xAssasin 1 Posted November 20, 2025 Posted November 20, 2025 Hello emby community! Have been using emby for quite a while, and i don't know if it is the most recent versions, but recently my emby is behavioring weirdly, it randomly spikes to 100% CPU even though i have a RTX 5060 as GPU for the transcoding process (yes, i did verify if it is set properly), can you guys enlighten me? i'm pretty sure i'm not the only one having trouble with it
blackcatTN 3 Posted November 20, 2025 Posted November 20, 2025 Having the same issue, I can't go more than 36 hours between restarting service since update. Was running fine.
adminExitium 355 Posted November 20, 2025 Posted November 20, 2025 8 hours ago, xAssasin said: randomly spikes to 100% CPU is it only CPU or memory too? If only CPU, it may be due to search queries: It will be helpful if you can try to isolate the spikes and provide the logs for the same.
xAssasin 1 Posted November 20, 2025 Author Posted November 20, 2025 6 hours ago, adminExitium said: is it only CPU or memory too? So, memory actually builds up to cap (about 16GB, running in ubuntu 24.04, but im gonna upgrade memory capacity later), but before that, the CPU goes from 2% to 100% pretty quickly, i've went through "top -H -p <emby's pid>" and it seems like a single thread is consuming the CPU and stopping from others running, including the emby-server itself. I can access the dashboard (runs slowly, but i can) while the machine is at about 80% CPU, but not the /Views (which runs when i access the home page), it keeps loading infinitely and gives me an error. Also it seems that it unclogs itself at a random point, and CPU goes back to 1-10% max, and that thread stops consuming that much CPU. .NET BGC and a single .NET TP Worker are the responsable for that
xAssasin 1 Posted November 20, 2025 Author Posted November 20, 2025 2 minutes ago, Neminem said: Devs will need server logs. There you go, pretty sure these have been collected when the server was clogging. embyserver (2).txt embyserver (1).txt embyserver.txt 1
xAssasin 1 Posted November 20, 2025 Author Posted November 20, 2025 Also if i could ask, is there a way to downgrade from 4.9.1.90 to 4.8.11.0 without breaking stuff or having to re-scan everything? tried stopping the server and downgrading through dpkg but seems like it broke and doesn't show any content, only part of the images and everything gives Internal Server Error. i can just upgrade back to 4.9.1.90 and it works perfectly fine
Happy2Play 9780 Posted November 20, 2025 Posted November 20, 2025 45 minutes ago, xAssasin said: Also if i could ask, is there a way to downgrade from 4.9.1.90 to 4.8.11.0 without breaking stuff or having to re-scan everything? Since there were database changes you would have to start over unless you by chance still have a 4.8.11.0 backup.
blackcatTN 3 Posted November 20, 2025 Posted November 20, 2025 I can make the server stay up longer if I disable and avoid emby's library updates and let the "darrs" update emby on new media. I suspect running update libraries might be causing memory leak. embyserver-63899193600.txt ffmpeg-transcode-09e675b7-0e88-4cc4-b5f0-3575523aadb9_1.txt ffmpeg-transcode-c988a4a3-97bf-48a8-b7e4-92866f84c8f4_1.txt ffmpeg-transcode-a13bb67e-54ab-49d5-9511-84061c14129d_1.txt embyserver-63899202587.txt ffmpeg-transcode-9e67fadf-d40f-48a2-a1e0-8f25fa22cc99_1.txt ffmpeg-transcode-3a7d66e5-9e7b-47cf-a87e-70abee1b6bfa_1.txt ffmpeg-transcode-080e4ec5-436a-4009-889d-eabcde64f486_1.txt embyserver2.txt
Happy2Play 9780 Posted November 20, 2025 Posted November 20, 2025 1 hour ago, blackcatTN said: I can make the server stay up longer if I disable and avoid emby's library updates and let the "darrs" update emby on new media. I suspect running update libraries might be causing memory leak. embyserver-63899193600.txt 15.7 MB · 0 downloads ffmpeg-transcode-09e675b7-0e88-4cc4-b5f0-3575523aadb9_1.txt 1.56 MB · 0 downloads ffmpeg-transcode-c988a4a3-97bf-48a8-b7e4-92866f84c8f4_1.txt 216.15 kB · 0 downloads ffmpeg-transcode-a13bb67e-54ab-49d5-9511-84061c14129d_1.txt 433.51 kB · 0 downloads embyserver-63899202587.txt 4.39 MB · 0 downloads ffmpeg-transcode-9e67fadf-d40f-48a2-a1e0-8f25fa22cc99_1.txt 84.99 kB · 0 downloads ffmpeg-transcode-3a7d66e5-9e7b-47cf-a87e-70abee1b6bfa_1.txt 15.51 kB · 0 downloads ffmpeg-transcode-080e4ec5-436a-4009-889d-eabcde64f486_1.txt 11.88 kB · 0 downloads embyserver2.txt 80.52 kB · 0 downloads I would first test reseting cache size as Emby changed to a multiconnection model. So the previous method of 1.5 to 2 times db size is obsolete. 2025-11-20 02:29:51.449 Info SqliteItemRepository: Initializing PooledDatabaseConnectionManager with pool size: 5 So it could be 5 x your cache size (cache_size=-3121152) Default is currently 128.
blackcatTN 3 Posted November 21, 2025 Posted November 21, 2025 Thank you. I set it to 256 but can't restart until off peak. Will monitor for the next couple of days and report back. Looking forward to seeing what the dev team officially endorses for optimal settings. Does it ever need to be changed now or not so much? That would be useful info to share.
Luke 42077 Posted November 21, 2025 Posted November 21, 2025 6 hours ago, blackcatTN said: I can make the server stay up longer if I disable and avoid emby's library updates and let the "darrs" update emby on new media. I suspect running update libraries might be causing memory leak. embyserver-63899193600.txt 15.7 MB · 0 downloads ffmpeg-transcode-09e675b7-0e88-4cc4-b5f0-3575523aadb9_1.txt 1.56 MB · 0 downloads ffmpeg-transcode-c988a4a3-97bf-48a8-b7e4-92866f84c8f4_1.txt 216.15 kB · 0 downloads ffmpeg-transcode-a13bb67e-54ab-49d5-9511-84061c14129d_1.txt 433.51 kB · 0 downloads embyserver-63899202587.txt 4.39 MB · 0 downloads ffmpeg-transcode-9e67fadf-d40f-48a2-a1e0-8f25fa22cc99_1.txt 84.99 kB · 0 downloads ffmpeg-transcode-3a7d66e5-9e7b-47cf-a87e-70abee1b6bfa_1.txt 15.51 kB · 0 downloads ffmpeg-transcode-080e4ec5-436a-4009-889d-eabcde64f486_1.txt 11.88 kB · 0 downloads embyserver2.txt 80.52 kB · 1 download Hi, please also try removing these plugins: 2025-11-19 14:21:08.017 Info App: Loading playback_reporting, Version=2.1.0.7, Culture=neutral, PublicKeyToken=null from /var/lib/emby/plugins/playback_reporting.dll 2025-11-19 14:21:08.017 Info App: Loading Emby.Bulky, Version=1.0.20.0, Culture=neutral, PublicKeyToken=null from /var/lib/emby/plugins/Emby.Bulky.dll 2025-11-19 14:21:08.017 Info App: Loading MediaBrowser.Plugins.TimeLord, Version=19.8.3.1105, Culture=neutral, PublicKeyToken=null from /var/lib/emby/plugins/MediaBrowser.Plugins.TimeLord.dll Then restart the server and see how things compare.
blackcatTN 3 Posted November 21, 2025 Posted November 21, 2025 1 hour ago, Luke said: Hi, please also try removing these plugins: 2025-11-19 14:21:08.017 Info App: Loading playback_reporting, Version=2.1.0.7, Culture=neutral, PublicKeyToken=null from /var/lib/emby/plugins/playback_reporting.dll 2025-11-19 14:21:08.017 Info App: Loading Emby.Bulky, Version=1.0.20.0, Culture=neutral, PublicKeyToken=null from /var/lib/emby/plugins/Emby.Bulky.dll 2025-11-19 14:21:08.017 Info App: Loading MediaBrowser.Plugins.TimeLord, Version=19.8.3.1105, Culture=neutral, PublicKeyToken=null from /var/lib/emby/plugins/MediaBrowser.Plugins.TimeLord.dll Then restart the server and see how things compare. I have been using those plugins for a long time without issue. Are you saying that plugins that have been working for years may now have a memory leak due to an Emby server code update?
Happy2Play 9780 Posted November 21, 2025 Posted November 21, 2025 10 hours ago, blackcatTN said: I have been using those plugins for a long time without issue. Are you saying that plugins that have been working for years may now have a memory leak due to an Emby server code update? Any plugin that is not created by Emby will always be asked to uninstall when there are user issues to eliminate this that are totally out of Emby's control. So if is identified as a plugin issue the appropriate plugin dev will have to look at the issue. But there are other similar topics out there as it is somewhat a platform issue. 1
Dan_86 5 Posted November 21, 2025 Posted November 21, 2025 (edited) I've noticed the same thing but I've gotten around it by scanning at 3am and 3pm. I have my config to automatically reboot the instance at 6 am and then I hit a manual reboot before peak hours start around 4:30 pm daily (after the 3 pm scan) This has kept things stable for myself and my users. Currently running 128mb cache, 800 row. Real time monitoring off for all libraries. I vacuum database once a week and do not scan intros. I only do Emby backups manually. Edited November 21, 2025 by Dan_86
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now