bluez dbus ble connection problem when a filter is set

bluez dbus ble connection problem when a filter is set

[Hakkı Göçeoğlu]

Hi All,

raspberry pi 3
bluez v5.39
testing platform client/bluetoothctl

I have a ble peripheral device advertising a custom uuid.
Within bluetoothctl when I try to connect to it after a filter is set,
one third of the time I get an error.
There is no problem if no filter is set, turning scanning off before
the connect command does not help.

Here is an example flow within bluetoothctl

[bluetooth]# set-scan-filter-uuids f655b638-fead-4268-b07c-e5d79d09b22b
SetDiscoveryFilter success
[bluetooth]# scan on
Discovery started
[CHG] Controller B8:27:EB:37:3F:CD Discovering: yes
[NEW] Device CF:05:30:F2:6F:5E CF-05-30-F2-6F-5E
[bluetooth]# connect CF:05:30:F2:6F:5E
Attempting to connect to CF:05:30:F2:6F:5E
[CHG] Device CF:05:30:F2:6F:5E Connected: yes
Failed to connect: org.bluez.Error.Failed
[CHG] Device CF:05:30:F2:6F:5E Connected: no


Any ideas? Thanks in advance, cheers

hakkig

Re: bluez dbus ble connection problem when a filter is set

[Vinicius Costa Gomes]

Hi hakkig,

Hakkı Göçeoğlu <hakki@bir618.com> writes:

> Hi All,
>
> raspberry pi 3
> bluez v5.39
> testing platform client/bluetoothctl

Which kernel version are you running?

>
> I have a ble peripheral device advertising a custom uuid.
> Within bluetoothctl when I try to connect to it after a filter is set,
> one third of the time I get an error.
> There is no problem if no filter is set, turning scanning off before
> the connect command does not help.
>
> Here is an example flow within bluetoothctl
>
> [bluetooth]# set-scan-filter-uuids f655b638-fead-4268-b07c-e5d79d09b22b
> SetDiscoveryFilter success
> [bluetooth]# scan on
> Discovery started
> [CHG] Controller B8:27:EB:37:3F:CD Discovering: yes
> [NEW] Device CF:05:30:F2:6F:5E CF-05-30-F2-6F-5E
> [bluetooth]# connect CF:05:30:F2:6F:5E
> Attempting to connect to CF:05:30:F2:6F:5E
> [CHG] Device CF:05:30:F2:6F:5E Connected: yes
> Failed to connect: org.bluez.Error.Failed
> [CHG] Device CF:05:30:F2:6F:5E Connected: no

I couldn't reproduce this, can you provide the btmon output when this
happens?

>
>
> Any ideas? Thanks in advance, cheers
>
> hakkig
> --
> To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html


Cheers,
--
Vinicius

Re: bluez dbus ble connection problem when a filter is set

[Hakkı Göçeoğlu]

Hi,

pi@raspberrypi:~ $ uname -a

Linux raspberrypi 4.1.19-v7+ #858 SMP Tue Mar 15 15:56:00 GMT 2016
armv7l GNU/Linux

would you need additional info?

On Sat, May 7, 2016 at 12:54 AM, Vinicius Costa Gomes
<vinicius.gomes@intel.com> wrote:
> Hi hakkig,
>
> Hakkı Göçeoğlu <hakki@bir618.com> writes:
>
>> Hi All,
>>
>> raspberry pi 3
>> bluez v5.39
>> testing platform client/bluetoothctl
>
> Which kernel version are you running?
>
>>
>> I have a ble peripheral device advertising a custom uuid.
>> Within bluetoothctl when I try to connect to it after a filter is set,
>> one third of the time I get an error.
>> There is no problem if no filter is set, turning scanning off before
>> the connect command does not help.
>>
>> Here is an example flow within bluetoothctl
>>
>> [bluetooth]# set-scan-filter-uuids f655b638-fead-4268-b07c-e5d79d09b22b
>> SetDiscoveryFilter success
>> [bluetooth]# scan on
>> Discovery started
>> [CHG] Controller B8:27:EB:37:3F:CD Discovering: yes
>> [NEW] Device CF:05:30:F2:6F:5E CF-05-30-F2-6F-5E
>> [bluetooth]# connect CF:05:30:F2:6F:5E
>> Attempting to connect to CF:05:30:F2:6F:5E
>> [CHG] Device CF:05:30:F2:6F:5E Connected: yes
>> Failed to connect: org.bluez.Error.Failed
>> [CHG] Device CF:05:30:F2:6F:5E Connected: no
>
> I couldn't reproduce this, can you provide the btmon output when this
> happens?
>
>>
>>
>> Any ideas? Thanks in advance, cheers
>>
>> hakkig
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
>
> Cheers,
> --
> Vinicius

Re: bluez dbus ble connection problem when a filter is set

[Hakkı Göçeoğlu]

[-- Attachment #1: Type: text/plain, Size: 2935 bytes --]

Hi again, please discard my previous mail I had not seen the second question:

> Which kernel version are you running?

pi@raspberrypi:~ $ uname -a

Linux raspberrypi 4.1.19-v7+ #858 SMP Tue Mar 15 15:56:00 GMT 2016
armv7l GNU/Linux

> I couldn't reproduce this, can you provide the btmon output when this
> happens?

I have attached two log outputs, one correct behaviour, and the other
erratic, on both the flow is the same until the line:

14:22 src/adapter.c:connected_callback() hci0 device EB:52:CE:B5:A3:52
connected eir_len 0

 on the correct flow the next four lines:
14:22 attrib/gattrib.c:g_attrib_ref() 0x1b7eb8: g_attrib_ref=1
14:22 src/gatt-client.c:btd_gatt_client_connected() Device connected.
14:22 src/device.c:load_gatt_db() Restoring EB:52:CE:B5:A3:52 gatt
database from file
14:22 No cache for EB:52:CE:B5:A3:52

and on the erratic flow the next six lines:
15:17 src/adapter.c:device_found_callback() hci0 addr
EB:52:CE:B5:A3:52, rssi -48 flags 0x0000 eir_len 21
15:17 src/device.c:device_set_legacy() legacy 0
15:17 src/device.c:device_set_rssi_with_delta() rssi -48 delta 16
15:17 src/device.c:att_connect_cb() connect error: Function not implemented (38)
15:17 src/device.c:btd_device_set_temporary() temporary 1
15:17 src/adapter.c:dev_disconnected() Device EB:52:CE:B5:A3:52
disconnected, reason 0

cheers

Hakki



On Sat, May 7, 2016 at 12:54 AM, Vinicius Costa Gomes
<vinicius.gomes@intel.com> wrote:
> Hi hakkig,
>
> Hakkı Göçeoğlu <hakki@bir618.com> writes:
>
>> Hi All,
>>
>> raspberry pi 3
>> bluez v5.39
>> testing platform client/bluetoothctl
>
> Which kernel version are you running?
>
>>
>> I have a ble peripheral device advertising a custom uuid.
>> Within bluetoothctl when I try to connect to it after a filter is set,
>> one third of the time I get an error.
>> There is no problem if no filter is set, turning scanning off before
>> the connect command does not help.
>>
>> Here is an example flow within bluetoothctl
>>
>> [bluetooth]# set-scan-filter-uuids f655b638-fead-4268-b07c-e5d79d09b22b
>> SetDiscoveryFilter success
>> [bluetooth]# scan on
>> Discovery started
>> [CHG] Controller B8:27:EB:37:3F:CD Discovering: yes
>> [NEW] Device CF:05:30:F2:6F:5E CF-05-30-F2-6F-5E
>> [bluetooth]# connect CF:05:30:F2:6F:5E
>> Attempting to connect to CF:05:30:F2:6F:5E
>> [CHG] Device CF:05:30:F2:6F:5E Connected: yes
>> Failed to connect: org.bluez.Error.Failed
>> [CHG] Device CF:05:30:F2:6F:5E Connected: no
>
> I couldn't reproduce this, can you provide the btmon output when this
> happens?
>
>>
>>
>> Any ideas? Thanks in advance, cheers
>>
>> hakkig
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
>
> Cheers,
> --
> Vinicius

[-- Attachment #2: correct.rtf --]
[-- Type: application/rtf, Size: 7657 bytes --]

[-- Attachment #3: error.rtf --]
[-- Type: application/rtf, Size: 4781 bytes --]

Re: bluez dbus ble connection problem when a filter is set

[Hakkı Göçeoğlu]

Hi Vinicius,

I have tested the same problem with the newer kernel
(Linux raspberrypi 4.4.9-v7+ #884 SMP Fri May 6 17:28:59 BST 2016
armv7l GNU/Linux)
the outcome is still the same, is there anything I can do, further
testing etc. ?

Thanks a lot, take care

Hakki

On Sat, May 7, 2016 at 8:12 AM, Hakkı Göçeoğlu <hakki@bir618.com> wrote:
> Hi again, please discard my previous mail I had not seen the second question:
>
>> Which kernel version are you running?
>
> pi@raspberrypi:~ $ uname -a
>
> Linux raspberrypi 4.1.19-v7+ #858 SMP Tue Mar 15 15:56:00 GMT 2016
> armv7l GNU/Linux
>
>> I couldn't reproduce this, can you provide the btmon output when this
>> happens?
>
> I have attached two log outputs, one correct behaviour, and the other
> erratic, on both the flow is the same until the line:
>
> 14:22 src/adapter.c:connected_callback() hci0 device EB:52:CE:B5:A3:52
> connected eir_len 0
>
>  on the correct flow the next four lines:
> 14:22 attrib/gattrib.c:g_attrib_ref() 0x1b7eb8: g_attrib_ref=1
> 14:22 src/gatt-client.c:btd_gatt_client_connected() Device connected.
> 14:22 src/device.c:load_gatt_db() Restoring EB:52:CE:B5:A3:52 gatt
> database from file
> 14:22 No cache for EB:52:CE:B5:A3:52
>
> and on the erratic flow the next six lines:
> 15:17 src/adapter.c:device_found_callback() hci0 addr
> EB:52:CE:B5:A3:52, rssi -48 flags 0x0000 eir_len 21
> 15:17 src/device.c:device_set_legacy() legacy 0
> 15:17 src/device.c:device_set_rssi_with_delta() rssi -48 delta 16
> 15:17 src/device.c:att_connect_cb() connect error: Function not implemented (38)
> 15:17 src/device.c:btd_device_set_temporary() temporary 1
> 15:17 src/adapter.c:dev_disconnected() Device EB:52:CE:B5:A3:52
> disconnected, reason 0
>
> cheers
>
> Hakki
>
>
>
> On Sat, May 7, 2016 at 12:54 AM, Vinicius Costa Gomes
> <vinicius.gomes@intel.com> wrote:
>> Hi hakkig,
>>
>> Hakkı Göçeoğlu <hakki@bir618.com> writes:
>>
>>> Hi All,
>>>
>>> raspberry pi 3
>>> bluez v5.39
>>> testing platform client/bluetoothctl
>>
>> Which kernel version are you running?
>>
>>>
>>> I have a ble peripheral device advertising a custom uuid.
>>> Within bluetoothctl when I try to connect to it after a filter is set,
>>> one third of the time I get an error.
>>> There is no problem if no filter is set, turning scanning off before
>>> the connect command does not help.
>>>
>>> Here is an example flow within bluetoothctl
>>>
>>> [bluetooth]# set-scan-filter-uuids f655b638-fead-4268-b07c-e5d79d09b22b
>>> SetDiscoveryFilter success
>>> [bluetooth]# scan on
>>> Discovery started
>>> [CHG] Controller B8:27:EB:37:3F:CD Discovering: yes
>>> [NEW] Device CF:05:30:F2:6F:5E CF-05-30-F2-6F-5E
>>> [bluetooth]# connect CF:05:30:F2:6F:5E
>>> Attempting to connect to CF:05:30:F2:6F:5E
>>> [CHG] Device CF:05:30:F2:6F:5E Connected: yes
>>> Failed to connect: org.bluez.Error.Failed
>>> [CHG] Device CF:05:30:F2:6F:5E Connected: no
>>
>> I couldn't reproduce this, can you provide the btmon output when this
>> happens?
>>
>>>
>>>
>>> Any ideas? Thanks in advance, cheers
>>>
>>> hakkig
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
>>> the body of a message to majordomo@vger.kernel.org
>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
>>
>> Cheers,
>> --
>> Vinicius

Re: bluez dbus ble connection problem when a filter is set

[Vinicius Costa Gomes]

Hi Hakkı,

Hakkı Göçeoğlu <hakki@bir618.com> writes:

> Hi Vinicius,
>
> I have tested the same problem with the newer kernel
> (Linux raspberrypi 4.4.9-v7+ #884 SMP Fri May 6 17:28:59 BST 2016
> armv7l GNU/Linux)
> the outcome is still the same, is there anything I can do, further
> testing etc. ?

Just something for futures emails, we are not big fans of top posting
around here.

The output of 'btmon' (it shows the HCI and MGMT traffic in the system)
should prove helpful.

I don't really follow the rpi scene, but I heard somewhere that there was an
update for the firmware of the Bluetooth controller.

>
> Thanks a lot, take care
>
> Hakki
>
> On Sat, May 7, 2016 at 8:12 AM, Hakkı Göçeoğlu <hakki@bir618.com> wrote:
>> Hi again, please discard my previous mail I had not seen the second question:
>>
>>> Which kernel version are you running?
>>
>> pi@raspberrypi:~ $ uname -a
>>
>> Linux raspberrypi 4.1.19-v7+ #858 SMP Tue Mar 15 15:56:00 GMT 2016
>> armv7l GNU/Linux
>>
>>> I couldn't reproduce this, can you provide the btmon output when this
>>> happens?
>>
>> I have attached two log outputs, one correct behaviour, and the other
>> erratic, on both the flow is the same until the line:
>>
>> 14:22 src/adapter.c:connected_callback() hci0 device EB:52:CE:B5:A3:52
>> connected eir_len 0
>>
>>  on the correct flow the next four lines:
>> 14:22 attrib/gattrib.c:g_attrib_ref() 0x1b7eb8: g_attrib_ref=1
>> 14:22 src/gatt-client.c:btd_gatt_client_connected() Device connected.
>> 14:22 src/device.c:load_gatt_db() Restoring EB:52:CE:B5:A3:52 gatt
>> database from file
>> 14:22 No cache for EB:52:CE:B5:A3:52
>>
>> and on the erratic flow the next six lines:
>> 15:17 src/adapter.c:device_found_callback() hci0 addr
>> EB:52:CE:B5:A3:52, rssi -48 flags 0x0000 eir_len 21
>> 15:17 src/device.c:device_set_legacy() legacy 0
>> 15:17 src/device.c:device_set_rssi_with_delta() rssi -48 delta 16
>> 15:17 src/device.c:att_connect_cb() connect error: Function not implemented (38)
>> 15:17 src/device.c:btd_device_set_temporary() temporary 1
>> 15:17 src/adapter.c:dev_disconnected() Device EB:52:CE:B5:A3:52
>> disconnected, reason 0
>>
>> cheers
>>
>> Hakki
>>
>>
>>
>> On Sat, May 7, 2016 at 12:54 AM, Vinicius Costa Gomes
>> <vinicius.gomes@intel.com> wrote:
>>> Hi hakkig,
>>>
>>> Hakkı Göçeoğlu <hakki@bir618.com> writes:
>>>
>>>> Hi All,
>>>>
>>>> raspberry pi 3
>>>> bluez v5.39
>>>> testing platform client/bluetoothctl
>>>
>>> Which kernel version are you running?
>>>
>>>>
>>>> I have a ble peripheral device advertising a custom uuid.
>>>> Within bluetoothctl when I try to connect to it after a filter is set,
>>>> one third of the time I get an error.
>>>> There is no problem if no filter is set, turning scanning off before
>>>> the connect command does not help.
>>>>
>>>> Here is an example flow within bluetoothctl
>>>>
>>>> [bluetooth]# set-scan-filter-uuids f655b638-fead-4268-b07c-e5d79d09b22b
>>>> SetDiscoveryFilter success
>>>> [bluetooth]# scan on
>>>> Discovery started
>>>> [CHG] Controller B8:27:EB:37:3F:CD Discovering: yes
>>>> [NEW] Device CF:05:30:F2:6F:5E CF-05-30-F2-6F-5E
>>>> [bluetooth]# connect CF:05:30:F2:6F:5E
>>>> Attempting to connect to CF:05:30:F2:6F:5E
>>>> [CHG] Device CF:05:30:F2:6F:5E Connected: yes
>>>> Failed to connect: org.bluez.Error.Failed
>>>> [CHG] Device CF:05:30:F2:6F:5E Connected: no
>>>
>>> I couldn't reproduce this, can you provide the btmon output when this
>>> happens?
>>>
>>>>
>>>>
>>>> Any ideas? Thanks in advance, cheers
>>>>
>>>> hakkig
>>>> --
>>>> To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
>>>> the body of a message to majordomo@vger.kernel.org
>>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>
>>>
>>> Cheers,
>>> --
>>> Vinicius


Cheers,
--
Vinicius

Re: bluez dbus ble connection problem when a filter is set

[Hakkı Göçeoğlu]

Hi again,

Thanks for the prompt reply, I will further investigate on my own. By
the way kernel, firmware all up to date.

Cheers

Hakki

On Tue, May 17, 2016 at 9:59 PM, Vinicius Costa Gomes
<vinicius.gomes@intel.com> wrote:
> Hi Hakkı,
>
> Hakkı Göçeoğlu <hakki@bir618.com> writes:
>
>> Hi Vinicius,
>>
>> I have tested the same problem with the newer kernel
>> (Linux raspberrypi 4.4.9-v7+ #884 SMP Fri May 6 17:28:59 BST 2016
>> armv7l GNU/Linux)
>> the outcome is still the same, is there anything I can do, further
>> testing etc. ?
>
> Just something for futures emails, we are not big fans of top posting
> around here.
>
> The output of 'btmon' (it shows the HCI and MGMT traffic in the system)
> should prove helpful.
>
> I don't really follow the rpi scene, but I heard somewhere that there was an
> update for the firmware of the Bluetooth controller.
>
>>
>> Thanks a lot, take care
>>
>> Hakki
>>
>> On Sat, May 7, 2016 at 8:12 AM, Hakkı Göçeoğlu <hakki@bir618.com> wrote:
>>> Hi again, please discard my previous mail I had not seen the second question:
>>>
>>>> Which kernel version are you running?
>>>
>>> pi@raspberrypi:~ $ uname -a
>>>
>>> Linux raspberrypi 4.1.19-v7+ #858 SMP Tue Mar 15 15:56:00 GMT 2016
>>> armv7l GNU/Linux
>>>
>>>> I couldn't reproduce this, can you provide the btmon output when this
>>>> happens?
>>>
>>> I have attached two log outputs, one correct behaviour, and the other
>>> erratic, on both the flow is the same until the line:
>>>
>>> 14:22 src/adapter.c:connected_callback() hci0 device EB:52:CE:B5:A3:52
>>> connected eir_len 0
>>>
>>>  on the correct flow the next four lines:
>>> 14:22 attrib/gattrib.c:g_attrib_ref() 0x1b7eb8: g_attrib_ref=1
>>> 14:22 src/gatt-client.c:btd_gatt_client_connected() Device connected.
>>> 14:22 src/device.c:load_gatt_db() Restoring EB:52:CE:B5:A3:52 gatt
>>> database from file
>>> 14:22 No cache for EB:52:CE:B5:A3:52
>>>
>>> and on the erratic flow the next six lines:
>>> 15:17 src/adapter.c:device_found_callback() hci0 addr
>>> EB:52:CE:B5:A3:52, rssi -48 flags 0x0000 eir_len 21
>>> 15:17 src/device.c:device_set_legacy() legacy 0
>>> 15:17 src/device.c:device_set_rssi_with_delta() rssi -48 delta 16
>>> 15:17 src/device.c:att_connect_cb() connect error: Function not implemented (38)
>>> 15:17 src/device.c:btd_device_set_temporary() temporary 1
>>> 15:17 src/adapter.c:dev_disconnected() Device EB:52:CE:B5:A3:52
>>> disconnected, reason 0
>>>
>>> cheers
>>>
>>> Hakki
>>>
>>>
>>>
>>> On Sat, May 7, 2016 at 12:54 AM, Vinicius Costa Gomes
>>> <vinicius.gomes@intel.com> wrote:
>>>> Hi hakkig,
>>>>
>>>> Hakkı Göçeoğlu <hakki@bir618.com> writes:
>>>>
>>>>> Hi All,
>>>>>
>>>>> raspberry pi 3
>>>>> bluez v5.39
>>>>> testing platform client/bluetoothctl
>>>>
>>>> Which kernel version are you running?
>>>>
>>>>>
>>>>> I have a ble peripheral device advertising a custom uuid.
>>>>> Within bluetoothctl when I try to connect to it after a filter is set,
>>>>> one third of the time I get an error.
>>>>> There is no problem if no filter is set, turning scanning off before
>>>>> the connect command does not help.
>>>>>
>>>>> Here is an example flow within bluetoothctl
>>>>>
>>>>> [bluetooth]# set-scan-filter-uuids f655b638-fead-4268-b07c-e5d79d09b22b
>>>>> SetDiscoveryFilter success
>>>>> [bluetooth]# scan on
>>>>> Discovery started
>>>>> [CHG] Controller B8:27:EB:37:3F:CD Discovering: yes
>>>>> [NEW] Device CF:05:30:F2:6F:5E CF-05-30-F2-6F-5E
>>>>> [bluetooth]# connect CF:05:30:F2:6F:5E
>>>>> Attempting to connect to CF:05:30:F2:6F:5E
>>>>> [CHG] Device CF:05:30:F2:6F:5E Connected: yes
>>>>> Failed to connect: org.bluez.Error.Failed
>>>>> [CHG] Device CF:05:30:F2:6F:5E Connected: no
>>>>
>>>> I couldn't reproduce this, can you provide the btmon output when this
>>>> happens?
>>>>
>>>>>
>>>>>
>>>>> Any ideas? Thanks in advance, cheers
>>>>>
>>>>> hakkig
>>>>> --
>>>>> To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
>>>>> the body of a message to majordomo@vger.kernel.org
>>>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>>
>>>>
>>>> Cheers,
>>>> --
>>>> Vinicius
>
>
> Cheers,
> --
> Vinicius