Cables for running in duplex mode

I'm about to get my hands on a dual channel 3200 controller, do these
specifically need a point-to-point 68pin cable or will they run with a
standard terminated or unterminated cable?

I'm guessing the cage terminates internally, so I would assume a cable
with a terminator is a no-no.

Regards,
Jason.

---
Replace nntp with my name to reply.
N0 5pAm H3r3: Include this tagline to pass my spam filter.

"Jason Arthurs" wrote in message
...
I'm about to get my hands on a dual channel 3200 controller, do these
specifically need a point-to-point 68pin cable or will they run with a
standard terminated or unterminated cable?

I'm guessing the cage terminates internally, so I would assume a cable
with a terminator is a no-no.

Regards,
Jason.


Uhm, Im not sure what duplexing has to do with it ...

but generally, yes. The 3200 was designed to be used with Proliant servers
and external Storage Subsystems. Its connected externally via 68pin-68pin
cable (point to point, as you say).

The termination is implimented via the storage systems.

Let me know if you have any additonal questions,

- LC

This is one of the things about the 3200 that I found
mystifying...though perhaps Nutcracker is used to this as being a
normal function of this card.

First of all, the 3000 has two drive cages...each one capable of
holding four drives. If the cage is wired in duplex mode, both trays
are connected and we are able to insert a total of seven drives (4 on
bottom, 3 on top). If the cage is wired in simplex mode, then we can
utilize all eight drives.

If I install one of my other cards, a 2DH for example, it provides me
with two internal channels...two seperate 68 pin connectors. If I cable
this over two the drive cage (in simplex mode) using two seperate
straight-through ribbon cables connected to the two connectors on the
back of the cage, the Array Configuration Utility gives me the option
of setting up two seperate arrays...with as many as four drives each.

But when the 3200 is installed and cabled EXACTLY as described above,
the ACU now allows me to set up a SINGLE array consisting of ALL EIGHT
drives! The seven drive limit has disappeared.

We can all agree that this aspect of the 3200 is far superior to the
previous Array Controllers...even BEFORE we talk about the greater
cache size, etc.

There is also a question about the long four-connector terminated cable
that is attached to the backplane behind the drive cage in the 3000.
Why would Compaq throw this spare cable in, if it was never to be used?
I can only guess that it is intended for us to use in connecting a
series of devices mounted in the 5.25" slots (like the one the CD drive
is mounted in). One of my 3000's, for example, has a 15/30 tape drive
mounted and cabled to the onboard controller, using this long
terminated cable.

I hope this all makes sense, I seem to be rambling.

Jeff

On Mon, 22 Aug 2005 10:40:47 -0500, "Nut Cracker"
wrote:


"Jason Arthurs" wrote in message
.. .
I'm about to get my hands on a dual channel 3200 controller, do these
specifically need a point-to-point 68pin cable or will they run with a
standard terminated or unterminated cable?

I'm guessing the cage terminates internally, so I would assume a cable
with a terminator is a no-no.

Regards,
Jason.


Uhm, Im not sure what duplexing has to do with it ...

but generally, yes. The 3200 was designed to be used with Proliant servers
and external Storage Subsystems. Its connected externally via 68pin-68pin
cable (point to point, as you say).

The termination is implimented via the storage systems.

Let me know if you have any additonal questions,

I think my original question wasn't clear enough. I currently have a
Smart 2SL single channel card and the bridging board installed between
the A & B cages on my 3000, this has the maximum of seven drives
installed. When I install the 3200 to go dual channel and install the
eighth drive, I will need a second SCSI cable for the second channel
(once I remove the bridging board on the rear of the cage). Now the
only spare cable I currently have is a terminated cable, if I connect
up cage A to the original cable will a terminated cable work for cage
B. Or do I need to buy an unterminated cable?

Regards,
Jason.

---
Replace nntp with my name to reply.
N0 5pAm H3r3: Include this tagline to pass my spam filter.

Sorry for dancing around the answer, Jason.

A nearby Compaq buddy and I often argue about "what is right" in terms
of SCSI cabling. He always seems to have a fear that the SCSI bus may
appear to be functioning properly now, but will drop a single bit
sometime in the unforeseen future. I do not share his obvious paranoia.

I have used the mutlti-connector cable to do just what you are saying,
by using the first available connector on the drive cage, and moving
the terminator to the second (next one up) connector on the cable. This
effectively terminates the cable as it should be. Of course, if you get
a straight through cable, you will make my paranoid friend happier.

I have a handful of the single connector straight-througfh cables, and
would be happy to send you one. They will probably be easy enough to
find locally, though.

Jeff