• Support
  • Power button action - shutdown

elcobi How about dmesg | grep -i button? Thanks for helping to debug.

SOLUSfiddler So it is receiving something when you press the power button. But what I'm wondering is if what it's receiving is not a power button event, but a sleep button event, which GNOME might not be using as a trigger to launch the "Power Button Action".

When I give dmesg | grep -i button, I have two power "buttons": PWRB and PWRF. If I understand correctly, PWRB is a normal power button press. When I short press it, this is the button that gets received and then GNOME is showing the shutdown dialog as desired. PWRF is for Force shutdown (?) when you hold down the power button.

Can anyone else confirm this? Also, if this was working for you before, I wonder if something in the kernel or ACPI domain changed causing this.

    tomocafe
    Hi!
    Just gave dmesg | grep -i button again and this is the output:

    [ 4.012866] input: Sleep Button as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0C0E:00/input/input3
    [ 4.029803] ACPI: Sleep Button [SLPB]
    [ 4.029901] input: Power Button as /devices/LNXSYSTM:00/LNXPWRBN:00/input/input4
    [ 4.031788] ACPI: Power Button [PWRF]
    [ 4.178622] input: ThinkPad Extra Buttons as /devices/platform/thinkpad_acpi/input/input6
    [ 5.496297] psmouse serio2: trackpoint: IBM TrackPoint firmware: 0x0e, buttons: 3/3

    And, yes - it used to work perfectly and every single time before.

    [Will be back tomorrow - stay safe and sleep tight!]
    SOLUSfiddler

    So I was finally able to reproduce this. Unfortunately, I have no idea what the root cause is. I powered off my machine at least thirty times and tested. About half the time it’s working, the other times not.

    When I run sudo journalctl -f in the terminal and then press the power button, I always see a message stating that the power button was pressed (i.e. the event is received) but on some boots, gnome-session-binary shows the shutdown dialog and it’s recorded in the journal; on other boots, nothing happens after the button event is recorded.

    I even compared the journal entries from the working and non-working boots, but didn’t see any obvious differences.

    Will keep digging around, but does anyone else have a hint or suggestion?