Server Version#:1.41.3.9314
Plex for Windows Version#:1.105.2.269-12a32491
Dear Plex community,
unfortunately I have the following problem with the Windows Server application of “Plex Media Server - 64-Bit”.
The problem is as follows: after about 4 to 6 hours after restarting the application, the entire “library” database becomes completely slow and only reacts sporadically to inputs.
This also applies to inputs or server settings, as I am then told to check the connection to the server, even though it is online.
After having tried and tested several solutions, which unfortunately did not improve the situation, I am at the end of my knowledge and hope that one of you has an idea
I also used the DBCheck tool, but unfortunately this didn’t work either.
The following settings are set for the library:
DB-Cache 8192MB
Create video preview thumbnails: Off
Search my media library automatically: On
Perform a partial scan when changes are detected: On
Scan my media library regularly: Off
Empty recycle bin automatically after each scan: On
Media may be deleted: On
Other Settings are default!
Scheduled tasks are all activated!
Hardware + virtualization:
i5 13600k
4TB NVME from WD
GPU: Nvidia P4 Tesla
VMware ESXi 8 - approx. 6 VMs - but no known VM problems
VM:
Windows Server 2022
GPU is passed through to VM
12GB RAM
100GB disk
Jan 07, 2025 13:45:42.414 [1524] WARN - Held transaction for too long (D:\gha\plex-media-server\plex-media-server\Statistics\StatisticsManager.cpp:284): 62.343750 seconds
Jan 07, 2025 13:45:42.415 [1016] WARN - [Req#1a21dc] Took too long (39.843750 seconds) to start a transaction on D:\gha\plex-media-server\plex-media-server\Library\MetadataItemSetting.cpp:548
Jan 07, 2025 13:45:42.415 [1016] WARN - [Req#1a21dc] Transaction that was running was started on thread 1524 at D:\gha\plex-media-server\plex-media-server\Library\Database\SqliteDB.h:100
Jan 07, 2025 13:45:42.415 [6424] WARN - [Req#1a2275] Took too long (15.000000 seconds) to start a transaction on D:\gha\plex-media-server\plex-media-server\Library\Tagging.cpp:252
Jan 07, 2025 13:45:42.415 [6424] WARN - [Req#1a2275] Transaction that was running was started on thread 1016 at D:\gha\plex-media-server\plex-media-server\Library\MetadataItemSetting.cpp:548
Jan 07, 2025 13:45:42.415 [1896] WARN - [Req#1a22a3] Took too long (12.062500 seconds) to start a transaction on D:\gha\plex-media-server\plex-media-server\Library\Database\SqliteDB.h:100
Jan 07, 2025 13:45:42.415 [1896] WARN - [Req#1a22a3] Transaction that was running was started on thread 6424 at D:\gha\plex-media-server\plex-media-server\Library\Tagging.cpp:252
Jan 07, 2025 13:45:42.415 [6940] WARN - [Req#1a2274] Took too long (0.765625 seconds) to start a transaction on D:\gha\plex-media-server\plex-media-server\Library\Tagging.cpp:252
Jan 07, 2025 13:45:42.415 [6940] WARN - [Req#1a2274] Transaction that was running was started on thread 1896 at D:\gha\plex-media-server\plex-media-server\Library\Database\SqliteDB.h:100
And see also this WARN too:
Jan 07, 2025 13:44:11.092 [3664] WARN - [Req#1a20f2] QueryParser: Invalid field 'sectionID' found, ignoring.
Jan 07, 2025 13:44:11.092 [3664] WARN - [Req#1a20f2] QueryParser: Invalid field 'contentDirectoryID' found, ignoring.
Jan 07, 2025 13:44:11.092 [3664] WARN - [Req#1a20f2] QueryParser: Invalid field 'pinnedContentDirectoryID' found, ignoring.
And:
Jan 07, 2025 13:39:29.975 [6360] WARN - Held transaction for too long (D:\gha\plex-media-server\plex-media-server\Statistics\StatisticsManager.cpp:284): 68.765625 seconds
Jan 07, 2025 13:39:29.975 [2848] WARN - Took too long (37.906250 seconds) to start a transaction on D:\gha\plex-media-server\plex-media-server\Statistics\Device.cpp:46
Jan 07, 2025 13:39:29.975 [2848] WARN - Transaction that was running was started on thread 6360 at D:\gha\plex-media-server\plex-media-server\Statistics\StatisticsManager.cpp:284
Set this back to default. It is not a file cache for the db file but relates to single db sessions. With your current setting, a single db session can clog up most of the RAM of your container.
