GPS Denied without reason

[costales]

You're using the GPS, and without reason, it stops. It doesn't return more positions.
If you cancel the route and set a new one, a popup will tell you that the GPS is denied.
I experienced this in OSM (From Alan) too.
You know uNav is not freeze because you can drag the map.

rc-proposed in Nexus 4 and BQ E4.5.

[taiebot65 (dedreuil)]

I confirm this behaviour disabling wifi (Agps) i have noticed much better stability and less gps stop without reason

[Jean-Baptiste Lallement (jibel)]

Thanks for your report. Could you please reproduce the issue and attach the following information:
- /var/log/ubuntu-location-service*
- /var/log/upstart/ubuntu-location-service.log
- /var/log/syslog
- The output of: sudo /system/bin/logcat
- The output of: ubuntu-location-serviced-cli --bus system --get --property visible_space_vehicles
- The output of sudo GLOG_v=1000 GLOG_logtostderr=1 ubuntu-location-serviced-cli --bus system --test

Thanks.

[Jean-Baptiste Lallement (jibel)]

I observed a similar behaviour with HERE, after some time the position stops updating. The user has to close and reopen HERE to have new updates.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Thomas Voß (thomas-voss)]

Also marking oxide as affected. I fail to reproduce with the simple command line clients, and a simple qml app does not suffer the symptoms either. Please also attach the app-specific logs from /home/phablet/.cache/upstart.

[Chris Coulson (chrisccoulson)]

Does anyone have a reduced HTML/JS test case that triggers this bug?

[Matthias Apitz (gubu)]

I don't know if this is related (or another bug): I can reproduce 100% on OTA-10 with the app SensorStatus, that the app opens successful the GPS service, issues some ioctls and without getting or waiting for a first fix, it just closes again the GPS, all this within 2 secs:

$ tail -f /var/log/syslog | fgrep GPS

Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.400623] (2)[1213:mtk_wmtd][WMT-CTRL][I]wmt_ctrl_gps_sync_set:ctrl GPS_SYNC(4)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.400669] (2)[1213:mtk_wmtd][WMT-FUNC][I]wmt_func_gps_pre_ctrl:ctrl GPS_SYNC_SET(0) ok
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.401061] (1)[1213:mtk_wmtd][WMT-FUNC][I]wmt_func_gps_pre_ctrl:set reg for GPS_SYNC function okay by chip ic_pin_ctrl
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.401191] (1)[1213:mtk_wmtd][WMT-CTRL][I]wmt_ctrl_gps_lna_set:ctrl GPS_LNA(2)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.403702] (3)[1074:mnld][GPS] [I]GPS_open: WMT turn on GPS OK!
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.403753] (3)[1074:mnld][GPS] [I]GPS_compat_ioctl: (8)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.403769] (3)[1074:mnld][GPS] [I]GPS_unlocked_ioctl: GPS co_clock_flag (0)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.404403] (3)[1074:mnld][GPS] [I]GPS_compat_ioctl: (7)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.404419] (3)[1074:mnld][GPS] [I]GPS_unlocked_ioctl: low power flag (0)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406411] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21505)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406429] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21505)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406445] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21506)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406459] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21506)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406474] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21515)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406487] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21515)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407859] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21505)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407880] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21505)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407896] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21506)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407910] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21506)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407925] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21515)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407938] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21515)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.418189] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21505)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.418211] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21505)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.418228] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21506)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.418264] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21506)
Apr 8 08:48:48 ubuntu-phablet kernel: [ 668...

[Thomas Voß (thomas-voss)]

On Mon, Apr 11, 2016 at 9:29 AM, Matthias Apitz <email address hidden> wrote:
> I don't know if this is related (or another bug): I can reproduce 100%
> on OTA-10 with the app SensorStatus, that the app opens successful the
> GPS service, issues some ioctls and without getting or waiting for a
> first fix, it just closes again the GPS, all this within 2 secs:
>

The app does not issue the ioctls, it is the service doing it (or
better: the Android GPS HAL implementation).
Did you put the app into background by any chance?

> $ tail -f /var/log/syslog | fgrep GPS
>
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.400623] (2)[1213:mtk_wmtd][WMT-CTRL][I]wmt_ctrl_gps_sync_set:ctrl GPS_SYNC(4)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.400669] (2)[1213:mtk_wmtd][WMT-FUNC][I]wmt_func_gps_pre_ctrl:ctrl GPS_SYNC_SET(0) ok
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.401061] (1)[1213:mtk_wmtd][WMT-FUNC][I]wmt_func_gps_pre_ctrl:set reg for GPS_SYNC function okay by chip ic_pin_ctrl
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.401191] (1)[1213:mtk_wmtd][WMT-CTRL][I]wmt_ctrl_gps_lna_set:ctrl GPS_LNA(2)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.403702] (3)[1074:mnld][GPS] [I]GPS_open: WMT turn on GPS OK!
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.403753] (3)[1074:mnld][GPS] [I]GPS_compat_ioctl: (8)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.403769] (3)[1074:mnld][GPS] [I]GPS_unlocked_ioctl: GPS co_clock_flag (0)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.404403] (3)[1074:mnld][GPS] [I]GPS_compat_ioctl: (7)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.404419] (3)[1074:mnld][GPS] [I]GPS_unlocked_ioctl: low power flag (0)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406411] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21505)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406429] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21505)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406445] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21506)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406459] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21506)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406474] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21515)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.406487] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21515)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407859] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21505)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407880] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21505)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407896] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21506)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407910] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21506)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407925] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21515)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.407938] (1)[8506:mnld][GPS] [I]GPS_unlocked_ioctl: unknown cmd (21515)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.418189] (1)[8506:mnld][GPS] [I]GPS_compat_ioctl: (21505)
> Apr 8 08:48:48 ubuntu-phablet kernel: [ 668.418211] (1)[8506:mnld][GPS] [I]GPS_unloc...

[Matthias Apitz (gubu)]

Thomas Voß (thomas-voss) wrote:

> The app does not issue the ioctls, it is the service doing it (or
> better: the Android GPS HAL implementation).

correct, my wording was not precise enough;

> Did you put the app into background by any chance?

no, I just clicked the SensorStat icon in unity8;

[Victor gonzalez (victor-gonzalez-0)]

This is also happening with Frieza. Some customers contacted BQ because of it, they say the GPS indicator switches on and off every second, it probably happens even more than in krillin/vegeta in some cases.

[Victor gonzalez (victor-gonzalez-0)]

This is the behaviour reported with Frieza:

http://makeagif.com/i/QlHXU7

[Thomas Voß (thomas-voss)]

@Victor: I think the issue you are mentioning is independent of the
one originally reported on this bug. I will open up another one if you
wouldn't mind.

On Thu, Apr 21, 2016 at 2:59 PM, Victor gonzalez <email address hidden> wrote:
> This is the behaviour reported with Frieza:
>
> http://makeagif.com/i/QlHXU7
>
> --
> You received this bug notification because you are a bug assignee.
> https://bugs.launchpad.net/bugs/1565469
>
> Title:
> GPS Denied without reason
>
> Status in Canonical System Image:
> Incomplete
> Status in Oxide:
> New
> Status in location-service package in Ubuntu:
> Confirmed
>
> Bug description:
> You're using the GPS, and without reason, it stops. It doesn't return more positions.
> If you cancel the route and set a new one, a popup will tell you that the GPS is denied.
> I experienced this in OSM (From Alan) too.
> You know uNav is not freeze because you can drag the map.
>
> rc-proposed in Nexus 4 and BQ E4.5.
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/canonical-devices-system-image/+bug/1565469/+subscriptions

[Thomas Voß (thomas-voss)]

@Victor: https://bugs.launchpad.net/ubuntu/+source/location-service/+bug/1573168,
also left a question for you.

On Thu, Apr 21, 2016 at 3:08 PM, Thomas Voß <email address hidden> wrote:
> @Victor: I think the issue you are mentioning is independent of the
> one originally reported on this bug. I will open up another one if you
> wouldn't mind.
>
> On Thu, Apr 21, 2016 at 2:59 PM, Victor gonzalez <email address hidden> wrote:
>> This is the behaviour reported with Frieza:
>>
>> http://makeagif.com/i/QlHXU7
>>
>> --
>> You received this bug notification because you are a bug assignee.
>> https://bugs.launchpad.net/bugs/1565469
>>
>> Title:
>> GPS Denied without reason
>>
>> Status in Canonical System Image:
>> Incomplete
>> Status in Oxide:
>> New
>> Status in location-service package in Ubuntu:
>> Confirmed
>>
>> Bug description:
>> You're using the GPS, and without reason, it stops. It doesn't return more positions.
>> If you cancel the route and set a new one, a popup will tell you that the GPS is denied.
>> I experienced this in OSM (From Alan) too.
>> You know uNav is not freeze because you can drag the map.
>>
>> rc-proposed in Nexus 4 and BQ E4.5.
>>
>> To manage notifications about this bug go to:
>> https://bugs.launchpad.net/canonical-devices-system-image/+bug/1565469/+subscriptions

[Victor gonzalez (victor-gonzalez-0)]

Hi @Thomas,

Thanks for opening that bug but this one also affects Frieza, screenshot attached sent by the user.

[Rüdiger Kupper (ruediger.kupper)]

From my experience I believe that the issue is triggered by several apps sharing GPS access.
For instance, GPS frequently breaks down with uNav, when other apps access GPS data concurrently. Like: Browser (geolocation, e.g. Google maps), nearby scope, cachemere, etc.
I tend to have open a lot of apps in background, many of them acessIng GPS. If I use uNav togetjer with those, it usually works for a few minuites, then freezes an loses GPS.
I can resolve this issue by closing all other apps using GPS. After that, uNav works perfect.

I.e.: uNav requires exclusive access to GPS in order to work. It breaks as soon as another process requires read access to GPS.

There us no logical reason, why not several processes at once should read GPS data (it is read-obky after all). Still, this reproducibly breaks.