>> I have a 1gb seagate scsi hd in my decstation 5000/120 running 1.4.1, and
>> /usr (/dev/rz3d) has a bad block at 2102406.
>>
>> [[ scsictl problems deleted ]]
>
> The problem is that scsictl using the MI SCSI layer, and 1.4.1 on the
> pmax (or even -current) doesn't support MI SCSI yet.
>
> Prehaps the easiest short-term solution is that if the block is in a
> known file is to move that file somewhere safe and make it mode 000.
>
> Another option is to try Tohru Nishimura's test MI SCSI kernel at
> ftp.aist-nara.ac.jp:~ftp/pub/personal/nisimura/newasc.tgz.  This isn't
> production code yet though...

I have updated the MI SCSI patchwork to reflect recent changes in
-current.  SCSI devices are now probed after configure() process has
well done, so debugging statements I placed in autoconf.c are harmful.
ELF and ECOFF kernel images are included.  No functional improvement,
yet.  I will switch to [thorpej-scsipi] new mi scsi layer soon. 

Tohru Nishimura
Nara Institute of Science and Technology