Database Corruption upgrading from 1.40.5.8921 to 1.41.6.9685

Server Version#: Fedora 31
Player Version#: 1.40.5.8921

Every time I attempt an upgrade (on failure, I revert to a snapshot as PMS is virtualized) to get significant SQLite errors in the log file and the Plex Media Service fails to launch with exit-code stop, exception 255. You can see the database errors in the plex log.

Oddly, running Plex SQlite on com.plexapp.plugins.library.db the PRAGMA integrity_check comes back “ok”. Doing a VACUUM and REINDEX have no effect.

I usually try for an hour or two to get it working, give up, revert the snapshot back to 1.40.5 and live a happy life again (logs are clean).

Any guesses on getting this upgrade so I can use Plex on my phone again (since I’m locked out not being able to upgrade to 1.41+)

PlexDBRepair: GitHub - ChuckPa/PlexDBRepair: Database repair utility for Plex Media Server databases

It does a more thorough job than the steps in the Repair a Corrupted Database article.

The Readme has instructions for various platforms.

Use the AUTO option.

If it cannot fix the database, try REPLACE to use a backup copy, if any are available.

Here’s an example running on a Synology NAS: GUI based database corruption identification/repair - #2 by FordGuy61

Thank you – I how PlexDBRepair creator a beer or two.

Pretty straight forward on Fedora; download, install, chmod +x, run on auto.

Found out it was the blob database NOT the main database that was damaged. Auto repair failed, but using the ignore command, it was able to complete (ignore one error on the blob repair) and the server is now running at 1.41.6 without issues.

It’s recommended to do a full library scan + metadata refresh, which I’ve done the first, but not the second.

Thanks again FordGuy61 for the links.

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