Timecode out of order [video]

Wowza Streaming Engine
1

Hi,

I’m running Wowza 3.5 with the transcoder module.

Our logs are generating the following error on a regular basis(and subsequently, losing the stream). The Wowza server’s CPU utilization stays high(when it’s not receiving a stream, it’s obviously nil). Yet we can’t view the video stream.

WARN server comment 2012-11-21 11:41:43 - - - - - 1014.133 – - - - - - - LiveStreamPacketizerSmoothStreaming.handlePacket[live/definst/myStream_720p]: Timecode out of order [video]: 122020000:122090000

We’re currently using FMLE and have adjusted the H.264 time interval to get rid of another error.

Any idea what is causing this? Thanks.

Update:

No additional errors in my log but suddenly the stream has stopped. Only information is this:

WARN server comment 2012-11-21 11:55:18 - - - - - 1777.407 – - - - - - - HTTPStreamerAdapterSanJoseStreamer.getAppInstance: Stream not found [live/ngrp:myStream_all/playlist_b5096000.abst]: myStream_all

Second update:

Stream is now frozen(while watching it) and receiving this new error:

WARN server comment 2012-11-21 11:57:28 - - - - - 1959.04 - – - - - - - LiveStreamPacketizerSmoothStreaming.handlePacket[live/definst/myStream_360p]: Timecode out of order [video]: 745100000:745110000

2

This is an encoder problem. There are many forum posts about this:

https://www.wowza.com/forums/google.php?query=timeccode%20out%20of%20order

Richard

3

This is an encoder problem. There are many forum posts about this:

https://www.wowza.com/forums/google.php?query=timeccode%20out%20of%20order

Richard

nothing comes up when I click the above link. I actually found this thread by searching for the same thing and so far I haven’t found how to address this on the encoder.

We have similar errors in our log - Wirecast encoding at 30fps and appears to be stable at that rate, keyframes set to 60 and timecode unchecked, going into transcoder for 3 lower bitrates. We see playback up and downshifting on Flash and iOS very frequently. Quality Monitor in JWplayer tells me there are occasionally dropped frames on the client end (100mbps client, server on 10gbps), but the encoder is running at low CPU utilization with a pretty generic profile.

this is what we are running for this stream:

<smil>
  <head /> 
	<body>
		<switch>
			<video src="mp4:myStream_source" system-bitrate="1941504" width="1280" height="720">
			<param name="audioBitrate" value="98304" valuetype="data" /> 
			<param name="videoBitrate" value="1843200" valuetype="data" /> 
			<param name="videoCodecId" value="avc1.77.31" valuetype="data" /> 
			<param name="audioCodecId" value="mp4a.40.2" valuetype="data" /> 
			</video>
			<video src="mp4:myStream_360p" system-bitrate="948304" width="640" height="360">
			<param name="audioBitrate" value="98304" valuetype="data" /> 
			<param name="videoBitrate" value="850000" valuetype="data" /> 
			<param name="videoCodecId" value="avc1.77.31" valuetype="data" /> 
			<param name="audioCodecId" value="mp4a.40.2" valuetype="data" /> 
			</video>
			<video src="mp4:myStream_240p" system-bitrate="448304" width="426" height="240">
			<param name="audioBitrate" value="98304" valuetype="data" /> 
			<param name="videoBitrate" value="350000" valuetype="data" /> 
			<param name="videoCodecId" value="avc1.66.21" valuetype="data" /> 
			<param name="audioCodecId" value="mp4a.40.2" valuetype="data" /> 
			</video>
			<video src="mp4:myStream_160p" system-bitrate="298304" width="284" height="160">
			<param name="audioBitrate" value="98304" valuetype="data" /> 
			<param name="videoBitrate" value="200000" valuetype="data" /> 
			<param name="videoCodecId" value="avc1.66.21" valuetype="data" /> 
			<param name="audioCodecId" value="mp4a.40.2" valuetype="data" /> 
			</video>
		</switch>
	</body>
</smil>

the problem is most pronounced during sequences with lots of motion, so it most likely is the encoder, even though it doesn’t appear to have a problem with the encode. However, in Wirecast you can’t force the frame rate - they state it isn’t “guaranteed”

The server runs at 23% CPU load with two such streams being transcoded. We had the issue even with just one transcode. Source encoder is an i7 running at a mild load as well.

Peter

4

pburke, did you ever find out how to fix this?

I’m running Wirecast too, and have a similar problem reported in this thread.