Hi,

On Tue, Oct 31, 2000 at 10:18:12AM -0500, Chris Mason wrote:

> > "partially done" means the same as "included in the log" as far as
> > disk updates are concerned, but yes, for writes you can't necessarily
> > make the same guarantees.  In ext3's case, that implies that writes
> > larger than a certain size are just not guaranteed to be atomic: they
> > may be broken into smaller chunks arbitrarily, but each such chunk
> > will itself be atomic and will result in a consistent filesystem.
> > 
> 
> Yes, I was less than clear ;-)  I meant that a crash could result in a
> fully consistent filesystem with file writes that are not reflected in the
> intermezzo log.

Not important for InterMezzo --- InterMezzo only propagates changes
onn a whole file basis.  If one node is modifying a file, then by
definition it has a lock and is the only owner of the file.  There is
no state held by InterMezzo about which parts of the file may have
changed --- as long as (a) the initial file creation gets logged, and
(b) the inode timestamp is updated on any write, InterMezzo should be
quite happy.

Peter, let me know if I've got any of this wrong, but I think
InterMezzo should still be safe in the presence of large, non-atomic
writes to cached files.

> This is not a huge issue, but it gets more interesting
> when you turn off data logging (and the ext3 code that flushes new blocks
> before transaction commit).

That flush code actually improves performance by eliminating thrashing
between the commit code and the VM, in measurements so far, so
typically it's not going to hurt if you don't turn it off!  

The biggest problem with turning off data-journaling is that the inode
modification timestamp would no longer be guaranteed to reflect any
data writes.  

Cheers,
 Stephen
-
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to [EMAIL PROTECTED]

Reply via email to