Hello!
I dig into an issue why EncodingClient is always restarting jobs (1.cmd) after it is finished.
I switched ffmpeg loglevel to debug and saw an error:
Code:
...
encoded 1440 frames, 95.87 fps, 2393.23 kb/s
Last message repeated 47 times
av_interleaved_write_frame(): Broken pipe
No more output streams to write to, finishing.
Error writing trailer of pipe:: Broken pipe
frame= 1442 fps= 95 q=-0.0 Lsize= 1945359kB time=00:01:00.14 bitrate=264972.9kbits/s speed=3.95x
video:755kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 257632.718750%
Input file #0 (\\USERX-MASTER\RipBot264temp\job1\Chunks\1.avs):
Input stream #0:0 (video): 1442 packets read (1993420800 bytes); 1442 frames decoded;
Total: 1442 packets (1993420800 bytes) demuxed
Output file #0 (pipe:):
Output stream #0:0 (video): 1442 frames encoded; 1442 packets muxed (772912 bytes);
Total: 1442 packets (772912 bytes) muxed
1442 frames successfully decoded, 0 decoding errors
[AVIOContext @ 0000027b54da5c00] Statistics: 0 seeks, 62006 writeouts
[AVIOContext @ 0000027b54da6f80] Statistics: 927 bytes read, 0 seeks
Conversion failed!
I splitted the commands in 1.cmd and wrote ffmpeg output to
1.pipe. Afterwards I piped the output to x264_x64.exe which finished without errors.
Last lines:
Code:
encoded 1440 frames, 20.82 fps, 2393.17 kb/s
Ein Prozess hat versucht, zu einer nicht bestehenden Pipe zu schreiben.
Is it possible that '
process tried to write to a nonexistent pipe' leads to the abortion?
I altered the opencl device (0 or 9) which leads to the same result but with cldev 9 (not exists) EncodingServer complete this chunk without restarting it.
Edit:
Appended
Full log.