HELP: mke2fs done instead of e2fsck :-(

tkmbe · 05-02-2008, 03:35 PM

Hello

Yes I now but ....
I have a raid ext3 fs (/dev/md0) and have done
an mke2fs /dev/md0 instead of e2fsck /dev/md0
The fs was not mounted and nothing more was done i.e.
there was no data overwritten.
How can I recover the cluster structure?

Thanks a lot

Beat

GushpinBob · 05-02-2008, 03:56 PM

What type of RAID level configuration did you have?

tkmbe · 05-02-2008, 04:53 PM

Only raid level 1 :-(
BTW: I have just verified from an old backup that the filesystem was created the same wayso the copies of the superblock should be intact or not?

Output of tune2fs of the orig filesystem was:

PHP Code:



Inode count:              768544 
Reserved block count:     76808 
Block size:               4096 
Fragment size:            4096 
Blocks per group:         32768 
Fragments per group:      32768 
Inodes per group:         16352

Output of the new one created was:

PHP Code:



mke2fs /dev/md0 
mke2fs 1.36 (05-Feb-2005) 
Filesystem label= 
OS type: Linux 
Block size=4096 (log=2) 
Fragment size=4096 (log=2) 
768544 inodes, 1536176 blocks 
76808 blocks (5.00%) reserved for the super user 
First data block=0 
47 block groups 
32768 blocks per group, 32768 fragments per group 
16352 inodes per group 
Superblock backups stored on blocks: 
        32768, 98304, 163840, 229376, 294912, 819200, 884736 
 
Writing inode tables: done 
Writing superblocks and filesystem accounting information: done 
 
This filesystem will be automatically checked every 25 mounts or 
180 days, whichever comes first.  Use tune2fs -c or -i to override.

tkmbe · 05-04-2008, 09:48 AM

Mhhhhh

Where are the linux gurus??
No hints???