So, I looked into it. The entry ingestion fails in the post convert phase, not the actual convert, which is why you DO see the flavour files on disk.
It fails here:
And indeed, $prdAud->_duration is set to 47ms, which is wrong because the entry duration is 40:32 seconds.
FFmpeg manages to generate 1 or 2 audio frames and stops there so the AUDIO duration for the generated flavours is 47ms, which is of course wrong and does not match the duration set for the source, hence the “post convert” job fails.
If you run the ffmpeg command listed in the log file, you will see:
[libfdk_aac @ 0xcedf80] Trying to remove 1024 samples, but the queue is emptyts/s speed=6.77x
[mp4 @ 0xe23f40] Timestamps are unset in a packet for stream 0. This is deprecated and will stop working in the future. Fix your code to set the timestamps properly
[mp4 @ 0xe23f40] Encoder did not produce proper pts, making some up.
So, there clearly is some issue with the sound track on this source video and indeed, when I use any player to test it [VLC for instance] I do not hear any sound, even though a sound track does exist… if you run the mediainfo util on the source file, you will see:
ID : 1
Format : AC-3
Format/Info : Audio Coding 3
Codec ID : 2000
Duration : 40s 32ms
Bit rate mode : Constant
Bit rate : 64.0 Kbps
Channel(s) : 2 channels
Sampling rate : 44.1 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 313 KiB (0%)
Alignment : Aligned on interleaves
The easiest workaround I can offer you is to remove the audio track from the original source and then ingest the resulting video. This can easily be done with FFMPEG like so:
$ ffmpeg -i /path/to/orig/video -vcodec copy -an /path/to/audio/stripped/video
Then you can ingest /path/to/audio/stripped/video and it will work. I tested that, of course.