lucid regression: does not resume from hibernation

Bug #577916 reported by Philippe Joyez
144
This bug affects 26 people
Affects Status Importance Assigned to Milestone
initramfs-tools (Ubuntu)
Invalid
Medium
Unassigned
linux (Ubuntu)
Invalid
Medium
Unassigned

Bug Description

I normally nearly always hibernate my system. This was working well before and after I upgraded from Karmic to lucid RC (except for a presumably unrelated cursor glitch https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/564423).

Then suddenly, on May 1st, my system stopped resuming from hibernate, always starting fresh (and with the network always disabled after an attempt to hibernate). From the Synaptic history, I suspect it could be related to the initramfs-tools updated from 0.92bubuntu77 to 0.92bubuntu78, or an accompanying change around that time.

I had initially reported this in https://bugs.launchpad.net/ubuntu/+source/linux/+bug/499940, from which I learned I am not the only one to experience this problem. I was asked to submit another bug, so here it is.

I'm willing to help debugging the resume process. What info should I collect?

Philippe

PS. this is ubuntu 10.04 on a dell dimension 9100 with ati video

== Regression details ==
Discovered in version: linux 2.6.32-21.32
Last known good version: linux 2.6.31-22.67

Revision history for this message
Philip Muškovac (yofel) wrote :

Hi, thank you for taking your time to report this bug, for hibernate issues you should follow
https://wiki.ubuntu.com/DebuggingKernelSuspendHibernateResume
and add the information you can collect to the report, thanks.

Changed in ubuntu:
status: New → Incomplete
tags: added: lucid
Revision history for this message
Philip Muškovac (yofel) wrote :

It would also help if you could run

apport-collect -p linux 577916

after a failed attempt to hibernate.

Revision history for this message
Beni Cherniavsky (cben) wrote :

I have the same situation - stopped hibernating from May 3, when I upgraded to lucid (initramfs-tools upgraded from 0.92bubuntu53 to 0.92bubuntu78).
The network is always disabled after boot (don't know if it's correlated with hibernating, as I didn't try any shutdowns since then).
It's a Dell Latitude D520, intel video.

I notice that I have two swap partitions, with /etc/initramfs-tools/conf.d/resume containing the UUID of one of them.
Could be hibernating to the other one? How can I check this?

dmesg says nothing at all about searching for a resume device. Will apport-collecting now.

Revision history for this message
Beni Cherniavsky (cben) wrote : apport information

AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.21.
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: beni 13196 F.... pulseaudio
 /dev/snd/pcmC0D0p: beni 13196 F...m pulseaudio
 /dev/snd/controlC1: beni 13196 F.... pulseaudio
CRDA: Error: [Errno 2] No such file or directory
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xefebc000 irq 21'
   Mixer name : 'SigmaTel STAC9200'
   Components : 'HDA:83847690,102801d4,00102201'
   Controls : 12
   Simple ctrls : 7
Card1.Amixer.info:
 Card hw:1 'J'/'A4 TECH A4 TECH USB2.0 PC Camera J at usb-0000:00:1d.7-4, high speed'
   Mixer name : 'USB Mixer'
   Components : 'USB0ac8:c40a'
   Controls : 3
   Simple ctrls : 2
DistroRelease: Ubuntu 10.04
HibernationDevice: RESUME=UUID=ce7212db-593c-409f-80c7-53fb37ad7cae
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release amd64 (20091027)
MachineType: Dell Inc. Latitude D520
Package: linux (not installed)
PccardctlIdent:
 Socket 0:
   no product info available
PccardctlStatus:
 Socket 0:
   no card
ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.32-22-generic root=UUID=d581b33c-711c-4998-a3cb-9db13d334ee4 ro quiet splash
ProcEnviron:
 LANGUAGE=en_US:en
 LANG=en_US.UTF-8
 SHELL=/usr/bin/fish
ProcVersionSignature: Ubuntu 2.6.32-22.33-generic 2.6.32.11+drm33.2
Regression: Yes
RelatedPackageVersions: linux-firmware 1.34
Reproducible: Yes
Tags: lucid hibernate resume regression-release needs-upstream-testing
Uname: Linux 2.6.32-22-generic x86_64
UserGroups: adm admin audio cdrom dialout dip fax fuse lpadmin netdev plugdev sambashare tape video
dmi.bios.date: 05/28/2007
dmi.bios.vendor: Dell Inc.
dmi.bios.version: A06
dmi.board.name: 0NF743
dmi.board.vendor: Dell Inc.
dmi.chassis.type: 8
dmi.chassis.vendor: Dell Inc.
dmi.modalias: dmi:bvnDellInc.:bvrA06:bd05/28/2007:svnDellInc.:pnLatitudeD520:pvr:rvnDellInc.:rn0NF743:rvr:cvnDellInc.:ct8:cvr:
dmi.product.name: Latitude D520
dmi.sys.vendor: Dell Inc.

tags: added: apport-collected
Revision history for this message
Beni Cherniavsky (cben) wrote : AlsaDevices.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : AplayDevices.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : ArecordDevices.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : BootDmesg.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : Card0.Amixer.values.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : Card0.Codecs.codec.0.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : Card1.Amixer.values.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : CurrentDmesg.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : IwConfig.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : Lspci.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : Lsusb.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : PciMultimedia.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : ProcCpuinfo.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : ProcInterrupts.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : ProcModules.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : RfKill.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : UdevDb.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : UdevLog.txt

apport information

Revision history for this message
Beni Cherniavsky (cben) wrote : WifiSyslog.txt

apport information

Philip Muškovac (yofel)
affects: ubuntu → linux (Ubuntu)
Revision history for this message
Philippe Joyez (ubuntu-5-pjoyez) wrote : Re: lucid regression: does not resume from hibernation, restarts fresh

I attach first the pm-suspend.log when resume was still working. You can see starting line 92 that its resumes fine.

Revision history for this message
Philippe Joyez (ubuntu-5-pjoyez) wrote :

Now, when it started not resuming. Now, you see starting line 96 the next hibernate attempt. No trace of a resume attempt in between.

Revision history for this message
Philippe Joyez (ubuntu-5-pjoyez) wrote :

Next, following https://wiki.ubuntu.com/DebuggingKernelSuspendHibernateResume, here is the dmesg after a failed resume from hibernate (resulting in a fresh start).

Revision history for this message
Philippe Joyez (ubuntu-5-pjoyez) wrote :

and finally:

cat /proc/cmdline
root=UUID=c1b03b1d-beab-4dd3-b31d-e936587ea210 ro quiet splash

cat /etc/initramfs-tools/conf.d/resume
RESUME=UUID=6c3068b5-d4b2-4447-bbe7-12b621ae4d84

both of which should be correct correct as far as I understand, since:

cat /etc/fstab
# /etc/fstab: static file system information.
#
# Use 'vol_id --uuid' to print the universally unique identifier for a
# device; this may be used with UUID= as a more robust way to name devices
# that works even if disks are added and removed. See fstab(5).
#
# <file system> <mount point> <type> <options> <dump> <pass>
proc /proc proc defaults 0 0
# / was on /dev/sdb1 during installation
UUID=c1b03b1d-beab-4dd3-b31d-e936587ea210 / ext3 relatime,errors=remount-ro 0 1
# swap was on /dev/sda2 during installation
UUID=6c3068b5-d4b2-4447-bbe7-12b621ae4d84 none swap sw 0 0

Changed in linux (Ubuntu):
status: Incomplete → New
Revision history for this message
Usama Akkad (damascene) wrote :

this bug effect me too.
I'm using Eee PC 1005HA and I've the acpi_osi=Linux in grub for the bug with function keys. bug #505452

maybe I'll test hibernating without it later.

A.Kromic (akromic)
Changed in linux (Ubuntu):
status: New → Confirmed
Revision history for this message
Beni Cherniavsky (cben) wrote : Re: [Bug 577916] Re: lucid regression: does not resume from hibernation, restarts fresh

Confirming that that networking is disabled when booting after after
hibernate, enabled when booting after shutdown. Don't know how it's related
to this.

Attaching dmesg of boot after hibernate (all my apport logs where taken then
too) and after shutdown.
You can see the diff with interesting parts highlighted at
http://pastebin.org/221892
Interestingly, BOTH dmesg logs show the line:

PM: Resume from disk failed.

The big difference is that after hibernate, I get:

EXT4-fs (sda6): INFO: recovery required on readonly filesystem
EXT4-fs (sda6): write access will be enabled during recovery
...
EXT4-fs (sda6): orphan cleanup on readonly fs
EXT4-fs (sda6): ext4_orphan_cleanup: deleting unreferenced inode 1146
...
EXT4-fs (sda6): ext4_orphan_cleanup: deleting unreferenced inode 12145
EXT4-fs (sda6): 44 orphan inodes deleted
EXT4-fs (sda6): recovery complete
EXT4-fs (sda6): mounted filesystem with ordered data mode

Both facts indicate that the failure is during hibernation, not during
resume.

The after-hibernate version is also missing some messages at the end about
sound, network & wireless. Now that I think of it, this computer has also
been booting (after failed hibernate attempts) with sound muted, and now
that I did a shutdown it came up with sound enabled. Does anybody else see
this?

--
Beni Cherniavsky-Paskin <email address hidden>

Revision history for this message
Daniel Hollocher (chogydan) wrote : Re: lucid regression: does not resume from hibernation, restarts fresh

I also have found network manager in disabled mode. I just noticed a couple other things: like others sound is disabled, and also I'm flat out unable to suspend-to-ram. I have to reboot to be able to suspend again.

Revision history for this message
mgiammarco (mgiammarco) wrote :

I have the same problem on my r50p just updated to lucid.

Hibernation works but the init.d scripts at resume "forgot" to see the past hibernation and do a cold start.
Network manager stays in disabled mode (because standby and hibernation scripts disable it). To reenable the quickest thing to do is putting the notebook in standby and then wake it.

summary: - lucid regression: does not resume from hibernation, restarts fresh
+ lucid regression: does not resume from hibernation
tags: added: hibernate regression-release resume
Revision history for this message
A.Kromic (akromic) wrote :

I believe Bug #578952 is a duplicate of this bug.

Revision history for this message
YJ Kim (youngjin) wrote :

Has anyone tried doing a downgrade to 0.92bubuntu77 of initramfs-tools to check if the resume from suspend works after a downgrade?

Revision history for this message
poundonu (poundonu) wrote :

I have the same problem. Just to reiterate:
Ubuntu 10.04 64bit upgraded from 9.10
Resume from Hibernate fails and does standard boot
Networking is disabled
dmesg reports two interesting lines:
PM: Checking image partition UUID=<uuid xxxxxx>
PM: Resume from disk failed.

Previous post recommends "downgrade to 0.92bubuntu77 of initramfs-toos". I don't know how to do that, or what that is really describing, but I'd be willing to try it if someone could describe it and the procedure.

Thanks.

Revision history for this message
A.Kromic (akromic) wrote :

Is anyone working on this?
This problem impairs several people using Lucid with laptops!
Until this is resolved, I cannot recommend anyone with laptop to upgrade, and this for sure is the last time I upgrade so soon... I cannot work normally since.

Revision history for this message
A.Kromic (akromic) wrote : apport information

AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.21.
Architecture: amd64
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: CONEXANT Analog [CONEXANT Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: arny 2422 F.... pulseaudio
                      arny 2430 F.... audacious
 /dev/snd/pcmC0D0p: arny 2422 F...m pulseaudio
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xfe100000 irq 17'
   Mixer name : 'Conexant CX20549 (Venice)'
   Components : 'HDA:14f15045,17aa20dc,00100100'
   Controls : 19
   Simple ctrls : 9
Card29.Amixer.info:
 Card hw:29 'ThinkPadEC'/'ThinkPad Console Audio Control at EC reg 0x30, fw 7KHT24WW-1.08'
   Mixer name : 'ThinkPad EC 7KHT24WW-1.08'
   Components : ''
   Controls : 1
   Simple ctrls : 1
Card29.Amixer.values:
 Simple mixer control 'Console',0
   Capabilities: pswitch pswitch-joined penum
   Playback channels: Mono
   Mono: Playback [on]
CheckboxSubmission: f36e81446aef1e927a74b9f40ed42e87
CheckboxSystem: 7f8062de7650099ca95bda200bc0be52
DistroRelease: Ubuntu 10.04
HibernationDevice: RESUME=UUID=12677cb8-c945-4d6a-a1a9-eece065a5f85
MachineType: LENOVO 7650A9G
Package: linux 2.6.32.22.23
PackageArchitecture: amd64
PccardctlIdent:
 Socket 0:
   no product info available
PccardctlStatus:
 Socket 0:
   no card
ProcCmdLine: root=UUID=93bea984-6682-48b6-b69e-3504833449be ro vga=792
ProcEnviron:
 PATH=(custom, user)
 LANG=en_IE.utf8
 SHELL=/bin/zsh
ProcVersionSignature: Ubuntu 2.6.32-22.33-generic 2.6.32.11+drm33.2
Regression: Yes
RelatedPackageVersions: linux-firmware 1.34
Reproducible: Yes
Tags: lucid hibernate resume regression-release needs-upstream-testing
Uname: Linux 2.6.32-22-generic x86_64
UserGroups: adm admin audio cdrom dialout dip fax fuse libvirtd lpadmin netdev plugdev sambashare scanner tape vboxusers video
WpaSupplicantLog:

dmi.bios.date: 05/13/2009
dmi.bios.vendor: LENOVO
dmi.bios.version: 7PETC2WW (2.22 )
dmi.board.name: 7650A9G
dmi.board.vendor: LENOVO
dmi.board.version: Not Available
dmi.chassis.asset.tag: No Asset Information
dmi.chassis.type: 10
dmi.chassis.vendor: LENOVO
dmi.chassis.version: Not Available
dmi.modalias: dmi:bvnLENOVO:bvr7PETC2WW(2.22):bd05/13/2009:svnLENOVO:pn7650A9G:pvrThinkPadR61e:rvnLENOVO:rn7650A9G:rvrNotAvailable:cvnLENOVO:ct10:cvrNotAvailable:
dmi.product.name: 7650A9G
dmi.product.version: ThinkPad R61e
dmi.sys.vendor: LENOVO

Revision history for this message
A.Kromic (akromic) wrote : AlsaDevices.txt

apport information

Revision history for this message
A.Kromic (akromic) wrote : AplayDevices.txt

apport information

Revision history for this message
A.Kromic (akromic) wrote : BootDmesg.txt

apport information

Revision history for this message
A.Kromic (akromic) wrote : CRDA.txt

apport information

21 comments hidden view all 101 comments
Revision history for this message
Marcus (mwillams73) wrote :

I have the exact same problem , unfortunately the apport thingy wouldnt work, also i tried tux on iceui kernal, it started then froze on the atomic cache writing or whatever the hell it was, also tried the apt-get install hibernate, same thing it freezes up at writing and copying the cache and restore. so i guess i dont get to hibernate, and since lucid has decided that my desktop isnt capable of suspend( even though in the bios my computer says it can and is enabled)
. Spec on computer_ Matsonic board with AMD Athlon XP 1900+ Single Core CPU- 1 gig DDR RAM PC2100 clk speed 2.5-Built in video 32 mb- All wrapped up in an AMD case

OS Spec- Ubuntu Lucid Lynx 10.04 current kernel is 2.6.32-22.33 I have all current updates

Revision history for this message
amid1999 (amidsin1999) wrote :

I had exact same problem. Since I had experience with hibernation problems before, I was able to fix it quickly. Since it is on my computer, I cannot provide system logs for it, but I would like to share particular aspect of my system, which might be causing the problem:
I have 2 Ubuntu installation on a single hard drive in different partitions:
dg@dg-laptop:~$ sudo blkid (with comments)
/dev/sda1: UUID="E8FC5740FC5707E8" TYPE="ntfs" - Windows NTFS partition
/dev/sda2: UUID="283c5318-77ab-4e4a-8c45-e2fb07372560" TYPE="ext4" - boot partition for Ubuntu instance 1
/dev/sda5: UUID="0ca9a4c4-7588-4600-bd2d-a62fe87ad4be" TYPE="ext4" - / partition for Ubuntu instance 1
/dev/sda6: UUID="0bd7cf97-fc45-4595-9199-47d5fa597515" TYPE="ext4" - /home partition for Ubuntu instance 1
/dev/sda7: UUID="cf8ed7a6-06ae-427d-9adf-75a876b3bb7d" TYPE="swap" - swap partition for Ubuntu instance 1
/dev/sda8: UUID="b841a114-8ae5-4349-adfc-420c4e4d1e2a" TYPE="reiserfs" - / partition for Ubuntu instance 2
/dev/sda9: UUID="8e42b73f-2df2-441b-be9c-ad3f56761415" TYPE="swap" - swap partition for Ubuntu instance 2.
/dev/sdb1: UUID="ECF8-5122" TYPE="vfat" - shared data partition

When I was installing Ubuntu instance 1, I selected manual partitioning and selected manually boot, root and home partitions. I have explicitly asked the system not to use second swap partition.

After clean reinstal of Ubuntu Lucid, system was hibernating fine, but was not resuming from the hibernate. It was restarting fresh with disabled networking.

/etc/fstab was showing that proper swap partition is used (swap partition for Ubuntu instance 1)
//etc/initramfs-tools/conf.d/resume contained UUID for swap partition of the Ubuntu instance 2. Which is incorrect.

I fixed this problem by changing /etc/initramfs-tools/conf.d/resume and then running update-initramfs.

It is working fine now.

In my situation, I suspect that population of /etc/initramfs-tools/conf.d/resume is working incorrectly specifically in the situation when there is more than one swap partition.

I hope it helps. PLease let me know if I can provide additional information to help resolve it.

Revision history for this message
mgiammarco (mgiammarco) wrote :

Good idea amid1999. It worked also for me. Apparently the swap partition got reformatted (and so uuid changed istantly).

Updated conf.d/resume and now it works.

The interesting thing is that I have swap on /dev/mapper/linux-swap which is a name that does not change. Why ubuntu uses UUID in my case with is a value that can change?

Revision history for this message
Usama Akkad (damascene) wrote :

replacing the uuid in /etc/initramfs-tools/conf.d/resume
with the actual swap partition uuid fixed my problem

maybe that happened because I've had 9.10 with 10.04 and then I removed 9.10 and it seems that the swap was not enabled on Lucid and I enabled it manually. I don't remember what happened exactly but I usually told the second system to use the same swap of the first one.

Revision history for this message
Juraj (juraj-hrcka) wrote : Re: [Bug 577916] Re: lucid regression: does not resume from hibernation

I'm running win/ubuntu dual boot system. After 10.04 was released I wiped
9.10 and performed clean install. No swap partition remained from the
previous installation of Ubuntu. Also resume, fstab and "sudo blkid" were
having the same UUID for swap partition.
As an next I re-formated swap partition. After that updated fstab and resume
files according the information I got from "sudo blkid" and ran sudo
update-initramfs -c -k all. Still the same behavior.

On Thu, Jun 10, 2010 at 8:15 PM, Usama Akkad <email address hidden> wrote:

> replacing the uuid in /etc/initramfs-tools/conf.d/resume
> with the actual swap partition uuid fixed my problem
>
> maybe that happened because I've had 9.10 with 10.04 and then I removed
> 9.10 and it seems that the swap was not enabled on Lucid and I enabled
> it manually. I don't remember what happened exactly but I usually told
> the second system to use the same swap of the first one.
>
> --
> lucid regression: does not resume from hibernation
> https://bugs.launchpad.net/bugs/577916
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in “linux” package in Ubuntu: Confirmed
>
> Bug description:
> I normally nearly always hibernate my system. This was working well before
> and after I upgraded from Karmic to lucid RC (except for a presumably
> unrelated cursor glitch
> https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/564423
> ).
>
> Then suddenly, on May 1st, my system stopped resuming from hibernate,
> always starting fresh (and with the network always disabled after an attempt
> to hibernate). From the Synaptic history, I suspect it could be related to
> the initramfs-tools updated from 0.92bubuntu77 to 0.92bubuntu78, or an
> accompanying change around that time.
>
> I had initially reported this in
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/499940, from which I
> learned I am not the only one to experience this problem. I was asked to
> submit another bug, so here it is.
>
> I'm willing to help debugging the resume process. What info should I
> collect?
>
> Philippe
>
> PS. this is ubuntu 10.04 on a dell dimension 9100 with ati video
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/577916/+subscribe
>

Revision history for this message
broe (erich-rupp) wrote :

i had the same problem starting with the upgrade to 10.04 and managed to get it going again with some tips from previous posters (thanks!). i also had 2 swap partitions.

had to:
1. remove 2nd swap partition from /etc/fstab
2. reformat 1st swap partition and add the new UUID in conf.d/resume, menu.lst, fstab + running "update-initramfs -u"

Revision history for this message
Daniel Hollocher (chogydan) wrote :

to the people who are having trouble with a second swap partition, did you get the same errors about lost inodes in your dmesg? Was hibernation working at the beginning of the lucid release, but stopped working? or has it been failed all along?

If you guys aren't getting the same error message than it is a different bug, and you should file a separate bug report.

FYI: I tried briefly in maverick, just before alpha 1, and hibernation was a total fail for me. I could not even shutdown, much less resume.

Revision history for this message
broe (erich-rupp) wrote :

yep, i had a 2nd swap partition and the same inode errors:

dmesg.0:[ 4.771221] ext3_orphan_cleanup: deleting unreferenced inode 10013
dmesg.1.gz:[ 5.778323] ext3_orphan_cleanup: deleting unreferenced inode 9982

i don't remember anymore if it stopped working immediatly after the upgrade to 10.04 or some days later.

Revision history for this message
munbi (gabriele) wrote :

Deleting /usr/share/initramfs-tools/conf.d/compcache and rebuilding initramfs solved the problem for me :-)
What I've done:

sudo swapoff /dev/ramzswap0
sudo rm /usr/share/initramfs-tools/conf.d/compcache
sudo update-initramfs -u

Pay attention that doing this disables compcache. I don't know if this could create problems, but definitely solved the problem for me.
Could it be that compcache is installed/activated during the update from 9.10 to 10.4 and left over?
Hope this helps. Bye.

Revision history for this message
Sean Simmons (sean-low-five) wrote :

Thanks, munbi! Compache turned out to be my problem as well. Rather than deleting /usr/share/initramfs-tool/conf.d/compcache, I edited it and set COMPCACHE_SIZE="0%" which had the effect of disabling compcache (after updating my initramfs).

It looks to me like the only affect of disabling compcache would be a decrease in performance on workloads that swap a lot.

Revision history for this message
munbi (gabriele) wrote :

A quick note just to say that the correct solution is the one posted by Sean (#71), not mine (#70) which lasts only until reboot. Sean solution always works,
Thanks Sean.

I'm wondering if the problem could depend on the swap partition being too small: on my system I have 2Gb of RAM and my swap partition is just little bigger (i.e. 2.1Gb). Is it possible that, when compcache is enabled, the compressed cache portion in ram could not fit into swap during hibernation (because it is uncompressed before flushing to disk or something similar?)
If this is the case, maybe leaving compcache active and growing the swap partition could be another solution?

Bye.

Revision history for this message
Daniel Hollocher (chogydan) wrote :

I'm a bit lost. I don't seem to have compcache activated in the first place, but now hibernation no longer works on lucid.

Revision history for this message
Juraj (juraj-hrcka) wrote :

The same here. My /usr/share/initramfs-tools/conf.d is empty

On Fri, Jun 25, 2010 at 10:09 PM, Daniel Hollocher <
<email address hidden>> wrote:

> I'm a bit lost. I don't seem to have compcache activated in the first
> place, but now hibernation no longer works on lucid.
>
> --
> lucid regression: does not resume from hibernation
> https://bugs.launchpad.net/bugs/577916
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in “linux” package in Ubuntu: Confirmed
>
> Bug description:
> I normally nearly always hibernate my system. This was working well before
> and after I upgraded from Karmic to lucid RC (except for a presumably
> unrelated cursor glitch
> https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/564423
> ).
>
> Then suddenly, on May 1st, my system stopped resuming from hibernate,
> always starting fresh (and with the network always disabled after an attempt
> to hibernate). From the Synaptic history, I suspect it could be related to
> the initramfs-tools updated from 0.92bubuntu77 to 0.92bubuntu78, or an
> accompanying change around that time.
>
> I had initially reported this in
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/499940, from which I
> learned I am not the only one to experience this problem. I was asked to
> submit another bug, so here it is.
>
> I'm willing to help debugging the resume process. What info should I
> collect?
>
> Philippe
>
> PS. this is ubuntu 10.04 on a dell dimension 9100 with ati video
>
> To unsubscribe from this bug, go to:
> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/577916/+subscribe
>

Revision history for this message
Steven Flintham (saf) wrote :

My hibernate stopped working after a fresh install of 10.04. Symptoms were that it would appear to hibernate correctly, but on restart would just boot to the login screen and once I logged in networking was disabled.

Following the above I had a look at /etc/initramfs-tools/conf.d/resume - it didn't exist. I ran:

sudo update-initramfs -u

and it didn't generate any errors but that file still didn't exist afterwards.

I manually populated the file with:

RESUME=UUID=7482c5cb-6168-45dd-bfce-041561c80512

and ran the update command again. It said:

update-initramfs: Generating /boot/initrd.img-2.6.32-23-generic
cryptsetup: WARNING: found more than one resume device candidate:
                     /dev/sdb5
                     UUID=7482c5cb-6168-45dd-bfce-041561c80512

These are both the same device, although my fstab refers to the swap device by UUID.

Anyway, I hibernated and this time it worked - I got a 'resuming' message on the boot screen after the grub menu.

So I suspect my problem was some sort of mismatch over the swap partition name, though I have no idea how it came about.

For what it's worth, I left my old partitions in place (originally created by an Ubuntu installed back in 2005/2006 - I did in-place upgrades until this time) when I installed 10.04. The installer refused to recognise my swap partition (I cannot remember the details, I'm afraid) and I had to do the install without swap (no problem as I have 2GB RAM) and manually add the swap myself afterwards. So maybe I added it badly, I suspect I copied the /etc/fstab line from the backup I had taken before the reinstall (left to my own devices I would probably just have called it /dev/sdb5) but I can no longer be sure.

Sorry this is a bit vague but I thought the information just might be useful. And thanks to the guys above for pointing me in the direction of a solution.

Revision history for this message
yell0w (yellowbloc+ubuntulaunchpad) wrote :

after reading this thread and
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/499940 and
https://bugs.launchpad.net/initramfs-tools/+bug/538534

"PM Resume from disk failed " is the only error I saw in syslog, so I look in /etc/initramfs-tools/conf.d/ and surprised to find that it's an empty directory.

Well I make /etc/initramfs-tools/conf.d/resume with RESUME=/dev/sda4 and then do sudo update-initramfs -u ( /dev/sda4 is my swap partition)

then hibernate works - Thanks to whoever pointed it out :D
Cheers!

Revision history for this message
Teo (teo1978) wrote :

I have the exact same problem.

After hibernating, when I power on the computer, instead of resuming it boots freshly as if never hibernated.
And network is disabled.

However in my case it happens most of the time but not always.
Indeed any of the following can happen:
- hibernates and resumes succesfully, just fine, OR
- during the hibernation process a few error messages appear and it fails to hibertate, meaning it gets stuck and never turns the power off; i then have to power it off manually, and when I power it on, it boots freshly as if nothing had happened; OR
- the same but with no error messages printed on screen: just a black screen with a blinking cursor, forever; OR
- just as described in the original post: seems to hibernate OK but then does not resume.

Importance should be set to the highest possible, this can cause data loss.

Revision history for this message
Damian Nadales (dnadales) wrote :

The network is disabled every time the computer doesn't resume properly. I suspended the laptop unsuccessfully today, and upon reboot I had the network disabled problem again.

Revision history for this message
Daniel Hollocher (chogydan) wrote :

I added initramfs-tools, since that is what some of you are pointing to. But, I tested a few weeks back in maverick, and it is working there. If it is fixed in mav, that's good enough for me

Revision history for this message
misGnomer (petrit) wrote :

@ Daniel Hollocher:

This resume bug affects Ubuntu 10.04 _LTS_ installs however, ie. it will be there for years to come - possibly even in OEM installations - unless it is fixed with the fix preferably included in an upcoming CD/ISO refresh.

Looking at some of the manual solutions it might not be an insurmountable task to fix the installation logic and/or update the failed component(s). Unless Long Term Support only applies to security updates.

As Matteo pointed above, this bug can cause data loss and only the most advanced users are able to locate and implement the manual fix.

Revision history for this message
Teo (teo1978) wrote :

I didn't change anything (apart accepting all automatic updates), but now, when I try to hibernate, it simply doesn't.
Instead of hibernating it does exactly the same as "lock screen", and never turns off the power.

So when battery gets critically low and the system automatically tries to hibernate, it just locks screen, then the power goes off not because the system turns it off but because the hardware does, and then obviously at reboot it boots freshly.

Note that any solution involving using the command line for hibernating is not only an unelegant solution: it is no solutoon at all, because it does not prevent data loss when the battery gets low.

I expect this to be fixed in such a way that one day, when the update manager automatically finds and downloads an update of whatever package is causing the issue, hibernate will start working fine. Isn't that how ALL bugs are supposed to be fixed???????

Revision history for this message
Ryan Thompson (rct86) wrote :

I would like to note that on my newly installed Lucid system, /etc/initramfs-tools/conf.d/resume did not even exist. I have populated it with what I believe to be the correct thing, and I will now try a reboot followed by a hibernate.

Revision history for this message
Ryan Thompson (rct86) wrote :

Well, hibernate just worked, but I don't know for sure what fixed it, because I just realized that with the last 5 or so fixes that I tried, I only remembered to do update-initramfs after the last one. So any of the fixes I tried could have been the one.

Revision history for this message
Daniel Hollocher (chogydan) wrote :

hey folks, spurred on by Ryan's success, I retried the compcache and initramfs things, and I managed to get a single good hibernate/restore. But only once! Now, my computer won't even shutdown when hibernation is selected.

I think it would be helpful if people could summarize the workarounds that have worked for them at the top of this bug. Just create a new section, and list out the steps that you did.

Revision history for this message
angus73 (angus73) wrote :

solution of amid1999 worked fine for me. I had cloned my lucid installation to a new hard disk, so the new uuid had to be entered in /etc/initramfs-tools/conf.d/resume

THanks a lot!
Andrea

Revision history for this message
Daniel Hollocher (chogydan) wrote :

I upgraded my install, which was having the issue, to maverick, and it now hibernates.

Revision history for this message
Daniel Hollocher (chogydan) wrote :

I guess I spoke too soon. Still having an issue with Maverick, with inodes being lost. But it just happens quite different, only when I have several programs open; and then hibernating just locks up my computer, it never shuts down

Revision history for this message
Teo (teo1978) wrote :

Now this is new: today battery power went low untill the power went off, and the system didn't even TRY to START the hibernation process.

USUALLY, the following happens: when battery power gets too low, the system starts hibernating, then one of two things can happen: (a) everything works fine; (b) the hibernation process hangs the system and never terminates, so power goes off and when i turn it on, it just boots from scratch.

Today however, the battery power went low, a warning appeared that the system would hibernate soon, but it didn't. Power just went off suddenly, while the sysyem was still running normally.

I can't understand why the importance of this bug is still undecided and why it is still unassigned. This should have the highest priority ever, and should have been fixed already. This makes the system completely unreliable.

Revision history for this message
Charlie Kravetz (cjkgeek) wrote :

Thanks for reporting this issue. For everyone except the original reporter, please file separate bugs using

ubuntu-bug linux

and include a comment referencing bug 577916. This is the kernel teams bug policy, since a bug affecting the kernel could require separate fixes for each reporter. Thank you for your cooperation and for helping to improve Ubuntu.

Changed in initramfs-tools (Ubuntu):
importance: Undecided → Medium
status: New → Triaged
description: updated
Changed in linux (Ubuntu):
status: Confirmed → Triaged
importance: Undecided → Medium
Revision history for this message
Teo (teo1978) wrote :

There seem to be 3 different issue, though any of them can happen, at random times:

1) Sometimes, when you suspend or hibernate (or when the battery goes low and the system is set to automatically suspend or hibernate), INSTEAD of hibernating/suspending it just locks screen. In this case, you can go back to normal operation by inserting the password and clicking "unlock"; however, it will never hibernate nor suspend (and in case the battery was low, power will ultimately go off and you'll lose everything)

2) Some other times, the suspend or hibernation process is started, but it HANGS at some point. In this case, there is nothing you can do except force power off physically by holding the power button.

3) Independently from all this, detection of the low battery condition does not work properly. I have my laptop set to automatically hibernate when battery is low. A lot of times, it simply happens TOO LATE, so the power goes off physically before hibernation is complete, even when hibernation itself is working properly. Sometimes, power goes off even before the system TRIES to hibernate.
Note that:
- the system DOES detect low battery, I can see it in the battery tray icon which is empty and red, and because popups appear a lot of times to warn me that the system will hibernate soon.
- Windows on the same machine hibernates properly on low battery, so it is NOT an issue with the hardware detecting low power too late (i.e. having a too-low threshold), it is an issue with the operating system that doesn't take action soon enough.

In my case, I've found out that issues (1) and (2) seem to happen only when I am running VirtualBox (a virtual machine software) or _after_ having run it. This probably means that there is a bug in VirtualBox which is preventing the host operating system to hibernate/suspend - however, this does NOT mean that there's not a bug in Ubuntu. Hibernation, suspending and shut down MUST always be bullet-proof and must always succeed even in presence of a buggy or even malicious program; if a buggy program can prevent hibernation to complete properly, the operating system has a bug in itself.

Revision history for this message
Teo (teo1978) wrote :

Importance medium??????????

That's crazy. It must be a mistake.
This is at the very least critical.

Please, anybody with the permission to do that, change the value of importance!

Revision history for this message
Teo (teo1978) wrote :

Is there any plan of fixing this?

Revision history for this message
Teo (teo1978) wrote :

By the way my suspects that Virtual Box had anything to do with this were wrong. I have experienced the issue a few times without ever running virtual box.

Also, I confirm that sometimes also the following happens: that the hibernation process actually _is_ completed (or at least, it does not hang forever and the pc does actually eventually turn itself off even if connected to power) but, when I reboot, it does a fresh boot as if it had been simply shut down.

Note that this and all the other already described buggy behaviours all happen at random times: sometimes one sometimes the other; and sometimes it succsesfully hibernates/suspends and resumes correctly.

Revision history for this message
Teo (teo1978) wrote :

PLEASE PLEASE PLEASE anybody that has the prvivileges to set the "importance" and is listening: it has been set to "medium", clearly by mistake. This is CRITICAL, not medium, and should be changed. A system that does not suspend and hibernate in a completely bulletproof way is a system that cannot be used safely.

Maybe this should be split into four separate issue, though I am not sure:
issue 1: suspend/hibernate sometimes hangs forever and can't either be completed nor cancelled
issue 2: sometimes intead of hibernating/suspending it just locks screen
issue 3: after a hibernation, the system sometimes does a fresh boot instead of resuming
issue 4: system sometimes "forgets" to hibernate when battery is low (doesn't even try)

I don't know how much they are related. Anyway they are all CRITICAL.

Revision history for this message
Teo (teo1978) wrote :

For the first time I have seen an error message that may shed some light on this, thought it doesn't always appear (I've seen it only once): it appear during the hibernation and said something like "not enough swap space" or something like that.

Indeed I have a swap partition of 2.2GB and I have 3GB of ram.

Must the swap partition be at least the size of the used RAM in order to hibernate? Does it depend on the amount of ram actually used? (that would explain that sometimes it does hibernate succesfully).

Anyway there still are a few things this doesn't explain:
1- even when hibernation fails, that error message doesn't always appears (indeed I saw it only once)
2- if there is not enough swap space to hibernate (or whatever condition necessary to hibernate is not met, for that matter), the system should definitely give a warning and refuse to hibernate instead of committing suicide
3- this doesn't explain when the hibernation completes apparently succesfully and then when you turn the computer on again it reboots instead of restoring
4- when I installed Ubuntu, I'm 100% sure I got no warning regarding the fact that I was setting a swap partition smaller than ram and that could cause problems.

Revision history for this message
Charlie Kravetz (cjkgeek) wrote :

@matteo: You listed three different bugs in your comment #90. Now, unless you have exactly the same errors and hardware as the original reporter, you need to file separate bug reports. Please file those separate bug reports using

ubuntu-bug linux

for you hardware. Not all hardware is going to fixed with the exact same fix, so by filing separate bug reports, and not marking them duplicates of this one, you have much greater odds of getting your own issues fixed.

Also, the fact that hibernate and suspend do not always work in VirtualBox will not make the bug critical, since VirtualBox must be run on real hardware. Does the hardware installation work?

Revision history for this message
Teo (teo1978) wrote :

Thank you for the clarifications, I'll follow your directions. And yes, the hardware I am using is always the same.

Regarding this:
"Also, the fact that hibernate and suspend do not always work in VirtualBox will not make the bug critical, since VirtualBox must be run on real hardware. Does the hardware installation work?"

I guess I din't explain myself clearly enough.
The only issues I am experiencing _are_ on the hardware installation. Ubuntu is my _host_ operating system, installed on real hardware, and it is the one that (more often than not) doesn't hibernate or suspend correctly. I have never tried running Ubuntu on a virtual machine.
I use VirtualBox as an application on Ubuntu (usually running Windows 7 as a guest OS on top of it), and I had been under the impression that the bug on the host system showed up when and only when I ran virtual box (as an application inside the host system) - so I thought that virtual box (perhaps its kernel module) may be triggering the issue. But as I have mentioned (I think) I found out that that wasn't true: I am experiencing the issue even without using VirtualBox.

The criticity (or criticalness?? or whatever) of the bug has nothing to do with using Virtual Box. The bug is critical by itself, in that an operating system that fails to hibernate and/or suspend cannot be used safely.

Revision history for this message
Daira Hopwood (daira) wrote :

The VirtualBox-related bug is on the VirtualBox tracker as http://www.virtualbox.org/ticket/7716 . I suggest not considering that issue to be part of this one, since it has different symptoms, and seems quite deterministic unlike this one. It almost certainly doesn't have anything to do with having two swap partitions (I have that bug with only one swap partition).

Revision history for this message
Ivan Zorin (iaz) wrote :

Similar issue - bug #957999

Revision history for this message
penalvch (penalvch) wrote :

Philippe Joyez, this bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? If so, could you please test for this with the latest development release of Ubuntu? ISO images are available from http://cdimage.ubuntu.com/daily-live/current/ .

If it remains an issue, could you please run the following command in the development release from a Terminal (Applications->Accessories->Terminal), as it will automatically gather and attach updated debug information to this report:

apport-collect -p linux <replace-with-bug-number>

Also, could you please test the latest upstream kernel available following https://wiki.ubuntu.com/KernelMainlineBuilds ? It will allow additional upstream developers to examine the issue. Please do not test the daily kernel folder, but the one all the way at the bottom. Once you've tested the upstream kernel, please comment on which kernel version specifically you tested. If this bug is fixed in the mainline kernel, please add the following tags:
kernel-fixed-upstream
kernel-fixed-upstream-VERSION-NUMBER

where VERSION-NUMBER is the version number of the kernel you tested. For example:
kernel-fixed-upstream-v3.12-rc2

This can be done by clicking on the yellow circle with a black pencil icon next to the word Tags located at the bottom of the bug description. As well, please remove the tag:
needs-upstream-testing

If the mainline kernel does not fix this bug, please add the following tags:
kernel-bug-exists-upstream
kernel-bug-exists-upstream-VERSION-NUMBER

As well, please remove the tag:
needs-upstream-testing

Once testing of the upstream kernel is complete, please mark this bug's Status as Confirmed. Please let us know your results. Thank you for your understanding.

Changed in linux (Ubuntu):
status: Triaged → Incomplete
Changed in initramfs-tools (Ubuntu):
status: Triaged → Incomplete
Revision history for this message
dino99 (9d9) wrote :

This version has expired

Changed in linux (Ubuntu):
status: Incomplete → Invalid
Changed in initramfs-tools (Ubuntu):
status: Incomplete → Invalid
Displaying first 40 and last 40 comments. View all 101 comments or add a comment.
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.