Yorkfield/Eaglelake system hang during boot (and no splashscreen)

Bug #553675 reported by Bryce Harrington
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
New
High
Unassigned
Lucid
New
High
Unassigned

Bug Description

[Problem]
With the -16, -18, and -19 kernels, system appears to freeze during boot (becomes completely unresponsive to input) without displaying any boot graphics. Some errors are shown regarding plymouth failing to start, but the screen clears before I can capture what they say.

I have an old -14 kernel installed, which boots fine.

[Discussion]
This is an unreleased developer box from Intel with the Yorkfield/Eaglelake chipset. I've been using it during Lucid as a development box, updating it every week or two, but it's been a few weeks since I rebooted. I updated to the -18 kernel and rebooted, and found that it was not booting except with the -14 kernel. I updated again to the -19 kernel, but same problem.

The boot process shows it going into KMS (the text on the console gets small anyway), and the screen refreshes a few times. I see listings of error messages printed out (DNS unavailable... NFS mount points couldn't be mounted... mountall freaking out... something about plymouth exiting...) and then the system does nothing further. If I'm quick I can do vt switches (they're all blank with one cursor blinking), or can do a ctrl-alt-del to reboot, but if I'm not quick it seems to lock up hard after a minute or so. It seems sometimes (every other boot maybe?) it locks up right away after printing the nfs mountall error messages and I can only cycle the power to recover.

Booting into recovery mode shows identical behavior (same error messages too).

Booting with the -14 kernel always works. (At least, X and gdm comes up, I've not seen the bootscreen stuff)

The system is connected to a KVM. I thought perhaps it could be filtering out the monitor's EDID, however while booted to the -14 kernel, running xrandr --verbose shows the EDID blob there, and appropriate resolutions listed and so on.

Since one of the big changes in -16 was the introduction of the 2.6.33 drm, I am wondering if it is a drm issue?
---
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.21.
Architecture: i386
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: bryce 11038 F.... pulseaudio
CRDA: Error: [Errno 2] No such file or directory
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xfe7f0000 irq 22'
   Mixer name : 'Intel G45 DEVELK'
   Components : 'HDA:10ec0885,10ec0885,00100103 HDA:80862803,80860101,00100000'
   Controls : 46
   Simple ctrls : 24
DistroRelease: Ubuntu 10.04
HibernationDevice: RESUME=UUID=6014ea95-bee9-4b30-b6f4-e5bb0891951b
InstallationMedia: Ubuntu 10.04 "Lucid Lynx" - Alpha i386 (20100113)
IwConfig:
 lo no wireless extensions.

 eth0 no wireless extensions.
MachineType: System Manufacturer To Be Filled By O.E.M. Product Name To Be Filled By O.E.M.
Package: linux (not installed)
ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.32-14-generic root=UUID=cc52aa3c-96d7-4e1e-b2d7-83b10e23c0d7 ro quiet splash
ProcEnviron:
 PATH=(custom, user)
 LANG=C
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.32-14.20-generic
Regression: Yes
RelatedPackageVersions: linux-firmware 1.33
Reproducible: Yes
RfKill:

Tags: lucid regression-potential needs-upstream-testing
Uname: Linux 2.6.32-14-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare
dmi.bios.date: 05/30/2008
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: CGELIA47.86P
dmi.board.asset.tag: BB Asset To Be Filled By O.E.M.
dmi.board.name: BB Name To be filled by O.E.M.
dmi.board.vendor: BB Manufacturer To be filled by O.E.M.
dmi.board.version: BB Version To be filled by O.E.M.
dmi.chassis.asset.tag: Chasis Asset Tag To Be Filled By O.E.M.
dmi.chassis.type: 3
dmi.chassis.vendor: Chasis Manufacturer To Be Filled By O.E.M.
dmi.chassis.version: Chasis Version To Be Filled By O.E.M.
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvrCGELIA47.86P:bd05/30/2008:svnSystemManufacturerToBeFilledByO.E.M.:pnProductNameToBeFilledByO.E.M.:pvrSystemVersionToBeFilledByO.E.M.:rvnBBManufacturerTobefilledbyO.E.M.:rnBBNameTobefilledbyO.E.M.:rvrBBVersionTobefilledbyO.E.M.:cvnChasisManufacturerToBeFilledByO.E.M.:ct3:cvrChasisVersionToBeFilledByO.E.M.:
dmi.product.name: Product Name To Be Filled By O.E.M.
dmi.product.version: System Version To Be Filled By O.E.M.
dmi.sys.vendor: System Manufacturer To Be Filled By O.E.M.

Revision history for this message
Bryce Harrington (bryce) wrote : AlsaDevices.txt

apport information

summary: - Yorkfield Eaglelake
+ Yorkfield/Eaglelake system hang during boot (and no splashscreen)
tags: added: apport-collected
description: updated
Revision history for this message
Bryce Harrington (bryce) wrote : AplayDevices.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : ArecordDevices.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : BootDmesg.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : Card0.Amixer.values.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : Card0.Codecs.codec.2.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : Card0.Codecs.codec.3.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : CurrentDmesg.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : Lspci.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : Lsusb.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : PciMultimedia.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : ProcCpuinfo.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : ProcInterrupts.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : ProcModules.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : UdevDb.txt

apport information

Changed in linux (Ubuntu):
importance: Undecided → High
Revision history for this message
Bryce Harrington (bryce) wrote : UdevLog.txt

apport information

Revision history for this message
Bryce Harrington (bryce) wrote : WifiSyslog.txt

apport information

Changed in linux (Ubuntu):
milestone: none → ubuntu-10.04
Revision history for this message
Bryce Harrington (bryce) wrote :

Note I did the apport-collect with the -14 kernel since the -16, -18, -19 kernels don't boot into recovery mode.

When I grep for 2.6.32 in /var/log, I see that I did successfully boot the -16 kernel at least once previously:

root@blumonc:/var/log# grep 2\.6\.32- * | grep -v 2\.6\.32-1[01234] | grep -v dpkg.log | grep 2.6.32-
dmesg.1:[ 0.000000] Linux version 2.6.32-16-generic (buildd@rothera) (gcc version 4.4.3 (Ubuntu 4.4.3-3ubuntu1) ) #25-Ubuntu SMP Tue Mar 9 16:33:52 UTC 2010 (Ubuntu 2.6.32-16.25-generic)
dmesg.1:[ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.32-16-generic root=UUID=cc52aa3c-96d7-4e1e-b2d7-83b10e23c0d7 ro quiet splash
kern.log.1:Mar 31 23:54:16 localhost kernel: [1317949.132309] Pid: 3744, comm: apparmor_parser Not tainted 2.6.32-16-generic #25-Ubuntu
kern.log.2:Mar 16 17:48:37 localhost kernel: [ 0.000000] Linux version 2.6.32-16-generic (buildd@rothera) (gcc version 4.4.3 (Ubuntu 4.4.3-3ubuntu1) ) #25-Ubuntu SMP Tue Mar 9 16:33:52 UTC 2010 (Ubuntu 2.6.32-16.25-generic)
kern.log.2:Mar 16 17:48:37 localhost kernel: [ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.32-16-generic root=UUID=cc52aa3c-96d7-4e1e-b2d7-83b10e23c0d7 ro quiet splash
messages:Mar 31 23:54:16 localhost kernel: [1317949.132309] Pid: 3744, comm: apparmor_parser Not tainted 2.6.32-16-generic #25-Ubuntu
messages.2:Mar 16 17:48:37 localhost kernel: [ 0.000000] Linux version 2.6.32-16-generic (buildd@rothera) (gcc version 4.4.3 (Ubuntu 4.4.3-3ubuntu1) ) #25-Ubuntu SMP Tue Mar 9 16:33:52 UTC 2010 (Ubuntu 2.6.32-16.25-generic)
messages.2:Mar 16 17:48:37 localhost kernel: [ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-2.6.32-16-generic root=UUID=cc52aa3c-96d7-4e1e-b2d7-83b10e23c0d7 ro quiet splash
syslog.1:Mar 31 23:54:16 localhost kernel: [1317949.132309] Pid: 3744, comm: apparmor_parser Not tainted 2.6.32-16-generic #25-Ubuntu

I think this may rule out my theory that the issue is drm related. But I'm not sure.

Revision history for this message
Bryce Harrington (bryce) wrote :

Tried booting -19 with nomodeset and without splash and quiet, and same thing (but console text was bigger and scrolled slowly)

tags: added: regression-potential
Revision history for this message
Bryce Harrington (bryce) wrote :

Tried "nomodeset video=vesafb" and "nomodeset video=disabled". The initial console output is larger and slow scrolling, then it flips to small font size and hangs.

Revision history for this message
Bryce Harrington (bryce) wrote :

Aha, Sarvatt figured it out. He indicated around -15 there was a change to how filesystem mounting worked, and some entries in fstab could cause this boot. Once I deleted all my chroot and nfs mounts out of fstab, it came up on the -19 kernel.

So that's fine, although this is an extraordinarily unfriendly way of handling issues with fstab formatting....

Revision history for this message
Bryce Harrington (bryce) wrote :

So the underlying issue is probably dupish of #510415, exacerbated by whatever reason that plymouth isn't working. If plymouth were running it presumably would give me the S&M options so I could skip mounting. In this configuration there's no solution other than to add nobootwait to the fstab entries.

tags: added: lucid
Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

I agree with Bryce that this appears to be the issue documented in bug 510415 which is triggered by the fact plymouth is failing. bug 510415 was marked Invaid because "this is fixed now in lucid with mountall and plymouth. When mountall is waiting for a disk to be available, a message is displayed, with the option to skip the disk." However, because pymouth appears to be failing, you're not present the option to skip. For now I'm marking this as a duplicate to bug 510415. Bryce, I presume you've opened a bug against plymouth as well to address why it's failing in the first place.

Revision history for this message
Bryce Harrington (bryce) wrote : Re: [Bug 553675] Re: Yorkfield/Eaglelake system hang during boot (and no splashscreen)

On Fri, Apr 02, 2010 at 02:36:19PM -0000, Leann Ogasawara wrote:
> For now I'm marking this as a duplicate to bug 510415. Bryce, I presume
> you've opened a bug against plymouth as well to address why it's failing
> in the first place.

For whatever reason, plymouth lacked its theme packages, and once I
manually installed them it worked. There may well be a plymouth bug at
the heart of that part of the issue but for now I'm just chalking it up
to a transitory issue that won't affect karmic upgraders.

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.