Reputation:
I have a pretty stock installation with authentication. The installation is only used for 1 to many broadcasting. 1 person shares their screen, everyone starts muted & hidden, follows that 1 person.
I have adaptive last n = true, channel last n = 1.
The screenshares begin to fail after a few minutes. The screen goes gray for participants. The broadcasting user has about an upload of 500kbps in the screenshare but everyone has a download of 11kbps or N/A.
I see this in the jvb logs intermittently
JVB 2017-01-24 14:17:56.506 INFO: [278] org.ice4j.ice.ConnectivityCheckClient.log() timeout for pair: 158.69.116.11:10000/udp/host -> 108.208.24.154:58802/udp/prflx (stream.RTP), failing.
JVB 2017-01-24 12:33:51.337 INFO: [77] org.ice4j.ice.harvest.AbstractTcpListener.readFromChannel() Failed to handle TCP socket Socket[addr=/108.208.24.154,port=60215,localport=4443]: End of stream! JVB 2017-01-24 12:33:51.367 INFO: [507] org.ice4j.ice.ConnectivityCheckServer.log() Failed to send BINDING-RESPONSE(0x101)[attrib.count=5 len=96 tranID=0x776264674249706335757139] through [2607:5300:60:850b:0:0:0:0]:10000/udp java.io.IOException: Network is unreachable (sendto failed) at java.net.PlainDatagramSocketImpl.send(Native Method) at java.net.DatagramSocket.send(DatagramSocket.java:693) at org.ice4j.ice.harvest.AbstractUdpListener$MySocket.send(AbstractUdpListener.java:613) at org.ice4j.socket.DelegatingDatagramSocket.send(DelegatingDatagramSocket.java:767) at org.ice4j.socket.DelegatingDatagramSocket.send(DelegatingDatagramSocket.java:767) at org.ice4j.socket.IceUdpSocketWrapper.send(IceUdpSocketWrapper.java:53) at org.ice4j.stack.Connector.sendMessage(Connector.java:327) at org.ice4j.stack.NetAccessManager.sendMessage(NetAccessManager.java:629) at org.ice4j.stack.NetAccessManager.sendMessage(NetAccessManager.java:575) at org.ice4j.stack.StunServerTransaction.retransmitResponse(StunServerTransaction.java:201) at org.ice4j.stack.StunServerTransaction.sendResponse(StunServerTransaction.java:178) at org.ice4j.stack.StunStack.sendResponse(StunStack.java:771) at org.ice4j.ice.ConnectivityCheckServer.processRequest(ConnectivityCheckServer.java:210) at org.ice4j.stack.EventDispatcher$RequestListenerMessageEventHandler.handleMessageEvent(EventDispatcher.java:511) at org.ice4j.stack.EventDispatcher.fireMessageEvent(EventDispatcher.java:268) at org.ice4j.stack.StunStack.handleMessageEvent(StunStack.java:987) at org.ice4j.stack.MessageProcessor.run(MessageProcessor.java:171) at java.lang.Thread.run(Thread.java:745) JVB 2017-01-24 12:33:51.367 INFO: [507] org.ice4j.stack.StunStack.handleMessageEvent() Received an invalid request. java.lang.RuntimeException: Failed to send a response at org.ice4j.ice.ConnectivityCheckServer.processRequest(ConnectivityCheckServer.java:222) at org.ice4j.stack.EventDispatcher$RequestListenerMessageEventHandler.handleMessageEvent(EventDispatcher.java:511) at org.ice4j.stack.EventDispatcher.fireMessageEvent(EventDispatcher.java:268) at org.ice4j.stack.StunStack.handleMessageEvent(StunStack.java:987) at org.ice4j.stack.MessageProcessor.run(MessageProcessor.java:171) at java.lang.Thread.run(Thread.java:745) Caused by: java.io.IOException: Network is unreachable (sendto failed) at java.net.PlainDatagramSocketImpl.send(Native Method) at java.net.DatagramSocket.send(DatagramSocket.java:693) at org.ice4j.ice.harvest.AbstractUdpListener$MySocket.send(AbstractUdpListener.java:613) at org.ice4j.socket.DelegatingDatagramSocket.send(DelegatingDatagramSocket.java:767) at org.ice4j.socket.DelegatingDatagramSocket.send(DelegatingDatagramSocket.java:767) at org.ice4j.socket.IceUdpSocketWrapper.send(IceUdpSocketWrapper.java:53) at org.ice4j.stack.Connector.sendMessage(Connector.java:327) at org.ice4j.stack.NetAccessManager.sendMessage(NetAccessManager.java:629) at org.ice4j.stack.NetAccessManager.sendMessage(NetAccessManager.java:575) at org.ice4j.stack.StunServerTransaction.retransmitResponse(StunServerTransaction.java:201) at org.ice4j.stack.StunServerTransaction.sendResponse(StunServerTransaction.java:178) at org.ice4j.stack.StunStack.sendResponse(StunStack.java:771) at org.ice4j.ice.ConnectivityCheckServer.processRequest(ConnectivityCheckServer.java:210)
I set org.jitsi.videobridge.ratecontrol.VideoChannelLastNAdaptor.MAX_STAY_AT_ZERO_MS=5000
in videobridge sip-properties to force the retry after 5 seconds.
The console logs from inspecting chrome in the frozen participants shows: Logger.js:125 [JitsiMeetJS.js] <Object.getGlobalOnErrorHandler>: UnhandledError: null Script: null Line: null Column: null StackTrace: Error: Ping timeout at strophe.ping.js:102 at u.TimedHandler.handler (strophe.js:2752) at u.TimedHandler.run (strophe.js:2095) at u.Connection._onIdle (strophe.js:3706) r @ Logger.js:125
What is the optimal config for this type of conference (screenshare 1 send to many receive)?
I'm honestly not sure why the video drops for some people and doesn't reconnect. Can anyone lend a helping hand?
Upvotes: 1
Views: 3322
Reputation: 1
You can also set users to be muted (and also prevent audio and video transmission) when a conference gets started (except the moderator who is presenting) by changing the code inside Jicofo. This is the method name to permanently disable a user from sending video or audio:
public boolean muteParticipant(ColibriConferenceIQ channelsInfo, boolean mute, MediaType mediaType)
Upvotes: 0
Reputation: 2010
How many viewers do you have? Note that adaptive last N is gone (as an option) it just behaves that way all the time now.
You could try to use meet.jit.si and set that everyone starts muted and hidden by the moderator just to check that it's not an infrastructure problem. Alternatively, this scenario is the perfect one for YouTube broadcasting: you can have the broadcaster(s) join the room and stream it to YouTube, where all viewers can watch it live.
Upvotes: 2