SOLVED: Playback gets stuck with high CPU usage

[robaj]

I'm experiencing an issue where playback gets stuck on my LG TV (using the webOS app). For certain videos, if I play beyond a particular time stamp (or seek immediately to the same time stamp), Jellyfin starts consuming lots of CPU and playback gets stuck. I don't think this is affecting all of the media that I'm attempting to play.

- Jellfin Version: 10.9.6
- OS: Ubuntu 24.04 LTS
- Linux Kernel: 6.1.0-1016-rockchip
- Virtualization: Docker
- Clients: LG TV webOS app
- Browser: N/A
- FFmpeg Version: ffmpeg version 6.0.1-Jellyfin
- Playback Method: Transcode audio only, not video
- Hardware Acceleration: none
- GPU Model: none
- Plugins: none
- Reverse Proxy: haproxy
- Base URL: yes - http://server.example-network:8081/jellyfin/ (local network DNS name, with base path handled via haproxy)
- Networking: Bridge
- Storage: local

I've turned on debug logging. From what I can see, there is a regular amount of logging until the problem occurs, after which a very large amount of logs are generated. It looks like DynamicHlsController.GetHlsVideoSegment is called very rapidly for the same two video segments. At this point the playback is frozen on the TV, and this logging continues to come out until the app is closed.

Logs link: https://pastebin.com/45nMAyg2

I can provide more logs if that would help. I'm not sure if there are logs I can get from the webOS app.

Any ideas what the issue is, or what I can do to further investigate this?