I’m running the latest Plex server (LinuxServer docker image) on Unraid 7.20. Over the last couple of weeks I have suddenly noticed the Plex UI being unresponsive or the libraries not loading at all. This is across mobile and TV clients. Killing and restarting often works. Sometimes, I hit to play a file and after a few seconds of loading it doesn’t play anything.
I’ve noticed this has coincided with seeing “Sqlite3: Sleeping for 200ms to retry busy DB.” errors in the log files. There is nothing else concerning or error related.
I’ve run the DBRepair script now several times and it doesn’t seem to make any difference. Likewise, restarting the docker container might sometimes help but the log messages and non-responsiveness soon return.
I am bemused because this has been running very stable for over 4 years and there has been no other changes to the system. Likewise, there is not any other significant disk or CPU activity that could cause this. The Plexx app data folder is also on an SSD cache.
Any ideas to what I can do to fix this? Thanks in advance.
I am in almost the exact same situation. Just upgraded Unraid to 7.2 and Plex has started losing connection to each library, with the log showing “Sqlite3: Sleeping for 200ms to retry busy DB“. Some times this persists indefinitely and some times the libraries ‘reconnect’ after a few minutes.
Hey, I found adjusting the cache helps. I set mine to 2000mb and in the last 4 days have only had to restart once. See here for more details.
Do you run Plex Meta Manager and have a large library? This seems to be somewhat common though why now and why when I have plenty of system resources is confusing.
I’m not sure what to do next. Other reports suggest rebuilding the database generally doesn’t fix it. I may increase the cache size another 1000mb.
@ChuckPa My DB size is ~2.5GB. I have a large library but have no idea if this size would be indicative of the bloat issue you mention. My server version is currently 1.42.2.10156.
Assuming I am impacted by this issue and from a quick search, I came across this. Is it the recommended way to fix the DB?
Hi @ChuckPa Thanks, I just did this. I think I misstated the DB size beforehand as I confused it with the blob db. It was 627mb and deflate reduced it to 617. I then did an auto clean/repair so will see how it goes.
