acpid: sleep and suspend no longer work in Edgy

 unsubscribe ubuntu-security

--
 - mdz

Matthew, do you have an idea here?

Does

sudo pmi action sleep

work?

I tied both hibernate and sleep actions, but using sudo as a user and from console as root. The result is the same in both cases:

Hibernate:

Successfully saves memory content to the swap partition and powers down. Then, out of the blue, about 2 minutes after that, it powers up. From the noise it makes, I'd say it reboots up to GDM, except that when it powers up, it returns to a pitch black display. Pressing any key or moving the mouse doesn't help. The only way to get something on the display again is to forcefully power off by holding the power button long enough for the hardware to notice and then power up manually to restart the system.

Sleep:

Starts saving memory to swap and then hangs with the drive LED remaining on. Left to its own devices and coming back 5 minutes later to check, it remained there. Again, the only way to get something out of the computer is to forcefully power off and reboot as described above.

One detail I forgot to add previously:

Hibernate and sleep previously worked fine with the Ubuntu 2.6.17 kernel.

A month ago, I had upgraded just Required packages and the kernel to Edgy, all while keeping the rest to Dapper (using APT preferences to pin * to Dapper), to be able to benefit from certain new kernel drivers added since 2.6.15. It only stopped working after I removed Dapper APT sources to completely upgrade to Edgy. This pulled in newer acpid, apci-utils and other software to manipulate ACPI from the GNOME logout panel.

I loose my fn key functions and acpi support (suspend to ram, hibernate) since the last update of my kernel (kernel-image-generic-2.6.17-8/9/10) works under 2.6.17-7 :-(

Unsurprisingly the change log for 2.6.17-8.22 says:

* Merged 154 commits from ACPI upstream.

It might be a good idea to revert these and instead backport only those few that made it into 2.6.18.

I would like to confirm this behaviour on an IBM thinkpad x31, upgraded from dapper to edgy.

suspending and hibernating blanks the screen (as per screensaver setup) and starts the "moon" LED blinking, but nothing firther happens. Only hard shutdown will bring the machine back.

As confirmed by some people in this bug's thread, the problem appears to come from recent ACPI backports in the 2.6.17 kernel sources. Reassigning.

This seems to have been fixed with recent upgrades to the edgy kernel. At least I can now suspend on my thinkpad x31

I confirm this behavior (see 64922).

Martin, when you're putting your laptop to sleep state can you switch to X (Ctrl+Alt+F7 or Alt+F7) ?

I confirm this: Edgy up to date has the same problem for a Thinkpad T60.

The sleep and hibernate sometimes work, and most times don't. Rather inconsistently. I have noticed that sometimes my swap partition is not on at all, I have to manually go to gparted and right-click and "swapon". Then 'top' correctly reports the swap. It's weird that in dmesg it says:

[17179653.888000] Adding 4425868k swap on /dev/sda7. Priority:-1 extents:1 across:4425868k

and later on in the boot messages I see that "mounting swap partition" reports "ok". But the swap is never on.

Even when manually activating swap, sleep and hibernate fail most of the time.

In thinkwiki.org there is a user reporting that disabling the powernowd (which controls CPU frequency scaling) allows for sleep to work. I've tried, and results are the same or worse: sleep works very little times.

A related issue is that when sleep works, sometimes it takes about 5 seconds, and most others something like 30 seconds. Independenlty of the number of open programs.

This little bug (when compared to an entire OS) makes ubuntu look very bad.

By the way when the modem is disabled, sleep NEVER works, perhaps because of the failing sound driver that also doesn't work without the modem enabled in the BIOS.

I'm using this kernel:
Linux pad 2.6.17-10-generic #2 SMP Tue Dec 5 22:28:26 UTC 2006 i686 GNU/Linux

This bug concerns an old release but I confirm that the issue still exists on Feisty with 2.6.20. As far as I can tell, changes made for 2.6.18 by upstream at Intel were backported to Ubuntu packages after 2.6.17-7, which is where it broke previous ACPI behavior on a number of ThinkPads.

I am assigning this bug to the 'ubuntu-kernel-team' per their bug policy. For future reference you can learn more about their bug policy at https://wiki.ubuntu.com/KernelTeamBugPolicies .

The 18 month support period for Edgy Eft 6.10 has reached it's end of life. As a result, we are closing the linux-source-2.6.17 Edgy Eft kernel task.

Hardy Heron 8.04 was recently released. It would be helpful if you could test the new release and verify if this is still an issue - http://www.ubuntu.com/getubuntu/download . You should be able to test your bug using the LiveCD. Please let us know your results. Thanks.

That's what one could call timely bug triaging. Pretty amazing how such bugs remain unattended for so long. Of course, by the time someone actually looked into it, we've already been through a number of Ubuntu releases... Yes, sleep and suspend work in Hardy.

Nice, so I put it as fixed.

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.