Hibernation problems with some hardware?

Bug #28149 reported by Celso Pinto
14
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Invalid
Low
Unassigned

Bug Description

I've tried the hibernate feature in Dapper and when resuming, apparently
everything loads but the graphical environment never comes up. It's possible to
hear the fan firing up (in normal usage this happens when processor frequency
reaches the maximum) so I guess it's trying to do something.

It's also impossible to switch to a console with any Alt+Function combinations
(e.g Alt-F1). The only way out is to cold-boot the laptop.

The logs say nothing, not even about a resume from hibernation.

Revision history for this message
Matthew Garrett (mjg59) wrote :

Did it work in Breezy? Which video drivers are you using?

Revision history for this message
Loic Pefferkorn (loic) wrote :

Doesn't work for me too.

Thinkpad T43

Tested in Breezy, at wakeup I got:

Stopping Tasks:==============================|
Freeing memory...done (91171 pages freed)
ACPI: PCI interrupt for device 0000:04:00.0 disabled
ACPI: PCI interrupt for device 0000:00:1e.2 disabled
resume= option should be used to set suspend device.............
swsusp: Need to copy 26579 pages.

Then nothing happens, no CPU activity (no fan activated), no hard drive activity (led doesn't blink).
All I can do is manual reboot.

In Dapper I got messages from usplash, then "Reading image data:x%", "ACPI: PCI interrupt(...)" messages, after I got a black screen, only thing to do is a manual reboot.

Breezy xorg's driver: i810
Dapper xorg's driver: vesa

Swap is on a lvm volume.

Revision history for this message
Celso Pinto (cpinto) wrote :

IIRC it did work in Breezy. It definitely worked in Hoary because that's what I installed on the laptop the first time it ever booted. Video driver is i810.

Revision history for this message
mtkuiken (matt-kuiken) wrote :

I had a similar issue in Breezy on a Vaio TX650P. The problem started after I installed Firefox 1.5.

After checking my disk drives, I found that my root drive only had 1.1Gb of storage available, while my main memory size is 1.5Gb. I have since re-partitioned the hard drive to give me more space on the root partition. I do not have the new install up and running, so I can't test my theory, but I think that the hibernation file was not completed because the file system ran out of space. When the system is booting, once the portion of the file that can be loaded has been loaded, the system hangs. Where does the hibernation file get stored? Could the people who are having this problem please check the drive that the file is on to ensure that they have enough space to store it? I'dlike to know if this is the problem, as hibernation was really sweet when it worked.

If this is indeed the problem, it would be really nice if either entering hibernation caused an error, or the error was correctly reported during the boot sequence, rather than hanging the computer.

-Matt Kuiken

Revision history for this message
mtkuiken (matt-kuiken) wrote :

Oops, I forgot to note, my chip set is an Intel i915 chipset. I was using the drivers that the install CD set up for it.

I post this in case my theory from above is incorrect.

-Matt Kuiken

Revision history for this message
Emmanuele Bassi (ebassi) wrote :

I can confirm this bug.

Hibernation (suspend to disk) stopped working after the last kernel upgrade into dapper, from 2.6.12 to 2.6.15: my notebook correctly shuts down but it doesn't come up. stand by (suspend to ram) still works.

I have a PB E6300, with a VIA VT8237 PCI bridge and a Unichrome Pro graphic card (running with the X.org vesa driver). kernel package is linux-image-2.6.15-8-k7

Revision history for this message
Emmanuele Bassi (ebassi) wrote :

oh, and there's no need to cold-boot the machine: just use the SysRq keys combo to reboot it safely: keep down Ctrl and SysRq, then press S, I, S, U, and B. this will sync, unmount and reboot without side-effects on the next boot.

this also proves that the kernel is still alive and kicking.

Revision history for this message
Celso Pinto (cpinto) wrote :

@mtkuiken

Hibernate, AFAIK, stores the data in the swap partition. In any case, I have 1Gb ram, swap size is 1.5 and both root and /home partitions have >10Gb available.

@Matthew Garrett

Why is a bug that may incur in data loss marked as Normal?

Revision history for this message
efittery (elmer-fwd) wrote :

I installed Ubuntu 8.04 about two weeks ago.

Normally I just turn do a "Shut Down" when I am finished using my machine.

Anyhow, I decided to try "Hibernate" for no real reason other than to just try it.

From that point on, whenever I power my machine up, I get stopped in the boot phase
and am informed that

Pri Master hard Disk S.M.A.R.T. status BAD, Backup and Replace

Up until I tried using the "Hibernate" option, I never got this message and the system just came up without any warnings.

By the way, if I just ignore the warning and hit the F1 key, the ubuntu comes up and seems to function as always.

Revision history for this message
Sebastian Nohn (sebastian-nohn) wrote :

- Worked with 7.04 on IBM T43p (ATI Driver)
- Did not work with 8.04 on same machine
- Works with 8.04.1 on same machine

Hope that helps to isolate the cause of that problem

Revision history for this message
Marco Diaz (zethabyte) wrote :

I have this problem with Ubuntu 8.10 and compaq presario CQ50-130LA

see this bug
https://bugs.launchpad.net/ubuntu/+source/acpi-support/+bug/35410

Revision history for this message
Danilo Piazzalunga (danilopiazza) wrote :

On a system with Intel i915 chipset, this problem has been caused by bug #302421.

Revision history for this message
penalvch (penalvch) wrote :

Celso Pinto, lease execute the following command, as it will automatically gather debugging information, in a terminal:
apport-collect 28149

affects: acpi-support (Ubuntu) → linux (Ubuntu)
Changed in linux (Ubuntu):
importance: Medium → Low
status: Confirmed → Incomplete
dino99 (9d9)
Changed in linux (Ubuntu):
status: Incomplete → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.