0 byte recording in mythtv after latest round of updates

[jhoyt4]

My working mythtv config stopped working after this latest round of updates. Here is the log file entry:

hauppauge2-E585-00-00D642B7.8095.log
2020-05-19T08:39:11.406560 CRIT <main> Logger.cpp:83 (setLogLevelFilter) Changing loglevel to NOTC
2020-05-19T08:39:11.406607 CRIT <main> hauppauge2.cpp:347 (main) Starting up
2020-05-19T08:39:11.415937 CRIT <main> hauppauge2.cpp:360 (main) Initializing [Bus: 3, Port: 1] E585-00-00D642B7
2020-05-19T08:39:30.505425 WARN <HAPI-cmd> Hapi.cpp:383 (cmdThreadFunc) pMsgWaitingRsp=0x7f5b9c005c90 waiting for 2647ms(limit=5000)
2020-05-19T08:39:31.637447 WARN <HAPI-cmd> Hapi.cpp:383 (cmdThreadFunc) pMsgWaitingRsp=0x7f5b9c005c90 waiting for 3779ms(limit=5000)
2020-05-19T08:39:32.769335 WARN <HAPI-cmd> Hapi.cpp:370 (cmdThreadFunc) SEVERE: pMsgWaitingRsp=0x7f5b9c005c90 appears stuck after 4911ms
2020-05-19T08:39:32.769426 ERRR <HAPI-cmd> Hapi.cpp:380 (cmdThreadFunc) SEVERE: pMsgWaitingRsp=0x7f5b9c005c90 removed from queue
2020-05-19T08:39:33.293754 ERRR <> Hapi.cpp:3254 (HAPI_Stop) HWAPI_MESSAGE_STOP fail!!! sem err:3, rspCode:0x01000000
2020-05-19T08:39:33.293864 ERRR <> CHapi.cpp:2829 (hapiStop) hapiStop() encoder returned hapiErr=0x1
2020-05-19T08:39:33.293894 WARN <> CHapi.cpp:1218 (hapiResetHW) hapiResetHW() gHWapiHandle=0x7f5b9c003dd0 gHapi=0x7f5b9c003e90 hapiErr=1 _API_ok=1

If I run /opt/Hauppauge/bin/hauppauge2 -s E585-00-00AF4321 -a 1 -d 2 -o /tmp/test.ts I can record video.

Any suggestions on what changed / how to fix?

Here's the config file (added the .txt to get it to upload):
colossus2-1.conf.txt

Myth runs the following command
/opt/Hauppauge/bin/hauppauge2 -c /opt/Hauppauge/etc/colossus2-1.conf

[jpoet]

I will replicate your configuration and try it later today.
Note that you should no longer have to use S/PDIF along side HDMI, and should ble able to use "-a 3" or "--audio 3".
I have occasionally had it fail the way you describe, and often it is because another copy of the hauppauge2 code is still running, but not always. Other times it can be some sort of source issue, but you say you can run it from the command line, so that is probably not it.

[jhoyt4]

Thanks John, I just tried the --audio 3 and it's still broken from myth-based recording. I did reboot just before trying to record and verified hauppauge2 was not running with ps.

An interesting thing I'll note is that even after stopping the recording, hauppauge2 now persists as a process until I shutdown myth-backend (even after trying to kill with kill-9).

[mdabbs]

Is it the "override-loglevel" setting in the config file? Since the logging changed? Just a guess.

[jpoet]

I think I found the problem and have pushed a fix. Please verify and let me know.

[jhoyt4]

That worked! Thank you for the quick turn.

Now to get rid of that S/PIDF cable...