2026-07-04, 08:24 PM
About a month ago I tried the new 12.0 unstable build using docker on a Qnap TS-853-Pro migrating in a separate instance from 10.11.11 and had a bunch of issues with doing a Scan all as it just got locked up over and over with nonstop "database is locked" and would freeze for long periods. I tried every single suggestion from adjusting parallel threads, to adding the environment to disable 2nd cache in SQL, and changing it from No Lock to Optimistic.
Here I am trying it again.
System:
QNAP - TS-853 Pro
Docker
Latest unstable build using jellyfin/jellyfin:unstable pull image (7/3/2026)
Environment: Added disable SQL 2nd cache to true
Parallel 1 thread
Optimistic - DB
The scan took 6 hours after the migration which took about 30 mins.
Still saw a few database is locked error messages.
I attached the full logs and yes I did stop the instance at around 30 and 70% a few times as it felt like it got stuck. After changing all the settings to help the database get passed the locked message it took a total of 6 hours to finish and was wondering if there was anything else I can do to alleviate the issue. 10.11.11 runs well with full scans taking about 20-30 mins and just the music videos taking 30 seconds to load which always puzzles me as I worked around it by making them movies with no scrape.
My setup is all media is on a full dual 1gb network pipe back to samba shares on a pc with about 120TB attached to it in various drive configurations.
The entire docker + system config everything is on Samsung 980 Pro and WD 1TB Blue SSD.
Is it just the way the first scan works and generally taking forever is normal?
Is it an incompatibility with plugins? (I switched to unstable repository and disabled almost all plugins not installed at setup)
There was a massive stall on "ratings" in the logs I noticed not sure what that's about.
Trying to figure out if something between 10.11.11 and 12.0-unstable I needed to configure something else to speed it up.
I absolutely love the improvements of 12.0-unstable and so many "it makes sense" new features.
Here I am trying it again.
System:
QNAP - TS-853 Pro
Docker
Latest unstable build using jellyfin/jellyfin:unstable pull image (7/3/2026)
Environment: Added disable SQL 2nd cache to true
Parallel 1 thread
Optimistic - DB
The scan took 6 hours after the migration which took about 30 mins.
Still saw a few database is locked error messages.
I attached the full logs and yes I did stop the instance at around 30 and 70% a few times as it felt like it got stuck. After changing all the settings to help the database get passed the locked message it took a total of 6 hours to finish and was wondering if there was anything else I can do to alleviate the issue. 10.11.11 runs well with full scans taking about 20-30 mins and just the music videos taking 30 seconds to load which always puzzles me as I worked around it by making them movies with no scrape.
My setup is all media is on a full dual 1gb network pipe back to samba shares on a pc with about 120TB attached to it in various drive configurations.
The entire docker + system config everything is on Samsung 980 Pro and WD 1TB Blue SSD.
Is it just the way the first scan works and generally taking forever is normal?
Is it an incompatibility with plugins? (I switched to unstable repository and disabled almost all plugins not installed at setup)
There was a massive stall on "ratings" in the logs I noticed not sure what that's about.
Trying to figure out if something between 10.11.11 and 12.0-unstable I needed to configure something else to speed it up.
I absolutely love the improvements of 12.0-unstable and so many "it makes sense" new features.
![[Image: GitHub%20Sponsors-grey?logo=github]](https://img.shields.io/badge/GitHub%20Sponsors-grey?logo=github){kind=icon}
