<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">Thanks Mikel.<br>
<br>
Am 04.06.2013 14:02, schrieb Mikel Astiz:<br>
</div>
<blockquote
cite="mid:CANT-zCV-W51bbxjKcccRu21Yfm9G3ndipaQquGW85fDg89QL4A@mail.gmail.com"
type="cite"><br>
<blockquote type="cite">
<pre wrap="">
parecord -r --device=bluez_source.00_1A_7D_11_C0_66 mm.wav
in my case. Creates a crackling sound in the earpiece and a mm.wav of zero
bytes, no matter how long I record.
</pre>
</blockquote>
<pre wrap="">
I tried to reproduce your issue without success, with two different
headsets. It could be an issue with your specific headset (any chance
you can test a different one?) or more likely, according to your
previous messages, your system is broken with multiple versions of the
same component or an incorrect configuration (PulseAudio, BlueZ).</pre>
</blockquote>
I re-copied a new wheezy-image on my sd-card. Now pactl list sources
short shows the bluez-device. But still, an empty file is the
output. So a broken system/incorrect configuration is unlikely now.<br>
<blockquote
cite="mid:CANT-zCV-W51bbxjKcccRu21Yfm9G3ndipaQquGW85fDg89QL4A@mail.gmail.com"
type="cite">
<pre wrap="">
A possible checklist you could follow is:
1. Make sure BlueZ's audio.conf is properly set. You might need to add
a line with "Disable=Socket".</pre>
</blockquote>
Done.<br>
<blockquote
cite="mid:CANT-zCV-W51bbxjKcccRu21Yfm9G3ndipaQquGW85fDg89QL4A@mail.gmail.com"
type="cite">
<pre wrap="">
2. Make sure the headset is connected (PA card exists) and profile is
set to "hsp".</pre>
</blockquote>
Done.<br>
<blockquote
cite="mid:CANT-zCV-W51bbxjKcccRu21Yfm9G3ndipaQquGW85fDg89QL4A@mail.gmail.com"
type="cite">
<pre wrap="">
3. During recording, make sure the profile's state is set to
"available: yes". You can see this with pacmd list-cards.</pre>
</blockquote>
This is what I get:<br>
<br>
<blockquote><i>pi@raspberrypi ~ $ parecord -r -d
"bluez_source.00_1A_7D_11_C0_66" mm.wav &</i><i><br>
</i><i>[1] 2828</i><i><br>
</i><i>pi@raspberrypi ~ $ pacmd list-cards</i><i><br>
</i><i> .....</i><i><br>
</i><i> index: 4</i><i><br>
</i><i> name: <bluez_card.00_1A_7D_11_C0_66></i><i><br>
</i><i> driver: <module-bluetooth-device.c></i><i><br>
</i><i> owner module: 24</i><i><br>
</i><i> properties:</i><i><br>
</i><i> device.description = "Iqua Slim"</i><i><br>
</i><i> device.string = "00:1A:7D:11:C0:66"</i><i><br>
</i><i> device.api = "bluez"</i><i><br>
</i><i> device.class = "sound"</i><i><br>
</i><i> device.bus = "bluetooth"</i><i><br>
</i><i> device.form_factor = "headset"</i><i><br>
</i><i> bluez.path =
"/org/bluez/1981/hci0/dev_00_1A_7D_11_C0_66"</i><i><br>
</i><i> bluez.class = "0x240404"</i><i><br>
</i><i> bluez.name = "Iqua Slim"</i><i><br>
</i><i> device.icon_name =
"audio-headset-bluetooth"</i><i><br>
</i><i> device.intended_roles = "phone"</i><i><br>
</i><i> profiles:</i><i><br>
</i><i> a2dp: High Fidelity Playback (A2DP)
(priority 10)</i><i><br>
</i><i> hsp: Telephony Duplex (HSP/HFP) (priority
20)</i><i><br>
</i><i> off: Off (priority 0)</i><i><br>
</i><i> active profile: <hsp></i><i><br>
</i><i> sinks:</i><i><br>
</i><i> bluez_sink.00_1A_7D_11_C0_66/#3: Iqua Slim</i><i><br>
</i><i> sources:</i><i><br>
</i><i> bluez_sink.00_1A_7D_11_C0_66.monitor/#5:
Monitor of Iqua Slim</i><i><br>
</i><i> bluez_source.00_1A_7D_11_C0_66/#6: Iqua
Slim</i><i><br>
</i><i> ports:</i><i><br>
</i><i><br>
</i></blockquote>
<br>
<blockquote
cite="mid:CANT-zCV-W51bbxjKcccRu21Yfm9G3ndipaQquGW85fDg89QL4A@mail.gmail.com"
type="cite">
<pre wrap="">
4. During recording, make sure the source is "state: RUNNING" (pacmd
list-sources). This confirms (3) meaning that SCO audio is streaming
fine.</pre>
</blockquote>
I used pactl list sources short. It showed <br>
<br>
<blockquote><i>1 bluez_sink.00_1A_7D_11_C0_66.monitor
module-bluetooth-device.c s16le 1ch 8000Hz
SUSPENDED</i><i><br>
</i><i>2 bluez_source.00_1A_7D_11_C0_66
module-bluetooth-device.c s16le 1ch 8000Hz <b>RUNNING</b></i><i><br>
</i></blockquote>
<br>
<blockquote
cite="mid:CANT-zCV-W51bbxjKcccRu21Yfm9G3ndipaQquGW85fDg89QL4A@mail.gmail.com"
type="cite">
<pre wrap="">If all this work fine but the issue persists, I'd have to see the
output of hcidump during recording.</pre>
</blockquote>
This is all hcidump gives:<br>
<br>
<blockquote><i>HCI sniffer - Bluetooth packet analyzer ver 2.4</i><i><br>
</i><i>device: hci0 snap_len: 1028 filter: 0xffffffff</i><i><br>
</i></blockquote>
<br>
and a waiting green square not returning to the command line. When I
disconnected the headset I got<br>
<br>
<blockquote><i>> HCI Event: Inquiry Result (0x02) plen 20</i><i><br>
</i><i> bdaddr 00:A0:02:00:0C:00 mode 19 clkoffset 0x1300 class
0x02000c</i><i><br>
</i><i> bdaddr 00:02:00:0C:01:05 mode 0 clkoffset 0x0000 class
0x000000</i><i><br>
</i><i> bdaddr 00:00:00:00:00:00 mode 0 clkoffset 0x0000 class
0x000000</i><i><br>
</i><i> bdaddr 00:00:00:00:00:00 mode 0 clkoffset 0x0000 class
0x000000</i><i><br>
</i><i> bdaddr 00:00:00:00:00:00 mode 0 clkoffset 0x0000 class
0x000000</i><i><br>
</i><i> bdaddr 00:00:00:00:00:00 mode 0 clkoffset 0x0000 class
0x000000</i><i><br>
</i><i>Stream-Fehler: Entität terminiert</i><i><br>
</i><i>> HCI Event: Disconn Complete (0x05) plen 4</i><i><br>
</i><i> status 0x00 handle 12 reason 0x13</i><i><br>
</i><i> Reason: Remote User Terminated Connection</i><i><br>
</i></blockquote>
<br>
<blockquote
cite="mid:CANT-zCV-W51bbxjKcccRu21Yfm9G3ndipaQquGW85fDg89QL4A@mail.gmail.com"
type="cite">
<pre wrap="">
You can also repeat the test with module-suspend-on-idle unloaded
before connecting the headset (e.g. disable it in default.pa). Perhaps
the headset has issues with audio suspend&resume.</pre>
</blockquote>
Done. Same result but this time <i>pactl list sources short</i>
showed IDLE instead of suspended.<br>
<blockquote
cite="mid:CANT-zCV-W51bbxjKcccRu21Yfm9G3ndipaQquGW85fDg89QL4A@mail.gmail.com"
type="cite">
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">I have the impression that you need to make the headset believe that a
phone call is coming in for it's mike to become active.
</pre>
</blockquote>
</blockquote>
<pre wrap="">
The audio stream (SCO) should start streaming as soon as parecord is
launched. No need for phone calls or anything.</pre>
</blockquote>
The guy from carwhisperer used the ring command for some reason. I
coded this:
<a class="moz-txt-link-freetext" href="http://stackoverflow.com/questions/16918659/dbus-exceptions-dbusexception-org-bluez-error-notavailable-operation-currently">http://stackoverflow.com/questions/16918659/dbus-exceptions-dbusexception-org-bluez-error-notavailable-operation-currently</a>
which made the headset actually ring in the hsp profile but play()
returns dbus.exceptions.DBusException: org.bluez.Error.NotAvailable:
Operation currently not available.<br>
<blockquote
cite="mid:CANT-zCV-W51bbxjKcccRu21Yfm9G3ndipaQquGW85fDg89QL4A@mail.gmail.com"
type="cite">
<pre wrap="">
</pre>
<blockquote type="cite">
<blockquote type="cite">
<pre wrap="">
I've never heard of such headsets. At least my headset allows PulseAudio
to use the microphone at any time.
</pre>
</blockquote>
<pre wrap="">By Headset I mean this
<a class="moz-txt-link-freetext" href="http://www.iqua.com/sites/default/files/images/slim_high5.jpg">http://www.iqua.com/sites/default/files/images/slim_high5.jpg</a>
So I talked to the carwhisperer author martin. He said that he couldn't help
me in this case as he doesn't know his code anymore well enough. If you are
interested, find his code at
<a class="moz-txt-link-freetext" href="http://trifinite.org/trifinite_stuff_carwhisperer.html#downloads">http://trifinite.org/trifinite_stuff_carwhisperer.html#downloads</a>
Maybe you / someone else can find the real cause.
</pre>
</blockquote>
</blockquote>
</body>
</html>