phone turned off on its own while idle

Bug #1511312 reported by Olivier Tilloy
28
This bug affects 6 people
Affects Status Importance Assigned to Milestone
Canonical System Image
Confirmed
High
John McAleely

Bug Description

(initially reported the issue on the ubuntu-phone ML)

I’m using the BQ aquaris E4.5, running rc-proposed (#159 as of yesterday).

This morning the alarm that normally wakes me up didn’t ring. When I woke up by myself an hour and a half later, the phone had turned off. I immediately thought "battery drain bug". When I plugged the phone to a power source, the charging screen showed 48%, pretty much the level it had before I went to sleep. So not a battery drain problem.

It took me a while to manage to get it to boot, after the initial "BQ, powered by Ubuntu" screen the device would turn off again, or go to the charging screen. I eventually managed to fully boot it after a handful of retries, and there’s nothing suspicious in there. Battery is indeed at 48%, the battery consumption graph shows a regular and slight slope downwards, no peak or anything, and there are no new crash files.

I’m also pretty sure I had closed all open apps before going to bed.

The only thing that I can think of is that the GPS was turned on, whereas it’s usually turned off (I had turned it on for navigation during the day, and had forgotten to turn it off again afterwards).

Tags: power-bugs
Revision history for this message
Olivier Tilloy (osomon) wrote :

Attaching the syslog from the device. It starts at 04:19:17 AM, and the last recorded event before the phone turned itself off is at 05:14:00. There are lots of events in between (214226 lines) for a supposedly idle device, I would say.

Revision history for this message
Olivier Tilloy (osomon) wrote :

I grabbed a copy of the contents of /var/log and /home/phablet/.cache/upstart from the phone, I can share selected files on demand if that helps investigate the issue.

Revision history for this message
Tomas Öqvist (tomoqv) wrote :

A similar thing happened to me last night (I was also on r159 on Aquaris 4.5). I suddenly noticed that the phone had turned off, and I wasn't able to turn it on until plugging it in to the charger. My charge level was about 35%, but it wouldn't turn back on again until it had charged for a while. Attaching syslog.

Revision history for this message
Pat McGowan (pat-mcgowan) wrote :

Similar to bug #1505241 the wakeup source is the MT662x and the phone does not suspend.

Revision history for this message
Jean-Baptiste Lallement (jibel) wrote :

It is similar but slightly different. In Oliver's case the system tries to suspend every second, then after almost an hour it suddenly stops.

According to syslog in #1505241, the system tries to suspend 559 times in 3h43min while in the case of this bug it tries to suspend 3273 times in 55min.

I'm wondering if the kernel watchdog could bring the system down if it fails to suspend repeatedly.

Revision history for this message
Olivier Tilloy (osomon) wrote :

What’s quite different from bug #1505241 is that no battery drain happened. The phone turned itself off, but it still had plenty of battery.

Revision history for this message
Pat McGowan (pat-mcgowan) wrote :

Unless this is a symptom of other reports where the reported battery level is incorrect. It acted a lot like it shutdown do to drained battery but no proof there.

The only shutdown triggers I know of are low battery and high temp.

Revision history for this message
Olivier Tilloy (osomon) wrote :

Then the reported battery level would be consistently wrong. I’ve been using the phone quite extensively since 10am this morning (which is when I resurrected it), it was plugged in at times for for short periods of time (when connected to fetch the logs). I’ve actively used the hotspot function for a couple of hours this morning to work from my laptop. If the battery had been empty, the device would have turned itself off long ago.

Attaching a capture of the battery consumption graph, the sudden drop around 10:30am corresponds to when I activated the hotspot while not plugged in to a power source.

Revision history for this message
Olivier Tilloy (osomon) wrote :
Revision history for this message
Olivier Tilloy (osomon) wrote :

Another (probably unrelated) comment: I keep disabling bluetooth, as I never ever use it, and it keeps re-enabling itself. Annoying, and I suppose it has an impact on battery life too.

Revision history for this message
Pat McGowan (pat-mcgowan) wrote :

Actually the description that it could not initially boot until it charged for a while is also consistent with the battery being depleted, or at least the device thinking that was so. What did the charging screen say it was?

Revision history for this message
Tomas Öqvist (tomoqv) wrote :

My battery reading was from the charging screen. I hadn't used the phone for a while when I picked it up and noticed it was "dead".

Changed in canonical-devices-system-image:
status: New → Confirmed
importance: Undecided → High
assignee: nobody → John McAleely (john.mcaleely)
milestone: none → backlog
Revision history for this message
Dan González (gonrodda) wrote :

Same behavior here. When happened, several times, battery levels range from ~30% to ~45%.
What I've noticed also is that while charging after these sudden power offs, if you turn on to the battery charge animation, wait the screen to turn off, turn on again ... the battery level displayed goes down 1% every turn on until it does it no more. This could be done 10 to 15 times, each time an one percent less than before.

tags: added: battery
tags: added: power-bugs
removed: battery
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.