boot problems ("invalid compressed format") / ("alloc magic is broken")/ (crc error)

[dino99]

Binary package hint: e2fsprogs

Having some problem to boot:

used grub-pc, os-prober on 3 hdd: 2 pata 1 sata (sdb xp ntfs, sdc jaunty32/karmic32 ext3, sda jaunty64 ext4)

The troubles does not exist before using grub2 & os-prober.

When i want to boot , it often fail: grub reboot or it freeze or exit with warning (for example: "invalid compressed format (err=1)" or "alloc magic is broken" or else) or begin to start & fall in a maintenace shell.

Same things if i erase splash or add acpi=off nolapic. Most of the time it boot at the third time but always fall in a maintenance shell: need to exit from root and then boot correctly from "normal boot option choice". Need to power off to try booting again.

Investigating, i found that os-prober check for all the non-existing fstype: don't know how to stop that.
/var/lib/os-prober/ only have 1 file (labels), not subfolder /mount/boot as seen in user.log & returned by 50mounted-tests

ProblemType: Bug
Architecture: i386
Date: Thu Jul 16 08:00:30 2009
DistroRelease: Ubuntu 9.10
NonfreeKernelModules: nvidia
Package: e2fsprogs 1.41.5-1ubuntu3
ProcEnviron:
 PATH=(custom, no user)
 LANG=fr_FR.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-3.19-generic
SourcePackage: e2fsprogs
Uname: Linux 2.6.31-3-generic i686

[dino99 (9d9)]

here is kernel.log

[dino99 (9d9)]

curiously, in kernel.log we see that jul 15 is ok and today jul 16 we have errors.

So, support work one time & not an other time :
what is the difference between: nothing except the updates avaliable in the last 12 hours, meaning nothing related to file system in that case.

[dino99 (9d9)]

thinking about the recent changes in karmic : become to use hal less & less but still use it, could it be a reason to explain those troubles ?

[Scott James Remnant (Canonical) (canonical-scott)]

I cannot find anything to justify why you believe this is a bug in e2fsprogs, frankly it looks a lot like a hardware problem to me - especially if you're fiddling with options like "acpi=off" and "nolapic" which disable components on your motherboard.

[Theodore Ts'o (tytso)]

Especially given that the symptoms reported are "failure to boot" and
"this didn't show up before grub2 or os-prober", this really doesn't
sound like an e2fprogs issue at all!

[dino99 (9d9)]

Architecture: i386
Dependencies:

DistroRelease: Ubuntu 9.10
NonfreeKernelModules: nvidia
Package: os-prober 1.29ubuntu2
PackageArchitecture: i386
ProcEnviron:
 SHELL=/bin/bash
 LANG=fr_FR.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-3.19-generic
Uname: Linux 2.6.31-3-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

[dino99 (9d9)]

Architecture: i386
DistroRelease: Ubuntu 9.10
NonfreeKernelModules: nvidia
Package: grub-pc 1.96+20090611-1ubuntu4
PackageArchitecture: i386
ProcEnviron:
 SHELL=/bin/bash
 LANG=fr_FR.UTF-8
ProcVersionSignature: Ubuntu 2.6.31-3.19-generic
Uname: Linux 2.6.31-3-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

[dino99 (9d9)]

May be an explanation:

At the very beginning when booting, i've seen a warning about acpi:
Acpi: aborted because broken padding

It's the first time i see this warning, actually run 2.6.28-15-generic (64 bits)

[dino99 (9d9)]

tried to reconfigure & got this:

oem@dubuntu:~$ sudo dpkg-reconfigure acpi-support
[sudo] password for oem:
 * Disabling power management... [ OK ]
update-rc.d: warning: /etc/init.d/acpi-support missing LSB information
update-rc.d: see <http://wiki.debian.org/LSBInitScripts>
 * Checking battery state...
/dev/sdc:
 setting Advanced Power Management level to 0xfe (254)
                                                                         [ OK ]

[dino99 (9d9)]

Architecture: amd64
DistroRelease: Ubuntu 9.04
NonfreeKernelModules: nvidia
Package: acpi-support 0.121
PackageArchitecture: amd64
ProcEnviron:
 SHELL=/bin/bash
 LANG=fr_FR.UTF-8
Uname: Linux 2.6.28-15-generic x86_64
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

[dino99 (9d9)]

these problems have gone now with the last upgrades of grub-pc

[Victor Vargas (kamus)]

This bug report is being closed due to your last comment regarding this being fixed with an update. For future reference you can manage the status of your own bugs by clicking on the current status in the yellow line and then choosing a new status in the revealed drop down box.

Thank you again for taking the time to report this bug and helping to make Ubuntu better. Please submit any future bugs you may find.

[Kip Warner (kip)]

I encountered the same problem under gNewSense metad running on mipsel. Rebooting the machine and it went away. I've only seen it happen once. Looks difficult to reproduce.