On Mon, Mar 06, 2006 at 09:44:47PM -0800, Elliott Mitchell wrote:
> I was in a hurry so I might of misread the symptoms I was seeing, but it
> sure looked like e2fsck 1.37-2sarge1 wasn't finding the journal. I didn't
> dare try reseting the field with tune2fs for fear of losing journal
> entries and data damage.
I just built and tried e2fsprogs 1.37-2sarge1, with the external
journal device original set up in /dev/loop0, and the filesystem in
/dev/loop1, and then moving the journal device from /dev/loop0 to
/dev/loop5, and indeed e2fsck was able to find it.
> You'll note that my first message said "...and correct the superblock
> field...", this is the main issue. Given that a filesystem with the
> journal device hint set wrong isn't mountable and tune2fs is dangerous on
> unclean filesystems, e2fsck MUST set the hint!
Actually, it would be _nice_ if e2fsck set the hint, but it's not the
only way to recover from the situation. You can also work around the
problem by using the journal_dev mount option. For example, if the
external journal device is now major 7, minor 5 (/dev/loop5, or
0x0705) and the filesystem is located at /dev/loop1, you can mount the
filesystem using the command:
mount -t ext3 -o journal_dev=0x0705 /dev/loop1
This will update the hint in the superblock.
> Isn't `mount` supposed to be a filesystem independant utility? At which
> point it has no business modifying flags in the superblock of an ext2
> filesystem.
No, it shouldn't be modifying flags in the superblock, but it should
be checking to see if an external journal is required, and then
specifying the journal_dev option if necessary. Mount does have some
filesystem dependent code for various filesystems, in particular
remote filesystems like nfs.
- Ted
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
