swap gone after hibernate

Bug #113232 reported by klerfayt
2
Affects Status Importance Assigned to Milestone
acpi (Ubuntu)
Fix Released
Undecided
Unassigned

Bug Description

loosing swap after hibernate without resume=/dev/sda1 boot option; using "nv" driver and 1.54GB swap with 512RAM

Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

I couldn't leave situation like this, reformated swap partition and tried hibernate with "nv" driver and I get exactly the same behaviour!
First hibernate does not resume, swap is here; second hibernate does not resume and swap is gone!

So I have ruled out nvidia driver being the reason here; the swap size is 1.16GB (should be enough for 512RAM).
I'm going to reformat swap again to re-enable it.

Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

what a weird solution I have found!
it seems that if I add
resume=/dev/sda3
to
/boot/grub/menu.lst after the line "# defoptions="
then suspend to disk works perfectly!

I would still like to hear the reasons why I need "resume=/dev/sda3" to get successful hibernate

Revision history for this message
Yuriy Kozlov (yuriy-kozlov) wrote :

Guessing a more appropriate package.

Revision history for this message
Sitsofe Wheeler (sitsofe) wrote :

Thank you for your bug report.

klerfayt:
When you reformatted the swap its UUID would have changed. I'm guessing that
cat /etc/initramfs-tools/conf.d/resume
is referencing a partition UUID that no longer exists (you can use vol_id /dev/sda3 to find the UUID for sda3). Can you include the output of
cat /etc/initramfs-tools/conf.d/resume
cat /etc/fstab
and
vol_id /dev/sda3
here please?

Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

I have reinstalled kubuntu feisty fawn and repartitioned harddrive, however I can attempt suspend to disk (hibernate) without "resume=/dev/sda3" boot option if you want to again.
ps. this time I choose to make only 384MB swap partition (during installation it asks at least 256MB; is there an unwritten rule about swap size to ram size ratio? )

Revision history for this message
Sitsofe Wheeler (sitsofe) wrote :

klerfayt:
There are quite a few suggestions as to the size of swap (some are out of date because memory can be in the 4Gbyte range). However because swap is used when you hibernate it's a good idea for it to be at least 1.5 times the size of real memory otherwise you may find yourself unable to hibernate due to lack of room to store the contents of RAM.

The missing /dev/sda3 is interesting (don't try and hibernate until that's resolved). Can you include the output of
uname -a
and
lscpi
?

Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

I did complete reinstall to increase the size of swap to 768MB, however installation didn't let me choose exactly 768MB for swap, closest I could get was 773MB; what's even more strange is the fact I see 737.35MB swap size in kinfocenter now - where did the rest 36MB go?
also I redid cat /etc/initramfs-tools/conf.d/resume cat /etc/fstab and vol_id /dev/sda3 (haven't done hibernate or suspend to ram this time, and won't do now unless you ask)

Revision history for this message
Sitsofe Wheeler (sitsofe) wrote :

klerfayt:
Hmm you're another sufferer of Bug #116996 so after the kernel upgrade to 2.6.20-16 your partitions will appear as /dev/hda1 etc. As all your partitions are referenced by UUID though this meant they could still be found despite this though! Can you again try a hibernate again with the nv driver?

Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

do I understand you correctly that I have to attempt hibernate (suspend to disk) without boot option "resume=/dev/..."?
at the moment my boot options are default, the way installer set them:
root=UUID=bd4f623f-d761-4127-b0a2-07cb3d611691 ro quiet splash locale=et_EE

Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

is bug #116996 related to failure of hibernation? or is it just another bug I have?

Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

gutsy gibbon - nothing has changed, swap gone after first attempt to hibernate (suspend to disk); I believe hibernate was successful and resume from hibernate failed.

#cat /etc/initramfs-tools/conf.d/resume
RESUME=UUID=ccba3413-90e7-40a9-84ff-05a04b40ff42

#cat /etc/fstab
# /etc/fstab: static file system information.
#
# <file system> <mount point> <type> <options> <dump> <pass>
proc /proc proc defaults 0 0
# /dev/sda3
UUID=01cae1b7-de96-410c-8224-73103cac3ed8 / ext3 defaults,errors=remount-ro 0 1
# /dev/sda4
UUID=5bde5cf9-c762-4d93-816c-42eddad9c9d0 /home ext3 defaults 0 2
# /dev/sda2
UUID=63ab3118-d3c8-401c-9c21-110d8b0c95b8 /media/sda2 ext3 defaults 0 2
# /dev/sda1
UUID=ccba3413-90e7-40a9-84ff-05a04b40ff42 none swap sw 0 0
/dev/scd0 /media/cdrom0 udf,iso9660 user,noauto 0 0

sudo vol_id /dev/sda1
[sudo] password for kristjan:
ID_FS_USAGE=other
ID_FS_TYPE=suspend
ID_FS_VERSION=s1suspend
ID_FS_UUID=ccba3413-90e7-40a9-84ff-05a04b40ff42
ID_FS_UUID_ENC=ccba3413-90e7-40a9-84ff-05a04b40ff42
ID_FS_LABEL=
ID_FS_LABEL_ENC=
ID_FS_LABEL_SAFE=

uname -a
Linux kohvi 2.6.22-14-generic #1 SMP Wed Oct 10 06:00:47 GMT 2007 i686 GNU/Linux

lspci
00:00.0 Host bridge: Intel Corporation Mobile 915GM/PM/GMS/910GML Express Processor to DRAM Controller (rev 03)
00:01.0 PCI bridge: Intel Corporation Mobile 915GM/PM Express PCI Express Root Port (rev 03)
00:1d.0 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #1 (rev 04)
00:1d.1 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #2 (rev 04)
00:1d.2 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #3 (rev 04)
00:1d.7 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB2 EHCI Controller (rev 04)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev d4)
00:1e.2 Multimedia audio controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) AC'97 Audio Controller (rev 04)
00:1e.3 Modem: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) AC'97 Modem Controller (rev 04)
00:1f.0 ISA bridge: Intel Corporation 82801FBM (ICH6M) LPC Interface Bridge (rev 04)
00:1f.2 IDE interface: Intel Corporation 82801FBM (ICH6M) SATA Controller (rev 04)
00:1f.3 SMBus: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) SMBus Controller (rev 04)
01:00.0 VGA compatible controller: nVidia Corporation NV43 [GeForce Go 6200/6400] (rev a1)
02:02.0 CardBus bridge: ENE Technology Inc CB-710/2/4 Cardbus Controller
02:02.1 FLASH memory: ENE Technology Inc ENE PCI Memory Stick Card Reader Controller
02:02.2 Generic system peripheral [0805]: ENE Technology Inc ENE PCI Secure Digital Card Reader Controller
02:03.0 FireWire (IEEE 1394): VIA Technologies, Inc. IEEE 1394 Host Controller (rev 80)
02:04.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10)
02:05.0 Network controller: Intel Corporation PRO/Wireless 2200BG Network Connection (rev 05)

Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

dmesg of boot with failed resume from hibernate

Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

this is a fix for anyone experiencing the same problem:

sudo fdisk -l #see where your swap is, in my case /dev/sda1
sudo swapoff -a #turning swap off to reformat it
sudo apt-get install qtparted #use qtparted to reformat swap
sudo blkid #checking the new UUID of swap
kdesudo kate /etc/fstab #changing new UUID because we reformated
sudo swapon -a #turning swap back on
kdesudo kate /boot/grub/menu.lst #now we add "resume=/dev/path/to/your/swap" after the line beginning with"# defoptions="
#^an example: # defoptions=locale=et_EE resume=/dev/sda1
sudo update-grub
and done, reboot and test hibernate (works for me)

description: updated
Revision history for this message
Daniel T Chen (crimsun) wrote :

Is this symptom still reproducible in 8.10 beta?

Changed in acpi:
status: New → Incomplete
Revision history for this message
klerfayt (klerfayt-deactivatedaccount) wrote :

No, it is not reproducible in 8.10 beta.

Changed in acpi:
status: Incomplete → Fix Released
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.