On Wed, Sep 10, 2003 at 10:52:50AM +0200, Norbert Preining wrote: > On Mit, 10 Sep 2003, Norbert Preining wrote: > > Restarting e2fscl from the beginning ... > > Ok, this didn't work at all. I could fix the biggest problems by: > * starting debugfs in write mode > * clearing inode 8 (which I found out later is the journal inode) > * fsck-ing and letting everything be fixed > * reboot, getting ext2 instead of ext3 (deleted journal, mounting as ext2) > * fsck again, fixing may duplicate blocks etc, (uuaaa) > * now it is working (at least what I see, surprises will soon hit me) It sounds like garbage got written into the the inode table. Without a detailed listing of the fsck messages, it's really hard to tell more, though. Note that when reporting problems like this, it's important to specify your hardware (which SCSI/IDE controller, motherboard, etc.) since although the filesystem code noticed the problem, it's at least as likely that the problem was caused a buggy device driver that scribbed to the disk in the wrong place. - Ted _______________________________________________ Ext3-users@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/ext3-users
