[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: root fs type in fstab



Stephen writes:
> On Tue, Jan 23, 2001 at 02:08:54PM -0700, Andreas Dilger wrote:
> > Stephen, you write:
> > > One thing is that since 0.0.5d, ext3 has been completely resetting the
> > > journal if it finds that NEEDS_RECOVERY is not clear.  Is e2fsprogs
> > > doing the same in fsck?
> > 
> > I was just checking e2fsck, and it appears it doesn't do a journal reset
> > unless it actually finds a corrupt journal superblock. 
> 
> It should do --- if it's going to make any changes to the fs, we
> _have_ to prevent any future accidental replay.  The kernel recovery
> code now has the ability to do a single pass over the journal to check
> on valid sequence numbers, so we can reset the journal to a value we
> know will never cause accidental replay.  It should be easy enough to
> do the same in e2fsck, given that recovery.c is designed to be
> compiled in both environments.

I guess it depends on what you mean by "journal reset".  The e2fsck code
does the journal recovery and it saves the last sequence number into
the journal superblock, and sets s_start = 0.  When I was talking about
"a journal reset", I meant that it is reinitializing all of the journal
fields, like a brand new journal.  It doesn't do any detection of the
last sequence number in this case, however.

> > It also aborts recovery (without any chance to correct) if there are
> > journal features which it doesn't understand.
> 
> Right, that's probably worth asking the user about.

I agree.  I'll have a look at that and send a patch to Ted.

Cheers, Andreas
-- 
Andreas Dilger  \ "If a man ate a pound of pasta and a pound of antipasto,
                 \  would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/               -- Dogbert





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]