[systemd-devel] [logind] retroactive lid close handle upon vt switch

[b3nmore]

On 01/23/2015 02:56 PM, Lennart Poettering wrote:
> Yeah, this is intentional, button inhibtiros are bound to the active
> session, as they are about processing input keys, and only the fg
> session should be able to do that.

One could argue here, that the lid close event has been already
processed by the than active session and a session, which gets activated
after the event, should not react retrospectively on it(?).

> Why precisely does your original session inhibit the lid switch? If
> you want to turn off the lid switch then turn it off properly,
> inhibition is not really about turning something fully off. It's about
> temporarily making logind not process it, for example, because you
> want to process it yourself or so.
> 
> GNOME for example never inhibits the lid switch, because there's
> really no reason to. Why does your DE inhibit it?

xfpm (the power manager of xfce) allows to configure how the system
should react to certain power events. In this case you can configure it
to either suspend or hibernate or lock the screen or just switch off the
screen, when the lid is closed. In order to do so, xfpm inhibits the
handle of the lid switch and initiates the configured pm action on its own.
It works as intended with one exception, when one uses a screen locker,
which switches vt's (other lockers are o.k.).