On Tue, Sep 27, 2005 at 07:35:47PM +0100, Dieter wrote:
> > You mean fsck background checking ? That is not possible in NetBSD and
> > is considered evil and dangerous.
> 
> Perhaps you could explain *why* background fsck is evil and dangerous,
> and what, exactly, Kirk and the gang got wrong?

My personal Point of View is that it's the wrong way to solve the
problem.

The problem that is trying to be solved is that "with large file
systems these days, fsck on reboot after an unclean filesystem
shutdown will take too long to run".

Background fsck addresses that by relying on soft updates to do this
more quickly, a kind of just-in-time fsck.

The other major solution for this problem is to use journalling, which
brings other benefits to the mix.

So, from a 10000 foot level, I see it as a question of "cleaning up
the pieces faster" with background fsck, as opposed to "not breaking
things in the first place" with journalling.

Please note here that I'm not advocating doing away with fsck
completely.

Does anyone know how well Scott Long's SoC project to add journalling
to ffs went?

Regards,
Alistair