Confirmed. This is a bug in /etc/acpi/events/ibm-sleepbtn, which is configured to invoke /etc/acpi/sleepbtn.sh instead of /etc/acpi/sleep.sh: the former calls acpi_fakekey, the latter is what will trigger a sleep event when g-p-m is not listening.
Confirmed. This is a bug in /etc/acpi/ events/ ibm-sleepbtn, which is configured to invoke /etc/acpi/ sleepbtn. sh instead of /etc/acpi/sleep.sh: the former calls acpi_fakekey, the latter is what will trigger a sleep event when g-p-m is not listening.