No Hibernation: '/etc/mkinitramfs/conf.d/resume' is not created

Not being able to do a hibernate from the GDM login menu has already been reported before: bug 3642

Yes, it is similar but just a little different. I post this here since Tormod Volden asked me to. Maybe I misunderstood, his message is below?

Re: Suspend (or hibernate) does nothing Posted by Tormod Volden at 2006-05-05 08:03:34 UTC

SonWon, please file a new bug for your issue, where you include more information on your hardware.

This bug is not about problems with suspend and hibernation, solely that the choices in the menu in the GDM greeter do not work.

I did some more testing and loaded all of the newest updates available.

I believe the swap file is corrupt. After a mkswap the laptop does shutdown during a hibernate but boots up like I never did the hibernate. After that the next hibernate will fail to shutdown or sometimes lock the laptop. I'll then have to force a reset with the power button.

Sleep still locks up the laptop when you attempt to awake the laptop with the power button.

Would testing with Flight 7 make a difference? I originally loaded with Flight 6.

Did you installed Flight 6 using the Live CD installer? Take a look at bug 38333 (and the recipe to solve the problem in one of the comment) I have a similar problem on my laptop: after creating a swap partition, I can try hibernate once, but this willl then corrupt the swap file.

Yes, I install Flight 6 from the Live CD installer.

I looked at bug # 38333 and tried the following:

1. echo RESUME=/dev/hda3 | sudo tee /etc/mkinitramfs/conf.d/resume

2. sudo /usr/sbin/update-initramfs -u

Now Hibernate shuts down okay but on reboot the laptop locks with alternating wide and narrow vertical bars. The interrupts are dead, at least the keyboard since the Caps Lock does not work. Also the swap space corrution is gone so that is some progress.

Any other ideas?

Also, the file /etc/mkinitramfs/conf.d/resume is there.

This morning I installed all of the available patches and hibernate still locks up on the power up cycle. I was hoping that the new HAL update would fix the problem but no luck.

I've loaded Flight 7 and all available updates as of today and the failure still occurs. Since I have all of the updates will Flight 8 make a difference?

Upgraded to major since the problem locks up the laptop on restart.

I've loaded all available updates as of today and the failure still occurs. I also upgrade the kernel (linux-k7) to match my processor (AMD Ahtlon) which made zero difference. I am wondering if Flight 8 will fix this bug?

I have all of the updates installed and suspend/hibernate still does not work? Are there any developers working on this bug? Seems like this would be a problem for a June release.

SonWon: can you include the output of:

  cat /proc/swaps

and does hibernate work for you if you do:

  sudo pmi action hibernate

from a Terminal. Does the same command succeed any better if you do if from a terminal (Ctrl-Alt-F1).

cat /proc/swaps was empty?

More...

sudo pmi action hibernate

No difference from terminal or Ctrl-Alt-F1. On return from hibernate, vertical bar wide and narrow locked up.

I wonder if it would work if gnome and X were not running?

SonWon: if '/proc/swaps' is empty, you have no Swap partition. The swap partition is used for hibernation and must be at least as large as the RAM in your machine.

Did you create a swap partition during install?

Can you post:

  cat /etc/fstab

  sudo fdisk -l /dev/sda (or /dev/hda, whichever is your main disk)

  sudo file -s /dev/sdaX (eg. /dev/sda3, or hda7, depending on what is mentioned in your 'fstab' or in the partition table shown by 'fdisk').

Yes, I created a swap but was hit by the swap partition corruption bug. I thought I had correctly repaired the swap after the bug?

I have 1GB of ram in my laptop, swap is 1.3 GB.

/etc/fstab
# /etc/fstab: static file system information.
#
# <file system> <mount point> <type> <options> <dump> <pass>
proc /proc proc defaults 0 0
/dev/hda2 / ext3 defaults,errors=remount-ro 0 1
/dev/hda1 /media/hda1 ntfs defaults,nls=utf8,umask=007,gid=46 0 1
/dev/hda3 none swap sw 0 0
/dev/hda4 /media/dos vfat iocharset=utf8,umask=000 0 0
/dev/hdc /media/cdrom0 udf,iso9660 user,noauto 0 0

sudo fdisk -l /dev/hda

Disk /dev/hda: 60.0 GB, 60011642880 bytes
255 heads, 63 sectors/track, 7296 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot Start End Blocks Id System
/dev/hda1 * 1 3298 26491153+ 7 HPFS/NTFS
/dev/hda2 3299 7122 30716280 83 Linux
/dev/hda3 7123 7295 1389622+ 82 Linux swap / Solaris
/dev/hda4 7296 7296 8032+ e W95 FAT16 (LBA)

sudo file -s /dev/hda3
/dev/hda3: Linux/i386 swap file (new style) 1 (4K pages) size 347404 pages

It looks okay to me?

Yes, problem (as described and fixed in bug #38333) can be manually worked around by doing a one off:

  sudo mkswap /dev/hda3
  sudo swapon -a

to start using the swap partition, and:

  echo RESUME=/dev/hda3 | sudo tee /etc/mkinitramfs/conf.d/resume
  sudo update-initramfs

followed by a reboot to activate it. Hibernate should then work.

No work?

Here is the output from the above.

sudo mkswap /dev/hda3
Setting up swapspace version 1, size = 1422966 kB
no label, UUID=b46c3b32-3ce4-4501-a780-c566684662a2

sudo swapon -a

echo RESUME=/dev/hda3 | sudo tee /etc/mkinitramfs/conf.d/resume
RESUME=/dev/hda3

sudo update-initramfs -u

I had to add -u because update-initramfs said, "You must specify at least one of -c, -u, or -d." Maybe I picked wrong choice?

Also, tried
sudo update-initramfs -c 2.6.15-23-k7

Still no go?

The /proc/swaps file is still empty?

Watching the hibernate process, it looks like the memory is dumped to the swap space. No go on start up just those vertical lines narrow and wide?

Sorry, yes:

  sudo update-initramfs -u $(uname -r)

to recreate it for your currently running kernel.

Can you do:

  sudo swapon -av

this should show verbose (extra output). BTW, you said that '/proc/swaps' was completely empty - could you try again and check that you did a:

  cat /proc/swaps

as opening this 'file' in a text-editor (such as 'gedit') will not work. It is a special file that doesn't actually exist; so it can't be 'opened', but the contents can be 'cat'ed.

My newbie mistake, sorry, here is the real cat /proc/swaps output.

Filename Type Size Used Priority/dev/hda3 partition 1389612 0 -1

sudo swapon -av
didn't product any output so I did a

sudo swapoff -av
swapoff on /dev/hda3

Then did,

sudo swapon -av
swapon on /dev/hda3

/proc/swaps output.

Filename Type Size Used Priority
/dev/hda3 partition 1389612 0 -1

Fixed formating.

Okay, now we've sorted that issue the infrastructure to hibernate/suspend should now be in place. Unfortunately, because each model of laptop has its own little quirks, we need to do a certain amount of workarounds. If your laptop isn't suspending/hibernating for some reason, can you file a new bug against 'acpi-support', including the output of:

  cat /var/lib/acpi-support/*-*

so that we can identify your laptop and work through the fresh issue.

Okay, thank you Paul!

I have the same problem here.
I repartitioned my machine consolidating a couple of partitions and moved the swap space around.

I had hibernate working perfectly before the repartitioning (even with a swap partition that was less than the amount of memory I have).

I recreated the new swap partition bigger than the available memory 2.1 G swap (2G memory).

I did all the step requested in this thread, changed the resume file, reconfigured the initramfs, everything seem to be correct.

Yet I have the laptop coming up and freezing with vertical bars.

Again, I haven't changed/updated any package I just moved around the partitions and it stopped working properly.

I have an IBM T43p if that makes any difference.

I never moved or changed any partition, other than rebuilding it when corrupted. Please add your comment to bug # 46680 so the devs can see it is not just a Mitac 8355 problem, thanks.

Please don't encourage people to create a giant bug full of information about lots of different systems! It would be much better for simo to file a *new* bug, following the same instructions as were given to you.

kamion: yes, simo and SonWon's problem have both been followed up in bug #46680 which is the new bug filed, as requested.

simo and SonWon: just a reminder that this report (swap not being setup) has been solved and is now closed.

Okay, sorry, I just thought it was the same bug since the problem was the same/similar.