On Tue, 2006-06-13 at 20:20 +0000, techlists@xxxxxxxxxxx wrote: > I had the electricity go out the other day. When my Centos 4.3 workstation came up, I said yes when it got to the prompt "Unclean shutdown, force filesystem check?" prompt. > > It ground away for awhile and then said something about a bad superblock. Yikes! I thought, that's a bad sign. > > After a reboot, I got nothing but a grub> prompt. > > I tried booting into rescue mode using the install CD, and got a message "No valid partitions found". I tried to mount /dev/hda, and got a message "no /etc/fstab found". > > I tried running the install CD, and went into Disk Druid to see if it could see any existing partitions on the drive, and got the message "Partition table corrupted, must re-initialize disk, and reformat". Not a good sign. I backed out at that point, not wanting to go beyond a point of no return. > > I have about 4 years of archived e-mails, and some other misc stuff on that drive that I don't want to lose. > > Anyone have any suggestions what steps to take next? The critical Q: do you *know* exactly what the partition settings were before? If so, there is certainly hope. Manual partitioning can be done using plain old fdisk or sfdisk. That gives some *possibilities*. I won't bore you with details yet. I have recovered from this situation with no special software (test and live) but I knew the layouts. I'll be alert for your message. Maybe there's some software that can help? Back in days of yore, I wrote some "C" that looked at some 3B FDs and HDs and detected the start of partition(s) (found super blocks, free inode lists, etc.). Maybe something like that in the FS world? > <snip sig stuff> > Paul HTH -- Bill
Attachment:
signature.asc
Description: This is a digitally signed message part
_______________________________________________ CentOS mailing list CentOS@xxxxxxxxxx http://lists.centos.org/mailman/listinfo/centos
