RE: sparse loop devices and ext3

["Peter J. Braam" <braam mountainviewdata com>]

Hi,

The problem I saw was that the allocation of blocks was not handled
all that well and seriously rolled back during recovery.

So shouldn't bmap in ext3 start a transaction to eliminate that?

I agree that journaling the data in the write is generally going
overboard.

- Peter -

 > -----Original Message-----
 > From: ext3-users-admin redhat com
 > [mailto:ext3-users-admin redhat com]On
 > Behalf Of Stephen C. Tweedie
 > Sent: Monday, February 05, 2001 12:01 PM
 > To: ext3-users redhat com
 > Cc: Trevor Hughes
 > Subject: Re: sparse loop devices and ext3
 >
 >
 > Hi,
 >
 > On Mon, Feb 05, 2001 at 10:02:27AM -0800, Peter J. Braam wrote:
 > >
 > > I noticed that using a sparse file as a loop device
 > leads to non
 > > journaled block allocation in Ext3.
 >
 > You mean running an ext3 sparse file as a loop device,
 > or running ext3
 > on top of a loop device?
 >
 > If the former, it's probably because loop doesn't use
 > the filesystem
 > calls for writing to the file --- it bmap()s the block
 > and writes it
 > directly.  It _is_ possible to get loop to use write,
 > but it will
 > always use bmap if that is available (which is the
 > wrong thing to do
 > for ext3 if you want journaled behaviour).
 >
 > > Is this something that should be fixed?
 >
 > Maybe, but probably not --- it would slow down the loop
 > device a great
 > deal, and it's not clear that we need data journaling
 > in this case.
 >
 > Cheers,
 >  Stephen
 >
 >
 >
 > _______________________________________________
 > Ext3-users mailing list
 > Ext3-users redhat com
 > https://listman.redhat.com/mailman/listinfo/ext3-users
 >