Updated to 1.19.4.2902 and the issue still persists.
Is there maybe a good third party solution to optimize my media? I really need the feature, because my CPU is just not fast enough to transcode on the fly
The sps_id errors in the first example aren’t necessarily an issue.
The “Unable to re-open … output file for the second pass (fast start)” message looks like a significant issue.
After the Plex Transcoder (ffmpeg) does the re-encoding, it rewrites the file to get the headers at the right place at the front of the file. If that’s failing, you’ll be left with a broken file.
Where are these files actually stored, and how are they made available to Ubuntu and Docker?
My first thought is permissions, or something odd about how the filesystem is mapped in.
My second thought is to try a simpler/smaller file, in case something else is going wrong inside the transcoder/ffmpeg. Can you grab this and add it to your library (doesn’t matter if it matches correctly) and try it?
Damn, I completely forgot to mention that I have my library stored on google drive. It’s then mounted using rclone. It’s mounted with code = drive so it has full read write access. The rclone flags are --allow-other --daemon --vfs-cache-mode writes
Also keep in mind that the optimization worked once. So it works in principle.
Setting a local directory as a storage for the optimized file makes it work, too, which I also could have mentioned earlier. Sorry about that