Stuttering/Skipping and Interruption during recordings this week

[gsulshski]

I have a mac mini has my dvr and have been using it for several years now. I am using a homerun prime and usually have a strong and consistent signal/connection.

I have had a couple of recordings this week that have been problematic and I examined the log file and see various warnings and errors.

for one TV show recording ...

2019/09/26 20:41:21 [WRN] Buffer for 1316952D ch786 is more than 50% full (clients=1, len=16777684)
2019/09/26 20:45:28 [WRN] Buffer for 1316952D ch786 is more than 50% full (clients=1, len=16777684)
2019/09/26 20:45:44 [WRN] Buffer for 1316952D ch786 is more than 50% full (clients=1, len=16777684)
2019/09/26 20:45:48 [WRN] Buffer for 1316952D ch786 is more than 75% full (clients=1, len=25165868)
2019/09/26 20:47:14 [WRN] Buffer for 1316952D ch786 is more than 50% full (clients=1, len=16777684)
2019/09/26 20:47:20 [WRN] Buffer for 1316952D ch786 is more than 75% full (clients=1, len=25165868)
2019/09/26 20:47:53 [WRN] Buffer for 1316952D ch786 is more than 50% full (clients=1, len=16777684)
2019/09/26 20:48:00 [WRN] Buffer for 1316952D ch786 is more than 75% full (clients=1, len=25165868)
2019/09/26 20:48:05 [WRN] Buffer for 1316952D ch786 is more than 95% full (clients=1, len=31877468)
2019/09/26 20:48:06 [WRN] Buffer for 1316952D ch786 is more than 99% full (clients=1, len=33219788)
2019/09/26 20:49:45 [WRN] Buffer for 1316952D ch786 is more than 50% full (clients=1, len=16777684)
2019/09/26 20:49:51 [WRN] Buffer for 1316952D ch786 is more than 75% full (clients=1, len=25165868)
2019/09/26 20:49:55 [WRN] Buffer for 1316952D ch786 is more than 95% full (clients=1, len=31877468)
2019/09/26 20:49:56 [WRN] Buffer for 1316952D ch786 is more than 99% full (clients=1, len=33219788)
2019/09/26 20:51:23 [WRN] Buffer for 1316952D ch786 is more than 50% full (clients=1, len=16777684)
2019/09/26 20:51:28 [WRN] Buffer for 1316952D ch786 is more than 75% full (clients=1, len=25165868)
2019/09/26 20:51:34 [WRN] Buffer for 1316952D ch786 is more than 95% full (clients=1, len=31877468)
2019/09/26 20:51:35 [WRN] Buffer for 1316952D ch786 is more than 99% full (clients=1, len=33219788)

For a football game.. I had an interruption...

2019/09/27 19:40:50 [WRN] Buffer for 1316952D ch811 is more than 50% full (clients=1, len=16777684)
2019/09/27 19:42:23 [WRN] Buffer for 1316952D ch811 is more than 50% full (clients=1, len=16777684)
2019/09/27 19:46:17 [WRN] Buffer for 1316952D ch811 is more than 50% full (clients=1, len=16777684)
2019/09/27 19:56:57 [WRN] Buffer for 1316952D ch811 is more than 50% full (clients=1, len=16777684)
2019/09/27 19:57:58 [ERR] Error during stream 1316952D ch811 Fox Spo: read tcp 192.168.1.19:65414->192.168.1.3:5004: read: connection reset by peer
2019/09/27 19:58:05 [TNR] Closed connection to 1316952D for ch811 Fox Spo
2019/09/27 19:58:20 [DVR] Job 1569628800-11 College Football ended prematurely: 3h1m39.971168s
2019/09/27 19:58:50 [DVR] Starting job 1569628800-11 College Football on ch=[6197 811 420 33]

Of course, I can send you the log if you give me some directions on how to do that.

[tmm1]

Buffer full errors while recording mean some sort of resource exhaustion is happening on your system. Usually it is because the disk isn't keeping up with the write speeds required, either cuz its getting full or its starting to fail.

Check the S.M.A.R.T status for your disk.

Also recommend checking Activity Monitor to make sure there are no runaway processes using CPU.

[gsulshski]

I am currently getting similar messages this afternoon at 50 to 95% on a recording...

2019/09/28 16:24:16 [DVR] Deleting /Volumes/MacIntosh HD 2/DVR/TV/Women's College Volleyball/Women's College Volleyball Michigan State at Penn State 2019-09-25-1900.mpg
2019/09/28 16:30:31 [WRN] Buffer for 1316952D ch801 is more than 50% full (clients=1, len=16777684)
2019/09/28 16:30:43 [WRN] Buffer for 1316952D ch801 is more than 50% full (clients=1, len=16777684)
2019/09/28 16:30:48 [WRN] Buffer for 1316952D ch801 is more than 75% full (clients=1, len=25165868)
2019/09/28 16:30:58 [WRN] Buffer for 1316952D ch801 is more than 95% full (clients=1, len=31877468)
2019/09/28 16:30:59 [WRN] Buffer for 1316952D ch801 is more than 99% full (clients=1, len=33219788)
2019/09/28 16:31:02 [WRN] Buffer for 1316952D ch801 is more than 50% full (clients=1, len=16777684)
2019/09/28 16:32:00 [WRN] Buffer for 1316952D ch801 is more than 50% full (clients=1, len=16777684)

I check the SMART status of the two drives (one for the OS and one for the recording) and they are verified.

I have 80ish GB available of a 1.5TB drive available and the other drive has 250GB free out of 1TB.

I updated to the latest MacOS yesterday and there is no rogue process this afternoon. The DVR was running about 30% for one of the four cores. Otherwise, not much going on. Have 16GB of memory and about 50% free.

So, I am scratching my head on what to look at.

Also, I see that one of the early programs recorded this week has corrupted data when played

019/09/28 06:58:20 [ENC] Starting encoder for A Million Little Things S02E01 2019-09-26 Coming Home 2019-09-26-2001.mpg in /Volumes/MacIntosh HD 2/DVR/Streaming/file2149-ipfe80--26-a566-1f4c-1528-en0/encoder-0-913174085 at 0 (0.000000) (encoder=h264_videotoolbox, resolution=720, deinterlacer=blend, bitrate=9488 segment_size=1)
[mpegts @ 0x7ff576001200] Dropped corrupted packet (stream = 1)
[mpegts @ 0x7ff576001200] Dropped corrupted packet (stream = 2)
2019/09/28 06:58:22 [ENC] Request for 309 is more than 10 from the last encoded segment of 6
2019/09/28 06:58:22 [ENC] Stopped encoder for A Million Little Things S02E01 2019-09-26 Coming Home 2019-09-26-2001.mpg in /Volumes/MacIntosh HD 2/DVR/Streaming/file2149-ipfe80--26-a566-1f4c-1528-en0/encoder-0-913174085 after encoding 0 to 7
2019/09/28 06:58:22 [ENC] Starting encoder for A Million Little Things S02E01 2019-09-26 Coming Home 2019-09-26-2001.mpg in /Volumes/MacIntosh HD 2/DVR/Streaming/file2149-ipfe80--26-a566-1f4c-1528-en0/encoder-309-846300262 at 309 (308.308000) (encoder=h264_videotoolbox, resolution=720, deinterlacer=blend, bitrate=9488 segment_size=1)
[mpegts @ 0x7faa19801200] Dropped corrupted packet (stream = 1)
[mpegts @ 0x7faa19801200] Dropped corrupted packet (stream = 2)
[mpegts @ 0x7faa19801200] Dropped corrupted packet (stream = 1)

[tmm1]

What version of macOS?

Is there some sort of disk speed test you can run? What kind of disk speed does Activity Monitor show while recording?

[gsulshski]

Mac OS 10.14.6

While recording, I see a consistent 1.65MB/s with peaks of 4ishMB/s and occasionally in the 678KB/s.

2019/09/28 21:54:19 [IDX] Generating video index for job 1569724200-ch800
2019/09/28 21:54:34 [WRN] Buffer for 1316952D ch800 is more than 50% full (clients=1, len=16777684)

[gsulshski]

I downloaded BlackMagic Disk speed test

Here are my results... the first one is the internal one I am using and I added the other two external ones.

[chDVRuser]

Are both the Mac Mini and the Prime connected to the same Gigabit network switch?
If not, it would be worth a try.

[gsulshski]

From a network perspective, I have an Apple Extreme (hooked to the wan) feeding directly to the mac mini while I have the homerun prime hooked to another apple extreme. It would be a pain for me to move connections around.

I am assuming that the warning message means that the mac mini can't keep up with the recording bandwidth vs the other way around. Maybe I am wrong in my assumption of what the client is.

This morning, I restarted the DVR via the interface and did three simultaneous recordings and I haven't seen any warning messages.

[djcastaldo]

I have an Apple Extreme (hooked to the wan) feeding directly to the mac mini while I have the homerun prime hooked to another apple extreme. It would be a pain for me to move connections around.

This is certainly going to cause issues. Hopefully, it's less of a pain to run a wire than it is to get messed up recordings.

[gsulshski]

One thing I have changed in the last couple of weeks...

I have added TV everything and have clicked on the new guide database .

I am not sure how much resources that the guide database has to deal with but this might be overkill configuration that I have.