>Also, aren't modern SCSI drives supposed to automatically remap bad blocks??

In some drives you have to specifically enable it, but in all drives, it
happens only if the drive knows what data to deposit in the mapped block;
i.e. if a write error occurs, or if the data is successfully read after
more recovery attempts than the drive believes is worthwhile (i.e. some
drives will make half-hearted recovery attempts, and if that fails, will
go to extreme lengths to recover the data, hoping to map it to a new
location if recovered).

> Also, the errors that I'm seeing a quite random, wouldn't a head crash tend
> to effect blocks in one specific location on the disk??  I'm getting at
> least 20 bad blocks, wouldn't at least a couple of them be contiguous, or
> atleast near each other.

You may have damaged only a couple of sectors on the disk, but a bent head
can no longer write data reliably, spreading corruption and disaster
wherever it goes.

An outside possibility is that a completely stray write managed to corrupt
drive parameters stored on an invisible block (hence the drive is using
ludicrous values for write compensation etc) but that also is not something
NetBSD should be able to screw up.

> But really, I don't care what's going on (I don't want to assign blame, or
> imply that you made a mistake or anything like that), I just want it not to
> happen.  Am I safe just running fsck and letting it take care of the bad
> blocks when it finds them or should I be doing something more drastic???

This sounds like a failing disk.  You will lose more and more blocks until
it fails entirely, and fsck is not going to help.

When you did the format, what utility did you use?  Did you do a media
certification pass, or just a format?  Format discovers only completely
destroyed sectors; a good media certification will uncover weak sectors
as well.