[FIXED]CS SCSI - Problems... [/FIXED]

[arnljot]

I have this CD ROM:
Manual for Plextor PX-40TSi

According to this, this CD ROM Drive has active termination if enabled with jumpers. Termination is documented on page 27 of the pdf.

There is also a seagate drive. Manual here

This drive is 50 pin and was placed after the CDROM. It provides termination. Without it the Amiga will just enter a cycle of indefinate reboots. With it, it will complain that one of the partitions have an write error because the device has an illegal state.

Page 37 of the PDF holds an overview of the configuration of the drive. I dont get the idea with the two TERMINATION POWER vs. the TERMINATION ENABLE.

Like the drive is now, it has a jumper on one TP and not enabled TE. That's wrong according to the manual...

From the manual:
TE (Applies only to "N" models)
On With the jumper installed, the On-board (non-removable) terminator circuits are enabled
(connected to the I/O lines). Default is jumper installed.
Off Terminator circuits not connected to I/O lines.

Now, this jumper is not enabled on the drive. And when everything boots up, but the drive complaints about an "illegal state" this jumper is not enabled.

From the manual:
TP  TP
Off Off No terminator power is connected to drive terminators or SCSI bus I/O pin 26.
On  Off Drive supplies its own terminator power only. Jumper on this position is factory default.
Off On Drive supplies power to I/O pin 26 of SCSI bus; none to internal terminators. When drives have
differential I/O circuits, a jumper on the right TP position may be needed to power external
terminators (see system documentation). The "ND" drive has differential I/O circuits which
have no terminator circuits on the drive.
On  On Drive supplies terminator power to itself (internal connection) and to I/O pin 26 of SCSI bus.
This is a legal jumper setting.

The drive is currently configured as "On Off" for Term power.

Both of these drives are after a uwscsi-2-fastscsi internal convertor.

On the other side of the cs scsi there is a uwscsi drive.

I could not find any online pdf manual for the Quantum Viking 4.5GB drive (68 pin uwscsi). But all jumpers are disabled. Also TE.

[arnljot]

So trying for a summary.


[Quantum]--[CS PPC]--[convertor]--[PLEXTOR]--[SEAGATE]

This setup "works", but the Seagate get A LOT of error messages that the device has an illegal state.

Quantum: No particular jumper settings.

Plextor: Just set it's ID to 5, no particular setting.

Seagate: Just id set. No other particular settings.

I will now try to set TE on the two drives, and leave the CD ROM as is.

[a1200freak]

Hi,

my SCSI chain looks like this:

T - CSPPC - HD (68pin with term power)- HD (68pin) - CDROM (adapter 50->68pin) - CDROM (adapter 50->68pin)- T

The terminators should be active UW (68 pins). One device in the chain should provide termination power (in my example the HD behind the CSPPC).

In your case the termination provided by the SEAGATE (50pins) only terminates the lower 8 Bits of the bus. That won't work.

Regards

a1200freak

[arnljot]

Duplicating you setup, using the HD's termination in one end (Quantum, which is 68pin and thus must terminate all bits), and using two converters for each of the FastSCSI2 devices with a 50-to-68pin-adapter and a active termination after the two did not remove the error.

The error I get from SFS is "The drive is in an illegal state, code=45, operation=3".

[Zac67]

If you use a wide-to-narrow-adapter, you'll have to make sure the upper data byte gets terminated there. Either use an adapter with 'half-terminator' or deactivate wide SCSI completely.

A good way to handle the problem is to run the wide cable to all devices and use separate wide/narrow adapters for each narrow device. Then terminate the bus at a wide device or with a separate cable terminator.

(Edit: looks like this is exactly what you tried last. Well, check if removing any single device solves the problem, swap the cable, ...)

Regarding termpower: I usually activate TP on the devices that supply termination as well or a device close to a cable terminator. In theory, you should have any number of termpower supplies (at least one), but I've had some devices play funny with too many TPs (e.g. some NCR adapters).

[arnljot]

Is there any way to "reset" what's in the SCSI list when you "esc" into the CS Menu.

It seems it's not affected from the different things I try.

[arnljot]

Okay.

Setup one: With the Seagate
Setup two: With the Quantum

Termination for both: One side of CS, 68->50->Active, other side; 68 pin active.

I'm quite sure. Almost positive that of the three 68->50 pins adapters I have, that one terminates the high bits.

Symptoms:
Both setups are seen by the early startup menu.

Setup one boots. Then is flooded with Error 45 on operation 3 from SFS.

Setup two seems to boot. Then reboots. And after another short read, it reboots and the powerled starts pulsing on and of endlessly.

My theory:
The Seagate drive is broken, setup one.
The Quantum drive has a bad boot sequence(software), setup two.

I really don't know... maybe reaching for straws here...

What I have to play with:
Quantum HD, UW-SCSI
Seagate HD, FastSCSI
Plextor CD ROM, Fast SCSI
2x50/68pin adapters
1x50/68pin adapter (suspected of having term)
1x50 pin internal/external adapter
1x68pin internal active termination
1x50pin external active termination
2x50 pin SCSI ribbon cable with two plugs
1x50 pin SCSI ribbon cable with three plugs
2x68 pin SCSI ribbon cable with five plugs

My goal:
Have atleast one working HD and CDROM.

Right now I'm at a loss how to achieve that...

edit:
With only the seagate on

edit 2:
Now I tried with both drives, no CD ROM.
Then the Seagate booted, so I changed boot pri on the Quantum, and copied the OS to that partition.
Then it booted from the Quantum, and entered the ever flashing powerled dance.

Question, could this be a 4GB barrier thing? I'll look up the SFS version, and see where the boot volume is located. Which SFS version should it be?

[a1200freak]

Did you copy the SFS in the RDB of the Quantum ?

Are you trying to install OS4 ? The version on OS4 CD is 1.277.

I had some problems with SFS installing OS4. After I copied the SFS to the RDB with MediaToolbox the Amiga bootet in a loop. I had to repartition the HD with a PC. So I used FFS from OS4 CD as described in the manual (set DOS type to DOS/07). Make sure you have the 68k filesystems in the RDB. Now it works.

EDIT: The boot partition has to be within the first 4GB.

[Framiga]

arnljot, that HD is gone :-(

Try this tool to check how many bad blocks you have (in the "Grown List" mainly)

http://aminet.net/disk/misc/q-device.lha

"Question, could this be a 4GB barrier thing?"

ah and of course, test with check4GB

http://aminet.net/disk/misc/check4gb.lha