Revision history [back]

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day by typing:with:

mcetool --disable-led-pattern=PatternCommunicationEmail

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me.me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

*Edit Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :( *:(

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

Update 2018.04.21

I got the charging led problem now with the community Sailfish X Compact (2.1.3.7) for the very first time.

The phone had run out of power previous night. I plugged it in charger and started the phone normally. When Sailfish had started it requested what to do with already plugged in USB-connection -> I chose 'Charge only'. After this I removed the cable once and plugged it back in just in case, and pressed power button once to close the screen and went to bed.

In the morning the phone showed white LED as it should. When I removed the device from the charging cable (from the end of the charger first), I saw that I had two unread emails in the status screen. When I pressed the power button to close the screen, the white LED was re-lit again. I do not remember whether I removed the cable from phone end after or before pressing the power button to close the screen.

I gathered dmesg, battery status and full journal from the device, first with the defaults and then with verbosity at notice level, both PatternBatteryCharging on and off. Luckily the journal is not very long, since the device had run out of juice in the previous night.

Journal_20180421.zip Journal_20180421.zip containing logs can be found here: https://perhonen.net/s/28P2o4FkHyoHJwr

Note:

I have PatternCommunicationEmail disabled and haptic and ffmemless lines deleted in email.ini to prevent reacting for emails. The phone screen still lights up though, as I haven't found a way to disable that.

I still suspect that receiving emails during charging phase may be triggering the problem, especially as it lights up the screen. I don't get emails often and I have email active hours from 7:00am to 10:00pm, but this time I woke up later and removed the device from the charger later than usual.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

Update 2018.04.21

I got the charging led problem now with the community Sailfish X Compact (2.1.3.7) (2.1.3.7) for the very first time.

The phone had run out of power previous night. I plugged it in charger and started the phone normally. When Sailfish had started it requested what to do with already plugged in USB-connection -> I chose 'Charge only'. After this I removed the cable once and plugged it back in just in case, and pressed power button once to close the screen and went to bed.

In the morning the phone showed white LED as it should. When I removed the device from the charging cable (from the end of the charger first), I saw that I had two unread emails in the status screen. When I pressed the power button to close the screen, the white LED was re-lit again. I do not remember whether I removed the cable from phone end after or before pressing the power button to close the screen.

I gathered dmesg, battery status and full journal from the device, first with the defaults and then with verbosity at notice level, both PatternBatteryCharging on and off. Luckily the journal is not very long, since the device had run out of juice in the previous night.

Journal_20180421.zip containing logs can be found here: https://perhonen.net/s/28P2o4FkHyoHJwr

Note:

I have PatternCommunicationEmail disabled and haptic and ffmemless lines deleted in email.ini to prevent reacting for emails. The phone screen still lights up though, as I haven't found a way to disable that.

I still suspect that receiving emails during charging phase may be triggering the problem, especially as it lights up the screen. I don't get emails often and I have email active hours from 7:00am to 10:00pm, but this time I woke up later and removed the device from the charger later than usual.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

Update 2018.04.21

I got the charging led problem now with the community Sailfish X Compact (2.1.3.7) for the very first time.

The phone had run out of power previous night. I plugged it in charger and started the phone normally. When Sailfish had started it requested what to do with already plugged in USB-connection -> I chose 'Charge only'. After this I removed the cable once and plugged it back in just in case, and pressed power button once to close the screen and went to bed.

In the morning the phone showed white LED as it should. When I removed the device from the charging cable (from the end of the charger first), I saw that I had two unread emails in the status screen. When I pressed the power button to close the screen, the white LED was re-lit again. I do not remember whether I removed the cable from phone end after or before pressing the power button to close the screen.

I gathered dmesg, battery status and full journal from the device, first with the defaults and then with verbosity at notice level, both PatternBatteryCharging on and off. Luckily the journal is not very long, long since the device had run out of juice in the previous night.

Journal_20180421.zip containing logs can be found here: https://perhonen.net/s/28P2o4FkHyoHJwr

Note:

I have PatternCommunicationEmail disabled and haptic and ffmemless lines deleted in email.ini to prevent reacting for emails. The phone screen still lights up though, as I haven't found a way to disable that.

I still suspect that receiving emails during charging phase may be triggering the problem, especially as it lights up the screen. I don't get emails often and I have email active hours from 7:00am to 10:00pm, but this time I woke up later and removed the device from the charger later than usual.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

Update 2018.04.21

I got the charging led problem now with the community Sailfish X Compact (2.1.3.7) for the very first time.

The phone had run out of power previous night. I plugged it in charger and started the phone normally. When Sailfish had started it requested what to do with already plugged in USB-connection -> I chose 'Charge only'. After this I removed the cable once and plugged it back in just in case, and pressed power button once to close the screen and went to bed.

In the morning the phone showed white LED as it should. When I removed the device from the charging cable (from the end of the charger first), I saw that I had two unread emails in the status screen. When I pressed the power button to close the screen, the white LED was re-lit again. I do not remember whether I removed the cable from phone end after or before pressing the power button to close the screen.

I gathered dmesg, battery status and full journal from the device, first with the defaults and then with verbosity at notice level, both PatternBatteryCharging on and off. Luckily the journal is not very long since the device had run out of juice in the previous night.

Journal_20180421.zip containing logs can be found here: https://perhonen.net/s/28P2o4FkHyoHJwr

Note:

I have PatternCommunicationEmail disabled and haptic and ffmemless lines deleted in email.ini to prevent reacting for emails. The phone screen still lights up though, as I haven't found a way to disable that.

I still suspect that receiving emails during charging phase may be triggering the problem, especially as it lights up the screen. I don't get emails often and I have email active hours from 7:00am to 10:00pm, but this time I woke up later and removed the device from the charger later than usual.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

Update 2018.04.21

I got the charging led problem now with the community Sailfish X Compact (2.1.3.7) for the very first time.

The phone had run out of power previous night. I plugged it in charger and started the phone normally. When Sailfish had started it requested what to do with already plugged in USB-connection -> I chose 'Charge only'. After this I removed the cable once and plugged it back in just in case, and pressed power button once to close the screen and went to bed.

In the morning the phone showed white LED as it should. When I removed the device from the charging cable (from the end of the charger first), I saw that I had two unread emails in the status screen. When I pressed the power button to close the screen, the white LED was re-lit again. I do not remember whether I removed the cable from phone end after or before pressing the power button to close the screen.

I gathered dmesg, battery status and full journal from the device, first with the defaults and then with verbosity at notice level, both PatternBatteryCharging on and off. Luckily the journal is not very long since the device had run out of juice in the previous night.

Journal_20180421.zip containing logs can be found here: https://perhonen.net/s/28P2o4FkHyoHJwr

Note:

I have PatternCommunicationEmail disabled and haptic and ffmemless lines deleted in email.ini to prevent reacting for emails. The phone screen still lights up though, as I haven't found a way to disable that.

I still suspect that receiving emails during charging phase may be triggering the problem, especially as it lights up the screen. I don't get emails often and I have email active hours from 7:00am to 10:00pm, but this time I woke up later and removed the device from the charger later than usual.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

Update 2018.04.21

I got the charging led problem now with the community Sailfish X Compact (2.1.3.7) for the very first time.

The phone had run out of power previous night. I plugged it in charger and started the phone normally. When Sailfish had started it requested what to do with already plugged in USB-connection -> I chose 'Charge only'. After this I removed the cable once and plugged it back in just in case, and pressed power button once to close the screen and went to bed.

In the morning the phone showed white LED as it should. When I removed the device from the charging cable (from the end of the charger first), I saw that I had two unread emails in the status screen. When I pressed the power button to close the screen, the white LED was re-lit again. I do not remember whether I removed the cable from phone end after or before pressing the power button to close the screen.

I gathered dmesg, battery status and full journal from the device, first with the defaults and then with verbosity at notice level, both PatternBatteryCharging on and off. Luckily the journal is not very long since the device had run out of juice in the previous night.

Journal_20180421.zip containing logs can be found here: https://perhonen.net/s/28P2o4FkHyoHJwr

Edit: I forgot to put battery state file into the zip. It showed following:

/run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:124816
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:301000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:272
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:99986
/run/state/namespaces/Battery/Voltage:4258550

Note:

I have PatternCommunicationEmail disabled and haptic and ffmemless lines deleted in email.ini to prevent reacting for emails. The phone screen still lights up though, as I haven't found a way to disable that.

I still suspect that receiving emails during charging phase may be triggering the problem, especially as it lights up the screen. I don't get emails often and I have email active hours from 7:00am to 10:00pm, but this time I woke up later and removed the device from the charger later than usual.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

Update 2018.04.21

I got the charging led problem now with the community Sailfish X Compact (2.1.3.7) for the very first time.

The phone had run out of power previous night. I plugged it in charger and started the phone normally. When Sailfish had started it requested what to do with already plugged in USB-connection -> I chose 'Charge only'. After this I removed the cable once and plugged it back in just in case, and pressed power button once to close the screen and went to bed.

In the morning the phone showed white LED as it should. When I removed the device from the charging cable (from the end of the charger first), I saw that I had two unread emails in the status screen. When I pressed the power button to close the screen, the white LED was re-lit again. I do not remember whether I removed the cable from phone end after or before pressing the power button to close the screen.

I gathered dmesg, battery status and full journal from the device, first with the defaults and then with verbosity at notice level, both PatternBatteryCharging on and off. Luckily the journal is not very long since the device had run out of juice in the previous night.

Journal_20180421.zip containing logs can be found here: https://perhonen.net/s/28P2o4FkHyoHJwr

Edit: I forgot to put battery state file into the zip. It showed following:

/run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:124816
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:301000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:272
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:99986
/run/state/namespaces/Battery/Voltage:4258550

Note:

I have PatternCommunicationEmail disabled and haptic and ffmemless lines deleted in email.ini to prevent reacting for emails. The phone screen still lights up though, as I haven't found a way to disable that.

I still suspect that receiving emails during charging phase may be triggering the problem, especially as it lights up the screen. I don't get emails often and I have email active hours from 7:00am to 10:00pm, but this time I woke up later and removed the device from the charger later than usual.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

Update 2018.04.21

I got the charging led problem now with the community Sailfish X Compact (2.1.3.7) for the very first time.

The phone had run out of power previous night. I plugged it in charger and started the phone normally. When Sailfish had started it requested what to do with already plugged in USB-connection -> I chose 'Charge only'. After this I removed the cable once and plugged it back in just in case, and pressed power button once to close the screen and went to bed.

In the morning the phone showed white LED as it should. When I removed the device from the charging cable (from the end of the charger first), I saw that I had two unread emails in the status screen. When I pressed the power button to close the screen, the white LED was re-lit again. I do not remember whether I removed the cable from phone end after or before pressing the power button to close the screen.

I gathered dmesg, battery status and full journal from the device, first with the defaults and then with verbosity at notice level, both PatternBatteryCharging on and off. Luckily the journal is not very long since the device had run out of juice in the previous night.

Journal_20180421.zip containing logs can be found here: https://perhonen.net/s/28P2o4FkHyoHJwr

Edit: I forgot to put battery state file into the zip. It showed following:

/run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:124816
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:301000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:272
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:99986
/run/state/namespaces/Battery/Voltage:4258550

Note:

I have PatternCommunicationEmail disabled and haptic and ffmemless lines deleted in email.ini to prevent reacting for emails. The phone screen still lights up though, as I haven't found a way to disable that.

I still suspect that receiving emails during charging phase may be triggering the problem, especially as it lights up the screen. I don't get emails often and I have email active hours from 7:00am to 10:00pm, but this time I woke up later and removed the device from the charger later than usual.

Update 2018.07.11

For the first time I got pulsing LED indicator (battery fully charged indicator) stuck after removing the device from the charger in the morning. Same symptoms otherwise, except this time I did not have any messages or emails during the night. So this was not triggered by any other notification event.

This was with official Sailfish X.

White LED did not go off when the charging cable was removed. It was blinking before taking the cable off and stuck on plain white. This happened after the first overnight full charge.

The light went out when the screen was open, but it turned back on when the screen went off (both by power button or by timeout).

I run csd and its LED test and all of the leds were ok. This did not clear up the problem, and eventually the restart of the phone fixed it.

Briefly testing with the power cable didn't bring the issue up again. I'll report here if I see it again, and try to find something in the dmesg/logs.

Might be important! I had disabled email notification LED with mcetool just previous day with:

mcetool --disable-led-pattern=PatternCommunicationEmail

Update 24.11.2017

I finally had this bug again. It may have been triggered by either: 1) Email arriving at night during the charging, which lights up the screen even when I have disabled the led pattern for it, or 2) timely removal of USB charging cable from the device (the moment when the pulsing LED cycle is at its brightest), or 3) both.

After removing the cable from the phone grep . /run/state/namespaces/Battery/* showed

run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:332792
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:1756000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:245
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:17136
/run/state/namespaces/Battery/Voltage:4258550

That looked pretty normal to me. State is discharging even though the led burns white (or very bright blue, I'm not sure which one it is).

Then I followed spiiroin's advice to increase logging, but grepping journalctl for modules/led or even just led did not show anything at all. I noticed that my phone had not logged _anything_ in journalctl since 12th of November, which explains that.

Current uptime was 22 days, so I lost 12 days for some reason. Last entries were:

Nov 12 15:49:04 Sailfish kernel: PM: suspend exit 2017-11-12 13:49:04.787473380 UTC
Nov 12 15:49:04 Sailfish kernel: PM: suspend entry 2017-11-12 13:49:04.787496391 UTC
Nov 12 15:49:04 Sailfish systemd[1]: Time has been changed
Nov 12 15:49:04 Sailfish mce[1411]: modules/proximity.c: report_proximity(): state: CLOSED -> OPEN
Nov 12 15:49:04 Sailfish systemd[2598]: Time has been changed

I noticed that when I had run mcetool --deactivate-led-pattern=PatternBatteryCharging, the led had gone off when the screen was off too. I activated it again with mcetool --activate-led-pattern=PatternBatteryCharging, and saw that the problem was still on. So at least we know that it is related to PatternBatteryCharging.

At this point I plugged the phone into my computer in developer mode, at which point the led went out and did not light up again.

I'll try to trigger the problem sooner by email and timely removal of USB charging cable.

Edit 2018.01.18: I simply haven't been able to reproduce the problem. It is very rare. Sry! :(

Update 2018.04.21

I got the charging led problem now with the community Sailfish X Compact (2.1.3.7) for the very first time.

The phone had run out of power previous night. I plugged it in charger and started the phone normally. When Sailfish had started it requested what to do with already plugged in USB-connection -> I chose 'Charge only'. After this I removed the cable once and plugged it back in just in case, and pressed power button once to close the screen and went to bed.

In the morning the phone showed white LED as it should. When I removed the device from the charging cable (from the end of the charger first), I saw that I had two unread emails in the status screen. When I pressed the power button to close the screen, the white LED was re-lit again. I do not remember whether I removed the cable from phone end after or before pressing the power button to close the screen.

I gathered dmesg, battery status and full journal from the device, first with the defaults and then with verbosity at notice level, both PatternBatteryCharging on and off. Luckily the journal is not very long since the device had run out of juice in the previous night.

Journal_20180421.zip containing logs can be found here: https://perhonen.net/s/28P2o4FkHyoHJwr

Edit: I forgot to put battery state file into the zip. It showed following:

/run/state/namespaces/Battery/Capacity:100.000000
/run/state/namespaces/Battery/ChargePercentage:100
/run/state/namespaces/Battery/ChargingState:discharging
/run/state/namespaces/Battery/Current:124816
/run/state/namespaces/Battery/Energy:8360000
/run/state/namespaces/Battery/EnergyFull:8360000
/run/state/namespaces/Battery/IsCharging:0
/run/state/namespaces/Battery/Level:normal
/run/state/namespaces/Battery/LowBattery:0
/run/state/namespaces/Battery/OnBattery:1
/run/state/namespaces/Battery/Power:301000
/run/state/namespaces/Battery/State:discharging
/run/state/namespaces/Battery/Temperature:272
/run/state/namespaces/Battery/TimeUntilFull:0
/run/state/namespaces/Battery/TimeUntilLow:99986
/run/state/namespaces/Battery/Voltage:4258550

Note:

I have PatternCommunicationEmail disabled and haptic and ffmemless lines deleted in email.ini to prevent reacting for emails. The phone screen still lights up though, as I haven't found a way to disable that.

I still suspect that receiving emails during charging phase may be triggering the problem, especially as it lights up the screen. I don't get emails often and I have email active hours from 7:00am to 10:00pm, but this time I woke up later and removed the device from the charger later than usual.

Update 2018.07.11

For the first time I got pulsing LED indicator (battery fully charged indicator) stuck after removing the device from the charger in the morning. Same symptoms otherwise, except this time I did not have any messages or emails during the night. So this was not triggered by any other notification event.

Xperia X device. SailfishOS version 2.2.0.29.