Our United Linux 1.0 server crashed because of power blackout.
After this, it wasn't bootable anymore but stopped with the error message
Kernel panic: vfs: unable to mount rootfs on 03:02

Usually we try to rescue the system by doing the following:

- boot from CD, select Rescue-System and login as root
- mkdir /sysimage
- insert new entry in /etc/fstab:
/dev/hda2 /sysimage reiserfs defaults 0 0
- mount /sysimage
- chroot /sysimage
- grub-install /dev/hda

But this failed this time at the grub-install command with the following error:
The file /boot/grub/stage1 not read correctly.

After this we tryed several things like rebooting again from CD or to install grub via dialog and it ended up with the following error at the grub-install command:
could not find device for /boot: not found or not a block device

This is the content of grub.conf:
root (hd0,1)
install --stage2=/boot/grub/stage2 /boot/grub/stage1 d (hd0) /boot/grub/stage2 0x8000 (hd0,1)/boot/grub/menu.lst
quit


To install grub via single commands also fails: if you enter grub and then root (hd0,1) it says that it cannot read the number. It's a pretty old grub version, lower than 1.

Does anyone has any idea?

Um. That kinda looks like it's a kernel problem to me. I think you might need to pass "root=/dev/hda3" (change /dev/hda3 to the location of the root filesystem) to it from GRUB.

Well I'd reckon on hardware (as in hard disk) problems.
Kernel panic means you got past grub, and had trouble loading the kernel.
I'd be looking to test that disk.

I wouldn't be too concerned about the grub version - given the versions of everything else UL shipped, I doubt grub is a problem.

i agree with syg00, same thing happened to me, problem was /boot wasn't able to meet booting constraints

Thanks to all. We also think now that it's a hardware problem.