$^*$()*! Kernel panic on os AND Livecd?!

violagirl23 · 2008-06-01 17:44:00

What the heck did I do? Oh crap oh crap oh crap. I am getting a kernel panic when I boot into Arch with the following:

Initramfs Completed - control passing to kinit
IP-Config: no devices to configure
Waiting 0s before mounting root device...
kinit: cannot open root device dev(0,0)
kinit: init not found!
kernel panic - not syncing: Attempted to kill init!

So I booted into the livecd I installed on to uncomment the Arch Fallback Kernel in my grub and try using it, but the CD kernel panics now too!

Code: 0f 0b eb fe e8 f1 ff ff 48 49 de 48 89 c7 31 d2 5b e9 0a
RIP     [<ffffffff8027955f>] unlock_page+0xf/0x30
  RSP  <ffffffff80656e00>
---[ end trace 237687beb75b2431 ]---
Kernel panic - not syncing : Aiee, killing interrupt handler!

Oh lord, what does that mean, what did I do? ;-; Did I break my laptop already? I've done a few things since I last rebooted. I followed the Arch laptop wiki and changed a lot of settings for my laptop, like turning off APM. I also installed Ubuntu because I PROMISED my friend I would on the 20 GB of free space at the end of the drive. I manually partitioned the drives, selecting only the free space and clicking on Advanced and unchecking "Make grub" (so I still have to manually enter the Ubuntu info into my Arch grub) and it SAID it was only going to make the free space partition Ubuntu and my Arch swap as swap.
But none of that should matter for an Arch LiveCD anyway! So what is going on? This is terrible!!!!!!!!!!

Update: The Ubuntu LiveCD boots just fine, so I booted into the LiveCD environment. Now what should I do? I've very concerned, since the Arch LiveCD also failed.

Last edited by violagirl23 (2008-06-01 17:59:08)

mrbug · 2008-06-01 17:59:09

It sounds like there's some piece of hardware that's causing something (udev or acpi?) to crash. I had the same problem with some bad memory.

Try using the "noacpi" boot parameter. You may also need to disable module autoload...

I could be wrong, though.

violagirl23 · 2008-06-01 18:20:22

I can boot into the fallback kernel with no problem. I became root and did

mkinitcpio -p kernel26

and tried rebooting into the regular Arch, but it did not work. I got the exact same error as before. What else can I try doing? I am so confused.

I read this somewhere while searching this problem:
'The problem is probably that mkinitcpio does filesystem autodetection - and it tries to access all block devices in /dev to do that. Maybe it hangs on trying to access one of them.'

So the fallback kernel is only different in that it does not have the autodetect hook. Arch was definitely booting before this, and that was WITH the default. So now possibly it may be that my hard drive or CD drive is causing it issues? This laptop is only 4 days old!

This is what really confuses me, however:
Ubuntu boots up fine. Wouldn't default Ubuntu ALSO use the autodetect hook in the kernel? So I don't understand.

Eep. Any suggestions? At the very least, how does one remove the autodetect hook, and could I have a normal system without this hook?

Oh, when it accesses all block devices in /dev... the only devices that have changed is /dev/sda5 and /dev/sda6. Is it possible that the information on the Ubuntu install is making it hang? (not that THAT makes anymore sense EITHER!)

Last edited by violagirl23 (2008-06-01 19:23:46)

violagirl23 · 2008-06-01 19:35:42

Maybe that IS the issue. Ubuntu seems to have done something terribly awkward, now that I look at cfdisk.
Here is my OLD fstab:

Name   Flags     Part Type   FS Type           [Label]        Size (MB)
----------------------------------------------------------------------------
                              Unusable                             1.05*
sda1    Boot      Primary     Unknown (27)                      6594.50*                        
sda2              Primary     NTFS               [^C]          31471.95*
sda3              Primary     Linux ext3                      101442.38*  
sda5              Logical     Linux swap / Solaris              2048.10*
                  Logical     Free Space                       18481.31*

Here is my new one, however:

Name   Flags     Part Type   FS Type           [Label]        Size (MB)
----------------------------------------------------------------------------
                              Unusable                             1.05*
sda1    Boot      Primary     Unknown (27)                      6594.50*                        
sda2              Primary     NTFS               [^C]          31471.95*
sda3              Primary     Linux ext3                      101442.38*  
sda5    NC        Logical     Linux swap / Solaris              2048.10*
                  Logical     Free Space                           7.34*
sda6              Logical     Linux ext3                       18473.98

Ubuntu put an NC flag on my swap, and when it made /dev/sda6, it put about 7 megabytes of free space before it! And the sda6 one doesn't have a * after the size like all the others. Could this be what is causing all my issues? What IS an NC flag anyway?

bgc1954 · 2008-06-01 19:52:27

Take a look at: http://bbs.archlinux.org/viewtopic.php?id=49309

If your kernel-fallback.img boots then check your /etc/mkinitcpio.d/kernel-fallback.conf against /etc/mkinitcpio.conf and see what the differences are and edit to suit.

This may answer some questions about the NC flag: http://www.ss64.com/bash/cfdisk.html

Looks like ubuntu did a number on you--or you did one on yourself.

Are you sure that ubuntu or you didn't mess with your grub? Recheck your /boot/grub/menu.lst maybe?

violagirl23 · 2008-06-01 20:08:54

I don't have a /etc/mkinitcpio.d/kernel-falllback.conf!

bash-3.2# cd /etc/mkinitcpio.d/
bash-3.2# ls
example.preset    kernel26.kver  kernel26.preset

I believe my grub is fine... I put in the Ubuntu part myself (but remember, that one boots) but didn't change anything else from before. And even before I added the Ubuntu part it wouldn't boot.

# general configuration:
timeout   5
default   0
color light-blue/black light-cyan/blue

# (0) Arch Linux
title  Arch Linux
root   (hd0,2)
kernel /boot/vmlinuz26 root=dev/disk/by-uuid/7a86282d-3dda-4065-ba32-edc157c85b71 ro vga=775
initrd /boot/kernel26.img

# (1) Arch Linux
title  Arch Linux Fallback
root   (hd0,2)
kernel /boot/vmlinuz26 root=/dev/disk/by-uuid/7a86282d-3dda-4065-ba32-edc157c85b71 ro
initrd /boot/kernel26-fallback.img

# (3) Ubuntu Linux
title  Ubuntu Linux
root   (hd0,5)
kernel /boot/vmlinuz-2.6.24-16-generic root=/dev/sda6 ro
initrd /boot/initrd.img-2.6.24-16-generic

# (2) Windows Vista
title Windows Vista
rootnoverify (hd0,1)
makeactive
chainloader +1

Last edited by violagirl23 (2008-06-01 20:13:38)

violagirl23 · 2008-06-01 21:25:51

I tried deleting Ubuntu and remade my swap, and my partition table looks as it did before, except it still won't boot, except into the fallback kernel.
;-;
This is terrible. I'm going to break my laptop if I have to shut down improperly anymore. It's been like 15 times since I got it 3 days ago now.
What exactly IS dev(0.0) anyway? That's what's the issue?

kinit: cannot open root device dev(0,0)
kinit: init not found!

Last edited by violagirl23 (2008-06-01 21:30:49)

tigrmesh · 2008-06-01 21:47:49

Have you tried something like Super GRUB Disk? That may help.

violagirl23 · 2008-06-01 21:51:10

My grub is working just fine, if that's what you mean. I've now tested, I could boot fine into Ubuntu (until I deleted it, of course), and I can boot fine into both Vista and the Arch fallback kernel. It is only the regular Arch kernel that is doing this.

Edit: Although I am not sure why this would pop up NOW and not earlier, do you think maybe this is the issue? The Configuring mkinitcpio Wiki says...

Troubleshooting
If you are having problems getting mkinitcpio to detect your hard drive giving errors akin to "Can't find device dev(0,0)" when switching to kinit, then this could be because of a conflict that the ata_piix and piix drivers have. The beyond kernel has some libata patches that cause ata_piix to *conflict* with piix.
[edit] Solution
Edit /etc/mkinitcpio.conf to only have ide or sata or scsi depending on what your system actually needs to boot.

Last edited by violagirl23 (2008-06-01 22:32:54)

bgc1954 · 2008-06-01 22:44:03

Maybe check this post: http://bbs.archlinux.org/viewtopic.php?id=49106

Or just remove the autodetect hook from /etc/mkinitcpio.conf and try that to see if you can get a proper kernel26.img.

Edit: Or for something right out of left field, maybe try changing your grub to /dev/sda3--or whatever arch is--instead of the disk-uid thingy. Maybe for some reason your kernel26.img doesn't recognize that naming scheme--like I said right out of left field.

Last edited by bgc1954 (2008-06-01 22:57:33)

ghostHack · 2008-06-01 22:58:23

Your Arch grub entry is wrong

For your main Arch kernel you have

root=dev/disk/by-uuid/7a86282d-3dda-4065-ba32-edc157c85b71

but for your fallback kernel you have

root=/dev/disk/by-uuid/7a86282d-3dda-4065-ba32-edc157c85b71

spot the difference?

You are missing a leading '/' on your main kernel (root=/dev/disk/by-uuid/7a86282d-3dda-4065-ba32-edc157c85b71). This is why the fallback boots but your main kernel doesn't

bgc1954 · 2008-06-01 23:09:50

ghosthack, you must have young eyes. I stared and stared at the grub entry and didn't see that missing /. That should definitely cure the problem.

violagirl23 · 2008-06-01 23:24:03

All right, as soon as I finish typing this paper I'll boot back into it.
I have one crucial question, however. Why would this also cause the FTP Install CD to have a kernel panic?

Update: YES! It booted! And I even managed to get my wireless working.... I forgot the crucial step if "ifconfig wlan0 up" and was wondering why iwlist wlan0 scan wasn't working.
*sweatdrop* I'm glad that's been figured out. And I thank you so much.

That only leaves me with one more question: I had to do a lot of improper shutdowns because of all this. Is there any way to check and/or fix any damage that might have been done by this? (I would estimate it happened about 15 times... eep).

THANK YOU! I seriously never would have noticed that in a million years! Once someone answers my question about checking about the improper shutdowns, I'll add [Solved] to the title.

Last edited by violagirl23 (2008-06-02 00:59:12)

Misfit138 · 2008-06-02 03:03:12

violagirl23 wrote:

That only leaves me with one more question: I had to do a lot of improper shutdowns because of all this. Is there any way to check and/or fix any damage that might have been done by this? (I would estimate it happened about 15 times... eep).
THANK YOU! I seriously never would have noticed that in a million years! Once someone answers my question about checking about the improper shutdowns, I'll add [Solved] to the title.

Certainly. Run a fsck. For ext2/3, man e2fsck.

Arch Linux

#1 2008-06-01 17:44:00