Description of problem: It happens quite often that I close the lid of my laptop (DELL latitude 600), and the system doesn't suspend - in fact, nothing happens. After opening and closing again, it suspends. Very peculiar. Further, it has happened twice so far, and is quite worrying, that the system failed to return after suspension - it was totally dead no X, no text terminals. Version-Release number of selected component (if applicable): 2.18.0 How reproducible: it happens quite often, but you can never tell if it will happen or not Steps to Reproduce: 1. close laptop lid 2. open it 3. close again Actual results: system doesn't suspend Expected results: system should suspend Additional info:
if you do: killall gnome-power-manager gnome-power-manager --verbose --no-daemon Can you please attach the output after you closed the lid, waited 30 seconds, then opened it again please. Thanks.
Created attachment 156405 [details] following the instructions given above, this is the output of 'gnome-power-manager --verbose --no-daemon'
Thanks. >[gpm_control_is_policy_timout_valid] gpm-control.c:127 (23:59:22): Skipping suppressed policy event Can you try setting /apps/gnome-power-manager/policy_suppression_timeout to 0 in gconf editor and trying again? Thanks.
Created attachment 156496 [details] output of '$ gnome-power-manager --verbose --no-daemon' unfortunately, still not works properly. what I've done: # yum install gconf-editor $ gconf-edit (required change done: /apps/gnome-power-manager/policy_suppression_timeout=0 $ gnome-power-manager --verbose --no-daemon what happened: the first attempt: system suspended succesfully 2nd attempt: nothing happened 3rd attempt: system suspended succesfully again (the log contains all these, I think)
Odd. You only get 1 lid-up events, not 3. Can you suspend, and then resume, and then do: cat /proc/acpi/button/*/*/* Also, run "sleep 5 && cat /proc/acpi/button/*/*/*" shut the lid, wait 10 seconds, and post the output. Also, "cat /proc/acpi/button/*/*/*" when you've just opened the lid would be good. Cheers. Richard.
here are the outputs: > Can you suspend, and then resume, and then do: > cat /proc/acpi/button/*/*/* type: Lid Switch state: open type: Power Button (CM) type: Sleep Button (CM) > sleep 5 && cat /proc/acpi/button/*/*/*" shut the lid, wait 10 seconds, type: Lid Switch state: closed type: Power Button (CM) type: Sleep Button (CM) > cat /proc/acpi/button/*/*/*" when you've just opened the lid would be good. type: Lid Switch state: open type: Power Button (CM) type: Sleep Button (CM) can it be the problem that I upgraded my previous fedora 6 to fedora 7? I mean, some rubbish just remained somewhere accidentally.
Created attachment 157120 [details] kernel patch Found the problem, it was a kernel bug. The attached patch fixes it, and is hopefully going to be in 2.6.22.
Cool! I applied the patch to the kernel, and now it's perfect.
Thanks for the quick testing. Could you add to the email thread the fact that it works for you? Thanks.
Dave, could you add this to the next build of fedora 7 updates please? I'm hoping it's in time for .22, but it might have missed it.
(In reply to comment #9) > Thanks for the quick testing. Could you add to the email thread the fact that it > works for you? Thanks. Sorry, which email thread?
It's okay, I've added to the lkml thread that this works for you, and that it should try and get into .22. Fingers crossed.
OK, I got it! By the way, the system failed to resume again, as I mentioned in the 1st comment. So it's suspended, and when I open the lid, the screen remain blank, and nothing can be done... It doesn't even respond to CapsLock or NumLock button. What I could save is a short part of /var/log/messages: Jun 16 18:01:23 ab686 gnome-power-manager: (eger) Suspending computer because the lid has been closed on ac power Jun 16 18:18:53 ab686 restorecond: Read error (Interrupted system call) Jun 16 18:18:53 ab686 gnome-power-manager: (eger) DBUS timed out, but recovering Jun 16 18:18:53 ab686 gnome-power-manager: (eger) Resuming computer Jun 16 18:20:52 ab686 syslogd 1.4.2: restart. Shall I create a new bug report or do you think this is still related to this kernel bug?
>It doesn't even respond to CapsLock or NumLock Then you need to have a look here: http://people.freedesktop.org/~hughsient/quirk/
-> kernel
oh, my goodness, it does the same in the 2.6.22 kernel...
It seems the problem has not disappeared with kernel 2.6.22.1-27.fc7. In order to get my ibm x40 to sleep, I must stillclose the lid twice. Otherwise, power management seems to work fine.
2.6.22.4-65.fc7 still has this issue. It goes away witht the patch.
Still not fixed in 2.6.23 testing. (tested on Acer Travelmate 8000)
FC8 same problem. Suspend manually works all the time.
Has the patch managed to make it into the kernel yet?
It's fixed in FC8. Check out bug: 275651.
Given that information, I'll go ahead and close this CURRENTRELEASE.
Patch went in 2.6.23.9-43
Took a while, didn't it? ;)