IBM/Lenovo Laptops crash when waking up from hibernation

[Don Cristóbal]

Several (or all?) IBM/Lenovo Laptop models don't hibernate properly. They go to sleep alright, but during the waking up process, they freeze totally. In many (all?) cases the caps lock LED keeps blinking. Nothing except physically turning off the computer can unblock the system. After that, the system is booted ordinarily and the hibernated session is lost.
This happens not every time, but very often (maybe one out of three times).

ProblemType: KernelOops
Annotation: This occured during a previous hibernate and prevented it from resuming properly.
Architecture: i386
DistroRelease: Ubuntu 9.04
ExecutablePath: /usr/share/apport/apportcheckresume
Failure: hibernate/resume
InterpreterPath: /usr/bin/python2.6
MachineType: IBM 2373WBZ
Package: linux-image-2.6.28-11-generic 2.6.28-11.37
ProcAttrCurrent: unconfined
ProcCmdLine: root=UUID=d8ef4b66-2553-41ee-b734-5eb4d2214127 ro quiet splash
ProcCmdline: /usr/bin/python /usr/share/apport/apportcheckresume
ProcEnviron: PATH=(custom, no user)
ProcVersionSignature: Ubuntu 2.6.28-11.37-generic
SourcePackage: linux
StressLog: Error: [Errno 2] No such file or directory: '/var/lib/pm-utils/stress.log'
Tags: resume hibernate
Title: [IBM 2373WBZ] hibernate/resume failure
UserGroups:

[Don Cristóbal (doncristobal)]

It happened again... Should I upload the apport details again? If yes: How can I do that?

[chaney (christopher-haney)]

Same issue with 9.04 for an IBM T61

[Don Cristóbal (doncristobal)]

It has happened again (see the attached crash file). I'll try some more times, always uploading the crash files.

chaney: You could also upload your crash files. I found mine in /var/crash, named susres[date][etc.], but I had to make them readable to the web browser (not root):
$ sudo cp /var/crash/sus* ./
$ sudo chown christoph sus*

[Mark K (mark-kempster)]

IBM T40 here. Happened on 8.10, and now for the first time on 9.04 after a few successful resumes.
Doubt I can add anything of benefit at this point.

[Don Cristóbal (doncristobal)]

one more crash report...

[Don Cristóbal (doncristobal)]

again...

[Mark K (mark-kempster)]

No documented proof, but seems to happen less when I choose 'Suspend' from the ubuntu menu, and more when I suspend by closing the lid of the laptop.

[Don Cristóbal (doncristobal)]

Mark: Does it mainly happen when /suspending/ or when /hibernating/, or when waking up from either of them? I've had it only once on suspend and many times on hibernate, always when waking up.

[snowstar1963 (snowstar1963)]

always on hibernate - suspend usually works. I have noticed a failed to thaw
error -16 message. There is additional text in the message but I can never
remember the entire text on the console. Seems something like "PM Device
0000.00.00.0 Failed to thaw error -16"

On Sat, Apr 4, 2009 at 11:38 AM, Christoph <email address hidden> wrote:

> Mark: Does it mainly happen when /suspending/ or when /hibernating/, or
> when waking up from either of them? I've had it only once on suspend and
> many times on hibernate, always when waking up.
>
> --
> IBM T42 crashes on wake up from hibernate
> https://bugs.launchpad.net/bugs/350680
> You received this bug notification because you are a direct subscriber
> of the bug.
>

[Don Cristóbal (doncristobal)]

Ah yes, this black screen error message. I managed to take a photograph. It reads:

[ 424.350061] pm_op(): pci_pm_thaw+0x0/0x50 returns -16
[ 424.350064] PM: Device 000:00:00.0 failed to thaw: error -16

And, by the way, one more crash report.

This seems to be a pretty common bug, with several people reporting it in a few days on a beta release.

[Mark K (mark-kempster)]

Christoph: Sorry for the late reply! I've had lockups from just about every possible non-shutdown - choosing 'suspend' from the ubuntu menu, shutting the lid (which I have configured to suspend), and also choosing 'hibernate' from the ubuntu menu.

I've also seen the failed-to-thaw errors resuming from a hibernate session that you posted on 2009-04-05.

[Don Cristóbal (doncristobal)]

It does not seem that a developer is looking into this? That's a major pity. Whatever, there's a new apport report.

[Bruno Girin (brunogirin)]

Confirmed on ThinkPad T42 with Intrepid and Jaunty.

[Bruno Girin (brunogirin)]

This seems to have been resolved for me by a recent update to Jaunty. Can others confirm?

[simon H. Rasmussen (simras)]

Yes Bruno, I can confirm!

My T42 seems to have had a lot issues with hibernation in Ubuntu, but I hope it's fixed for all eternity and no future updates or upgrades will break our newly found bliss ;-)

Thanks to the Ubuntu people, great work - GREAT OS!

[Mark K (mark-kempster)]

Update manager tells me that I'm current, but it happened to me today after a hibernate.

mark@osiris:~$ uname -a
Linux osiris 2.6.28-11-generic #42-Ubuntu SMP Fri Apr 17 01:57:59 UTC 2009 i686 GNU/Linux

[Mark K (mark-kempster)]

I was able to successfully suspend / wake my laptop 3x in a row, by having the lid-close wired to 'suspend' instead of to 'hibernate'.

I'm tentatively hopeful that suspend at least works reliably... Seems that hibernate has the problem. I can live with that, for now.

When it did hang coming back from a hibernate (my last post), there was no crash report created in /var/crash. The only thing I've run across is this, from /var/log/kern.log when I attempted a wake-after-hibernate

...
May 20 07:27:44 osiris kernel: [ 3.805758] usbcore: registered new interface driver usbhid
May 20 07:27:44 osiris kernel: [ 3.805762] usbhid: v2.6:USB HID core driver
May 20 07:27:44 osiris kernel: [ 3.856539] PM: Starting manual resume from disk
May 20 07:27:44 osiris kernel: [ 3.856544] PM: Resume from partition 8:5
May 20 07:27:44 osiris kernel: [ 3.856546] PM: Checking hibernation image.
May 20 07:27:44 osiris kernel: [ 3.856826] PM: Resume from disk failed.
May 20 07:27:44 osiris kernel: [ 3.898317] kjournald starting. Commit interval 5 seconds
...

I imagine there's somewhere else I should look to view the inner workings of the 'checking hibernation image', but I'll need guidance on what to provide and where to find it.

[Don Cristóbal (doncristobal)]

I confirm: NOT FIXED. Do not close this bug report. The report is about IBM T42 crashing on wake up from /hibernation/.

Mine has suspended fine (aka standby, power not completely off) from the beginning in Jaunty. But hibernation still fails 1 out of 2 or 3 times. Blinking caps lock led and nothing moving as before.
I attach the latest error log.

[Mark K (mark-kempster)]

And to contradict my own comment from the other day, I can confirm that suspending still has intermittent problems for me. I had a string of good luck there with several successful resumes, but I still have problems when suspending for longer periods of time (overnight). That's probably a red herring, but I figure I'd mention it.

As Christoph notes, since this bug is specifically for resuming from hibernation. My suspending problems are probably better tracked elsewhere.

[zardoz (hyphos)]

Same problem here on Lenovo T61. Hibernate doesn't work. I crashes when waking up (blinking caps lock). I have to do a hard reset then. Any solutions in sight?

[Mark K (mark-kempster)]

zardoz -

I don't think there's been any real work on _this_ particular bug. I
hope it's gone on unnoticed by me, attached to another bug. But
regardless, the behavior still exists. I've basically given up on
hibernating or suspending with 9.04 on my T40.

One of these days I hope to nuke my machine and do a complete
re-install, to see if something got knocked out of whack (along with a
number of other things that happened during the upgrade), but I
haven't had the time to deal with that just yet.

On Sat, Aug 15, 2009 at 6:47 PM, zardoz<email address hidden> wrote:
> Same problem here on Lenovo T61. Hibernate doesn't work. I crashes when
> waking up (blinking caps lock). I have to do a hard reset then. Any
> solutions in sight?
>
> --
> IBM T42 crashes on wake up from hibernate
> https://bugs.launchpad.net/bugs/350680
> You received this bug notification because you are a direct subscriber
> of the bug.
>

[Don Cristóbal (doncristobal)]

Indeed there is no visible sign that a developer is working on this. However, we should at least keep them updated about the newest releases. Karmic Koala (Ubuntu 9.10) Alpha4 has been released - does anyone have the time to test hibernating it on a IBM/Lenovo that suffers from the blinking LED disease with 9.04? (I don't have time, unfortunately). You can get it from http://www.ubuntu.com/testing and install it on a spare hard drive or on a partition of the same hard drive (create partitions with gparted or directly in the installation wizard; it has greatly improved). As far as I know, hibernating is impossible with Live CDs and USB sticks (because they have to work even if there is no valid swap partition for the hibernation mirror). So you really have to install it for this test.

[Don Cristóbal (doncristobal)]

Could someone set this bug's priority to medium at least, please? It's not just annoying, it is really b a d .
Thank you!

[Bruno Girin (brunogirin)]

@Christoph: I've got a spare test partition on my T42 for just that purpose so I'll download a fresh version of Karmic and re-test during the week. I'll report my findings back here. Note however that with Jaunty I only have issues when hibernating, suspending works fine. So if anybody has issues with suspending, it may be worth pointing out in case it's actually a different bug.

[Mark K (mark-kempster)]

Just to run the groundwork, i can confirm the hibernate freeze with a
fresh 9.04 install (no updates) on an empty HD with a T40.

Suspending worked a couple times, but it's been on-again-off-again for
me on this hardware.

I'm off to download Karmic, and try the same test as brunogirin.

On Sun, Aug 16, 2009 at 6:13 PM, Bruno Girin<email address hidden> wrote:
> @Christoph: I've got a spare test partition on my T42 for just that
> purpose so I'll download a fresh version of Karmic and re-test during
> the week. I'll report my findings back here. Note however that with
> Jaunty I only have issues when hibernating, suspending works fine. So if
> anybody has issues with suspending, it may be worth pointing out in case
> it's actually a different bug.
>
> --
> IBM T42 crashes on wake up from hibernate
> https://bugs.launchpad.net/bugs/350680
> You received this bug notification because you are a direct subscriber
> of the bug.
>

[Mark K (mark-kempster)]

I installed Karmic on a spare HD, and have suspended and hibernated my
T40 half a dozen times now, without incident.

Upon resume from hibernate, I'm presented with _exactly_ where I was
before - logged in, and all. I thought hibernate brought me back to a
password prompt, but I may not be remembering that properly (it's been
a while since my last hibernate).

So far, things are headed in the right direction. I hesitate to say
yet that Karmic (or perhaps, the new kernel) fixes this hibernate bug,
but I plan on using this system for a couple days to see what happens.

Mark

On Tue, Aug 18, 2009 at 9:33 PM, Mark Kempster<email address hidden> wrote:
> Just to run the groundwork, i can confirm the hibernate freeze with a
> fresh 9.04 install (no updates) on an empty HD with a T40.
>
> Suspending worked a couple times, but it's been on-again-off-again for
> me on this hardware.
>
> I'm off to download Karmic, and try the same test as brunogirin.
>
> On Sun, Aug 16, 2009 at 6:13 PM, Bruno Girin<email address hidden> wrote:
>> @Christoph: I've got a spare test partition on my T42 for just that
>> purpose so I'll download a fresh version of Karmic and re-test during
>> the week. I'll report my findings back here. Note however that with
>> Jaunty I only have issues when hibernating, suspending works fine. So if
>> anybody has issues with suspending, it may be worth pointing out in case
>> it's actually a different bug.
>>
>> --
>> IBM T42 crashes on wake up from hibernate
>> https://bugs.launchpad.net/bugs/350680
>> You received this bug notification because you are a direct subscriber
>> of the bug.
>>
>

[Don Cristóbal (doncristobal)]

(Xubuntu 9.04) There was a kernel update this morning, with a very long list of changes. Since then, I've restarted, then hibernated and resumed fine a couple of times, no blinking CapsLock LED, no freezing... I'm continuing the test and reporting back later.
Meanwhile, the bug # 121737 (no password prompt after suspend and hibernate) is still not solved.

[Mark K (mark-kempster)]

I did a final hibernate last night on Ubuntu 9.04, and upon resume
this morning, saw the blinking caps-lock. I did a forced shutdown and
restart, and it hung for a few minutes, but then came up.

dmesg has pauses after 'Trying to unpack rootfs image as initramfs...'
with time waiting from 0.16 to 0.43 when it says 'Freeing initrd
memory: 7050k freed'. There's another pause here

0.48 isapnp: Scanning for PnP cards...
0.83 isapnp: No Plug & Play device found

I'm not sure if this bit is normal, or due to
startup-after-failed-resume-from-hibernate
2.11 usbhid: v2.6:UST HID core driver
2.69 PM: Startign manual resume from disk
2.69 PM: Resume from partition 8:5
2.69 PM: Checking hibernation image.
2.69 PM: Resume from disk failed.

Then it goes on about EXT4-fs recovery on sda1.

It talks about failing to read my thumb drive, but that happens now
with 9.04 (I suspect a hardware issue there), so I think is unrelated.

On the upside, ubuntu is now telling me about the crash report again,
so that's something.

Mark

On Thu, Aug 20, 2009 at 3:44 AM, Christoph<email address hidden> wrote:
> (Xubuntu 9.04) There was a kernel update this morning, with a very long list of changes. Since then, I've restarted, then hibernated and resumed fine a couple of times, no blinking CapsLock LED, no freezing... I'm continuing the test and reporting back later.
> Meanwhile, the bug # 121737 (no password prompt after suspend and hibernate) is still not solved.
>
> --
> IBM T42 crashes on wake up from hibernate
> https://bugs.launchpad.net/bugs/350680
> You received this bug notification because you are a direct subscriber
> of the bug.
>

[Bruno Girin (brunogirin)]

I finally got round to installing Karmic Alpha 4 on my T42 today. I then did the following tests:

1. Suspend by closing the lid: OK, password prompt on wake up
2. Suspend through hot key (Fn+F4): OK, password prompt on wake up
3. Suspend through CTRL+ALT+DEL menu: OK, password prompt on wake up
4. Hibernate: OK but NO password prompt on wake up, as in bug 121737 and as reported by Mark above

[Bruno Girin (brunogirin)]

I forgot to add: I did an apt-get update / apt-get dist-upgrade before doing the tests so that's with the very latest kernel (2.6.31-6-generic).

[Don Cristóbal (doncristobal)]

Mark, have you updated your Karmic? Maybe you're still missing THE crucial kernel update.

In the meantime, my Jaunty is still behaving although I suspend and hibernate it all the time.

If anyone has the time: There is a half-automatic stress test for suspend and resume on http://testcases.qa.ubuntu.com/System/SuspendResume

[Mark K (mark-kempster)]

I installed Karmic on a fresh HD, but I haven't done any updates since
that install from the iso - perhaps it's running a stale kernel as
well. I've been putting off Jaunty's kernel updates since they showed
up - I hope to get to that this weekend and see if there's any change.

I'll check out that stress-test, though. If nothing else, hopefully it
will help to prove / disprove that the recent kernel updates actually
have an effect.

On Fri, Aug 21, 2009 at 1:49 AM, Christoph<email address hidden> wrote:
> Mark, have you updated your Karmic? Maybe you're still missing THE
> crucial kernel update.
>
> In the meantime, my Jaunty is still behaving although I suspend and
> hibernate it all the time.
>
> If anyone has the time: There is a half-automatic stress test for
> suspend and resume on
> http://testcases.qa.ubuntu.com/System/SuspendResume
>
> --
> IBM T42 crashes on wake up from hibernate
> https://bugs.launchpad.net/bugs/350680
> You received this bug notification because you are a direct subscriber
> of the bug.
>

[Don Cristóbal (doncristobal)]

It happened again :'-(
blinking caps lock LED when resuming from hibernation
(Xubuntu 9.04, newest kernel 2.6.28-15-generic)

sad sad sad

[Bruno Girin (brunogirin)]

@Christoph: yes, the issue still exists with (K|X)Ubuntu Jaunty 9.04, kernel 2.6.28-15. However, it seems to be resolved in Ubuntu Karmic 9.10 Alpha 4, with kernel 2.6.31-6.

@Mark: there are a lot of updates on top of the Karmic Alpha 4 ISO, including a new version of the kernel, so it would be worth updating and trying again to confirm that the fix is in that update.

[Don Cristóbal (doncristobal)]

Again...
Maybe it's a coincidence, but it started happening only when I started using the xfce-power-manager icon from the panel. That icon allows me to have a locked screen after resuming, which is nice. But maybe it causes (part of) the problem by sending some wrong messages somewhere?

[Bruno Girin (brunogirin)]

My T42 now hibernates and wakes up properly in Karmic. Can other people check and confirm that the bug has been resolved and can be closed please?

[Don Cristóbal (doncristobal)]

No, sorry Bruno to disappoint you, it is still here. My T42 froze on the very first try to hibernate on Karmic :-(

Isn't there anything one could do to help find the cause?

[zardoz (hyphos)]

Same problem still here also (on a Lenovo T61).

[Michael Rooney (mrooney)]

I'm marking this as Triaged since there is a recent crash report attached, and assigning importance to Medium as it has a significant impact on a portion of users.

[linfidel (linfidel)]

I have never been able to get any computer to wake up with Ubuntu using suspend. I put off installing Ubuntu on my old Dell Inspiron 8500 laptop for this reason (it worked fine with Windows XP). I finally got tired of XP and reformatted, did a fresh install of Karmic using all the defaults, and it will not suspend or hibernate.

I have an nVidia card with proprietary drivers, 1GB Ram, and I tried increasing the swap file from 1.7GB default to 3GB, with no luck. Suspend appears to work, but doesn't ever wake up; hibernate gives the "failed to thaw" error, and won't wake up. The 2nd reboot after hibernate says it's "Waking up. Please wait...", but it still didn't wake up.

Then, after following the directions (for versions up to 8.10) at https://help.ubuntu.com/community/NvidiaLaptopBinaryDriverSuspend, suspend still fails to wake up, but the "failed to thaw" error didn't display, and hibernate actually came alive after the 2nd reboot (only once so far, but I'm an optimist).

I may play around with settings in /etc/default/acpi-support, but it's pretty time-consuming to test; I'd be willing to do it if I had more of a clue what to try rather than using the random change method (I feel like one of an infinite number of monkeys).

[MoarningSun (slowinthedark)]

This problem affects me too. My laptop is Thinkpad T60, I'm using ubuntu 9.10, wake up from suspend works good and quick, but after hibernation it fails to wake up most of the time and shows the blinking caps lock.

[LeoRochael (leorochael)]

This problem (crash with blinking caps-lock after resume from hibernation) also affected me on a Thinkpad R500 after karmic installation until one of the two kernel upgrades sine then. Now running on 2.6.31-16-generic for x86_64 and hibernation seems to work properly.

[Don Cristóbal (doncristobal)]

No, sorry, not fixed for me. I hibernated once, and crash! - the blinking LED. IBM T42, Xubuntu 9.10, up to date 2009-12-23.

[Detlef Lechner (detlef-lechner)]

This is a regression. My T42 thawed all right when I used Ubuntu 8.04.3.
I installed Ubuntu 9.10 from Desktop Edition CD keeping my ~/home from Ubuntu 8.04.3. Now my T42 crashes most times when thawing from hibernation. The Caps lock indicator keeps blinking. I could not yet isolate what component causes this crash.
'~$ uname -a; Linux T42 2.6.31-17-generic #54-Ubuntu SMP Thu Dec 10 16:20:31 UTC 2009 i686 GNU/Linux'

[rmcd (rmcd1024)]

I use a Lenovo X200s and have consistent problems with hibernation in Karmic. It will usually hibernate successfully once or twice but then it crashes on resume. Suspend usually works okay, but I have had suspend crashes as well, albeit less frequently. I have tried the default hibernation scheme as well as uswsusp (changed by editing /etc/pm/config.d/00sleep_module. Hibernation seemed much more reliable on the same machine in Jaunty.

I am presently running 2.6.31-18-generic #55-Ubuntu SMP Fri Jan 8 14:55:26 UTC 2010 i686 GNU/Linux . I tried the pre-release kernel in the hopes that hibernation would work better. It doesn't.

I also have a Lenovo R61 with a fresh install of Karmic; it also has hibernation crashes. Hibernation seems broken to me.

[Detlef Lechner (detlef-lechner)]

Complementing comment #50:
The hard disk keeps spinning. The display image shows in black-and-white the Ubuntu logo and the text: "Waking up. Please wait..." ad infinitum. The computer does not respond to The key combinations Alt+Print+K or Alt+Proint+REISUB.

[rmcd (rmcd1024)]

It appears to me that using the hibernate package (1.99-1) may reduce crashes associated with hibernation. I invoke it using a script where I first check for remote mounted drives, then I just call hibernate-disk. (I wondered if failing to unmount might be causing problems.) In limited testing, this is considerably more stable. I would be curious to know if anyone else has tried the hibernate package and if so whether if helps. I also wonder if unmounting helps anyone else.

[sneakypete (sneakypete81)]

I have 9.10 installed on a Thinkpad T42.. I hibernate every day, and have never had any problems, although I do see the "failed to thaw error -16" message every time.

[Don Cristóbal (doncristobal)]

rmcd: Could you post your script here in order for us to test if your procedure helps?

However, I'd like to have said that unmounting all partitions (or all except /) can't be a real solution, unless they're automatically remounted after hibernation. There are many people who use internal/external disks, network drives, several partitions on the primary disk etc.

[rmcd (rmcd1024)]

Christoph: I think the important change is that my very basic script uses hibernate-disk (from the hibernate package) rather than the gnome menu. The script only unmounts specific network drives and ensures that virtualbox is not running (something that guaranteed a crash pre-karmic; I haven't tested it recently).

I completely agree that unmounting is not a satisfactory fix, only at best a temporary workaround.

[Don Cristóbal (doncristobal)]

hibernate-disk seems to work (tried a couple of times without a crash), even without unmounting any partitions. However:
a) the power management setting of the hard disk is wrong after thawing out (normal: 128 for battery, 254 for line; after hibernation: 128; $ sudo hdparm -B /dev/sda).

b) in contrast to traditional hibernation, grub's OS selector is displayed. This would probably lead to crashes when you don't choose the system that you hibernated.

[rmcd (rmcd1024)]

Your finding about the hard disk power management setting raises a couple of questions. 1. Would the power setting change cause issues if the user hibernated on battery and restored with line power or vice versa. 2. Apart from the power setting change, what about the use case where the user hibernates, changes batteries (one could imagine this on a long plane flight), and then resumes. Is that a situation covered by the hibernation code? (I ask because the one time in Karmic that I did change batteries --- I changed from one size battery to another --- on resume I got an ACPI error message and terrible things happened.)

Regarding your comment (b), I can unfortunately confirm that terrible things happen if you choose a different system than the one you hibernated.

[rmcd (rmcd1024)]

I want to withdraw my earlier speculation that hibernate-disk fixes the crashing problem. It doesn't. I think I got lucky initially but over the last several weeks have crashed a lot when resuming from hibernation. Typically I get no error message --- the screen freezes as it is switching graphics modes. Today I got this (had to write it down, it didn't make it into the logs):

init[1]: segfault at 0 ip 003f8790 sp bfc70c0c
error 6 in libdbus-1.so.3.4.0 [3ea000+37000]

and then there were two lines, one containing "kernel panic" and one referring to "drm:intelfb_panic"

[alejandro (laimagendelmundo)]

suspending problems on lenovo s10

report:
never waking up after suspension when closing the lid.
no problems experienced during first month or so after karmic installation
-----
waking up after hybernation seems to work though, but i have only tried it a few times.

[Don Cristóbal (doncristobal)]

alejandro, I suppose you have a different issue than the one here. Maybe open a separate bug if the problem persists.

Xubuntu 10.04 beta1 has hibernated and waken up several times now on my T42. I'll go on testing it - hoping the new upstart procedure has fixed our bug.

[Liam Proven (lproven)]

This also seems to affect my Thinkpad X31 model 26723G0: Bug #370935 in 10.04 beta. It was also an issue on 9.04 and 9.10 & in Linux Mint 7 & 8 (based on 9.04 & 9.10 respectively).

[rCX (rcx)]

Replying to rmcd's comment #53. I also only experience this problem when I have external (USB) drives mounted. I am using a t43p with Jaunty.

[Liam Proven (lproven)]

Just to note that this is still happening in 10.10 (release version) on my Thinkpad X31 & I have observed a related problem on a friend's Thinkpad R400 in 10.04. It appears to resume OK but the keyboard does not respond.

[Liam Proven (lproven)]

Updated information as of Ubuntu 10.10.

Even a fully-updated 10.10 system still displayed the same problem; if the machine was suspended or hibernated, it never resumed. When the system goes into sleep mode, the backlight of the display does not turn off unless the lid is closed or the lid-close sensor button in the hinge is depressed with a pen or screwdriver.

However, adding "nomodeset" to the kernel parameters enables the machine to suspend and resume correctly, including hibernation. Unlike 10.04, in 10.10 the X server starts normally in 1024×768/16M colours and the machine is perfectly usable.

I amended the following line by adding "nomodeset" to the end in /etc/default/grub:
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash nomodeset"

I then ran update-grub to enable the change.

However, the nomodeset line means that the plymouth graphical loading screen no longer works. It is replaced with a text-based one.

In the hope of at least running in the native resolution of the LCD and getting a smoother text loading screen, I added a vga mode parameter to the same line, as follows:

GRUB_CMDLINE_LINUX_DEFAULT="quiet splash nomodeset vga=792"

This not only worked and reset the console resolution to 1024×768, but it also re-enabled the graphical loading screen, which furthermore now appeared with the correct colour palette, removing the coloured fringes that had affected the default install.

I then dropped the X screen from 16M colours to 65K colours, which gives a sharp increase in display performance & enables me to use dual-head with an external monitor running at 1280×1024.

I have documented how I did this in a blog post here:
http://liam-on-linux.livejournal.com/20637.html

I then amended the GRUB line so that the bootup screen mode was the same as the X one:

GRUB_CMDLINE_LINUX_DEFAULT="quiet splash nomodeset vga=791"

The result is that everything now works: graphical startup & shutdown, suspend & resume, hibernation, and a fast X display which supports dual-head operation.

This workaround is entirely successful for me, but I don't know if it counts as closing the bug.

[Brad Figg (brad-figg)]

This bug was filed against a series that is no longer supported and so is being marked as Won't Fix. If this issue still exists in a supported series, please file a new bug.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.