Am Sat, 08 Dec 2012 22:09:54 +0000 schrieb xfce-request@lists.fedoraproject.org:
Date: Sat, 8 Dec 2012 13:52:27 -0700 From: Kevin Fenzi kevin@scrye.com To: xfce@lists.fedoraproject.org Subject: Re: Standby and suspend buttons greyed out Message-ID: 20121208135227.1993073f@jelerak.scrye.com Content-Type: text/plain; charset="utf-8"
On Fri, 7 Dec 2012 16:30:28 +0100 (CET) "Raphael Groner" raphgro@web.de wrote:
{…}
I'm not sure what you are asking here... can you rephrase?
We cannot directly modify the defaults in logind.conf... that would be something for the end user. Our choices are to add a inhibit to session (which I don't like), or get xfce4-power-manager to handle inhibiting (which I think is the correct solution).
kevin
Yeah, I just wanted to sum up what's the possible solution so far for the end user. But you can do it obviously much shorter. :)
I don't understand how the session is related to that issue in any case
xfce4-power-manager uses a dbus interface to instruct the power management component. Is upower already merged to systemd, will it be in some future version (any plans to do that)?
KDE (and GNOME) have similiar issues. The main bug seems to be. https://bugzilla.redhat.com/show_bug.cgi?id=859227
- and then for GNOME in particular: https://bugzilla.redhat.com/show_bug.cgi?id=859224
I tend to give up on that hibernate/suspend/sleep well/simulate death thingy when reading through all those comments. WTF external monitors need special handling? Cause it gets more and more complex. Yeah, it should be routed to the end user that can configure it fully like she wants. SCNR to do some trolling here.
Raphael
On Sun, 9 Dec 2012 12:09:20 +0100 Raphael Groner raphgro@web.de wrote:
Yeah, I just wanted to sum up what's the possible solution so far for the end user. But you can do it obviously much shorter. :)
Yeah. We might want to make a note in release notes or have a wiki page to point people to...
I don't understand how the session is related to that issue in any case
xfce4-power-manager uses a dbus interface to instruct the power management component. Is upower already merged to systemd, will it be in some future version (any plans to do that)?
I don't know off hand.
KDE (and GNOME) have similiar issues. The main bug seems to be. https://bugzilla.redhat.com/show_bug.cgi?id=859227
Yeah, I am seeing that same thing here with inhibit... ;(
- and then for GNOME in particular:
https://bugzilla.redhat.com/show_bug.cgi?id=859224
I tend to give up on that hibernate/suspend/sleep well/simulate death thingy when reading through all those comments. WTF external monitors need special handling? Cause it gets more and more complex. Yeah, it should be routed to the end user that can configure it fully like she wants. SCNR to do some trolling here.
Yeah, ideally users only need to configure things in xfce4-power-manager and not change random config files and hunt around. ;(
kevin
On 12/09/2012 03:53 PM, Kevin Fenzi wrote:
On Sun, 9 Dec 2012 12:09:20 +0100 Raphael Groner raphgro@web.de wrote:
Yeah, I just wanted to sum up what's the possible solution so far for the end user. But you can do it obviously much shorter. :)
Yeah. We might want to make a note in release notes or have a wiki page to point people to...
I don't understand how the session is related to that issue in any case
xfce4-power-manager uses a dbus interface to instruct the power management component. Is upower already merged to systemd, will it be in some future version (any plans to do that)?
I don't know off hand.
KDE (and GNOME) have similiar issues. The main bug seems to be. https://bugzilla.redhat.com/show_bug.cgi?id=859227
Yeah, I am seeing that same thing here with inhibit... ;(
- and then for GNOME in particular:
https://bugzilla.redhat.com/show_bug.cgi?id=859224
I tend to give up on that hibernate/suspend/sleep well/simulate death thingy when reading through all those comments. WTF external monitors need special handling? Cause it gets more and more complex. Yeah, it should be routed to the end user that can configure it fully like she wants. SCNR to do some trolling here.
Yeah, ideally users only need to configure things in xfce4-power-manager and not change random config files and hunt around. ;(
kevin
Xfce Power Manager handles the suspend for me (when in Xfce) and I have no issues. Do you guys have trouble with it?
On Sun, 09 Dec 2012 16:04:22 -0200 Sergio secipolla@gmail.com wrote:
Xfce Power Manager handles the suspend for me (when in Xfce) and I have no issues. Do you guys have trouble with it?
Did you modify /etc/systemd/logind.conf or are you running systemd-inhibit?
If not, then it's likely systemd that's doing the suspend, which is fine, but in the case where you set say 'do nothing' on lid close in xfce4-power-manager, systemd will not know or care about that and suspend on lid close.
kevin
On 12/09/2012 04:08 PM, Kevin Fenzi wrote:
On Sun, 09 Dec 2012 16:04:22 -0200 Sergio secipolla@gmail.com wrote:
Xfce Power Manager handles the suspend for me (when in Xfce) and I have no issues. Do you guys have trouble with it?
Did you modify /etc/systemd/logind.conf or are you running systemd-inhibit?
No.
If not, then it's likely systemd that's doing the suspend, which is fine, but in the case where you set say 'do nothing' on lid close in xfce4-power-manager, systemd will not know or care about that and suspend on lid close.
Ok, no lid here :-)
The feature to lock the screen when going for suspend/hibernate works properly (but I suspect it actually locks the screen after resuming). Just mentioning, I know it may well have nothing to do with this issue.
On 12/09/2012 04:16 PM, Sergio wrote:
On 12/09/2012 04:08 PM, Kevin Fenzi wrote:
On Sun, 09 Dec 2012 16:04:22 -0200 Sergio secipolla@gmail.com wrote:
Xfce Power Manager handles the suspend for me (when in Xfce) and I have no issues. Do you guys have trouble with it?
Did you modify /etc/systemd/logind.conf or are you running systemd-inhibit?
No.
First time I looked at that file was right now. But I think it's being inhibited by something (not systemd-inhibit) because the behaviour hasn't changed from F17. My guess is that it could be acpid that inhibits it.
I run acpid and it's set to run 'systemctl 'Poweroff'' on power button press and 'pm-suspend' on sleep button press. This works just like that out of X, in Lightdm's screen or in IceWM. In Xfce I have xfce4-power-manager running and it takes precedence. How do I know? Because in Xfce the power button is configured to 'Ask' and the sleep button is configured to suspend but I know it's xfce4-power-manager managing it because it locks the screen on resume (suspending from the log-out dialogue doesn't lock the screen probably because I haven't set it up in 'Session and Startup' settings or something else).
So either acpid or xfce4-power-manager inhibits systemd's power managing functions.
If not, then it's likely systemd that's doing the suspend, which is fine, but in the case where you set say 'do nothing' on lid close in xfce4-power-manager, systemd will not know or care about that and suspend on lid close.
Then in my case it must be acpid inhibiting it. You know that, of course, but /etc/systemd/logind.conf has LidSwitchIgnoreInhibited=yes
Ok, no lid here :-)
The feature to lock the screen when going for suspend/hibernate works properly (but I suspect it actually locks the screen after resuming). Just mentioning, I know it may well have nothing to do with this issue.
Actually it does (answering myself).
Just givin' further feedback.
On Mon, 10 Dec 2012 09:27:54 -0200 Sergio secipolla@gmail.com wrote:
First time I looked at that file was right now. But I think it's being inhibited by something (not systemd-inhibit) because the behaviour hasn't changed from F17.
How do you reach that conclusion? I suspect rather that systemd is doing the same job you expect, so it doesn't look like it changed, even though it has.
My guess is that it could be acpid that inhibits it.
I run acpid and it's set to run 'systemctl 'Poweroff'' on power button press and 'pm-suspend' on sleep button press. This works just like that out of X, in Lightdm's screen or in IceWM.
If you change that value in acpid, does it take affect?
In Xfce I have xfce4-power-manager running and it takes precedence. How do I know? Because in Xfce the power button is configured to 'Ask' and the sleep button is configured to suspend but I know it's xfce4-power-manager managing it because it locks the screen on resume (suspending from the log-out dialogue doesn't lock the screen probably because I haven't set it up in 'Session and Startup' settings or something else).
So either acpid or xfce4-power-manager inhibits systemd's power managing functions.
xfce4-power-manager has no ability to do this.
I would be pretty surprised if acpid did.
Then in my case it must be acpid inhibiting it. You know that, of course, but /etc/systemd/logind.conf has LidSwitchIgnoreInhibited=yes
Thats not default.
does: rpm -V systemd show the file has been modified?
kevin
- systemd-inhibit isn't running - xfce4-power-manager manages for sure the 'power' and the 'sleep' keys in Xfce for me just as in F17. I know that because of what I wrote before: when I hit the power key the logout dialogue pops-up and when I hit the sleep key the system suspends and locks the screen on resume (only xfce4-power-manager is set to do that).
So I said that it could be acpid inhibiting it because of you saying that xfce4-power-manager doesn't do that. But one other possibility is simply that the 'power', 'suspend', 'hibernate' and 'lidswitch' keys are not the usual 'power' and 'sleep' keys in the keyboard. I'm on a desktop PC, not a laptop.
This is the default, untouched, /etc/systemd.logind.conf
# This file is part of systemd. # # systemd is free software; you can redistribute it and/or modify it # under the terms of the GNU Lesser General Public License as published by # the Free Software Foundation; either version 2.1 of the License, or # (at your option) any later version. # # See logind.conf(5) for details
[Login] #NAutoVTs=6 #ReserveVT=6 #KillUserProcesses=no #KillOnlyUsers= #KillExcludeUsers=root #Controllers= #ResetControllers=cpu #InhibitDelayMaxSec=5 #HandlePowerKey=poweroff #HandleSuspendKey=suspend #HandleHibernateKey=hibernate #HandleLidSwitch=suspend #PowerKeyIgnoreInhibited=no #SuspendKeyIgnoreInhibited=no #HibernateKeyIgnoreInhibited=no #LidSwitchIgnoreInhibited=yes
What I said is that 'LidSwitchIgnoreInhibited=yes' just does what you and Raphael were talking about: if the user don't change that value then xfce4-power-manager's settings are ignored (in case it inhibits systemd-logind.service).