So after Bill's experimental fixes got my raid1 partition performing
correctly, I started to feel better about my 1.5 machine.

I woke up this morning to a "panic: dup ialloc" (from memory) and
after waiting an hour to rebuild the raid parity, I did an fsck at
which point the filesystem commenced eating itself with "excessive dup's"
and other spooge.

Fortunately, I did an 'fsck -n' first to see how bad it was going to
be. I then failed one of my raid components and then did the fsck to
watch it eat itself.  I should theoretically be able to boot off the
other component and recover most of my data.

I don't think Bill's fix had anything to do with it.  However, I don't
know what it was.

thinking about going back to 1.4.3 or installing Solaris.  Definitely
not having a good time here.