[mythtv] MythTV audio initialization
Warpme
warpme at o2.pl
Wed Nov 7 19:45:08 UTC 2012
On 11/7/12 9:51 AM, Greg G. wrote:
> Hello,
>
> I'm seeing the following strange effect:
>
> - system starts from S3 suspend.
> - NVidia X server still running ok
> - cat /proc/asound/card#0/eld lists an invalid ELD
> - speaker-test runs, but produces no sound
> - cat /proc/asound/card#0/eld still invalid
> - now I start mythfrontend, but no playback
> - now /proc/asound/card#0/eld is valid!
>
> So the good news is there is no mythtv bug, actually more of a feature!
>
> But I would really like to know what magic it is that mythfrontend does
> here. Can I reproduce that on the command line?
>
> Reason I'm asking is that I would like to keep mythfrontend running
> while suspending to S3 to keep the start-up time shorter. But after
> wakeup I want a working audio device to be able to run non-mythtv PCM
> based applications. Any magic alsctl incantation I could try?
>
> Thanks for any insights!
>
Greg,
I have bad news for You: while it is fully possible to keep mythfrontend
running across s3 cycles I believe You will quit this idea (as I'm) when
You discover that BE->FE event connection is not surviving when sleep is
long enough (over 30min I believe). User notices this as i.e.:
-lack of EPG program rec.status changes when You programing given
program to record on EPG.
-sockets timeouts in BE (in 0.25 leading quickly to BE deadlocks)
During FE sleep BE reports things like this:
2012-11-07 19:22:09.396113 E MythSocket(2645fc0:133): writeStringList:
Error, No data written on writeBlock (882 errors)
starts with: 1389 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE
UPDATE[]:[...
2012-11-07 19:22:10.398837 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (881 errors)
starts with: 1382 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE
UPDATE[]:[...
2012-11-07 19:22:11.401671 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (882 errors)
starts with: 770 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE
UPDATE[]:[...
2012-11-07 19:22:12.404289 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 764 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE
UPDATE[]:[...
2012-11-07 19:22:13.408262 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 829 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE
UPDATE[]:[...
2012-11-07 19:22:14.412064 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 1469 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE
UPDATE[]:[...
2012-11-07 19:22:15.414907 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 1022 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE
UPDATE[]:[...
2012-11-07 19:22:16.418580 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 1463 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE
UPDATE[]:[...
2012-11-07 19:22:17.419073 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (879 errors)
starts with: 84 BACKEND_MESSAGE[]:[]UPDATE_FILE_SIZE 21101
2012-11-0...
2012-11-07 19:22:18.421166 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 85 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE DELETE
238...
2012-11-07 19:22:19.423220 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (879 errors)
starts with: 85 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE DELETE
238...
2012-11-07 19:22:20.424839 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (878 errors)
starts with: 84 BACKEND_MESSAGE[]:[]UPDATE_FILE_SIZE 23809
2012-11-0...
2012-11-07 19:22:20.697378 I Reschedule requested for MATCH 11169 0 0
- SaveRule Komisarz Alex 2
2012-11-07 19:22:21.427100 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (846 errors)
starts with: 84 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE DELETE
804...
2012-11-07 19:22:22.099349 I Scheduled 450 items in 1.4 = 0.01 match
+ 0.00 check + 1.35 place
2012-11-07 19:22:22.960106 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 84 BACKEND_MESSAGE[]:[]UPDATE_FILE_SIZE 21101
2012-11-0...
2012-11-07 19:22:23.961688 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 70 BACKEND_MESSAGE[]:[]SYSTEM_EVENT SCHEDULER_RAN
SENDE...
2012-11-07 19:22:24.962162 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (875 errors)
starts with: 45 BACKEND_MESSAGE[]:[]SCHEDULE_CHANGE[]:[]empty
2012-11-07 19:22:25.965975 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (879 errors)
starts with: 85 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE DELETE
238...
2012-11-07 19:22:27.849471 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 84 BACKEND_MESSAGE[]:[]UPDATE_FILE_SIZE 23809
2012-11-0...
2012-11-07 19:22:29.417627 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 85 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE DELETE
238...
2012-11-07 19:22:32.962204 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (879 errors)
starts with: 84 BACKEND_MESSAGE[]:[]UPDATE_FILE_SIZE 21101
2012-11-0...
2012-11-07 19:22:33.964137 E MythSocket(2645fc0:133):
writeStringList: Error, No data written on writeBlock (880 errors)
starts with: 85 BACKEND_MESSAGE[]:[]RECORDING_LIST_CHANGE DELETE
238...
Fortunately it seems like BE is not deadlocking on 0.26 by errors like
above, but event connection is still present :-(
By above issues I have to exit/reload mythfrontend across S3 cycles on
all my FEs. With such reload 0.26 is ROCK STABLE (1800rec so far, no
single 0B rec nor deadlock).
I believe MythTV hasn't yet full implementation for s3 support (I mean
it is not reacting for system s3 sleep/s3 resume events by
closing/reopening network connections).
I wish it will be fully implemented someday.
There was even ticket for this (#7847) - but it was closed -
unfortunately rather by lack of full solution than by full resolution...
br
-------------- next part --------------
A non-text attachment was scrubbed...
Name: warpme.vcf
Type: text/x-vcard
Size: 83 bytes
Desc: not available
URL: <http://www.mythtv.org/pipermail/mythtv-dev/attachments/20121107/78b710d0/attachment.vcf>
More information about the mythtv-dev
mailing list