superblock is corrupt after attempted resize & move of ext3 partition
Hi All,
I could really use some help here (I hope I am posting in correct place), I have made my /home partition inaccessible :(
Problem started when I used gparted to resize and move /dev/sda3 (where all my personal data is). This partition is about 250G and the process was going to take some hours to complete so I went to bed before it finished. When I checked it in the morning my pc had shut itself down, so I rebooted and started up gparted again. Looking at the partition table in gparted I noticed that /dev/sda6 had been resized, however it hadn't been moved. I tried to move it and the following error came up...
Code:
GParted 0.5.2
Libparted 2.2
Check and repair file system (ext3) on /dev/sda3 00:00:01 ( ERROR )
calibrate /dev/sda3 00:00:01 ( SUCCESS )
path: /dev/sda3
start: 39054829
end: 474914343
size: 435859515 (207.83 GiB)
check file system on /dev/sda3 for errors and (if possible) fix them 00:00:00 ( ERROR )
e2fsck -f -y -v /dev/sda3
e2fsck: Group descriptors look bad... trying backup blocks...
e2fsck: going back to original superblock
Filesystem mounted or opened exclusively by another program?
e2fsck 1.41.11 (14-Mar-2010)
e2fsck: The ext2 superblock is corrupt when using the backup blocks
e2fsck: Device or resource busy while trying to open /dev/sda3
This is the dmesg output...
Code:
karlos@ubuntu-10:~$ dmesg|tail
[ 4588.270555] EXT3 FS on sda6, internal journal
[ 4588.270565] EXT3-fs: mounted filesystem with ordered data mode.
[ 4669.197392] kjournald starting. Commit interval 5 seconds
[ 4669.197633] EXT3 FS on sda6, internal journal
[ 4669.197643] EXT3-fs: mounted filesystem with ordered data mode.
[ 6401.853257] kjournald starting. Commit interval 5 seconds
[ 6401.853573] EXT3 FS on sda6, internal journal
[ 6401.853584] EXT3-fs: mounted filesystem with ordered data mode.
[ 7895.924569] EXT3-fs error (device sda3): ext3_check_descriptors: Block bitmap for group 1 not in group (block 0)!
[ 7895.925081] EXT3-fs: group descriptors corrupted!
Then I did some playing around with the partition table, deleting partitions and then rescuing them with parted. This resulted in sda3 becoming sda1. To continue
## backing up sda1
Code:
sudo dd if=/dev/sda1 of=/home/karlos/backup/backup-sda1.img
435859512+0 records in
435859512+0 records out
223160070144 bytes (223 GB) copied, 3623.43 s, 61.6 MB/s
## then after trying a number of different superblocks.. (found using mke2fs -n /dev/sda1)
Code:
karlos@ubuntu-10:~$ sudo e2fsck -f -b 163840 /dev/sda1
e2fsck 1.41.11 (14-Mar-2010)
Resize inode not valid. Recreate<y>? yes
Pass 1: Checking inodes, blocks, and sizes
Inode 161, i_blocks is 3440, should be 0. Fix<y>? yes
Inode 417, i_blocks is 48, should be 0. Fix<y>? yes
Inode 32769 is a unknown file type with mode 00 but it looks like it is really a directory.
Fix<y>? yes
Inode 98337, i_blocks is 16, should be 0. Fix<y>? yes
Inode 98338, i_blocks is 72, should be 0. Fix<y>? yes
Inode 114849 is a unknown file type with mode 00 but it looks like it is really a directory.
Fix<y>? yes
Inode 147489, i_blocks is 8, should be 0. Fix<y>? yes
Inode 147490 is a unknown file type with mode 00 but it looks like it is really a directory.
Fix<y>? yes
Inode 180225 is a unknown file type with mode 00 but it looks like it is really a directory.
Fix<y>? yes
Inode 180225, i_size is 0, should be 4096. Fix<y>? yes
Inode 213025, i_blocks is 4168, should be 0. Fix<y>? yes
Inode 213026, i_blocks is 4240, should be 0. Fix<y>? yes
Inode 213027, i_blocks is 3368, should be 0. Fix<y>? yes
Inode 213028, i_blocks is 4448, should be 0. Fix<y>? yes
Inode 213029, i_blocks is 3512, should be 0. Fix<y>? yes
Special (device/socket/fifo) inode 2116299 has non-zero size. Fix<y>?
.....
Inode 5179684 has illegal block(s). Clear<y>? yes
Illegal block #0 (4040140531) in inode 5179684. CLEARED.
Illegal block #1 (3902218074) in inode 5179684. CLEARED.
Illegal block #2 (771429886) in inode 5179684. CLEARED.
Illegal block #3 (935633403) in inode 5179684. CLEARED.
Illegal block #4 (1327107087) in inode 5179684. CLEARED.
Illegal block #5 (3680397508) in inode 5179684. CLEARED.
Illegal block #6 (1350728668) in inode 5179684. CLEARED.
Illegal block #7 (846466656) in inode 5179684. CLEARED.
Illegal block #8 (3803158996) in inode 5179684. CLEARED.
Illegal block #9 (346235357) in inode 5179684. CLEARED.
Illegal block #10 (2055782526) in inode 5179684. CLEARED.
Too many illegal blocks in inode 5179684.
Clear inode<y>?
......
Illegal block #135267340 (1295874896) in inode 5179471. CLEARED.
Error storing directory block information (inode=5179471, block=0, num=126821203): Memory allocation failed
e2fsck: aborted
### try again...
Code:
root@PartedMagic:~# e2fsck -f -b 163840 /dev/sda1
e2fsck 1.41.11 (14-Mar-2010)
Resize inode not valid. Recreate<y>? yes
Pass 1: Checking inodes, blocks, and sizes
Inode 5179471 has illegal block(s). Clear<y>? yes
Illegal block #135267340 (1295874896) in inode 5179471. CLEARED.
Error storing directory block information (inode=5179471, block=0, num=126821236): Memory allocation failed
e2fsck: aborted
So now I am stuck with an error stating "memory allocation failed", does anyone know what this means and how I might work around it? |
