SOLVED: Meaning of "Current HLS implementation doesn't support non-keyframe breaks..."

[kt7brwahvw]

In my logs I see "Current HLS implementation doesn't support non-keyframe breaks but one is requested, ignoring that request" - what does that mean? Google is failing me and only seems to be related to questions when transcoding fails, but my transcoding is fine. Hardware acceleration is fine:


[2024-08-21 22:33:10.610 +00:00] [INF] [21] Jellyfin.Api.Controllers.DynamicHlsController: Current HLS implementation doesn't support non-keyframe breaks but one is requested, ignoring that request
[2024-08-21 22:33:10.610 +00:00] [INF] [21] MediaBrowser.MediaEncoding.Transcoding.TranscodeManager: "/usr/lib/jellyfin-ffmpeg/ffmpeg" "-analyzeduration 200M -probesize 1G -ss 00:02:03.000 -noaccurate_seek -init_hw_device vaapi=va:/dev/dri/renderD128,driver=iHD -hwaccel vaapi -hwaccel_output_format vaapi -noautorotate -i file:\"whatever" -noautoscale -map_metadata -1 -map_chapters -1 -threads 0 -map 0:0 -map 0:1 -map -0:s -codec:v:0 h264_vaapi -low_power 1 -rc_mode VBR -b:v 4026216 -maxrate 4026216 -bufsize 8052432 -sei -a53_cc -force_key_frames:0 \"expr:gte(t,n_forced*3)\" -vf \"setparams=color_primaries=bt709:color_trc=bt709:colorspace=bt709,scale_vaapi=format=nv12:extra_hw_frames=24\" -codec:a:0 libfdk_aac -ac 2 -ab 256000 -af \"volume=2\" -copyts -avoid_negative_ts disabled -max_muxing_queue_size 2048 -f hls -max_delay 5000000 -hls_time 3 -hls_segment_type fmp4 -hls_fmp4_init_filename \"0f9cc6b0086d0b09cec6cf268e9a8ed2-1.mp4\" -start_number 41 -hls_segment_filename \"/cache/transcodes/0f9cc6b0086d0b09cec6cf268e9a8ed2%d.mp4\" -hls_playlist_type vod -hls_list_size 0 -y \"/cache/transcodes/0f9cc6b0086d0b09cec6cf268e9a8ed2.m3u8\""
[2024-08-21 22:33:55.679 +00:00] [INF] [37] MediaBrowser.MediaEncoding.Transcoding.TranscodeManager: Transcoding kill timer stopped for JobId "f4dc7ae673824a19ae56f7d9f2ed48a0" PlaySessionId "a570164dff57461ab84c007dd53b409b". Killing transcoding
[2024-08-21 22:33:55.679 +00:00] [INF] [37] MediaBrowser.Controller.MediaEncoding.TranscodingJob: Stopping ffmpeg process with q command for "/cache/transcodes/b4fdee81f93b7b70535705116a206f4f.m3u8"
[2024-08-21 22:33:55.835 +00:00] [INF] [37] MediaBrowser.MediaEncoding.Transcoding.TranscodeManager: FFmpeg exited with code 0
[2024-08-21 22:33:55.836 +00:00] [INF] [37] MediaBrowser.MediaEncoding.Transcoding.TranscodeManager: Deleting partial stream file(s) "/cache/transcodes/b4fdee81f93b7b70535705116a206f4f.m3u8"
[2024-08-21 22:37:49.432 +00:00] [INF] [23] Emby.Server.Implementations.HttpServer.WebSocketManager: WS "192.168.1.124" request
[2024-08-21 22:37:49.722 +00:00] [INF] [23] Emby.Server.Implementations.HttpServer.WebSocketManager: WS "192.168.1.124" closed


The closing messages are just my starting/stopping the video for testing.

[TheDreadPirate]

That message appears only once in the Jellyfin server code with this comment.

Code:

// FIXME: this is actually a workaround, as ideally it really should be the client which decides whether non-keyframe
//        breakpoints are supported; but current implementation always uses "ffmpeg input seeking" which is liable
//        to produce a missing part of video stream before first keyframe is encountered, which may lead to
//        awkward cases like a few starting HLS segments having no video whatsoever, which breaks hls.js

The code comment indicates it is a workaround to prevent something else from breaking.

It can be safely ignored.