Transcoder error - EAE Timeout?

Plex Media Server Desktops & Laptops
1

Hello

I’m getting this weird error from my plex server log, and the CPU is running high for some reason, until I restart the PMS on my debian system.

Jun 25, 2022 07:47:24.000 [0x7f0cdb717b38] ERROR - [Transcoder] [eac3_eae @ 0x7f16433d2b40] EAE timeout! EAE not running, or wrong folder? Could not read ‘/tmp/pms-937f4984-1209-4807-a1c4-d548707aa375/EasyAudioEncoder/Convert to WAV (to 8ch or less)/80188f1e-5a01-4d18-94a7-af16ca672549_7514-0-99.wav’
Jun 25, 2022 07:47:24.000 [0x7f0cdb588b38] ERROR - [Transcoder] [eac3_eae @ 0x7f16433d2b40] error reading output
Jun 25, 2022 07:47:24.000 [0x7f0cdb27ab38] ERROR - [Transcoder] Error while decoding stream #0:1: I/O error

Server Version#: 1.27.1.5916

Any ideas how to fix this ?
Plex Media Server Logs_2022-06-25_07-47-46.zip (6.2 MB)

2

In your logs, I see where the scanner is processing a very large number of directories.

In total (total of all directories and subdirectories under all media top-level directories), do you have?

(Can be counted by find /top-dir1 /top-dir2 <add more here as needed> -type d -print | wc -l

The default max inotify directory count is either 8192 or 65536 (distro and kernel dependent)

I am asking this because the transcoder needs 2 inotify slots for the EAE.

This error will happen when all the inotify slots have been allocated to “update my library automatically when changes detected”.

Jun 25, 2022 07:42:15.000 [0x7f0cdb717b38] ERROR - [Transcoder] [eac3_eae @ 0x7f2533886b00] EAE timeout! EAE not running, or wrong folder? Could not read '/tmp/pms-937f4984-1209-4807-a1c4-d548707aa375/EasyAudioEncoder/Convert to WAV (to 8ch or less)/ce3952ab-2c6a-4a9e-8cef-7bcdaa151da8_28389-0-3816.wav'
Jun 25, 2022 07:42:15.000 [0x7f0cdbb50b38] ERROR - [Transcoder] [eac3_eae @ 0x7f2533886b00] error reading output
Jun 25, 2022 07:42:15.001 [0x7f0cdbd5bb38] ERROR - [Transcoder] Error while decoding stream #0:1: I/O error

The solution is simple:

  1. Count the number of directories in use
  2. Round up to the next integral multiple of 32768 (or 65536 if you have ample memory)
  3. Increase the default table size and restart the host.
3

Thanks alot, yes that solved the problem instantly!

4

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.