Bug 273873

Still trying to fix this. Applied today's update of Bluedevil but no change. Could udev be the problem?

At boot time there are a zillion messages stating
SYSFS{}= will be removed in a future udev version, please use ATTR{}= to match the event device, or ATTRS{}= to match a parent device, in /etc/udev/rules.d/51-hso-udev.rules

When I look at hso-udev.rules I see it says:
     # usb_device switch need for kernel 2.6.24 and newer, which does no longer support usb_device directly

Bluedevil WAS working on my previous machine. On this machine I switched to using the pae kernel:
     2.6.37.6-0.7-pae

The hso file has a number of USB devices listed, including a line commented out in [Old Syntax]

#SYSFS{idVendor}=="05c6", SYSFS[idProduct}=="1000"

My device is 

  Vendor: usb 0x148f "Ralink Technology, Corp."
  Device: usb 0x1000 

so the device matches, although the vendor doesn't.

I'm getting towards the deep end of the pool so I'll stop, but perhaps you would have some insight into whether this might be the problem.

I have several different kernel booting options, so I think I'll just try booting different versions to see if it works on ANY of them.

TIA, Andy

OK, more investigation:

There are several places in the boot sequence where things fail for bluetooth:

bluetoothd[4344]: Parsing /etc/bluetooth/input.conf failed: No such file or directory
bluetoothd[4344]: Parsing /etc/bluetooth/audio.conf failed: No such file or directory
bluetoothd[4344]: Parsing /etc/bluetooth/serial.conf failed: No such file or directory

Using 'locate' fails to show them anywhere.

Indeed, bluedevil sees the device and I can turn it on and off and have bluedevil (and blueman) see it, but it doesn't transmit or receive.

TIA, Andy
===================================================
/var/log/messages contains the following entries regarding bluetooth:
Aug  4 13:44:27 hp2t bluetoothd[4326]: Bluetooth deamon 4.88
Aug  4 13:44:27 hp2t bluetoothd[4344]: Starting SDP server
Aug  4 13:44:28 hp2t kernel: [   93.009523] Bluetooth: L2CAP ver 2.15
Aug  4 13:44:28 hp2t kernel: [   93.009527] Bluetooth: L2CAP socket layer initialized
Aug  4 13:44:28 hp2t ifup:     wlan0     device: RaLink RT3090 Wireless 802.11n 1T/1R PCIe
# ======== Is this the problem: ======== #
Aug  4 13:44:28 hp2t bluetoothd[4344]: Parsing /etc/bluetooth/input.conf failed: No such file or directory
Aug  4 13:44:28 hp2t bluetoothd[4344]: Parsing /etc/bluetooth/audio.conf failed: No such file or directory
# ==================================== #
Aug  4 13:44:28 hp2t kernel: [   93.518866] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
Aug  4 13:44:28 hp2t kernel: [   93.518871] Bluetooth: BNEP filters: protocol multicast
Aug  4 13:44:28 hp2t kernel: [   93.726762] Bluetooth: SCO (Voice Link) ver 0.6
Aug  4 13:44:28 hp2t kernel: [   93.726766] Bluetooth: SCO socket layer initialized
Aug  4 13:44:29 hp2t bluetoothd[4344]: HCI dev 0 up
# ======== Is this the problem: ======== #
Aug  4 13:44:29 hp2t bluetoothd[4344]: Parsing /etc/bluetooth/serial.conf failed: No such file or directory
# ==================================== #
Aug  4 13:44:29 hp2t bluetoothd[4344]: Adapter /org/bluez/4326/hci0 has been enabled
Aug  4 13:44:29 hp2t kernel: [   94.169367] Bluetooth: RFCOMM TTY layer initialized
Aug  4 13:44:29 hp2t kernel: [   94.169381] Bluetooth: RFCOMM socket layer initialized
Aug  4 13:44:29 hp2t kernel: [   94.169388] Bluetooth: RFCOMM ver 1.11
Aug  4 13:47:45 hp2t bluetoothd[4344]: Discovery session 0xb77eccb0 with :1.53 activated

Well, if BlueDevil (BlueZ) detects your adapter, then it is not a udev problem at all since the adapter is already detected.

Can you execute hcitool scan  ? you should see discoverability results.

Hi thanks.

hcitool scan returns NOTHING:

    tm2t:/Desktop # hcitool scan
    Scanning ...
    tm2t:/Desktop #
 
The Bluetooth symbol is in the system tray.

Clicking it brings up a menu of all the Bluedevil functions.

"Configure adapter" shows that the adapter is seen with a MAC address,
it is "Powered" and "Always visible."

I'll be glad to do any other experimentation you direct...

Saludos, Andy

On 10/08/2011 02:24 PM, Alex Fiestas wrote:
> https://bugs.kde.org/show_bug.cgi?id=273873
>
>
> Alex Fiestas <afiestas@kde.org> changed:
>
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>              Status|UNCONFIRMED                 |NEEDSINFO
>                  CC|                            |afiestas@kde.org
>          Resolution|                            |WAITINGFORINFO
>
>
>
>
> --- Comment #3 from Alex Fiestas <afiestas kde org>  2011-10-08 18:24:24 ---
> Well, if BlueDevil (BlueZ) detects your adapter, then it is not a udev problem
> at all since the adapter is already detected.
>
> Can you execute hcitool scan  ? you should see discoverability results.
>

The "hcitool scan" command returning no discovered devices tell us that this is not a BlueDevil issue but rather something below the stack (Bluez).

Just to point the obvious, are you certain that you have your device (cellphone, printer, gps etc...) in discoverability mode ?

As last test, I'd say you should test a recent kernel, for example a Ubuntu/Kubuntu oneiric livecd  will do (they ship with 3.0).

Thanks for all the testing.

Hi, thanks.

On 10/09/2011 04:50 AM, Alex Fiestas wrote:
> https://bugs.kde.org/show_bug.cgi?id=273873
>
> --- Comment #5 from Alex Fiestas <afiestas kde org>  2011-10-09 08:50:27 ---
> The "hcitool scan" command returning no discovered devices tell us that this is
> not a BlueDevil issue but rather something below the stack (Bluez).
Mmmm. Here are my installed files:

tm2t:~ # zypper se -s blue|grep "i |"

S | Name                     | Type    | Version    | Arch   | Repository
--+--------------------------+---------+------------+--------+------------------------------------

i | bluedevil                                      | package    |
1.0.2-3.8.1   | i586   | openSUSE-11.4-Oss                            
i | bluez                                            | package    |
4.88-2.1      | i586   | openSUSE-11.4-Oss                            
i | libbluedevil-devel                        | package    |
1.8-3.1       | i586   | openSUSE-11.4-Oss                            
i | libbluedevil1                                | package    |
1.8-3.1       | i586   | openSUSE-11.4-Oss                            
i | libbluetooth3                              | package    |
4.88-2.1      | i586   | openSUSE-11.4-Oss                            
i | pulseaudio-module-bluetooth  | package    | 0.9.22-6.11.1 | i586   |
Updates for openSUSE 11.4 11.4-0             
tm2t:~ #
> Just to point the obvious, are you certain that you have your device
> (cellphone, printer, gps etc...) in discoverability mode ?
Yes, they can all connect to each other, and the other (Fujitsu p1630)
laptop, just not this one.
> As last test, I'd say you should test a recent kernel, for example a
> Ubuntu/Kubuntu oneiric livecd  will do (they ship with 3.0).
Actually, I have, several 2.6.37.6-0.7 and 3.0.4-43 versions:  Desktop,
Default, Vanilla, Failsafe...

But I've got several live DVDs from Linux Format, so will try them out
and see...

I suppose it could be a hardware thing, except that BlueDevil is seeing
the Ralink adapter and the WiFi part is working...
> Thanks for all the testing.
Thank YOU. I'll be glad to do more...

I haven't given up. I did some system browsing and came across a YaST module for adding Kernel Settings for the Bluetooth adapter:

     yast2 > system > Kernel Settings > PCI ID Setup 

offers the ability to add specific hardware setups either manually or from a list. Choosing From List indeed offers our contentious

     Ralink RT3090 Wireless 802.11n 1T/1R PCIe (0000:03:00.0)

So I try to add it, but the wizard demands a SysFS Directory and driver name...

Now, 
     locate ralink

only offers drivers for the pae and vanilla kernels, not the mainstream kernel. Does that mean that the ralink driver is included in the kernel, not as a module?

What answer do I give for SysFS directory and driver name???

Also, 

     http://en.opensuse.org/index.php?title=HCL:Network_%28Wireless%29&redirect=no#Ralink

says for the RT2800 (on which the RT3090 is based)
 
	'ralink-firmware' package required -> auto installed by YaST

but searching for ralink-firmware returns nothing...

Does this mean that the firmware module is included in the kernel? Or does it mean that something else is afoot, since the module doesn't exist as a module...

I'd love to help fix this. It has been broken for several KDE point releases...

As for "upstream" issues (bluez) where would you suggest I post this elsewhere... I've gone to Novell bugs...

Cheers, Andy

Well this doesn't seem to be a problem in kde but rather in the kernel or some place downstream.

Can you execute rfkill list ? depending on what you see there you may want to try to "set bluetooth on". 

If the rfkill thing fixed the issue, then is our fault for not supporting rfkill as we should.

On 04/27/2012 09:47 PM, Alex Fiestas wrote:
> https://bugs.kde.org/show_bug.cgi?id=273873
>
> --- Comment #8 from Alex Fiestas <afiestas@kde.org> ---
> Well this doesn't seem to be a problem in kde but rather in the kernel or some
> place downstream.
>
> Can you execute rfkill list ? depending on what you see there you may want to
> try to "set bluetooth on". 
The first result is this:
    tm2t:~ # rfkill list
    0: phy0: Wireless LAN
        Soft blocked: no
        Hard blocked: no
    1: hci0: Bluetooth
        Soft blocked: no
        Hard blocked: no

I then repeat after set bluetooth on (no change):
    tm2t:~ # set bluetooth on
    tm2t:~ # rfkill list
    0: phy0: Wireless LAN
        Soft blocked: no
        Hard blocked: no
    1: hci0: Bluetooth
        Soft blocked: no
        Hard blocked: no
> If the rfkill thing fixed the issue, then is our fault for not supporting
> rfkill as we should.
Apparently no fault there... It still does not discover any devices, nor
can they discover it. They CAN discover each other.

Cheers, Andy

BT hasn't worked on this machine (HP Touchsmart TM2T, OpenSuSE 12.1, 
Ralink BT/WIFI adapter) since very soon after I got it.

Today I bought a Targus Micro USB BT ADB10US1 (ADB73) adapter. It doesn't work either.

The Internal Ralink is recognized by YaST Hardware Info in extensive detail reported above, but also after my signature for ease.

The Targus (Broadcom) USB is not recognized at all, neither by yast2 hwinfo nor by dmesg.

hcitool scan returns... nothing.

hcitool dev returns
	devices: hci0 CC:52:AF:50:0D:FE
This is the MAC of the internal adapter. If you repeat this with the 
Targus removed you get the same result.
So the USB is not being recognized.

tm2t:~ # hcitool name CC:52:AF:50:0D:FE
Device is not available.

hcitool inq returns nothing.

=====

The stack is bluez plus BlueDevil for KDE. Standard OpenSuSE 12.1.

But I've tried this with the Gnome stack and it doesn't do any better.

I thought that maybe getting the new outboard adapter would solve 
things. The BlueDevil developer Alex says the problems are "upstream", so it 
appears to be a problem with the Bluez stack.

Regardless, neither Broadcom nor Ralink work.

Cheers, Andy
=========Output from yast2 hwinfo================
39: USB 00.0: 11500 Bluetooth Device
  [Created at usb.122]
  Unique ID: JPTW.tGttqvOhWG9
  Parent ID: FKGF.0j9+vWlqL56
  SysFS ID: /devices/pci0000:00/0000:00:1d.0/usb2/2-1/2-1.3/2-1.3:1.0
  SysFS BusID: 2-1.3:1.0
  Hardware Class: bluetooth
  Model: "Ralink Bluetooth Device"
  Hotplug: USB
  Vendor: usb 0x148f "Ralink Technology, Corp."
  Device: usb 0x1000 
  Revision: "52.76"
  Driver: "btusb"
  Driver Modules: "btusb"
  Speed: 12 Mbps
  Module Alias: "usb:v148Fp1000d5276dcE0dsc01dp01icE0isc01ip01"
  Driver Info #0:
    Driver Status: btusb is active
    Driver Activation Cmd: "modprobe btusb"
  Config Status: cfg=no, avail=yes, need=no, active=unknown
  Attached to: #36 (Hub)

Well, this clearly seems to be a upstream issue of your hardware not being supported :/

You should report a bug at https://bugzilla.kernel.org/ (or at your distro) to try to get support for it.

Another option would be to test http://en.opensuse.org/Portal:Tumbleweed to get the latest kernel and see if you have luck.

So atm, I'm going to close this bug as upstream since everything points that this is actually kernel fault.

If I'm wrong please, re-open this bug!

Thanks!