Re: [Live-devel] MilestoneXProtect playing problem

19.06.2013 0:20, Ross Finlayson пишет: On Jun 18, 2013, at 8:39 AM, Chris Richardson (WTI) > wrote: That is correct. You can verify the server will keep sending data for each broken stream by simply killing the client process and restarting it. No, it doesn't do

Re: [Live-devel] MilestoneXProtect playing problem

Hi Ross, I'm relieved to hear I was not mistaken in my original posting. I also forgot to say thanks for including that patch in the new software. Thanks! Chris Richardson WTI From: live-devel-boun...@ns.live555.com [mailto:live-devel-boun...@ns.live555.com] On Behalf Of Ross Finlays

Re: [Live-devel] MilestoneXProtect playing problem

> if you set ‘reclamationTestSeconds’ to 0 that it does stop the server from > killing non-active client sessions, including those that have been closed in > a non-graceful manner. Isn’t this the expected behavior? Yes - it is the intended behavior. I never intended to imply otherwise. Ross F

Re: [Live-devel] windows winsock error 10038

You are right about the "removeSocketDescription" my explication was not good, forget about this. I should not have waste time trying to find an explanation because I don't have a clear view why this is happening. However this is what I have found so far, if this can help anybody track this dow

Re: [Live-devel] MilestoneXProtect playing problem

Hi Ross, I do understand what a process is and believe my definition to be the same as most other people. Regardless of that, I can tell you that if you set 'reclamationTestSeconds' to 0 that it does stop the server from killing non-active client sessions, including those that have been closed

Re: [Live-devel] MilestoneXProtect playing problem

On Jun 18, 2013, at 8:39 AM, Chris Richardson (WTI) wrote: > That is correct. You can verify the server will keep sending data for each > broken stream by simply killing the client process and restarting it. No, it doesn't do any such thing (unless the word "process" means something very dif

Re: [Live-devel] windows winsock error 10038

> The problem I found is in RTPInterface.cpp file > > -> SocketDescriptor::tcpReadHandler1 line 422 > If readSocket fails (-1) it triggers a deletion of the SocketDescriptor object > Correct. > > However this is leaving a dangling pointer in the HashTable in RTPInterface > (lookupSocketDescript

Re: [Live-devel] MilestoneXProtect playing problem

That is correct. You can verify the server will keep sending data for each broken stream by simply killing the client process and restarting it. The network usage will continue to rise. The best solution is to either subclass RTSPServer and handle OPTIONS as a keep-alive request, or apply the

Re: [Live-devel] windows winsock error 10038

Hi Ross and Andrey, I got the exact same bug on my Linux server, a random "bad file descriptor" error that kills my server every 24 hours or so using RTP over TCP streaming. Hopefully I have an auto restart mechanism that respawn my server but that's not really great solution. Same thing for m

Re: [Live-devel] MilestoneXProtect playing problem

18.06.2013 12:09, Andrey пишет: Unfortunately I have a week waiting for a response from the Milestone, but it is not. What if I use the reclamationTestSeconds = 0 when creating RTSPServer. In theory this should solve the problem. If I set reclamationTestSeconds = 0, the session with the Mileston

Re: [Live-devel] Full HD Contents. trickplay question (with VLC)

Hi Ross I had already conducted a test with VLC Media Player before testing with STB. (VLC v.2.0.6 and VLC v2.0.7) As you can see original file, 2x file is not animated as original file and 4x file is somehow animated but looks quite unnatural. Could you please check it out again?

Re: [Live-devel] Full HD Contents. trickplay question

> I’m testing a Trick-play function using Live555 and turned out most of the > contents tested are working well in STB. > But some 2x, 4x converted to testMPEG2TransportStreamTrickPlay are not played > properly in STB. > > I’d like to know if there’s something I did wrong. > > I used content

[Live-devel] MilestoneXProtect playing problem

Unfortunately I have a week waiting for a response from the Milestone, but it is not. What if I use the reclamationTestSeconds = 0 when creating RTSPServer. In theory this should solve the problem. ___ live-devel mailing list live-devel@lists.live555.c

Re: [Live-devel] windows winsock error 10038

> Thanks for the quick reply. However, the error occurs even if the test code > testOnDemandRTSPServer reuseFirstSource set to True. I have not changed > anything in the code, except the reuseFirstSource. If that's the case, then unfortunately you're going to have to track down the specific pro

[Live-devel] Full HD Contents. trickplay question

Hello everyone, I’m testing a Trick-play function using Live555 and turned out most of the contents tested are working well in STB. But some 2x, 4x converted to testMPEG2TransportStreamTrickPlay are not played properly in STB. I’d like to know if there’s something I did wrong. I

[Live-devel] windows winsock error 10038

Thanks for the quick reply. However, the error occurs even if the test code testOnDemandRTSPServer reuseFirstSource set to True. I have not changed anything in the code, except the reuseFirstSource. The error occurs on the socket fClientInputSocket of RTSPServer.cpp. Closing the socket occurs a