On 2002-10-29 09:51, Chuck Yerkes <chuck+nbsd@2002.snew.com> wrote:
> Quoting Richard Rauch (rauch@rice.edu):
> > When I asked about FFS fragmentation in the past, I was assured
> > that FFS doesn't substantially suffer from fragmentation-related
> > performance problems.  (Due, apparently, to the scattering of
> > files on the disk.)
>
> And yet a client just spent a very expensive night doing overnight
> emergency consulting when his 87% full disk wouldn't write any more
> files due to excessive fragmentation.  It's rare and had he used
> VxFS as we recommended (solaris), he wouldn't have had the problem.

You don't need to use VxFS for this sort of thing.  The maximum number
of fragments before the filesystem switches from time-optimization to
space-optimization is determined by the "reserved percentage" AFAIK.
You can either try to reduce fragmentation by dump / newfs with some
different block:fragment ratio / restore, or temporarily change the
amount that fragments can occupy before space optimization kicks in by
tweaking the reserved percentage value.