2025-04-28, 09:34 PM
(This post was last modified: 2025-04-28, 09:40 PM by random_human_being_. Edited 1 time in total.)
I'm having the same issue with Jellyfin 10.10.7 on Docker (though it might have started sometimes before that).
Jellyfin is running on a NUC with an Intel J5005 and 8GB RAM. It is connected via ethernet to a 100Gb connection up/down. Nothing else is using the bandwidth.
The server is shared by me, my brother, and my partner. First, my brother told me he kept having issues on the Android TV app (accessing the server via Caddy, on the standard https port). Any direct stream/remuxed video of ~20Mbps would start fine, but then after 10-15 minutes would start buffering, and after than keep buffering every minute or so. Launching the video with Kodi as an external player would give him the "source too slow" error. Forcing a transcode to about ~15Mbps would usually avoid stuttering, though sometimes he would have to go as low as 5Mbps, without any consistency.
Now my partner is also having a similar issue. She's trying to watch a TV show from WebOS (rooted), but this device is connected to the local network of the Jellyfin server, and points to its http://192.xxx.x.x:yyyy, bypassing Caddy. Yet every 10 minutes or so the playback freezes for a few minutes. I cannot see anything meaningful in the logs, expect some:
MediaBrowser.MediaEncoding.Transcoding.TranscodeManager: FFmpeg exited with code 0
... and...
Emby.Server.Implementations.Session.SessionWebSocketListener: Sending ForceKeepAlive message to 1 inactive WebSockets.
... which sometimes happen when the stream freezes, sometimes not.
I will post an issue on GitHub if it makes sense, but for the time being I would like some idea on how to troubleshoot this further.
Edit: what I see is similar to what is description in this issue, albeit in my case the host OS is libreec and docker is not in bridge mode, but set up with its own network
Jellyfin is running on a NUC with an Intel J5005 and 8GB RAM. It is connected via ethernet to a 100Gb connection up/down. Nothing else is using the bandwidth.
The server is shared by me, my brother, and my partner. First, my brother told me he kept having issues on the Android TV app (accessing the server via Caddy, on the standard https port). Any direct stream/remuxed video of ~20Mbps would start fine, but then after 10-15 minutes would start buffering, and after than keep buffering every minute or so. Launching the video with Kodi as an external player would give him the "source too slow" error. Forcing a transcode to about ~15Mbps would usually avoid stuttering, though sometimes he would have to go as low as 5Mbps, without any consistency.
Now my partner is also having a similar issue. She's trying to watch a TV show from WebOS (rooted), but this device is connected to the local network of the Jellyfin server, and points to its http://192.xxx.x.x:yyyy, bypassing Caddy. Yet every 10 minutes or so the playback freezes for a few minutes. I cannot see anything meaningful in the logs, expect some:
MediaBrowser.MediaEncoding.Transcoding.TranscodeManager: FFmpeg exited with code 0
... and...
Emby.Server.Implementations.Session.SessionWebSocketListener: Sending ForceKeepAlive message to 1 inactive WebSockets.
... which sometimes happen when the stream freezes, sometimes not.
I will post an issue on GitHub if it makes sense, but for the time being I would like some idea on how to troubleshoot this further.
Edit: what I see is similar to what is description in this issue, albeit in my case the host OS is libreec and docker is not in bridge mode, but set up with its own network