@tmm1 The only real issue(s) I have remaining in CC4C are:
- Multi-hour shows/movies recordings will fail after three hours with an inactivity error from the website with an error like “due to inactivity, playback has been stopped. To resume playback, reload the page.” Is there any way to simulate some activity to stop the sites from timing us out?
To replicate: record a 4 hour block on any NBC station…wait for hour 3 to begin…timeout will occur.
- The recordings themselves are sometimes flaky - there’s no consistency to when issues occur. I’ll occasionally see some errors like:
2023/07/19 07:45:21.411631 [HLS] ffmpeg: file30372-89d927ab5889: [h264 @ 0x158e08370] vt decoder cb: output image buffer is null: -12909
2023/07/19 07:45:21.411703 [HLS] ffmpeg: file30372-89d927ab5889: [h264 @ 0x158e33750] hardware accelerator failed to decode picture: Unknown error occurred
Though I’ve seen others as well like:
2023/07/17 21:00:00.381731 [HLS] ffmpeg: chrome-Bravo (East): [matroska,webm @ 0x152f04770] Invalid length 0x72 > 0x8 for element with ID 0xE7 at 0xad4224a2
2023/07/17 21:00:00.383188 [HLS] ffmpeg: chrome-Bravo (East): [matroska,webm @ 0x152f04770] Seek to desired resync point failed. Seeking to earliest point available instead.
2023/07/17 21:00:04.766938 [HLS] ffmpeg: chrome-Bravo (East): [matroska,webm @ 0x152f04770] Unknown element 89 at pos. 0xad7340c6 with length 0x5acdf23 considered as invalid data. Last known good position 0xad72ac8f, 1 unknown elements in a row
2023/07/17 21:00:04.766962 [HLS] ffmpeg: chrome-Bravo (East): [matroska,webm @ 0x152f04770] Seek to desired resync point failed. Seeking to earliest point available instead.
2023/07/17 21:05:41.793187 [HLS] ffmpeg: chrome-Bravo (East): [matroska,webm @ 0x152f04770] Unknown element 643E at pos. 0xbffab3d9 with length 0x17fba3 considered as invalid data. Last known good position 0xbff9d720, 2 unknown elements in a row
2023/07/17 21:05:41.793450 [HLS] ffmpeg: chrome-Bravo (East): [matroska,webm @ 0x152f04770] Seek to desired resync point failed. Seeking to earliest point available instead.
2023/07/17 21:08:00.093377 [ERR] Error during stream M3U-ChromeCaptureforChannels ch7005 Bravo (East): read |0: file already closed
2023/07/17 21:08:00.093409 [TNR] Closed connection to M3U-ChromeCaptureforChannels for ch7005 Bravo (East)
2023/07/17 21:08:00.131927 [HLS] ffmpeg: chrome-Bravo (East): av_interleaved_write_frame(): Broken pipe
2023/07/17 21:08:00.132653 [HLS] ffmpeg: chrome-Bravo (East): Error writing trailer of pipe:: Broken pipe
2023/07/17 21:08:00.132666 [HLS] ffmpeg: chrome-Bravo (East): Error closing file pipe:: Broken pipe
My setup is an M2 Mac mini, using hardware acceleration and HEVC when streaming remote, running the latest beta DVR server.
Thoughts? I suspect the first issue (recordings timing out due to inactivity) is reasonably straightforward to address…but who knows.