Busy DB error - 1.15.2.793 (unRAID)

Server Version#: 1.15.2.793
Player Version#: 3.91.0

I’ve been experiencing an issue I’ve been trying to pin down the last few weeks. I haven’t payed close attention to server version until recently but I’m thinking this may have crept up since 1.15.x

I’m running the official Plex docker on my unRAID server.
My appdata folder located on mirrored SSD cache that is otherwise performing normally for other docker containers.

I’m experiencing an issue where, I would say every day at this point Plex is slowing to a crawl. Transcodes slow down to the point where video playback gets stuck at buffering.
unRAID log indicates: Sqlite3: Sleeping for 200ms to retry busy DB.

The issue persists for as long as I leave it sit. Restarting the container seems to clear up the issue and things go back to normal for several hours to the better part of the day (seems to depend on utilization?).

I’ve been able to successfully optimize my database and clean bundles so I don’t believe there is a database corruption issue? Pulling a database backup, size is around 274mb (which I don’t feel seems to crazy for the amount of media I have).

CPU and Memory utilization during these errors seem fine.
System is Dual Xeon E5-2650 CPU’s with 128gb memory for what it’s worth

Attached are a log pull, I can attach database as well if necessary.

So far I have, as stated, successfully ran a database cleanup as well as it being scheduled normally. I have changed my Library scan interval to daily. I’ve also tried a few random things I’m not specifically recalling at the moment but they did not have an improvement.

Looking for ideas as to what may be causing my troubles, unless they’re related to 1.15.x?

For the time being I’m resorting to restarting my container daily to help mitigate the issue.

Thanks in advance

Plex Media Server Logs_2019-03-26_21-23-12.zip (4.0 MB)

Quick update to my situation.

I discovered that my appdata share for unRAID had changed from only cache to cache preferred. I adjusted the setting back so that the share is running exclusively off the cache. However, I still noticing busy DB messages in the container log.

Still looking for advice on how to proceed in getting this corrected.

Thanks

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