Boot failure: hard disk does not exist (Hardy beta)

[Gerv]

Binary package hint: linux-source-2.6.20

I just upgraded my Thinkpad X40 to the Hardy beta (using update-manager -d or similar) and now it won't boot by default.

It turns out that only the latest installed kernel (2.6.24-12-386) fails. Previous available kernels back to 2.6.17-10 will; before that, there's a kernel panic which is probably unrelated.

The symptoms when booting the normal kernel are that the boot sequence gets to:

Loading, please wait...

and stops. After about three minutes, it drops to a busybox shell, saying:

    Check root= bootarg cat /proc/cmdline
    or missing modules, devices: cat /proc/modules ls /dev
  Reading all physical volumes. This may take a while...
ALERT: /dev/disk/by-uuid/ff3536b7... does not exist. Dropping to a shell!

...

(initramfs) _

/proc/cmdline is:

root=UUID=ff3536... ro quiet splash

Doing ls from the (initramfs) prompt, there is no /dev/disk, /dev/hd*, or /dev/sd*.

The symptoms when booting the recovery kernel are:

Mounting root file system ...
Done.
...
Waiting for root file system ...
<hang>

The following comment in an older bug may be the same thing, although I don't get the mount errors he gets:
https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/106063/comments/97

Please let me know what additional information I can provide to help debug this problem.

Gerv

[Gerv (gerv)]

<cough> I know you guys are busy and it's just been a weekend, but I would have thought that "your soon-to-be-released operating system makes my machine unbootable" bugs might attract a certain amount of interest :-) Or at least being allocated an Importance...

Gerv

[Malcolm Parsons (malcolm-parsons)]

> Please let me know what additional information I can provide to help debug this problem.

I think the output from the following commands might help:

dmesg
lspci -vvv

[Gerv (gerv)]

You didn't say whether from the broken boot or the good boot. lspci isn't available on the broken boot, and I have no way of getting the dmesg output out of the computer because I can't access my disks or mount a USB drive. So here are the files from a good boot.

Gerv

[Gerv (gerv)]

Let me know if there's anything else you need. I could probably manually copy out the last few lines of dmesg output from the bad boot, or grep it for particular strings, if that would help.

Gerv

[Malcolm Parsons (malcolm-parsons)]

> or grep it for particular strings

anything to do with "sd", "ata" or "scsi" might help.

[Gerv (gerv)]

I've just updated to the latest packages, and 2.6.24-14 boots fine for me.

Are you still interested in the dmesg output from the boot of 2.6.24-12?

Gerv

[Malcolm Parsons (malcolm-parsons)]

Not really, I just read your blog.

[Gerv (gerv)]

There doesn't seem to be a "works for me" status, so I'll go with "fix committed".

Gerv

[Gerv (gerv)]

Reopening. The beta kernels seem to alternate between working and not working. At the moment (2.6.24-16 and 2.6.24-15) they are not working, so I'm reopening this bug. 2.6.24-14 does work.

Gerv

[Gerv (gerv)]

Ah, maybe that's why no-one was looking at my bug. Switching package to linux-source-2.6.24.

Gerv

[Gerv (gerv)]

"'linux-source-2.6.24' is a binary package. This bug has been assigned to its source package 'linux' instead."

Huh? How can linux-source be a binary package? :-| Oh well, hope it's in the right place now.

Gerv

[Gerv (gerv)]

...and it works again, with the final Hardy kernel. <shrug>

Gerv

[Leann Ogasawara (leannogasawara)]

Per the last comment, I'm marking this Fix Released for Hardy. However the 2.6.20 task will be closed since Feisty has reached it's End of Life - http://www.ubuntu.com/news/ubuntu-7.04-end-of-life . Thanks.

[Fixxser (steve-333)]

For me, this was the problem I had. Just installed Ubuntu 8.10 - the Intrepid Ibex on a AMD64.
There was a big delay on booting from bios to os as well. Turned out I had a jumper on the extra data drives I had in the system, (Western Digital), on the master w/slave jumpers. Diagram showed for single or master it should be all NO jumbers, blank. It works and boots great now.