C Sights wrote: > Package: kpowersave > Version: 0.7.2-3 > Severity: normal > > Hi, > There seems to be some conflict between whether kpowersave or > powersaved > handles button presses. After suspending to RAM, then pushing the power > button, the computer wakes up then immediately suspends to disk. > I think powersaved is intercepting the power button event and > hibernating the > computer. The computer wakes up normally if powersaved is not installed. > One oddity is that if I suspend to RAM using the kpowersave icon, then > push > the power button, the computer wakes up normally without immediately suspend > to disk -ing. > Here is a table summarizing the observations: >
Your observation is correct. Beginning with the 0.7 series, kpowersave does not depend on powersaved anymore, but defines the policy (what to do on button presses etc.) itself and calls the corresponding HAL methods on its own. This will interere with powersaved < 0.15, which also defines its own policy, so you get to processes reacting on the same event. The one which reacts first, wins. Beginning with powersaved 0.15, this will be solved properly. powersaved claims the D-Bus name org.freedesktop.Power, when it starts and is then responsible for the policy. Whenever you start a graphical session, with kpowersave, it will take over the D-Bus interface org.freedesktop.Power and then is solely responsible for the powermanagement. When you log out again, powersaved takes over again. Unfortunately, powersaved-0.15 depends on a package called PolicyKit in a version, which has been abandoned upstream. So powersaved 0.15 will not enter unstable in the current form. Maybe I can patch out the PolicyKit requirement in powersaved and use other authorization mechanisms, based on group memberships e.g. This will take time though. Sorry for the inconvenience. Michael -- Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth?
signature.asc
Description: OpenPGP digital signature
