filesystem error unable to boot, how to restore the boot
 

hi guys,

i have got system with fedora core 2 installed,the problem is that due power cut my filesyemt started to give errors, so i tried fsck to check the filesystem , i got huge bunch of errors and the system got hanged so i had to restart , due which the filesystem of the root is not mounting now and i am not able to mount the root filesystem.it is giving following error while booting.


Red Hat nash ver 35.22 staring
mkrootdev : label / not found
mount : error 2 mounting ext 3
pivot root : pivot_root(/sysroot /sysroot/initrd failed 2
umount /initrd/proc failed 2
kernel panic - not syncing : no init found
try passing init= option to kernel

can any of the linux gurus help me out

Was your FC2 system in a logical volume? If so, you probably killed it by running fsck on it instead of on /dev/mapper/<logical volume name>. A "logical volume" is not an ext3 file system, even if the files systems in the volume are all ext3.

Hopefully you have a recent backup, in which case you can repartition, reinstall you system and restore from the backup.

Aside: Why are you using an old, unsupported, Fedora release? If you need to reinstall, I'd suggest that you consider installing a newer release.

hi Richard F. Davidson's
thank for the reply ,
the fedora core 2 was installed by the earlier linux administrator, even i am having the same idea of yours to migrate to newer version, i have rescued the secondry hdd that was there in the server , now i want to rescue the primary hdd (/root) but i am not able to mount it on another linux sytem of same version(we have 2 linxu servers) now can u tell how can i recover the root or resuce the the linux server to its old stable state. and i if am going for a newer version of the linux the which version of fedora core do u recomend or should i go for the RHEL 4 version. again thank u for the reply.

from
nesargha

As I said, it all depends:

If /root was in a logical volume, then you have probably destroyed the HD by running fsck on /dev/hda, and you will not be able to recover anything.
If /root was not in a logical volume, fsck should have done as much as possible to recover the file system.

So, what happened when you put the HD into your other Linux system?
Did you see anything when you ran fdisk -l ?

If you didn't see anything, then I'd suspect that the whole drive is unusable.

If you did see anything, what did pvscan tell you? If it shows the same LV names for /dev/hda and /dev/hd?, then your problem with mounting it on your second system is that LVM can't mount two LVs with the same name.

Suggestion: Look here for a list of some recovery tools you might find helpful.

hi PTrenholme ,
sorry i thought that the quote was writen by u and addressed to u as hi Richard F. Davidson's, yeah i have done the fsck on the primary now,and i was able to recover the data , the thing is it has gone into lost+found directory , each data again in some number format like #1525230 , the pain is that i have to grep each of the folders which i need to take back up and then retrive the data, no doubt linux / unix is great and stable, it takes lot of man hours to do that. and it teach my how to be in patience while working with the system. so if you happen to know any eassier idea to retrive the data from the rescued data.

Well, the easiest way is, of course, to have good, current, backups.

If you've got the resources available, consider a RAID system.

At a minimum, make sure you're using a journalizing file system -- ext3 works well.

If fsck put the files into lost and found, then what happened was that the directory file was "hit" in your crash. So the names were lost. Not much you can do at this point but what you're doing. Sorry.

hi PTrenholme,
thank you for the reply.
we had backup up of some of the data and since the secondary (hdb) had not affected by this, our loss was of the configuration files and some of the recent data on the primary hdb, some how we have recovered about 40% of the primary (hda) using the "grep" command to find out the directory where the fsck command has put the data and we are making up the lost data by the backup data.

any way i was a good experience coming by the "hard way", now we have put the backup server into the operation and i am thinking of implementing RAID using SATA hard disk, i am right now trying it on a red hat 9 server, since i couldnt get hold of a fedora core 4 or 5 version CDs,

can u tell how is fedora core 5, my only worry is that since it is development distro, lot of patches has to be loaded and when a newer version comes in a year or so we would be forced to upgrade to higher version 6 or 7 what ever it comes and we have to go through the process again and for a small organization like ours it too much overhead.