Plex on Linux Docker - Slowness, not loading media, generic errors on web player

Plex Media Server
1

Server Version#: 1.42.2.10156
Player Version#: 4.147.1
<If providing server logs please do NOT turn on verbose logging, only debug logging should be enabled>

Plex Media Server Logs_2025-10-29_11-32-46.zip (2.4 MB)

Hello all, I’m running Plex on Linux Docker, in a Linux Mint VM running on Proxmox. Currently passign through a GTX 1070ti for transcoding, if it makes any difference.

For about a month now Plex has started being slower and slower showing any content and starting play anything, to the point of not being able to load any media and showing an empty library (whether TV shows or movies). Sometimes in the Docker logs I see the error udc input failed with 1001 spammed over and over again, some other times the error is Sqlite3: Sleeping for 200ms to retry busy DB, some other times there’s no error showing in Docker at all.

I’m getting this no matter what device I’m playing on (Android TV, Android phone app, web interface), what network I’m playing through (over LAN or remotely with Plex Pass), or what quality the media is (480i, 1080p, 4k).

So far what I tried (without any success) was:

  • Removing the Docker container and pulling it from scratch
  • Rebuilding the database following this guide
  • Using this tool to rebuild again the DB

The only thing that temporarily work is just restarting the container, but it’s not an acceptable workaround, and it only lasts 2-3 hours before it stops working again.

Is there anything I overlooked here? Anything else that I should try?

2

immagine
immagine1916×910 36.4 KB
Point in case, this is a screenshot of what happens a couple of hours after I restart the Docker container. The message reads:

No content available
Check your Internet connection and make sure the media servers are connected.

Internet connection is avsolutely fine, as well as my media folders.

3

Never mind, I decided to go full ballistic and completely nuked the old Plex installation, deleted every single file related to it (database and all) and rebuilt the container from scratch. So far it seems to be working as expected, let’s see how long it’ll last.