Hardware Support Account wrote:
Actually upon looking at it again, it looks like the drive is not set
in either the SCSI BIOS to be bootable or the BIOS is not set to boot
from the SCSI card. Not sure how it sounds like a geometry issue,
however its midnight and I should go to bed soon >
I was referring to the situation where the driver’s notion of heads,
track and cylinders does not match BIOS/partition table numbers.
I have never been more than fuzzy about this; there is an fsys -h
option for SCSI drives, and for IDE drives an eide -b option as well.
I have had this error message with IDE drives when the numbers were
wrong, and I seem to recall the same (long ago) when, say, the
“greater than 1KB” support was mis-checked on a SCSI drive.
E.
Richard R. Kramer <> rrkramer@kramer-smilko.com> > wrote:
With respect, Erick, doesn’t this sound more like a geometry
mismatch?
Richard
Hardware Support Account wrote:
Pass the device ID and vendor ID to the driver and it should be detected
properly. This is a known problem and an updated driver has been done
and should make the next patch.
Erick.
Shawn McNamara <> smcnamara@toptech.com> > wrote:
I’ve also had this same problem in getting QNX to recognize the SCSI when
trying to install the OS. I used the parameters you gave for the
Fsys.aha8scsi driver. Got QNX loaded fine. However, upon a restart, QNX
tries to boot to the Partition and it reports “Missing Operating System”.
Any suggestions on what it might be happening here? No matter what I try, it
doesn’t seem to be able to find the hard drive on startup (thought I can go
back with a boot disk and get it loaded back up through the Fsys.aha8scsi
line and a few mount commands).
“Richard R. Kramer” <> rrkramer@kramer-smilko.com> > wrote in message
news:> 3BD89182.9E47A5A1@kramer-smilko.com> …
OK, the following works:
Fsys.aha8scsi aha8scsi -V 9005 -D 80 &
Notice the spaces after “V” and “D”. Not life threatening, but
not elegant, either. Without the spaces, the driver reports parsing
errors and quits. As I recall, Fsys.sym8scsi behaves the same way.
Richard
Hardware Support Account wrote:
Hi Richard,
Richard R. Kramer <> rrkramer@kramer-smilko.com> > wrote:
Qdn says Fsys.aha8scsi should work with an Adaptec 29160N, however
show_pci reports the vendor ID is 9005 - not the traditional 9004.
The 4.24Q version of Fsys.aha8scsi we have did not find the card in
the first trials, and we were obliged to quit (more testing tonight).
In the mean time, can anyone provide suggestions as to how it can
be made to work? We will try various combos of chipset, vendor ID
and device ID; other hints are welcome.
Not sure if this means you tried passing the vendor ID to the driver
yet or not. If you do so, also pass the device ID.
Erick.
TIA
Richard