[NTLUG:Discuss] Hardware / kernel question

[Bryan J. Smith]

The kernel panic is occurring because your LILO/GRUB says load root from /dev/hda.
In GRUB, you can dynamically edit this by hitting "e" and changing  anything that says hda to hdc.

You'll still probably boot into a recovery mode because /etc/fstab also needs to be changed.
I.e., even when you make the GRUB change to prevent the kernel panic, you'll get mount errors.
But you can do that with vi from there - modify both the fstab and grub.conf.

Understand while 16-bit DOS kernels use the 16-bit BIOS disk assignment,
32-bit NT and 32/64-bit Linux kernels do not.
So after the BIOS loads the kernel, 
the kernel's devices may not match.

In your case, hdc is still GRUB disk hd0 (BIOS disk 80h),
but it is no longer hda so the root=hda won't work (this causes the kernel panic),
and the fstab also still points to hda (which causes your mount errors).

-- Bryan

P.S.  I no longer build desktops, much less servers, without a 3Ware Escalade 7006-2 (ATA) or 8006-2 (SATA).

-- 
Bryan J. Smith (currently mobile)
b.j.smith at ieee.org

-----Original Message-----
From:  Douglas King 
Date:  04-12-26 10:25
To:  discuss at ntlug.org
Subj:  [NTLUG:Discuss] Hardware / kernel question

I have a mail server that is "biting the dust".  It APPEARS that the ide1 
controller may be what is failing.  After changing the ribbon over to the 
ide2 connector...I try to boot the OS and get a kernel panic.  Says 
something about the init missing...or something.  Can I change something 
while booted to fix this before changing it over?  The server is 
dying...won't last much longer as is....


_______________________________________________
https://ntlug.org/mailman/listinfo/discuss