Re: [orca-list] Orca not starting under 4.17.2 kernel

[Janina Sajka <janina rednote net>]

Thanks, Peter, All, for your suggestions and your reports of success
using the latest revs.

I can now confidently agree the issue is neither in the kernel nor in
pulseaudio as downgrading each of these did nothing--at least on Machine
1. As of this morning Machine 1 is working with Orca, though I can't
explain why. I do know I tried various things, including moving to
speech-dispatcher-git, which I am now usin. As of this morning Machine 1
is working with Orca, though I can't explain why. I do know I tried
various things, including moving to speech-dispatcher-git, which I am
now using. .  I was unable however, to get espeak-ng to build, but in
the course of trying ended up reinstalling espeak itself.

Peter, I agree that the first step in these kinds of problems should be
to ascertain the card is actually sounding. I keep a .wav file in my
$HOME just for this purpose, actually. I prefer to hear a few notes from
it as opposed to listening to test tones! <grin>

What's up on Machine 2 remains a mystery. I tried not killing Speakup
after issuing startx to see whether that might yield some clue and found
a lot of xkb error messages that seemed possiblyto even be looping. So,
I don't know. It could be something relating to drivers on Apple
hardware far away from Orca and the audio systems. More investigation
needed.

I would note that spd-say is working on Machine 2, so the problem may
indeed be the graphical environment itself.

Thanks again.

Janina


Peter Vágner writes:
> Hello,
>
>
> On the first machine speech-dispatcher is not working thus it would be
> helpfull to see its log if possible.
>
> On the second machine gstreamer is failing.
>
> Given the fact gstream is not considered hard dependency of orca we should
> try to patch such a condition to fail gracefully without gstreamer specific
> features.
>
>
> I don't have a clue on what's going on. I am suspecting this might be a
> kernel driver change possibly related to some apple specific hardware.
>
>
> In order to try to troubleshoot it further have you tried some more apps
> producing audio to see what these will end up with? For example speaker-test
> from alsa-utils?
>
>
>
> Greetings
>
>
> Peter
>
>
>
>
>
> Dňa 21.06.2018 o 19:56 Janina Sajka via orca-list napísal(a):
> > Hi, Alex:
> >
> > I'm not sure the problem is the kernel, but neither am I convinced it's
> > the new pulseaudio. Recalling I am failing on two machines, let me again
> > point out that Machine 1 has a failing speech-dispatcher. My
> > speechd.conf has the following two settings:
> >
> >   AudioOutputMethod "alsa"
> >   AudioALSADevice "plughw:1,0"
> >
> > Machine 2, the Apple Airbook is a very new installation where I'm trying
> > hard to stay as close as possible to installation defaults.
> > Consequently, my speechd.conf is unedited and defaults to pulseaudio. On
> > this machine spd-say works.
> >
> > There is an additional difference on Machin 2, however. I'm running
> > speech-dispatcher-git on this machine in order to avail myself of
> > espeak-ng. So,
> >
> > In both machines I'm also running orca-git, so I guess I'm not fully
> > default install.
> >
> > Let me know anything else I should try. Thanks for your interest and
> > your help.
> >
> > Janina
> >
> >
> > Alex ARNAUD writes:
> > > Are you sure it's related to kernel ? In the last day, PulseAudio 12.0 has
> > > been released. The reported error seems related to PulseAudio.
> > >
> > > Can you try to revert PulseAudio and let me know ?
> > >
> > > Best regards,
> > > Alex.
> > >
> > > Le 21/06/2018 à 17:44, Janina Sajka via orca-list a écrit :
> > > > Hi,
> > > >
> > > > In the past 24 hours I've updated two Arch installations to latest
> > > > release packages which includes the 4.17.2 kernel. In both cases I'm
> > > > unable to get Orca to run, though the errors are a bit different in each
> > > > case. I suspect I'm going to resolve this by down versioning my
> > > > kernel--at least on one machine.
> > > >
> > > > Here's what I get ...
> > > >
> > > > Machine 1 gets speech-dispatcher errors, so that spd-say does not work.
> > > > Attempting to restart Orca shows:
> > > > $export DISPLAY=:0
> > > > janina@concerto 11:32:24 hwdev$orca -r
> > > >
> > > > Traceback (most recent call last):
> > > >     File "/usr/lib/python3.6/site-packages/orca/speech.py", line 106, in
> > > >     init
> > > >         settings.speechServerInfo)
> > > >           File "/usr/lib/python3.6/site-packages/orca/speech.py", line 95,
> > > >         in _initSpeechServer
> > > >             raise Exception("ERROR: No speech server for factory: %s" %
> > > >             moduleName)
> > > >             Exception: ERROR: No speech server for factory:
> > > >             orca.speechdispatcherfactory
> > > >
> > > >             During handling of the above exception, another exception
> > > >             occurred:
> > > >
> > > >             Traceback (most recent call last):
> > > >               File "/usr/lib/python3.6/site-packages/orca/speech.py",
> > > >               line 112, in init
> > > >                   _initSpeechServer(moduleName, None)
> > > >                     File
> > > >                     "/usr/lib/python3.6/site-packages/orca/speech.py",
> > > >                     line 95, in _initSpeechServer
> > > >                         raise Exception("ERROR: No speech server for
> > > >                         factory: %s" % moduleName)
> > > >                         Exception: ERROR: No speech server for factory:
> > > >                         speechdispatcherfactory
> > > >
> > > > Machine 2: spd-say works as expected. This is on a late 2013 vintage
> > > > Apple Airbook running Arch off a USB 3.0 flash drive.
> > > > $export DISPLAY=:0
> > > > janina@toccata 11:42:40 ~$orca -r
> > > > Traceback (most recent call last):
> > > >     File "/usr/sbin/orca", line 44, in <module>
> > > >         from orca import orca
> > > >           File "/usr/lib/python3.6/site-packages/orca/orca.py", line 81,
> > > >         in <module>
> > > >             from . import sound
> > > >               File "/usr/lib/python3.6/site-packages/orca/sound.py",
> > > >               line 163, in <module>
> > > >                   _player = Player()
> > > >                     File
> > > >                     "/usr/lib/python3.6/site-packages/orca/sound.py",
> > > >                     line 55, in __init__
> > > >                         self.init()
> > > >                           File
> > > >                           "/usr/lib/python3.6/site-packages/orca/sound.py",
> > > >                           line 123, in init
> > > >                               self._pipeline.add(self._sink)
> > > >                               TypeError: Argument 1 does not allow None
> > > >                               as a value

-- 

Janina Sajka

Linux Foundation Fellow
Executive Chair, Accessibility Workgroup:       http://a11y.org

The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI)
Chair, Accessible Platform Architectures        http://www.w3.org/wai/apa