Red Hat Bugzilla – Bug 121061
"apm -s" is the default suspend command even with ACPI
Last modified: 2013-03-05 22:40:52 EST
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040312
Description of problem:
The default suspend command of the applet is: /usr/bin/apm -s
Doesn't work because:
[kvdb@t-7-182 kvdb]$ apm --version
apm version 3.0.2
[kvdb@t-7-182 kvdb]$ apm
No APM support in kernel
Since the default kernels don't have APM enabled, the applet should
have another default (some ACPI sleep command?). Not?
Gnome battstat 0
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. read above
*** This bug has been marked as a duplicate of 121376 ***
Wrong bug, sorry.
Sounds reasonable. Just need to figure out what the ACPI command is.
A problem is that the default suspend command is specified in a GConf
schema - hard (or at least unusual) to change a schema default
Might want to have two GConf keys - suspend_command_apm and
*** Bug 125562 has been marked as a duplicate of this bug. ***
Mark, a good suspend command would be:
"echo 3 > /proc/acpi/sleep"
*** Bug 130685 has been marked as a duplicate of this bug. ***
I suggest that we add a command with the following properties:
- it works for both apm and acpi systems
- users have permission to run it by default (probably using
consolehelper so it's only console users)
and then set that as the command. Dynamically changing gconf stuff
is wonky. And we don't want to have to go around fixing everything
next time the /proc "interfaces" change ;-)
Perhaps this is just an extra flag to "shutdown"?
Unfortunately the command "suspend" already exists
Random thought: Would "standby" be a suitable name for such a command?
Bill do you have any objection to adding such a command?
Go for it. The problem is that the semantics change, of course; the
apm suspend returns, the acpi suspend doesn't return until you resume,
Taking off FC3 target tracker list
Okay, spliting out a separate bug for the "we need a standby command"
into bug #143397
*** Bug 140667 has been marked as a duplicate of this bug. ***
David is actually working on a better solution to this (don't ask for a command
at all). I'm going to reassign this bug to him.
Actually, I recently filed bug 155613 for inclusion of suspend scripts. Said
scripts are a prequisite for resolving this bug.
I also got a patch for battstat-applet-2 to remove the text entry field and
provide some combo boxes for actions to take when a) lid is closed; b) suspend
button is pressed; and c) power button is pressed.
For FC5 we'll probably just ship gnome-power-manager instead of the battstat
applet; see firstname.lastname@example.org archives for details (it's under development).
Moving to FC5Target from discussion during desktop core team meeting.
Suspend was moved to the menu, so I guess you could call this fixed.