Can data be recovered? md0 bad superblock?
Greetings!
Sitting on my bench are two 80 gig drives from a Fedora Core 2.6.10-1.771_FC2 machine. When I try to boot from these drives I get: Code:
JBD: IO error reading journal superblock Code:
[root@qpak /]# mount -t ext3 /dev/md0 /mnt/hda/ The disks are partitioned like this: Code:
Disk /dev/hda: 80.0 GB, 80026361856 bytes Code:
[root@qpak /]# cat /proc/mdstat Any ideas? Thanks! |
Try using the -sb option for mount or the -b option for fsck and indicate an alternate superblock such as 8193.
You could also try the mount option "errors=remount-ro" It seems that the md device is OK but the ext3 filesystem on it isn't. Good Luck! |
Thanks for the quick reply!
I'm not too clear on the -sb option for mount. This is what I get: Code:
[root@qpak mnt]# mount -t ext3 -sb 294912 /dev/md0 /mnt/hda/ Code:
[root@qpak mnt]# e2fsck -b 294912 /dev/md0 Code:
[root@qpak mnt]# mount -t ext3 -o errors=remount-ro /dev/md0 /mnt/hda/ |
What does "sudo file -s /dev/md0" say.
Do you get an output from "sudo /sbin/tune2fs /dev/md0". Look at the "blocks per group". That should indicate what the spare superblocks are. Your mount command was wrong. mount -t ext3 /dev/md0 /mnt/hda/ -o sb=<sparesuperblock> try: mount -t ext3 /dev/md0 /mnt/hda/ -o sb=8193 I just did a bit of a test/demo where I created an image of an ext3 filesystem to play with. Code:
tune2fs -l test.iso For the mount command I was able to mount it with the command: mount -t ext3 test.iso testdirectory/ -o loop,sb=8193 ( I like to try things out myself, if possible as I post a reply) I seem to remember another command that examined the filesystem and listed the spare superblocks, but I don't remember what it was, or even if I was looking at an ext3 filesystem. |
Here are some results:
Code:
[root@qpak mnt]# sudo file -s /dev/md0 It appears that the RAID 0 is approx 157Gb of which 867Mb are in use. Too bad they didn't set up a RAID 1, they didn't need the space. Thanks for your help! |
Try "/sbin/dumpe2fs /dev/md0"
Look at the parts that look like this: Code:
Group 1: (Blocks 8193-9999) Code:
Group 9: (Blocks 294912-327679) Code:
echo 'scale=3; 38551936*4096/1024/1024/1024' | bc |
OK,
The locations of backup superblocks are the same as the locations shown in "mke2fs -n /dev/md0". The output of "/sbin/dumpe2fs /dev/md0" is too big to post here. I sent the output to a file and here are some parts. It starts like this: Code:
Filesystem volume name: <none> Ends like this: Code:
Group 1176: (Blocks 38535168-38551935) Thanks for your help! |
I was afraid that the assumptions made by mk2fs -n would be wrong, and supplied the dumpe2fs info to provide the actual backup superblocks instead of assumed ones.
Try each of them in the options to mount or fsck.ext3. This link may help you recover data if you can't get the filesystem repaired: http://www.geocities.com/CapeCanaver.../7731/lde.html Also check for Live distro's with tools to repair disks or recover files. You also might consider making an image copy of the drive before trying to modify it. |
All times are GMT -5. The time now is 04:06 AM. |