error on StartStream

[SteveTyrakowski]

This has been an intermittent issue where some recordings have started late.  It seems that the server is trying to start a stream but fails and then goes on the another tuner to try.  My first 4 tuners are on an internal hauppauge quad tuner card and the last is an HDHomerun tuner.  I am thinking that the internal card tuners are all failing when this happens because right now I have 2 recordings scheduled but only one eventually started running 5 minutes late.

When I try to look at device status in the web interface it says no devices active even though it is recording.

To complicate things, locally in Chicago, the Willis tower where several local stations have their antenna has been without power for several days due to flooding in its basement, so I have had failed recordings due to that so it has muddied the water a little.  

Here is a portion of the log:

2020-05-22 14:29:00.066 [DEBUG][8] DefaultDirectoryChoice: spread
2020-05-22 14:29:00.329 [DEBUG][8] allocating recording target filename: /recordings/Manual Recordings/Pritzker_20200522_14301540.ts
2020-05-22 14:29:00.332 [DEBUG][8] About to start recording (166132 on 23): /recordings/Manual Recordings/Pritzker_20200522_14301540.ts...
2020-05-22 14:29:00.332 [DEBUG][8] (Recording is due to be stopped at 5/22/2020 3:40 PM)
2020-05-22 14:29:00.332 [DEBUG][8] http://localhost:37649/stream/start?slip...4301540.ts
2020-05-22 14:30:40.330 [ERROR][8] Unexpected error on LocalRecorder.StartStream(): System.Net.WebException: The operation has timed out.
  at System.Net.HttpWebRequest.GetResponse()
  at System.Net.WebClient.GetWebResponse(WebRequest request)
  at System.Net.WebClient.DownloadBits(WebRequest request, Stream writeStream)
  at System.Net.WebClient.UploadBits(WebRequest request, Stream readStream, Byte[] buffer, Int32 chunkSize, Byte[] header, Byte[] footer)
  at System.Net.WebClient.UploadDataInternal(Uri address, String method, Byte[] data, WebRequest& request)
  at System.Net.WebClient.UploadString(Uri address, String method, String data)
  at NShared.LocalRecorder.StartStream(ChannelMapping channelMapping, String target, String& reason)
2020-05-22 14:30:40.331 [DEBUG][8] Failed to start recording on allocated device (166132:/recordings/Manual Recordings/Pritzker_20200522_14301540.ts): 'The operation has timed out.'
2020-05-22 14:30:40.331 [DEBUG][8] Looking for second chance device to perform recording...
2020-05-22 14:30:40.343 [DEBUG][8] Found possible backup device
2020-05-22 14:30:40.348 [DEBUG][8] http://localhost:32811/stream/start?slip...4301540.ts
2020-05-22 14:31:31.630 [DEBUG][70] Got request [::ffff:192.168.9.120]: /services/service (recording.list)
2020-05-22 14:31:31.630 [DEBUG][44] Got request [::ffff:192.168.9.120]: /services/service (setting.get)

I can probably clear it up by rebooting, but I don't want to lose the recording that is in process right now.

So as far as anything that might need your attention, the only thing is the web interface not showing device status.  That would help me in diagnosing the hardware if I can see which tuners are actually being used.

As far as the rest, any suggestions as to what might be the problem on my end?  Do you think the tuner is ready to fail?  It is several years old carried over from my old v4 system.  Any ways to automatically  detect the problems ahead of time so I can reboot the box?  I've rebooted several times the last few days with that other issue in the web interface, but I have not done a full power down.  I'm inclined to think it is hardware issue so mostly looking for advice.

Thanks
Steve T

[mvallevand]

Yes this shows two internal tuners but don't bother with snippets send the zipped logs.

I have an internal Haupauge dual tuner device on linux and only one tuner works. Did you copy one tuner or did you scan and merge?

Martin

[SteveTyrakowski]

OK Here is the log.  I tried another recording, and when it failed I powered down and then started up again and it looks like it worked.

And I just tried a new recording that worked immediately on the local tuner and it does show up in the device status page so perhaps it was just waiting for the local tuner and hadn't timed out yet when I didn't see anything earlier.

[mvallevand]

How often does Stephen Colbert record for you it doesn't look very good in these logs? I don't know if it is the tuner but I suspect it is a signal issue.

Martin

[SteveTyrakowski]

How often does Stephen Colbert record for you it doesn't look very good in these logs?  I don't know if it is the tuner but I suspect it is a signal issue.

Martin

That is the issue I mentioned with the Willis Tower (formerly Sears tower, tallest USA building at one point) having no power.  Our CBS station and our CW stations have not been available OTA for about a week now because of it.

Steve

[mvallevand]

Ok looking at the logs for adapter3log.1 that tuner didn't shutdown correctly after a Colbert "fail". Sub might need to look into what happens on DeviceHostLinux in the scenario where it doesn' t lock.

22:34:36:606 [DEBUG] status 00 | quality 00 | signal fffeb010 | snr 0000 | ber 00000000 | unc 00000000 | pre_ber 00000000 | per 00000000 |
22:34:36:654 [INFO] Creating writer for: /recordings/The Late Show With Stephen Colbert/Season 05/The.Late.Show.With.Stephen.Colbert.S05E137.Joe.Biden;.Milky.Chance;.Jack.Johnson.ts
17:10:13:082 [DEBUG] Ctrl-C...shutting down

22:42:57:289 [DEBUG] status 00 | quality 00 | signal fffeb010 | snr 0000 | ber 00000000 | unc 00000000 | pre_ber 00000000 | per 00000000 |
22:42:57:315 [INFO] Creating writer for: /recordings/The Late Show With Stephen Colbert/Season 05/The.Late.Show.With.Stephen.Colbert.S05E136.Keegan-Michael.Key;.Wilco.ts
23:54:22:702 [DEBUG] Ctrl-C...shutting down

22:41:17:173 [DEBUG] status 00 | quality 00 | signal fffeac28 | snr 0000 | ber 00000000 | unc 00000000 | pre_ber 00000000 | per 00000000 |
22:41:17:199 [INFO] Creating writer for: /recordings/The Late Show With Stephen Colbert/Season 05/The.Late.Show.With.Stephen.Colbert.S05E136.Keegan-Michael.Key;.Wilco.ts
23:54:22:702 [DEBUG] Ctrl-C...shutting down

Martin