Centos recovering from Kernel panic on startup

[Thomas]

>   1. Get a backup of the actual bits on the drive if you can

http://www.garloff.de/kurt/linux/ddrescue/
might come in handy.

Ideally, slurp everything into an image file, copy the file and then
work on the copy of the image file (i.e. loopback mount as ext2, run FS
recovery tools etc. etc.)
("dd noerror ..." could/should work as well, but this seems neater.)

>   2. Attempt to mount the driver WITHOUT the journal (it sounds like it's
>      ext3; you might be able to mount it as ext2)
> 
>      BIG WARNING: Doing this step _may_ / _will_ lose data. Please do NOT
>      complain, blame, sue or otherwise tell me I didn't tell or warn you.

That's why you do it on a loopback-mounted COPY of the original bits.
If it works, fine.
If it noodles the data, no problem; grab another copy of the dd-ed bits
and try something else.

> Realistically, I'd get a rescue disk or an install disc with a 
> sufficiently good set of tools (Gentoo is good, so are the modern 
> Debian's) and quickly get all the important information off the drive.
> 
> Then see if it's still under warranty because it's probably going to get 
> more bad blocks and worse.

Also.... have you tried running SMART utilities on the disk?
The manufacturer may have a bootable ISO that you can use to test the
disk (warning! if the disk is _really_ bad then running the self-test or
SMART diagnostics may nuke what little is left of your filesystem. get
the DD copy first if at all possible!)


Thomas