Server Version#: Version 1.18.3.2129
Player Version#: Version 4.16.4
Hello there, been recently having issues transcoding TrueHd video, they used to work and just updated the server recently but am unsure if it started after or not. Video will not start and just sit with a buffer wheel. Other formats seem to transcode properly and notcied this error:
Dec 10, 2019 13:07:37.000 [0x7fac227f8700] Error — [Transcoder] [truehd_eae @ 0x2d55280] error reading output
Dec 10, 2019 13:07:37.001 [0x7fac22ff9700] Error — [Transcoder] Error while decoding stream #0:2: Input/output error
Dec 10, 2019 13:07:39.000 [0x7fac237fa700] Error — [Transcoder] [truehd_eae @ 0x3321280] EAE timeout! EAE not running, or wrong folder? Could not read ‘/tmp/pms-216adebb-34bd-4916-a3c6-00ab35e6607f/EasyAudioEncoder/Convert to WAV (to 8ch or less)/6d7zsur2hwl35xos08fa1eaf_12036-0-57.wav’
Ive verified permissions of the /tmp folder and found them to be okay. I noticed the file its looking for to not be a .wav extension as the log illustrates but actually a .mlp extension with the same name ex. 6d7zsur2hwl35xos08fa1eaf_12036-0-57.mlp
im unsure if thats the issue or not, any insight would be appreciated.
May I hace the full DEBUG (settings - server - general - show advanced) logs please?
Enable the logging (VERBOSE OFF)
Restart PMS
Wait 1 minute (so the startup can also be captured)
Recreate the issue
Wait 30 seconds
Download the ZIP log files
The error shown is because the linux inotify table was full (monitoring directories).
The number of monitored directories isn’t going to change from version to version.
Hmm, i slightly follow. I dont seem to be having an issue now with the couple of videos that were failing before. Is there something i can do to invoke this to occur again?
If you would turn on the debug logging (it doesn’t use more disk space than it does when off but do keep verbose off), we can look the next time it fails.
“Debug” tells us what it’s about to / trying to do.
“Error” on tells us the error code (which isn’t enough most of the time).
I happen to recognize this problem because the only time it fails as you show is when the Linux inotify table is full.
If the kernel were still busy clearing the old watches and you jumped back in early, it’s possible (not sure how likely) it failed for that reason.
I already had debug enabled, I restarted the server and began watching on of the videos failing before. The video didnt fail this time around but uploaded the logs incase theres something you can see
Edit: Just realized you said next time it happens, will monitor.
I think I’m hit by this too. Not sure if it’s TrueHD or 7.1 channel audio but since a recent upgrade (last or previous) movies with TrueHD are not played anymore. They are buffering infinitely. It has been test on Samsung TV and Plex Madia Player on Linux.
Note that transcoding from TrueHD 7.1 to Opus 7.1 is working with a Plex Media Player on Linux, but not a transcoding from TrueHD 7.1 to AAC.
I have 7.1 here for my Shield and ATV 4K testing. It’s transcoding fine so I’d like to see the DEBUG logs of a recreation please to determine if a server or app problem.
