filesystem not found ext3 Kernel Panic ! attempted to kill init
 

hi,

last night i was creating swap sapce on my linux enterprises 4

i creat swap and label /dev/hdc1 to root by e2label command, after that i tried same on /dev/HDC2 which was swap partion, but this command did not work on it,

i then tried mke2fs-L <label name> on /dev/hdc2, mistakenlly during running this command (mke2fs-L) i was in root@/etc directory.

after running this command it format that swap partion, and within one minute i was unable to do anything on my PC. i just restart it and it is not booting

it is giving errors like:


kernel/vmlinuz-2.6.9-5 EL or root=/dev/volgroup00/logvol00 rhgb quiet
[Linux -bzimage,setup=0x1400,size=0x15cb84]
initrd/initrd-2.6.9-5-EL.img
[linux_initrd@0xf6f2000,0xed7f2 bytes]
uncompressing linux ok booting the kernel
audit(1221088194 . 860:0): initialized

redhat nash version 4.1.18 starting

File discriptor 3 left open
filetype is ext2 partion type ox83


reading all physical volumes. this may take a while..

no volumegruop found

File discriptor 3 left open

reading all physical volumes. this may take a while..

no volumegruop found
File discriptor 3 left open

reading all physical volumes. this may take a while..

no volumegruop found

Mount: error6 mounting ext3
mount: error2 mounting none

switch root : mount failed : 22
unmount/initrd/dev failed :22




Will some one give me solution that how to recover such a problen

i could go only in Grub now.
Kernel panic - not syncing : attempted to kill init !

Which distro? There are many "enterprise" flavours.

So /dev/hdc1 is your root partition?
So /dev/hdg2 is your swap partition?

Note - e2label vill not work to label a swap partition because swap is not an ext file system.

So now, /dev/sdc2 is some ext file system and not swap?

Looks like you have Redhat Enterprise Linux 4, is this correct?

Try booting to rescue mode and running fsck.
You can probably also use this to restore your swap partition.

However - this is a guess... need more info. For eg. Why were you creating a new swap? Why did you feel you needed to label your root partition - is it new?

Use "mkswap /dev/sdc2" to recreate your swap partition. Did you change any of the other swap partitions?

I'm wondering if there is a "resume=/dev/sdc2" entry in grubs kernel line entry. This might cause a problem because the resume device should be a swap partition. Another possibility is if you changed the partitioning scheme on the drive and a device such as the root partition has changed numbers. However from your description, it doesn't seam to be the case.

while you are in the rescue disk, explore around with your devices. For example, use fdisk -l to examine the partitions.

Also run lvscan.

I think what happened has little to do with trashing your swap partition. Maybe running e2label on your root partition did something to the lvm volume or volume label. You may need to run lvrename, but I'm just guessing now.

Well, if you executed mke2fs on /, then / may be gone. mke2fs creates a brand new file system, unlike tune2fs -L, which only assigns a label. Still, if / was mounted, you should have been warned.

"attempted to kill init" actually means, "the 'init' process died." (Basically, this process is not allowed to die.)

It seems to me that filesystem-support for 'ext3' is not properly included in your kernel, and/or the startup-sequence is not loading whatever kernel-module is needed to support it.

Check all of the messages carefully, working backward from the point where 'init died.'

Background Information:
When Linux (or any Unix-like system) finishes booting, it auto-creates one process: 'init,' the so-called "process #1." This process has a central role in that it starts all other processes and monitors their execution. Without it, Linux could not function. Therefore, if for any reason this process does die, the kernel panics.