Subject: 2 Problems with SCSI drives
To: None <port-hp300@netbsd.org>
From: Sageev T. George <sageev@usc.edu>
List: port-hp300
Date: 01/29/2001 12:19:17
About one week ago, I was able to get NetBSD 1.5 installed
on a 425t.  The system had only one Rodime 200MB hard
drive.  Everything was working great until the drive
would stop working intermittently, which I'm sure isn't
a surprise to anybody subscribed to this mail list.   Initially,
resetting the system was enough, and the drive was
repaired (?) by fsck.  Eventually, however, the drive got worse,
and it looks like kernel file "netbsd" got corrupted.
The system refused to bring me to single-user mode
or normal run levels.

So, I'm going to reinstall.  First problem:  What can
I do to minimize the problems that these deteriorating
Rodime drives present?  Can I do a low-level format?
Even better, is there any way that I can make sure
that any file system creation does a very, very unforgiving
bad block check?  I'm assuming that the file system
creation program can mark blocks as bad so that the
file system won't include them.  Any other ideas?

Second problem:  I'd love to use another SCSI drive,
but SCSI-II drives that are recognized by my 386 with
NetBSD are not recognized by this hp300 system.
The drive I'd like to install is a Seagate ST1480N.
I've really only tried accessing this drive with the
SYS_INST program, which refused to see it.  If
I can get the Rodime up and running, would a running
NetBSD system be able to see the Seagate drive?
Is there something I can do with the boot-PROM
menu to make it see this drive?  I've read in the
archives that pressing the spacebar makes the boot
program scan for the SCSI drives, but I wasn't
sure if that was correct.

Thanks
Sageev