Hi, short introduction (for Matthew mainly): hal has been modified to rely on acpi-support if the package is installed (cf #381708). Now when gnome-power-manager is installed, acpi-support's scripts do nothing and thus the change doesn't have the expected result. That's the point where Sjoerd cced me and asked:
On Wed, 23 Aug 2006, Sjoerd Simons wrote: > On Tue, Aug 22, 2006 at 01:05:39AM +0200, Clément Stenac wrote: > > acpi-support detects if gnome-power-manager ok its kde counterpart is > > running, and if so, will disable its sleep/hibernate scripts, unless you > > call the scripts with the "force" argument. > > > > I don't know exactly who should do what, either you add the force > > argument, or acpi-support changes its behavior, or g-p-m does > > something,... > > Hopefully the acpi-support maintainer can tell us something more about this ;) > I've cc'd him in this mail. > > I suppose the sleep/hibernate scripts are disabled so there is no conflict > with > a power-manager and the scripts. Although i don't think detecting random > power-managers is actually scalable... > > At some point we really need a good generic power-management structure in > debian, adding support for all kinds of programs/scripts to the hal scripts > just doesn't scale :( I'm maintaining acpi-support but I'm really not much involved in all this stuff and I can't answer you. However Matthew, the upstream author of acpi-support can enlighten us on the right way to go forward. When I packaged acpi-support, Matthew made me clear that it's only a temporary hack to make things work until a better solution is worked out. As such I'm wondering if it's wise to make hal rely on acpi-support at all... Matthew, what do you think? Can you also tell us how Ubuntu is going forward on that front? Cheers, -- Raphaël Hertzog Premier livre français sur Debian GNU/Linux : http://www.ouaza.com/livre/admin-debian/
