Server Version#: Plex Media Server v1.42.1.10060-4e8b05daf - Docker Docker Container (LinuxServer.io) x86_64 - build: linux-x86_64 debian
Player Version#: Plex Web Version 4.147.1
<If providing server logs please do NOT turn on verbose logging, only debug logging should be enabled>
PMS is running in a Docker container on Linux Debian 9 VM hosted on latest version of Proxmox. TrueNAS is running as a second VM on same Proxmox host and all HBAs are passed through to TrueNAS. TrueNAS is used to manage pools, datasets, and Samba shares.
The Samba shares are mounted in the Linux VM and mapped to containers as needed. All folder and file permissions are set correctly, both at the host VM level and inside the Docker containers.
Today I confirmed the Plex database is not corrupt by creating a temporary container to get access to the Plex specific SQLITE3 libraries and dependencies, and copying over the database to this temp container so that I could run the database integrity test. I could not run in the Plex container due to database locks.
Iām still having issues with a couple of things now. 1) Plex crashes daily on me. Restarting the Plex container brings it back online. 2) Newly added content video and audio sources are missing in Plex details. That is, the media and related meta data get added to the Plex UI, the media is in the folder where itās expected to be, and the all permissions are correct. As a result, I cannot play anything recently added to Plex.
I did verify that the manually checking the video and audio using the ffmpeg tools. I forgot the exact commands now, but everything checked out when I manually checked earlier today. Logs are showing a lot of the following errors which would explain why I canāt play recently added media. Also, I have tried to manually analyze the media thatās having issues and forcing meta data updates to no avail.
Sep 14, 2025 19:16:51.919 [140603226745656] ERROR - [Req#cb17b/Transcode] MDE: video has neither a video stream nor an audio stream
Iāve been running a PMS since like 2013 without hardly any issues. This move from TrueNAS on bare metal to Proxmox has been a series of constant hurdles. Is this a SMB issue? I tested my SMB shares today as well and the tests all passed. I had NFS shares setup for a while before realizing that I needed SMB, after which I switched back. That presented a number of issues and Iām wondering if this is related.
What kind of help are you expecting? Debian 9 has been EOL for around 6 years.
Many better options for running Plex
LXC on the Proxmox host
Debian 12/13 VM without docker
Even Docker on Proxmox
The database looks like it has severe logical corruption which canāt be checked or corrected via SQLITE pragma integrity_check or vacuum.
Find your database backups and check when the last one was created, if more than 3 days ago the corruption has existed for a while. You can copy that over the current one (with Plex not running) and try again. I would do the pair com.plexapp.plugins.library.db and com.plexapp.plugins.library.blobs.db.
Iām running Debian 12 VM, not 9, apologies. Proxmox is Debian 13.
Docker on Proxmox isnāt officially supported and Iām heavily invested in Docker at this stage. The VM with PMS also hosts my media stack, so there are 2 dozen docker containers. Itās only PMS thatās Iām having issues with. And the whole reason I moved to Proxmox (from TrueNAS) was for isolation of services I run. LXC shares the kernel with the Proxmox host defeating the want for isolation.
Running the SQLITE3 integrity check against my Plex database didnāt return any issues. Iām beginning to wonder if itās how I have my data organized. Data is temporarily stored in the incomplete folder, then moved to the completed folder when successfully extracted, and then it gets moved to the media folder. I used this config before, but on TrueNAS bare metal install and there was no noticeable lag in moving data between the completed folder and the media folder. Could this be a factor?
Enter command # -or- command name (4 char min) : 5
Exporting current databases using timestamp: 2025-09-15_22.09.12
Exporting Main DB
Exporting Blobs DB
Successfully exported the main and blobs databases.
Start importing into new databases.
Importing Main DB.
Runtime error near line 1830451: database is locked (5)
Error 1 from Plex SQLite while importing from ā./dbtmp/library.plexapp.sql-2025-09-15_22.09.12ā
Cannot continue.
DBRepair checks to make certain PMS is stopped. (see below)
Why not use option 2 / āautoā ? (check, repair, reindex)
# Repair (Same as optimize but assumes damaged so doesn't check)
5|repa*)
# Check if PMS running
if IsRunning; then
WriteLog "Repair - FAIL - PMS runnning"
Output "Unable to repair databases. PMS is running."
continue
fi
# Is there enough room to work
if ! FreeSpaceAvailable; then
WriteLog "Import - FAIL - Insufficient free space on $AppSuppDir"
Output "Error: Unable to repair database. Insufficient free space available on $AppSuppDir"
continue
fi
Option 2 was the first one I tried which returned same error as with option 5. I made a copy of my appdata/plex folder then created a new one. Running new PMS setup now and there is definitely something wrong with my configuration as it is super slow adding movies and updating metadata. Iāve been using Plex since 2012 and Iāve never had an issue like this. Iām sure itās something to do with my configuration. Thank you for your help.
Isolation, networking, and Docker is not officially supported on LXCs. I have several VMs running Docker without issue. Even the VM/Docker with PMS has a bunch of other media related containers running without issue. Plex is the only one giving me a hard time. (and Iām not suggesting this is a Plex issue, very much doubt it actually).
Just went and checked my VMs in PVE. All but the one hosting PMS are using CPU Type āHostā. It should have also been using āHostā. Will change and see if that helps any.
I have one ProxMox machine (4x4 box for support purposes only)
It created the LXC using TTECK helper scripts
I can then place DBRepair in that container nd run it.
[chuck@lizum plexmediaserver.2035]$ ssh root@192.168.0.25
Linux pve 6.8.4-2-pve #1 SMP PREEMPT_DYNAMIC PMX 6.8.4-2 (2024-04-10T17:36Z) x86_64
The programs included with the Debian GNU/Linux system are free software;
the exact distribution terms for each program are described in the
individual files in /usr/share/doc/*/copyright.
Debian GNU/Linux comes with ABSOLUTELY NO WARRANTY, to the extent
permitted by applicable law.
Last login: Tue Sep 16 23:41:09 2025 from 192.168.0.13
root@pve:~# ./DBRepair.sh stop auto start exit
Database Repair Utility for Plex Media Server (Debian GNU/Linux 12 (bookworm))
Version v1.11.09
[2025-09-16 23.41.39] PMS already stopped.
[2025-09-16 23.41.39] Automatic Check,Repair,Index started.
[2025-09-16 23.41.39]
[2025-09-16 23.41.39] Checking the PMS databases
[2025-09-16 23.41.40] Check complete. PMS main database is OK.
[2025-09-16 23.41.40] Check complete. PMS blobs database is OK.
[2025-09-16 23.41.40]
[2025-09-16 23.41.40] Exporting current databases using timestamp: 2025-09-16_23.41.39
[2025-09-16 23.41.40] Exporting Main DB
[2025-09-16 23.41.42] Exporting Blobs DB
[2025-09-16 23.41.42] Successfully exported the main and blobs databases.
[2025-09-16 23.41.42] Start importing into new databases.
[2025-09-16 23.41.42] Importing Main DB.
[2025-09-16 23.41.47] Importing Blobs DB.
[2025-09-16 23.41.47] Successfully imported databases.
[2025-09-16 23.41.47] Verifying databases integrity after importing.
[2025-09-16 23.41.48] Verification complete. PMS main database is OK.
[2025-09-16 23.41.48] Verification complete. PMS blobs database is OK.
[2025-09-16 23.41.48] Saving current databases with '-BACKUP-2025-09-16_23.41.39'
[2025-09-16 23.41.48] Making repaired databases active
[2025-09-16 23.41.48] Repair complete. Please check your library settings and contents for completeness.
[2025-09-16 23.41.48] Recommend: Scan Files and Refresh all metadata for each library section.
[2025-09-16 23.41.48]
[2025-09-16 23.41.48] Backing up of databases
[2025-09-16 23.41.48] Backup current databases with '-BACKUP-2025-09-16_23.41.48' timestamp.
[2025-09-16 23.41.48] Reindexing main database
[2025-09-16 23.41.50] Reindexing main database successful.
[2025-09-16 23.41.50] Reindexing blobs database
[2025-09-16 23.41.50] Reindexing blobs database successful.
[2025-09-16 23.41.50] Reindex complete.
[2025-09-16 23.41.50] Automatic Check, Repair/optimize, & Index successful.
[2025-09-16 23.41.50] Starting PMS.
[2025-09-16 23.41.50] Started PMS
root@pve:~#
Make absolute certain PMS stops.
ps -ef | grep -i plex
If itās not stopping then thatās a problem Iāve been seeing on some hosts with PMS.
You can shutdown with kill -15 $(pidof 'Plex Media Server')
This is the manual stop.
If itās the case with you too, Iāll see if I can improve dealing with stubborn PMS machines.
Iāve got PMS running in an LXC now. Used the helper script to do the installation and figured out the user mapping to bind an smb share to the container. Everything is loading super fast now and as expected. Thanks for your help.
