Subject: AHA 1740 driver: ahb0: host_stat 18
To: None <port-i386@NetBSD.ORG, current-users@NetBSD.ORG>
From: Sean Berry (most of the time) <spberry@iastate.edu>
List: current-users
Date: 11/07/1996 11:24:50
We acquired a nice HP server with 4G of disk and 40M of core.  Problem
being: that 4G of disk is served by a 1740.  The disks are Seagate ST1200Ns
set to run synchronously (four of them) presumably for speed.  (they're
labeled as HP D2076Bs)

It is my belief that this machine previously ran Netware.  It is an HP
Vectra 486/25T, (modular memory and CPU boards), and it would appear the
only way to access the bios is through a special boot floppy.  

The 1740 does have a 'ctrl-a' bios as far as I can determine.  It has only
one block of jumpers, that being a four jumper set, with no jumpers in
place.

The machine doesn't like having an IDE disk attached.  

I was able to put a BSD install onto one of the disks (set to scsi id 0),
and the machine successfully loaded bootblocks, and kernel.  Unfortunately,
after the probe, where it should probe for disks on the controller, I got
the error message: ahb0: host_stat 18.

The machine runs DOS and installed NT fine (although when rebooted, reported
No bootable partition, although DOS tells me there's an active FAT
partition, and there appears to be the correct NTLDR etc).

Is it possible that the 1740 is set to look like a standard IDE (ST506?)
interface at x1f0 and thus isn't acting like a 'real' 1740?

I can throw an Ultrastor 14F in this machine if I need to to get it running,
but it would a: be a shame to throw away a decent card which probably
outperforms the 14F greatly, and b: the ultrastor is mine, not work's.  :)

I welcome any comments, suggestions, crazy ideas, and hunches.

spb (off to search HP for the third time)
--
Sean Berry is an ISTJ cellist who works in netbsd.   (must have more toys)
I imagine someone is likely to misinterpret	    
my opinions as those of my various employers.  This is not the case.