[Expired for pulseaudio (Ubuntu) because there has been no activity for
60 days.]
** Changed in: pulseaudio (Ubuntu)
Status: Incomplete => Expired
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/
I spoke too soon. It crashed out twice in a row on me, the second
crash within five minutes of the first. I have attached two new log
files showing this.
I wonder, could the runaway speech-dispatcher process cause the sound
input to crash sooner the second time? Usually when it crashes, I
unplu
Yes, it looks like tsched=0 might have lengthened the time between
crashes.
I bet speech-dispatcher is being used by Mumble, which has
text-to-speech messaging capabilities. It is possible that killing
pulseaudio, as I do after an input crash, is causing speech-dispatcher
to go nuts.
--
pulseau
So tsched=0 improved the situation, but did not really eliminate it?
Speech-dispatcher is something text-to-speech related, sometimes used by
visually impaired users to play back text. Are you running it on purpose
and do you know what is triggering it? Perhaps Luke can fill in a bit on
this issue
Well it took a while, but I finally managed to get a verbose log of the
voice input crashing while "tsched=0" has been set.
I have also noticed that after the crash a there is a runaway "speech-
dispatcher" process hanging around. speech-dispatcher runs up to 100%
CPU utilization, and stays there
Hmm. This seems like a difficult one to trace down. I see nothing
obvious in the PulseAudio logs, and I'm still unsure exactly where in
the logs that the problem actually happens. My guess is that it is
hardware or driver related.
Perhaps tsched=0 could help as a workaround? To try that, replace t
Here is another verbose log file, but this time I unplugged the headset
*before* terminating the log program. Perhaps the file will still be of
use.
** Attachment added: "Pulseaudio detailed log #2"
http://launchpadlibrarian.net/44337832/pulseverbose.log.1
--
pulseaudio input crashes after a
Here is a verbose pulseaudio log file using the instructions you posted.
The first file is fine: I stopped logging, then unplugged the headset.
** Attachment added: "Pulseaudio detailed log"
http://launchpadlibrarian.net/44337671/pulseverbose.log.0
--
pulseaudio input crashes after an hour o
Right, so the strange pulseaudio errors come from unplugging the headset rather
than the input silence.
Hmm, perhaps a pulseaudio log would help? https://wiki.ubuntu.com/PulseAudio/Log
Please stop the logging after the input has stopped working, but before you
physically disconnect the headset.
Sound output through the both the USB headphones and onboard speakers
still works fine, even after the input via the USB headset microphone
has crashed.
Just to be clear, the "USB disconnected/connected" messages in the log I
posted are me physically unplugging and replugging the headset in order
It looks like your computer thinks that the headset has become
disconnected. Can we rule out hardware errors, by e g testing the
headset in another computer?
Also, you say that sound output continues to work - is that sound output
through the headset or through the on-board sound card?
** Changed
** Description changed:
Binary package hint: pulseaudio
- Pulseaudio sound input will crash after an hour of continuous use, such
- as a Skype call, or Mumble voice chat. On the Input tab of the Sound
- Preferences panel, all of the controls are normal, but the Input Level
- meter no longer
** Attachment added: "Excerpt of input crash cycling from /var/log/syslog"
http://launchpadlibrarian.net/44185306/pulseaudio-input-crash-syslog.txt
** Attachment added: "AlsaDevices.txt"
http://launchpadlibrarian.net/44182908/AlsaDevices.txt
** Attachment added: "AplayDevices.txt"
http:
13 matches
Mail list logo