Recordings ending prematurely

I've send in some logs c5eeb94e-c3b6-4423-bc48-ee63a71acf67.

2021/05/21 00:00:02.764116 [TNR] Opened connection to M3U-XfinityBackend for ch1405 AMC
2021/05/21 00:00:02.764634 [DVR] Recording for job 1621576800-ch1405 from M3U-XfinityBackend ch1405 into "Movies/Tombstone (1993) 2021-05-21-0000.mpg" for 2h59m59.994100707s
2021/05/21 00:00:03.101043 [IDX] Generating video index for job 1621576800-ch1405
2021/05/21 00:15:49.002736 [TNR] Closed connection to M3U-XfinityBackend for ch1405 AMC
2021/05/21 00:15:49.002762 [DVR] Job 1621576800-ch1405 Tombstone (1993) ended prematurely: 2h44m10.997240706s

This sequence keeps repeating in the logs.

Backend logs 13f2e15f-b6e0-42c0-979e-e07c435fe084

2021/05/20 23:00:02.761186 [TNR] Opened connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/20 23:15:49.001893 [TNR] Closed connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/20 23:15:50.422418 [TNR] Opened connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/20 23:17:19.749210 [TNR] Closed connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/20 23:17:20.346201 [TNR] Opened connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/20 23:18:17.619643 [TNR] Closed connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/20 23:18:18.126516 [TNR] Opened connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/21 00:04:14.874482 [TNR] Closed connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/21 00:04:16.352928 [TNR] Opened connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/21 00:51:14.960864 [TNR] Closed connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/21 00:51:18.530949 [TNR] Opened connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/21 00:52:28.537282 [TNR] Closed connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/21 00:52:29.330185 [TNR] Opened connection to TVE-Comcast_SSO for ch6086 AMC
2021/05/21 02:00:03.807504 [TNR] Closed connection to TVE-Comcast_SSO for ch6086 AMC

What about the logs from the original stream, including the MTS lines?

It looks like you're feeding Channels some intermediate source, but omitting the logs from that intermediate source. When you use Channels in novel and unique ways, you're making it more difficult to offer support, and things breaking are more likely to happen.

Mmm, good point. Let me get a hold of the backend system. It's just another channels instance.

Yep, Interrupted TVE recordings, I was wondering what would happen...

P.S. Looks like your two servers are set to two different time zones as your frontend server time is an hour ahead of your backend server time.

1 Like

I'll look into it! Thanks.

baa9f5c4-8931-4004-b2b4-693cdab957a9

I've submitted another set of logs from my docker backend instance. Lots of different errors, but it's making the recording super unstable.

So I did a test recording using regular TVE instead of my split setup. I'm still seeing the recording ending prematurely error. So I don't think this has to do with the split setup itself.

I've submitted diagnostics again for this particular recording. 622117bc-406d-4a81-a53c-5a6e8f3342cc This is also on version 2021.05.21.0113 of the DVR. Says up do date when I check for update.

2021/05/24 07:46:29.806574 [TNR] Closed connection to TVE-Comcast_SSO for ch6082 FXX
2021/05/24 07:46:29.806603 [DVR] Job 1621863000-ch6082 Home (2015) ended prematurely: 1h43m30.193401925s
2021/05/24 07:46:29.899725 [DVR] Starting job 1621863000-ch6082 Home (2015) on ch=[6082 1410.5]
2021/05/24 07:46:31.391438 [TVE] stream timestamps: fxx: start_at=2021-05-24T07:45:35-06:00 current_at=2021-05-24T07:45:55-06:00 end_at=2021-05-24T07:46:03-06:00
2021/05/24 07:46:31.392423 [TNR] Opened connection to TVE-Comcast_SSO for ch6082 FXX
2021/05/24 07:46:31.392519 [DVR] Recording for job 1621863000-ch6082 from TVE-Comcast_SSO ch6082 into "Movies/Home (2015) 2021-05-24-0730.mpg" for 1h43m30.10024589s

So it looks like it stops for 1.5s. Based upon my own personal experiences, this is most likely an error in the network website's stream having a missing segment, failing to send the next segment, or even perhaps segments coming from different CDNs.

However, the quick re-establishment of the stream seems to me like this is more a network/external issue, and not really something Channels has control over.

Looks like internet blip around 7:59 with multiple repeated timeout errors

err="dial tcp: i/o timeout" time="10.002814904s"

2 Likes

Ah, I obviously missed that. My bad. Time to start looking at the reasons that might be happening.

1 Like