Re: A single SCSI disk scanned multiple times
- From: dnichols@xxxxxxxxxxx (DoN. Nichols)
- Date: Sat, 25 Feb 2006 04:21:36 +0000
According to <wcui@xxxxxxxxxx>:
DoN,
Many thanks for your detailed explanation. In fact, this machine is
connected by a SCA board to the SCSI bus,so the SCA board has the SCSI
ID setting for the disks. I think the the SCSI ID setting should be
correct.
O.K. Then there is a chance that there are jumpers on the other
end (of the drive) from the SCA connector. Or perhaps elsewhere in the
SCSI bus.
What is the machine in which the drive is installed? This may
not matter, but again it may. It might have been enough to tell me that
the drive would have an SCA interface.
I don't quite understand the difference between SCSI internal and
external disks, could you explain it?
The drives could well be identical.
However (assuming the 68-pin interface which the Seagage web
page showed me), a drive mounted as an internal one would have jumpers
on the drive selecting the SCSI ID (not normally so on a drive with an
SCA interface, of course).
An external drive (again 68-pin or 50-pin) would be more likely
to have a connector block with a ribbon cable going to a switch on the
back of the external housing to allow selection of the SCSI ID without
having to open the case.
This latter situation is the one in which the block could be
inserted upside down, thus letting the ground bus of the drive connect
all the SCSI-ID pins together, and the ground bus of the connector
connect all of the drive's SCSI-ID pins together, thus making the drive
occupy SCSI-ID 7 (for a 50-pin interface), and SCSI-ID 15 (for a 68-pin
interface). In any case, no matter which SCSI ID is selected by the
computer, the drive will respond as though it had been selected
explicitly.
Now -- I just reached out and picked up a bad 18GB Seagate
Cheetah (ST831203LC), and looking at its label, I see documented both
J5 (which is only present on a 68-pin interface drive), and J6 (which is
present on both types, and which has six pairs of exposed pins (of which
the first four pair can select the SCSI ID), and a remaining four pairs,
covered by a black protective block (to prevent anyone from using them).
If there were anything connected to J6, it could override the
SCSI ID selected by the SCA interface.
Or -- if there is a block connector on the board carrying the
SCA connectors into which the drives plug, and if that has been removed
from the system, it is possible that a connector on that is reversed,
with the same effect as is found in the case of the reversed connector
on a drive in an external housing.
Or -- if a tape drive, or a CD-ROM drive, or anything else is in
an external case with a SCSI selection switch on the back, it could be
tying the SCSI selection pins together -- but that would mean that *that*
drive, instead of the disk drive, would appear at all SCSI IDs. (If it
were a typical 50-pin CD-ROM or tape drive, it would only be in a
position to tie the first three of the four SCSI ID pins together, so
only SCSI IDs 0-7 would appear as one. IIRC, your description had it
responding to all 16 addresses.
So -- examine the disk drive for any jumpers or connectors on
J6.
Try the current disk drive in the other SCA drive connector
(most Sun boxen which have internal SCA drives have sockets for two,
which normally appear at different SCSI IDs.) The older ones put the
internal drives at SCIDs 3 and 1 (in that order), while the newer ones
put the internal drives at SCSI IDs 0 and 1 (again, in that order).
And -- try a different drive (even a very small one) in the same
socket that the troublesome drive is in, and see what that does (with
the troublesome drive missing.)
Of course -- it *could* be problems in the SCSI host adaptor in
the CPU board or motherboard.
I hope that this leads you to identifying the source of the
problem.
Best of luck,
DoN.
--
Email: <dnichols@xxxxxxxxxxx> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---
.
- References:
- A single SCSI disk scanned multiple times
- From: wcui
- Re: A single SCSI disk scanned multiple times
- From: DoN. Nichols
- Re: A single SCSI disk scanned multiple times
- From: wcui
- A single SCSI disk scanned multiple times
- Prev by Date: Re: 6 bay scsi enclosure. overheating?
- Next by Date: Re: 6 bay scsi enclosure. overheating?
- Previous by thread: Re: A single SCSI disk scanned multiple times
- Next by thread: E3500 FC problem
- Index(es):
Relevant Pages
|
