Kernel 3.2.0-31 fails to boot from grub

Bug #1058969 reported by Tim Taylor
12
This bug affects 2 people
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Incomplete
High
Unassigned

Bug Description

I recently updated my kernel packages to 3.2.0-31 (I'm using the generic-pae flavour). I have a dual boot laptop with grub2. This has always worked fine, but having upgraded to 3.2.0-31, grub fails to boot the system. There are no error messages on screen - after a couple of seconds the screen just goes blank and stays like that. The machine is still alive to some extent - I cannot do CTRL-ALT-F1 to get to a command prompt, but I can do CTRL-ALT-DELETE which will reboot the machine. If I choose "Older versions of Linux" from the Grub menu and select 3.2.0-30 (or older versions), then the system boots up as normal. If I remember correctly, this problem is intermittent - I think I have occasionally managed to boot up 3.2.0-31 successfully, but, normally, it crashes as described above. I have currently uninstalled the 3.2.0-31 kernel until this problem is fixed.

ProblemType: Bug
DistroRelease: Ubuntu 12.04
Package: linux-image-3.2.0-31-generic-pae (not installed)
ProcVersionSignature: Ubuntu 3.2.0-30.48-generic-pae 3.2.27
Uname: Linux 3.2.0-30-generic-pae i686
NonfreeKernelModules: wl
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.24.
ApportVersion: 2.0.1-0ubuntu13
Architecture: i386
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 1: SB [HDA ATI SB], device 0: STAC92xx Analog [STAC92xx Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC1: tim 1700 F.... pulseaudio
 /dev/snd/controlC0: tim 1700 F.... pulseaudio
CRDA: Error: command ['iw', 'reg', 'get'] failed with exit code 1: nl80211 not found.
Card0.Amixer.info:
 Card hw:0 'Generic'/'HD-Audio Generic at 0xf0344000 irq 43'
   Mixer name : 'ATI R6xx HDMI'
   Components : 'HDA:1002aa01,00aa0100,00100200'
   Controls : 6
   Simple ctrls : 1
Card0.Amixer.values:
 Simple mixer control 'IEC958',0
   Capabilities: pswitch pswitch-joined penum
   Playback channels: Mono
   Mono: Playback [on]
Card1.Amixer.info:
 Card hw:1 'SB'/'HDA ATI SB at 0xf0340000 irq 16'
   Mixer name : 'IDT 92HD87B2/4'
   Components : 'HDA:111d76d9,103c3387,00100107'
   Controls : 17
   Simple ctrls : 10
Date: Sun Sep 30 10:49:16 2012
HibernationDevice: RESUME=UUID=f337aa1d-0c6a-4072-85e1-f784e31f760e
InstallationMedia: Kubuntu 12.04 LTS "Precise Pangolin" - Release i386 (20120423)
MachineType: Hewlett-Packard HP Pavilion dm1 Notebook PC
ProcEnviron:
 SHELL=/bin/bash
 TERM=xterm
 PATH=(custom, user)
 LANG=en_GB.UTF-8
 LANGUAGE=en_GB:en
ProcFB: 0 radeondrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.2.0-30-generic-pae root=UUID=fd5dd7d3-238a-4d8b-80a0-964f9d484351 ro quiet splash thermal.off=1 vt.handoff=7
RelatedPackageVersions:
 linux-restricted-modules-3.2.0-30-generic-pae N/A
 linux-backports-modules-3.2.0-30-generic-pae N/A
 linux-firmware 1.79.1
SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 08/04/2011
dmi.bios.vendor: Hewlett-Packard
dmi.bios.version: F.01
dmi.board.asset.tag: Base Board Asset Tag
dmi.board.name: 3387
dmi.board.vendor: Hewlett-Packard
dmi.board.version: 36.0A
dmi.chassis.type: 10
dmi.chassis.vendor: Hewlett-Packard
dmi.chassis.version: Chassis Version
dmi.modalias: dmi:bvnHewlett-Packard:bvrF.01:bd08/04/2011:svnHewlett-Packard:pnHPPaviliondm1NotebookPC:pvr0697100000204600000320100:rvnHewlett-Packard:rn3387:rvr36.0A:cvnHewlett-Packard:ct10:cvrChassisVersion:
dmi.product.name: HP Pavilion dm1 Notebook PC
dmi.product.version: 0697100000204600000320100
dmi.sys.vendor: Hewlett-Packard

Revision history for this message
Tim Taylor (drtimt) wrote :
Brad Figg (brad-figg)
Changed in linux (Ubuntu):
status: New → Confirmed
Revision history for this message
Joseph Salisbury (jsalisbury) wrote :

Can you follow the "Boot options" instructions on the following wiki to enable additional output on boot:

https://wiki.ubuntu.com/DebuggingKernelBoot

As mentioned on the wiki, it would be great if you can attach a log file which may have captured any messages you see. If you are unable to capture a log file, a digital photo will work just as well. As a last resort you can even copy messages down by hand.

Changed in linux (Ubuntu):
importance: Undecided → High
tags: added: regression-update
Changed in linux (Ubuntu):
status: Confirmed → Incomplete
Revision history for this message
Tim Taylor (drtimt) wrote :

I reinstalled kernel 3.2.0-31-generic-pae and changed the boot options as described on the wiki.

The first thing to note is that when I edit the boot options, the system boots up successfully most of the time (it seems to boot successfully more often than if I just leave grub to run without interruption, but maybe that is just down to luck when I tried both methods a number of times this morning).

As per the wiki, is removed the quiet, splash, and vt.handoff options, and set gfxmode to text.

Having done this, on an occasion when the problem reoccurred, the screen just went blank (and stayed that way) immediately after pressing Ctrl+x from the grub edit options menu. So I was unable to see any error messages.

I retried booting with acpi=off (as suggested in the Wiki) in addition to the other changes. This time I got screensfull of messages, and, eventually (after about 50s) got dropped into an Initramfs shell. The final screen of messages showed some errors, as shown in the attached screenshot. Whether these are related to the original problem, or just due to me turning off acpi, I don't know.

Revision history for this message
Tim Taylor (drtimt) wrote :

I upgraded a number of packages on my laptop this morning, and I think these updates may have solved the problem. I've booted the system 6-7 times today without any problems. I'll report back if I spot a re-occurrence of the issue, but I would assume it is fixed unless I post anything else. The packages that got updated today are all related to the Nvidia driver, so it looks like the problem was video driver related.

The packages updated today were:

 - nvidia-common (updated to 0.2.44.2)
 - jockey-common (updated to 0.9.7-0ubuntu7.3)
 - jockey-kde (updated to 0.9.7-0ubuntu7.3)
 - dh-modaliases (updated to 0.9.7-0ubuntu7.3)

Revision history for this message
Tim Taylor (drtimt) wrote :

Unfortunately the problem re-emerged when booting this morning, so the package updates listed above have not completely solved the issue.

The failure to boot does seem to be happening less frequently now, though. I imagine that this is something to do with how the video driver interacts with the kernel.

Revision history for this message
Tim Taylor (drtimt) wrote :

I think this problem is now fixed. I was previously having problems with CPU lock-ups on my laptop (an HP Pavilion dm1-4020sa), and had fixed the problem by blacklisting the sp5100_tco driver, and disabling the thermal module (as discussed at https://bbs.archlinux.org/viewtopic.php?id=133541).

It seems that subsequent updates to the kernel have now fixed this CPU lock-up problem, and, in addition, the blacklisting of the driver and disabling of the thermal module was causing my most recent problems with failing to boot from grub from kernel 3.3.0-31 onwards.

I have now removed the blacklisting of the sp5100_tco driver, and re-enabled the thermal module. I'm no longer getting the CPU lockups, and the new kernels are now successfully booting. So everything seems to be OK.

Thanks to everyone who has helped looking in to this.

Revision history for this message
Tim Taylor (drtimt) wrote :

Unfortunately I spoke too soon again - this is an intermittent problem and it occurred again this morning (I have now upgraded to Quantal with kernel 3.5.0-17-generic).

I have collected dmesg log files from an occasion when the problem occurred (the screen went blank during the boot process), and also from an occasion when the boot process was successful. I can't see anything particularly worrying in the first file, but I am attaching them here in case someone else can figure out what's going wrong.

I've called the log file from the failed case dmesg-bad, and from the successful case, dmesg-good.

To help with comparing the files, I am also attaching versions with the initial timestamps on each line removed (these have a suffix -stripped), and also versions of the stripped files where the lines are sorted alphabetically to make it easy to spot which lines are different (with suffix -stripped-sorted).

Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :

I am also attaching the Xorg.0.log files, from a failed and successful boot, again also with versions which have the timestamps stripped and the lines sorted alphabetically.

Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :
Revision history for this message
Tim Taylor (drtimt) wrote :

This appears to be an issue with the xorg video driver, which arose for some reason when running in combination with 3.2.0-31 and subsequent kernels, but not with earlier kernels.

It turns out that the laptop is not dead, but just that the screen does not work at all. I can do CTRL-ALT-F1 to get to a text terminal (although I can't see it), log in with my user name and password, then issue "sudo shutdown -r now" (all blind), to reboot the laptop.

I have now installed the proprietary fglrx driver for the Radeon 6320M, and the problem has gone away.

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.