.html
This does solve the issue, and I only get in trouble now while testing if I start and stop the application too many times in a short time (due to a socket limitation, creating sockets returning a -1 socket handle that were not properly handled), but you already responded to this :) -----Original Message----- From: live-devel-boun...@ns.live555.com [mailto:live-devel-boun...@ns.live555.com] On Behalf Of Ross Finlayson Sent: Monday, March 08, 2010 10:06 AM To: LIVE555 Streaming Media - development & use Subject: Re: [Live-devel] openRTSP > >OK, the problem is that the *client* port numbers are the same for both >>streams. I don't know why the server does this, but in any case it's >>not something that we currently support in our code (to do so would >>require that we demultiplex based on SSRC, which is something that we >>don't do). > >Just to make things clear, it is the client (openRTSP) that initially >proposes the client port numbers to the server, not the other way around OK, now I understand the point you're making - sorry. Yes, it seems that you've run across a bug (which your code is more likely to trigger, because it creates a large number of client subsessions (and thus sockets) for a single RTSP stream). Please try the following: In the file "liveMedia/MediaSession.cpp", add the line NoReuse dummy; at line 638 - i.e., just after the statement Boolean success = False; Please let us know if this seems to prevent the problem from occurring. (If so, I'll add this to a future release of the code.) -- Ross Finlayson Live Networks, Inc. http://www.live555.com/ _______________________________________________ live-devel mailing list live-devel@lists.live555.com http://lists.live555.com/mailman/listinfo/live-devel _______________________________________________ live-devel mailing list live-devel@lists.live555.com http://lists.live555.com/mailman/listinfo/live-devel
