I’ve ran in to an issue which really stumps me. Playback on clients freezes from time to time, or buffers for a long time (>1m) before continuing. This mostly happens remote (two different Chromecast users, one Ultra, one V2), but also happens locally (Nvidia Shield), but much less frequent.
I’m running the official plexinc/pms-docker image (PMS Version 1.25.2.5319) on a Debian Buster host machine (with an Intel G4560 CPU). I have HW transcoding enabled, but the issue also occurs with it disabled.
No errors are logged in the debug PMS.log, but I’ve attached the file (containing a remote chromecast stream) nevertheless.
The stream info is shown below:
To find it in the log: stream started on 14:26 and froze on 14:37.
When it freezes CPU usage shoots up:
Afterwards the client quality was reduced from 1080p (direct play) to 480p, this did not resolve the buffering issue. I have a 500mbit/50mbit (down/up) internet connection available, so this should be the bottleneck. When transcoding Tautulli reports the transcode to be ‘throttled’ so this indicates that is is more than fast enough.
I have been running a Plex server for years, and never had an issue with streaming to both Chromecasts. It seems that this issue has started in the last few weeks, but I’m not exactly sure on the timeline.
Is there any way to figure out what causes this issue?
Plex Media Server.log (7.6 MB)
