Hey Everyone!
I have just spent the past three hours trying to figure out how to undo some damage I may have caused to my EXT3 file system. I am running SME 6.0.1 with a spare mounted data drive. No problems for the past few months. Great Distro, by the way!
This is what I was trying to do:
(1) Change directory to “/mnt/160gig/stuff” in Midnight Commander. It reported an error “Can’t read ‘/mnt/160gig/stuff/whatever’” but proceeded normally by showing the rest of the directories under “./stuff”. Even "du –hs" had errors reading some directories.
(2) So I looked this up on Google to learn what the problem could be. I also read the man page on fsck.ext3
(3) I unmounded the spare drive, and ran the command as root “nohup fsck.ext3 -pyvf /dev/hde3” (-f: force –p: repair –y: assume yes to all questions, -v: verbose) and let it run for a while.
(4) tail nohup.out:
Illegal block #9 (1078196562) in inode 16777067. CLEARED.
Illegal block #10 (1076175792) in inode 16777067. CLEARED.
Illegal block #-1 (607695078) in inode 16777067. CLEARED.
Illegal block #7240 (2905473982) in inode 16777067. CLEARED.
Illegal block #7241 (45696543) in inode 16777067. CLEARED.
Too many illegal blocks in inode 16777067.
Clear inode? yes
Restarting e2fsck from the beginning...
fsck.ext3: A block group is missing an inode table while checking ext3 journal for /1
(5) I started to sweat by now
(6) #mount /mnt/160gig/
mount: wrong fs type, bad option, bad superblock on /dev/hde3,
or too many mounted file systems.
(7) Back to Google for more answers!
#e2fsck /dev/hde3
e2fsck 1.27 (8-Mar-2002)
e2fsck: A block group is missing an inode table while checking ext3 journal for /1
...so now I am stuck and I don't know what to do next. Do I have to rebuild the inode table?
Thanks for any help!
--Brad D.
1 Replies
5443 Views