I've run some more tests but not very scientifically -- Without mounting
my second drive it doesn't seem to crash as frequently but still does
eventually.  I seem to be able to crash it by running a Makefile that
calls ar(1) -- No reason why it's that specific program, but perhaps it
does a sequence of I/O events that aggrivate the problem.  I haven't had
a chance to disassemble my storage pack to see what happens if I
completely made my second drive 'disappear'...

BTW, is there a key sequence like the C-A-D and S-A sequences on
NetBSD/i386 and NetBSD/Sparc, respectively, to drop to the kernel
debugger?

Thanks,
-bc