2024-10-09, 08:09 PM
(This post was last modified: 2024-10-09, 10:46 PM by TheDreadPirate. Edited 2 times in total.)
One thing I'm noticing now that I didn't see before was that the database errors are always preceded by an access to the "Resume" endpoint. I'm assuming that is the api endpoint that populates the continue watching row. And is not related to the playback reporting plugin.
My guess is that table is borked for some reason. Possibly from your shifting things around. Possibly not. Hard to say.
Hypothetically, if the move is "transparent" it shouldn't affect Jellyfin. Transparent meaning that the mount location, while going to different destinations, stays at the same location on the host running jellyfin. Ideally, the switch from one NAS to the other would have happened while Jellyfin was shutdown to ensure it didn't attempt to access that path while you were migrating.
I'm not sure what the least intrusive way you would "fix" this. If you know what WAS in continue watching, you could try going to each movie/show/episode and marking it as "watched".
My guess is that table is borked for some reason. Possibly from your shifting things around. Possibly not. Hard to say.
Hypothetically, if the move is "transparent" it shouldn't affect Jellyfin. Transparent meaning that the mount location, while going to different destinations, stays at the same location on the host running jellyfin. Ideally, the switch from one NAS to the other would have happened while Jellyfin was shutdown to ensure it didn't attempt to access that path while you were migrating.
I'm not sure what the least intrusive way you would "fix" this. If you know what WAS in continue watching, you could try going to each movie/show/episode and marking it as "watched".