Is the GattServices property still updated?

Is the GattServices property still updated?

[François Beaufort]

I'm not sure to understand what is happening and I'd love your help there.

Is the "GattServices" device property still supposed to be updated
once all remote GATT services of the device
have been discovered and exported over D-Bus as mentioned in
http://git.kernel.org/cgit/bluetooth/bluez.git/tree/doc/device-api.txt#n215

If so, does this DBUS messages trace looks good to you:
http://pastebin.com/emxWB4w1?

Re: Is the GattServices property still updated?

[François Beaufort]

For info, here's some background:
https://code.google.com/p/chromium/issues/detail?id=560277

On Tue, Dec 1, 2015 at 11:03 AM, François Beaufort
<beaufort.francois@gmail.com> wrote:
> I'm not sure to understand what is happening and I'd love your help there.
>
> Is the "GattServices" device property still supposed to be updated
> once all remote GATT services of the device
> have been discovered and exported over D-Bus as mentioned in
> http://git.kernel.org/cgit/bluetooth/bluez.git/tree/doc/device-api.txt#n215
>
> If so, does this DBUS messages trace looks good to you:
> http://pastebin.com/emxWB4w1?

Re: Is the GattServices property still updated?

[Luiz Augusto von Dentz]

Hi François,

On Tue, Dec 1, 2015 at 12:06 PM, François Beaufort
<beaufort.francois@gmail.com> wrote:
> For info, here's some background:
> https://code.google.com/p/chromium/issues/detail?id=560277
>
> On Tue, Dec 1, 2015 at 11:03 AM, François Beaufort
> <beaufort.francois@gmail.com> wrote:
>> I'm not sure to understand what is happening and I'd love your help there.
>>
>> Is the "GattServices" device property still supposed to be updated
>> once all remote GATT services of the device
>> have been discovered and exported over D-Bus as mentioned in
>> http://git.kernel.org/cgit/bluetooth/bluez.git/tree/doc/device-api.txt#n215
>>
>> If so, does this DBUS messages trace looks good to you:
>> http://pastebin.com/emxWB4w1?
> --

It is a regression caused by restoring the attributes from storage,
Ive send a patch for that. Btw, it is not recommended to mess up with
the storage manually if someone is doing this please advise to remove
the device properly.


-- 
Luiz Augusto von Dentz

Re: Is the GattServices property still updated?

[François Beaufort]

On Tue, Dec 1, 2015 at 12:55 PM, Luiz Augusto von Dentz
<luiz.dentz@gmail.com> wrote:
> Hi François,
>
> On Tue, Dec 1, 2015 at 12:06 PM, François Beaufort
> <beaufort.francois@gmail.com> wrote:
>> For info, here's some background:
>> https://code.google.com/p/chromium/issues/detail?id=560277
>>
>> On Tue, Dec 1, 2015 at 11:03 AM, François Beaufort
>> <beaufort.francois@gmail.com> wrote:
>>> I'm not sure to understand what is happening and I'd love your help there.
>>>
>>> Is the "GattServices" device property still supposed to be updated
>>> once all remote GATT services of the device
>>> have been discovered and exported over D-Bus as mentioned in
>>> http://git.kernel.org/cgit/bluetooth/bluez.git/tree/doc/device-api.txt#n215
>>>
>>> If so, does this DBUS messages trace looks good to you:
>>> http://pastebin.com/emxWB4w1?
>> --
>
> It is a regression caused by restoring the attributes from storage,
> Ive send a patch for that. Btw, it is not recommended to mess up with
> the storage manually if someone is doing this please advise to remove
> the device properly.
>

What a relief! Thank you Luiz.

For info, I experienced this with Bluez 5.35.
Will it be part of Bluez 5.37 release?

>
> --
> Luiz Augusto von Dentz

Re: Is the GattServices property still updated?

[François Beaufort]

Could that be related to
https://code.google.com/p/chromium/issues/detail?id=541606 as well?

On Tue, Dec 1, 2015 at 1:04 PM, François Beaufort
<beaufort.francois@gmail.com> wrote:
> On Tue, Dec 1, 2015 at 12:55 PM, Luiz Augusto von Dentz
> <luiz.dentz@gmail.com> wrote:
>> Hi François,
>>
>> On Tue, Dec 1, 2015 at 12:06 PM, François Beaufort
>> <beaufort.francois@gmail.com> wrote:
>>> For info, here's some background:
>>> https://code.google.com/p/chromium/issues/detail?id=560277
>>>
>>> On Tue, Dec 1, 2015 at 11:03 AM, François Beaufort
>>> <beaufort.francois@gmail.com> wrote:
>>>> I'm not sure to understand what is happening and I'd love your help there.
>>>>
>>>> Is the "GattServices" device property still supposed to be updated
>>>> once all remote GATT services of the device
>>>> have been discovered and exported over D-Bus as mentioned in
>>>> http://git.kernel.org/cgit/bluetooth/bluez.git/tree/doc/device-api.txt#n215
>>>>
>>>> If so, does this DBUS messages trace looks good to you:
>>>> http://pastebin.com/emxWB4w1?
>>> --
>>
>> It is a regression caused by restoring the attributes from storage,
>> Ive send a patch for that. Btw, it is not recommended to mess up with
>> the storage manually if someone is doing this please advise to remove
>> the device properly.
>>
>
> What a relief! Thank you Luiz.
>
> For info, I experienced this with Bluez 5.35.
> Will it be part of Bluez 5.37 release?
>
>>
>> --
>> Luiz Augusto von Dentz

Re: Is the GattServices property still updated?

[Luiz Augusto von Dentz]

Hi François,

On Tue, Dec 1, 2015 at 2:07 PM, François Beaufort
<beaufort.francois@gmail.com> wrote:
> Could that be related to
> https://code.google.com/p/chromium/issues/detail?id=541606 as well?

This one seems to be related to suspend/resume, maybe it is related to
firmware loading? It usually works for me in Fedora in the same
conditions.

> On Tue, Dec 1, 2015 at 1:04 PM, François Beaufort
> <beaufort.francois@gmail.com> wrote:
>> On Tue, Dec 1, 2015 at 12:55 PM, Luiz Augusto von Dentz
>> <luiz.dentz@gmail.com> wrote:
>>> Hi François,
>>>
>>> On Tue, Dec 1, 2015 at 12:06 PM, François Beaufort
>>> <beaufort.francois@gmail.com> wrote:
>>>> For info, here's some background:
>>>> https://code.google.com/p/chromium/issues/detail?id=560277
>>>>
>>>> On Tue, Dec 1, 2015 at 11:03 AM, François Beaufort
>>>> <beaufort.francois@gmail.com> wrote:
>>>>> I'm not sure to understand what is happening and I'd love your help there.
>>>>>
>>>>> Is the "GattServices" device property still supposed to be updated
>>>>> once all remote GATT services of the device
>>>>> have been discovered and exported over D-Bus as mentioned in
>>>>> http://git.kernel.org/cgit/bluetooth/bluez.git/tree/doc/device-api.txt#n215
>>>>>
>>>>> If so, does this DBUS messages trace looks good to you:
>>>>> http://pastebin.com/emxWB4w1?
>>>> --
>>>
>>> It is a regression caused by restoring the attributes from storage,
>>> Ive send a patch for that. Btw, it is not recommended to mess up with
>>> the storage manually if someone is doing this please advise to remove
>>> the device properly.
>>>
>>
>> What a relief! Thank you Luiz.
>>
>> For info, I experienced this with Bluez 5.35.
>> Will it be part of Bluez 5.37 release?

Yes, it will be part of the next release.

>>>
>>> --
>>> Luiz Augusto von Dentz



-- 
Luiz Augusto von Dentz