GORRIEDE@UREGINA1.BITNET (Dennis Robert Gorrie) (09/11/89)
WARNING: This is a LONG message about my probs with a Micropolis 1674-7
hard disk.
Some of you may remember my earlier complaints about UPS. I ordered a
drive from Cache Systems, CA. by UPS Same Day Service. I received it 10
days later, damaged. The shipper sent me a second drive, that was lost by
UPS Sana Cruz. The shipper sent me a third drive, which UPS also cannot
find. I am realy not sure what to do. I am wondering what consumer groups
I might contact. I live in Regina, Saskatchewan, Canada, so that makes
dealing with UPS and this company a bit dificult. It has already been over
a month since the first drive was sent (Aug 8).
I am not sure what is wrong with the drive. Here are the details:
Amiga 1000
Insider 1 Meg RAM
C-LTD SCSI 1000 Controller
Micropolis SCSI 1674-7 150 Meg drive, with case and pwr by Cache Systems.
Drive device indicator at 0
1 SCSI terminator on second drive SCSI connector, SCSI cable to controller on
first connector.
/*- DevSetup generated by CLtd's MLed, by Steve Tibbett -*/
Device CLtd.device
Define Handler SCSI.Handler
Load Handler SCSI.Handler
Define Driver HardDisk.driver
Load Driver HardDisk.driver
Initialize Handler 100
HandlerCode SCSI.Handler
TaskName SCSI_Bus_1
TaskPriority 15
StackSize 250
HostAddress 7
ExitInit
Initialize Unit 1000
DriverCode HardDisk.driver
HandlerNumber 100
TaskName Unit_1000
TaskPriority 12
StackSize 250
AutoRetries 4
BufferSize 8
UnitAddress 0
Unit_LUN 0
Flags BlockRead
Flags AdaptiveBuffer
ExitInit
Exit
/*- Mountlist generated by MLed by Steve Tibbett -*/
DH0: Device = CLtd.device
FileSystem = L:Fastfilesystem
DosType = 0x444F5301
Unit = 1000
StackSize = 4000
Surfaces = 7
Flags = 0
BlocksPerTrack = 36
Reserved = 2
Interleave = 0
LowCyl = 0
HighCyl = 1245
Buffers = 5
GlobVec = -1
BufMemType = 0
#
***** I would realy appreciate the correct specs for this hard disk! *****
The SCSI INQUIRE gives me 305216 blocks, where as above we would get 7x36x1245=
313740. So I figure the specs they gave me are obviously wrong. At any
rate, I have tried values from 26 to 54 from the blocks per track, and I
have tried small partitions all over the the drive with no luck. It always
stops at cyl 0, or soon after, during the AmigaDOS format.
The FEdX formatter does not work. It does an "Initial Format" and stops
with a "Unit Select Failure"
The GenFormat ... universal format program, runs and finishes with no
messages, no errors.
Then I tried to do an AmigaDOS format. "system/format drive dh0: name
test ffs" It got to cyl 160 or there abouts, and then quit. I tired a
gain, and it got to cyl 20 and quit. I tried again, and it quit at cyl 2.
I tried again, it reached 6, I tried again, it made it to 50. You get the
picture. I re-ran gen-format several times. Then I tried an AmigaDOS
format, and it quit on cyl 0, several times.
I talked to a tech at C-LTD and he said the drive must be trashed, and
maybe a head was ditched.
I talked to Cache Systems, and they acted like I was nuts. After I phoned
them several times they sent me the second drive (or so they tell me).
I then tried the drive at a friends place:
Amiga 1000
Comspec SA controller for Amiga 1000
My hard disk
NO SCSI terminator used.
He set the specs to be 7 heads, 35 tracks, 1245 cyls. (notice the only
change was to 35 from 36 tracks).
He did a low level format. While doing a disk test, he got the error "SET
A FAILURE" or some such thing. He went on to do an Amiga DOS format. It
stoped at cyl 160. Whe moved the mountlist partition 200 cyls ahead. Then
the format (FFS) completed to cyl 1245.
We tried to format the 0 to 200 cyl partition again, for OldFileSystem. It
bombed at 160 again. We moved the partition end down to 140. Whe tried to
format it again and it bombed at cyl 114.
We gave up.
Can anyone tell me what is going on here for sure? Can anyone tell whether
it is the drive, the controller, or BOTH that are at fault?
Please respond directly, via E-mail, if you can. Thanks in Advance
|============================================================================|
| "The man who dropped a 3270 terminal on his foot... |
| |
| ... and lived to tell about it..." |
| |
| DENNIS GORRIE (GORRIEDE AT UREGINA1.BITNET) |
|============================================================================|GORRIEDE@UREGINA1.BITNET (Dennis Robert Gorrie) (09/12/89)
OK. In the devsetup file, I changed BlockRead to NoBlockRead, and things
seem to be working, ALMOST...
Do you (or anyone, especialy Steve Tibbet, who wrote FEdX) have experience
FEdX?? It does not seem to be creating proper .hdft format files for the
format program.
I start the format program. It goes through pattern tests 0 to 6.
Then :
'Beginning final format'
'Error - 26h - Invalid field in parameter list'
'Program aborted - Device error'
'CLTD format failed return code 100'
In the format file the next thing is an errors list of the errors I entered, in
Cyl, Head, Offset Format. For the heads, when I entered the errors, I entered
them with head numbers 0 to 6, like the errors sheet I recieved.
I am thinking that this is the cause of the 'Invalid field in parameter list'
error. Should I number the heads 1 to 7? Or, should I be using BLOCK format
for entering the errors?
I ran another format, WITH 0 pattern tests, and I CLEARED all of the errors
list. THAT TIME, the format completed successfully.
SO, it seems the formater will complete OK if I don't enter the errors list,
but it bombs if I do enter the errors list.
SO it is still next to useless , as far as reliability goes.
Any ideas? Should the heads be numbered 0 to 6 or 1 to 7 in the errors list?
Also, is the ADAPTIVE BUFFERS flag crucial in my devsetup? Should I remove
it altogether from my DEVSETUP?
Thanks for all of your help
Dennis Gorrie GORRIEDE AT UREGINA1.BITNET