Jump to content


Photo

Strange transcoding error: "Requested segment past EOS, returning empty segment"


  • Please log in to reply
1 reply to this topic

#1 aresando

aresando

    Member

  • Members
  • PipPip
  • 15 posts

Posted 28 March 2017 - 10:32 AM

Hi

 

Today for the first time I encountered a weird error...

This only happens on one specific file.

The file has nothing special, it loads and plays normally on the server (with VLC)

It is part of a series that were all encoded the same. All other files in the series play normally.

 

Only, this one file keeps "stalling" in air video HD at various positions: the player displays "Loading..." and the server live log reads "Requested segment past EOS, returning empty segment".

 

The error happens with both the latest stable version (2.2.1) and beta (2.2.4) (the server is a Mac with OSX 10.11 (El Capitan)

 

Thanks!

 

More log from the server (v2.2.4-beta2):

 

18:23:54 DEBUG 1C transcoding.manager TranscodingManager.cpp:1267 - Created transcoding session: /live-transcoding/playlist/session27-a4f72a60-13a0-11e7-87c0-34159e2e6afa/master.m3u8
18:23:54 INFO  1B transcoding.transcoder GstLiveTranscoder.cpp:31 - Creating transcoder for '/Volumes/[path]/[file].mkv' with initial offset 3250.000000s
18:23:54 WARN  1B gst.basesrc gstbasesrc.c:3480 - - - - - - - - - - pad not activated yet
18:23:54 FIXME 1B gst.default gstutils.c:3697 - - - - - - - - - - - Creating random stream-id, consider implementing a deterministic way of creating a stream-id
18:23:54 WARN  1B gst.typefind gsttypefindelement.c:782 - - - - - - failed to query peer uri
18:23:54 FIXME 1B gst.default gstutils.c:3697 - - - - - - - - - - - Creating random stream-id, consider implementing a deterministic way of creating a stream-id
18:23:54 FIXME 1B gst.default gstutils.c:3697 - - - - - - - - - - - Creating random stream-id, consider implementing a deterministic way of creating a stream-id
18:23:54 INFO  1B transcoding.transcoder BaseGstTranscoder.h:79 - - Added pad video_0 - No Decoder (bufferskip153), video/x-h264,level=4.1,profile=high,stream-format=avc,alignment=au,w=1280,h=716,PAR=179/180,fps=24000/1001,parsed=true,codec_data=buffer
18:23:54 INFO  1B transcoding.transcoder BaseGstTranscoder.h:79 - - Added pad audio_0 - VLCAudioDecoder (vlcaudiodecoder_ac3-13), audio/x-raw,format=F32LE,layout=interleaved,rate=48000,channels=6,channel-mask=0x000000000000003f
18:23:54 ERROR 1B gst.ebmlread ebml-read.c:164  - - - - - - - - - - Invalid EBML ID size tag (0xb) at position 566245468 (0x21c0385c)
18:23:54 WARN  1B gst.matroskademux matroska-demux.c:3922 - - - - - parse error, trying next cluster expected at 566351873
18:23:54 WARN  1B gst.basesrc gstbasesrc.c:2391 - - - - - - - - - - processing at or past EOS
18:23:54 WARN  1B gst.ebmlread ebml-read.c:154  - - - - - - - - - - peek failed, ret = eos
18:23:54 FIXME 1B gst.basesink gstbasesink.c:3064 - - - - - - - - - stream-start event without group-id. Consider implementing group-id handling in the upstream elements
18:23:54 WARN  1B transcoding.stateRegular TranscodingState.cpp:601 Requested segment past EOS, returning empty segment
18:23:59 WARN  1B transcoding.stateRegular TranscodingState.cpp:601 Requested segment past EOS, returning empty segment (repeated 8 times)
18:24:02 WARN  1B transcoding.stateRegular TranscodingState.cpp:601 Requested segment past EOS, returning empty segment
18:24:08 WARN  1B transcoding.stateRegular TranscodingState.cpp:601 Requested segment past EOS, returning empty segment (repeated 8 times)
18:24:13 WARN  1B transcoding.stateRegular TranscodingState.cpp:601 Requested segment past EOS, returning empty segment (repeated 8 times)
18:24:19 WARN  1B transcoding.stateRegular TranscodingState.cpp:601 Requested segment past EOS, returning empty segment (repeated 8 times)
18:24:19 WARN  1B transcoding.stateRegular TranscodingState.cpp:601 Requested segment past EOS, returning empty segment (repeated 6 times)

 



#2 aresando

aresando

    Member

  • Members
  • PipPip
  • 15 posts

Posted 28 March 2017 - 10:41 AM

Hi.

I made additional verifications with other players and it looks like the file is actually corrupted in some weird way.

Let's consider the problem solved.






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users