>> Actually, a more general tool which defragments and re-optimizes ufs
>> filesystems might be more useful in that regard.

> Have you (or someone else) actually measured the "fragmentation" of
> real filesystems?

> Writing a defragmenter that doesn't work with O(n^2) running time is
> hard and may be slower then dump/newfs/restore.

Of course, it may be worth waiting, if you don't have anywhere to put
the backup while you're newfsing.

> Or a tool that gives the names of the handful of fragmented files.

Yeah.  As I understand it, the code is good enough that fragmentation
is not a problem _if_ you allow it the 5%-10% free space it wants and
keep turning files over, creating and removing them.

A defragger/reoptimizer would primarily be useful for filesystems (like
/export/swap) which do not have the free space and constant turnover
(ie, are very full and/or more or less frozen) and you want good
performance out of.

					der Mouse

			    mouse@collatz.mcrcim.mcgill.edu