On Sun, Jun 13, 2010 at 3:11 AM, Zane H. Healy <healyzh at aracnet.com> wrote:
What version of RSTS/E, and does it support the RQZX1?
?I think that support
might have come with 10.1.
Yes, I am using 10.1
I'm not sure if anyone has actually attempted what
you're doing, some of us
were discussing this a year or two ago, but I know I've not had time to try
it.
I tried all the obvious combinations for imaging the hard disk, the
final one being:
SIMH using the SCSI drive as a block device (Linux: attach rq
/dev/sdc) and confirming that SIMH boots the drive successfully after
the install, however on re-attaching to the 11/93 it still throws the
Drive not ready error when trying to boot.
Of course you may simply have some sort of hardware
issue going on there as
well, I'm not familiar with the RQZX1.
The drives passes the built-in RQZX1 diagnostics, which includes
read/write/verify tests etc. I formatted the SCSI drive via the RQZX1
too (prior to using the drive on Linux via SIMH to install RSTS/E),
just in case the format process was doing something special (although
I can't imagine what since I understand MSCP masks drive specifics).
A next posible step is to use this source-code for a MSCP boot so I
can catch the initial boot load and see what is in the first disk
block (containing the next level bootstrap):
http://www.slowdeath.com/AK6DN/PDP-11/M9312/23-767A9/23-767A9.lst
I will also try a different model of SCSI drive too - perhaps the
Quantum Fireball is too much of a good thing (4200RPM 2.1GB); I have
an old clunker of a Quantum ProDrive 80S (3600 RPM 80MB!)
Does anyone know how SCSI ID's are mapped to MSCP LUNs? so far the
RQZX1 appears to map the first SCSI ID (in my case #4) to LUN 0 (which
is what I want) - I am assuming it just does them in sequence.