>>> Dec 10 23:59:09 acheron /netbsd: sd0(aic0:1:0):  Check Condition on CDB: 0x2a 00 00 69 05 7e 00 00 01 00
>>> Dec 10 23:59:09 acheron /netbsd:     SENSE KEY:  Illegal Request
>>> Dec 10 23:59:09 acheron /netbsd:      ASC/ASCQ:  Logical Block Address Out of Range
>>   To me this looks more like a size problem [...]
> This happened right at the very beginning of the newfs, on a
> partition that, while it did run to the end of the disk, didn't quite
> start at the beginning of the disk.  The only possibility for overrun
> would be on the "far end."  Does newfs do anything down there before
> proceeding from the beginning?  It's possible I did something funky
> with the partition size...

Yes, it does.  Quoting from sbin/newfs/mkfs.c:

        /*
         * Validate the given file system size.
         * Verify that its last block can actually be accessed.
         */
        if (fssize <= 0)
                printf("preposterous size %d\n", fssize), exit(13);
        wtfs(fssize - 1, sectorsize, (char *)&sblock);

The CDB 0x2a 00 00 69 05 7e 00 00 01 00 is a 10-byte write operation,
attempting to write 0x0001 sectors at address 0x69057e (6882686
decimal).  This is at approximately the 3.28G point from the beginning
of the disk.  Since you say the drive is "a two gig Quantum Fireball",
this is almost certainly just due to a partition table that places that
partition off the end of the disk.

					der Mouse

			       mouse@rodents.montreal.qc.ca
		     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B