Jan 122018
 

 

File ALL_QA.ZIP from The Programmer’s Corner in Category HD Utilities 
Lots of good hard drive information
File Name File Size Zip Size Zip Type
ESDI_Q&A.TXT 273012 90427 deflated
IDE_Q&A.TXT 143288 48470 deflated
MFM_Q&A.TXT 146802 49412 deflated
RLL_Q&A.TXT 123175 41651 deflated
SCSI_Q&A.TXT 43924 15903 deflated
TOPTEN.TXT 23476 7574 deflated
TPCREAD.ME 199 165 deflated

 

Download File ALL_QA.ZIP Here

Contents of the ESDI_Q&A.TXT file


Questions and Answers on ESDI drives from the Ontrack BBS - 10/24/90
-------------------------------------------------------------------------
QUESTION:

I've got a NEC 5655 with the UltraStor controller. Hardware = DTK 25
mhz, '386 with VGA, DOS 3.3, 2 meg RAM. I boot to DOS, enter SWBIOS,
then DM /M. When I finally get to 'Prepare a partition', the small
(3 meg C:) first partition with 'system' and volume name, the error
message "No record found I/O error Drive 1" shows up, then for each
cylinder and head, an additional error message is shown, incremented by
a value of one, i.e. "cylinder 1, head 2; cylinder 2, head 3, cylinder
3 head 4, etc.", until the 24 cylinders of first partition are
completed. Then the partition is still shown as unprepared. The drive
works perfectly otherwise. I have exchanged cables, checked jumpers and
switches - no change. Any ideas???

ANSWER:

With the Utrastor controller, you must low level initialize the
drive thru the controller's BIOS using DEBUG, then partition the drive
only with DM. When you come into DM use DM/M, then go DIRECTLY to the
partitioning menu, create your partitions and prepare them. With an
ESDI controller, if you init the drive thru the controller's BIOS, it
appears as a standard drive from that point on, also you probably will
have a translation mode available to make the drive appear to have less
than 1024 cyls so you don't need to use SWBIOS.
-------------------------------------------------------------------------
QUESTION:

Am trying to find out if a new version of Disk Manager is available for
Imprimis (CDC) Wren VI 630Mb hard drives which is compatible with 3Com
3+ Share. In desprite need. As of now, am initializing with v4.02 and
using 3.2 as driver.

ANSWER:

At this time, Disk Manager & 3COM don't work together. We
have offered to help them figure out why, but they don't seem interested
in making it work. So at this time it doesn't work.
-------------------------------------------------------------------------
QUESTION:

Our division is running OS/2 V1.2 and plan to install LAN Manager as
soon as Microsoft delivers it to us. Our file server is a Micronics
25Mhz 386 with a Maxtor XT-8760E hard disk controlled by a Western
Digital WD1007V-SE2 controller. For backup purposes we are using the
FAT file system on all our machines, but even though we can format and
partition the drive with Ontrack's Disk Manager, OS/2 will only
recognize the first partition (the C: drive only). DOS 4.01 will
recognize the entire disk, but will not allow us to create one large
partition. As the machine sits now, we have a 430Mb C: drive that both
DOS and OS/2 can access, and a 250Mb D: drive that only DOS can see.
Any and all information about how to get OS/2 1.2 to see large
drives would be greatly appreciated, as OS/2's FDISKPM will only
recognize drives as large as about 280Mb.

ANSWER:

At this point, Ontrack doesn't have a product to support OS/2
installations. If you enable the BIOS on the 1007V and init the drive
thru the controller BIOS, the drive will appear as a standard drive to
the system. Unfortunatly the drive will still appear as haveing more
than 1024 cylinders, which is a problem for both DOS and OS/2. You will
only be able to access this portion of the drive thru a Disk Manager
Write/Read partition with SWBIOS. And again, the device driver that
mounts this partition is only supported under DOS. There may be a time
in the future when we offer a driver that works under OS/2, but at this
time we don't, and there are no immediate plans to write one.
-------------------------------------------------------------------------
QUESTION:

I AM RUNNING A ZEOS 386/20 WITH AN INPRIMIS 94166-182 DRIVE AND AN
ADAPTEC ACB-2322B CONTROLLER. I USED V3.10 OF YOUR SOFTWARE THAT I
PURCHASED FROM ZEOS TO FORMAT THE DRIVE UNDER NOVELL ELS II 2.15
VERSION C. THE SERVER IS RUNNING IN NON DEDICATED MODE. THE NETWORK
IS UP BUT I GET A TERRIBLE KEYBOARD DELAY WHICH DOES NOT WORK UNDER DOS.
WHEN I DID THE ELSGEN I USED THE STANDARD AT-COMPATABLE RESOURCE WITH
THE WESTERN DIGITAL CONTROLLER & INTERUPTS. DO YOU KNOW WHY I HAVE THE
DELAY? DO I HAVE TO CREATE ANOTHER RESOURCE FOR YOUR SOFTWARE TO DO THE
GENERATION

ANSWER:

With this question, you should really call our tech support line
so that we could get more info concerning your installation process. In
general, with this controller, you need to init the drive thru debug,
then partition and prepare the partition (scan) thru DMN as a standard
drive, then run MODELSII as if you were installing a nonstandard drive
(because of the way the Adaptec card works), then run ELSGEN. If you
followed that procedure, then I guess I don't know why the server
keyboard would be that slow. Possibly a conflict between the controller
and the BIOS of the server? Also remember that you are running this
server non-dedicated, and should expect it to be somewhat slow.
-------------------------------------------------------------------------
QUESTION:

I am attempting to configure a Club-AT system, with a WD1007V-SE1 ctrlr
and an Imprimis (CDC Wren) Model 94186-383H ESDI Drive. I am using DMN
v3.04. The problem is that the system frequently hangs
during the boot process. The cold boot loaders message appears, but the
system does not progress to 'Mounting volume SYS'. This occurs approx
50% of the time. I believe that I have properly used Modspt and Modnet
to set things up. (Also, this system has two NICs, SMC Arcnet at IRQ
2, addr 2E0 and Novell NE2000 at IRQ3, addr 300).
Any help would be welcome.

ANSWER:

The fact that you mentioned MODSPT & MODNET leads me to believe
that you are using 2.0a adv ntwr. Are you letting DMN truncate the
drive to 255Meg? Are you keeping your volume sizes to 128M or less?
The best way to handle this one is to disable the BIOS & translation on
the 1007 and to let DMN do the whole job as a nonstandard drive. Then
run MODSPT 34, GENOS, then MODNET. Be sure to run DMN/I if you reboot
during the process or if COMMAND.COM is reloaded for any reason prior to
running INSTALL. When you do run INSTALL, make sure you select option 1
from the menu and then answer Y to the question "initialize?".
-------------------------------------------------------------------------
QUESTION:

I have two questions. The first: I am formatting an Impremis 320MB
ESDI hard drive (model # 94186-383H) in an AST 386/25 with a Western
Digital Hard Drive Controller 1007-AWAH. When we attempt to perform a
low level format, we get an error "SWBIOS NOT LOADED". Any advice?

Second question: We have some systems that are running disk intensive
operations. We have some that are running with DM and some without.
The systems without DM are running 10x faster than those with DM. Any
advice? All units are running the same 320mb impremis with the same
configuration as above. Is there a slowdown with the >1024 translation,
or is there something in DMDRVR.BIN thats slowing disk operations?

ANSWER:

With the 1007A, you have the ability to low level format this
drive thru debug using g=c800:5. Initing the drive thru this method
gives you two advantages:
1. initing the drive thru debug makes the drive appear as a standard
drive to your system from that point on.
2. you have a 63 sector per track translation mode available that will
make this drive appear as having 600 and some odd cylinders by 16 heads
by 63 sectors per track. Using this translation mode makes it
unnessesary to use SWBIOS, as Disk Manager then sees the drive as having
less than 1024 cylinders.
If you init the drive thru debug, using 63spt translation, then go into
DM in manual mode and go directly to the partitioning menu and partition
the drive only, you won't have to use SWBIOS at all.
As far as a general slowdown, if you follow this method there should be
no slowdown using DMDRVR at all. Also, if you use this method of
initing the drive (thru debug), you could just partition the drive with
FDISK and keep DM out of the picture if you want.
-------------------------------------------------------------------------
QUESTION:

I am trying to install a Miniscribe 9380E ESDI drive in an IBM AT using
a Data Technology controller card DC6280. When I use the DM program
it gives me a message about the drive not being compatible with the
set. My problem is, I have no idea what I should use as a setup drive
number. I would appciate some help.

ANSWER:

With the 6280, you should be low level formatting your drive thru
debug (g=c800:5). Your CMOS drive type should be set to type 1, and the
parameters of this drive type will be overridden at boot time by the
controller its self which will present the true parameters of the drive
to the system. You should low level the drive thru debug using Native
mode of the controller, then run SWBIOS, then run DM in manual mode
(DM/M) and go to the configuration menu. Press S for standard
parameters, then press W to write them to the disk. Return to the main
menu and then go to the partitioning menu, setup your partitions and
prepare them.
-------------------------------------------------------------------------
QUESTION:

I am trying to reformat a SEAGATE drive (ESDI) model 94166-182 but
can not determine its head and cylinder configuration.
Does any one know the details of this one? HELP!

ANSWER:

The 94166-182 has 969 cylinders, 9 heads, and 34-36 sectors per
track (depending on how the drive is jumpered). Your best answer though
is probably low level formatting the drive thru debug instead of Disk
Manager. If you low level the drive thru debug, the controller will
automatically know the parameters of the drive, and will make the drive
appear as a standard drive to the system from that point on. You would
be using Disk Manager for partitioning only in this case.
-------------------------------------------------------------------------
QUESTION:

WE WOULD LIKE TO GET A CURRENT COPY OF DIMANAGER TO FORMAT A 330 MED
CDC HARDISK. FOR SOME REASON, WHEN WE COPIED A 70 MEG DRIVE ONTO A 330
MEG DRIVE THE INFORMATION SIZE DOUBLE. FOR EXAMPLE, 67 MEGS OF THE 70 MEG
DRIVE WERE FILLED, WHEN WE USED "GETC" A FILE TRANSFER APPLICATION LIKE
BROOKLYN BRIDGE OR LAPLINK, THE SPACE OCCUPIED ON THE 330 MEG DRIVE
AMOUNTED TO NEARLY 140 MEGS. I NEED DISK MANAGER TO DO A LOW LEVE
FORMAT AND REFORMAT WITH DOS 4.01.

ANSWER:

The reason for this is probably that you have a very large
partition on the CDC. With a large partition under 3.x DOS, you
increase the cluster size (it's a tradeoff for the large partition
size). If you have a 300Meg partition, you have gone to a 32k cluster
size. A cluster is your minimum allocation unit, meaning that even if a
file is only 10 bytes long, it will take up 32k of disk space. You are
correct in thinking that DOS 4.01 will help the problem. In this
situation, you should low level format the drive thru debug, using the
BIOS on your ESDI controller, and envoke the 63 sector per track
translation mode that your controller offers, thus making the drive
appear to have about 600 cylinders, 16 heads, and 63 sectors per track.
This gets around DOS's 1024 cylinder limitation. You can then use DOS
4.01's FDISK and FORMAT C:/S commands to do the rest of the job. You
really don't need Disk Manager for this situation. If you would still
like a current copy of the universal Disk Manager, I'm sure sales would
love to sell you one. If you have further questions about the low level
format part of this procedure using the controller BIOS, your best bet
is to contact the controller manufacturer directly for tech support.
-------------------------------------------------------------------------
QUESTION:

Using DMN Version 3.1 Installing a Maxtor XT-4170E 150 megs with
2322b Adaptec Controller in a Northgate Elegance 386/20. Trying to
install with ANW 2.0a. Went through the install, and it booted up fine,
but now, 24 hours later, things started with "Checksum in Load file"
error on bootup. Have conflicting advice from Northgate re: disabling
the BIOS on the card. They said, to disable, and also to install the
drive as a type 41 (appearing in the computer's BIOS) but this has
different parameters, heads, sectors, etc. than the drive said.
(slightly different, only.) The weird thing is that the thing booted
up perfectly, and I had restored everything from tape and all was
working fine. Ideas here? Thanks.

ANSWER:

I guess the best thing I could do here is to give you the
recommended install procedure for this combination of hardware. With
the ACB2322B, you should leave the BIOS enabled, init the drive thru
debug (g=c800:5), go into the special options menu and select 63 sector
per track translation mode, and exit from the debug routine. At this
point the machine will reboot and you run DMN in manual mode. Once
you're at the main menu make sure the parameters are appearing as
standard and that those params are aprox 300 cyls by 16 heads by 63
sectors per track. If that is not the case, you haven't done something
correctly within debug. If the parms are correct, press F6 to create
your Netware partition and the F7 to prepare it. Let DMN get thru 1
pass and then press ESC to terminate the scanning. Exit out of DMN and
run MODSPT 63, followed by GENOS, followed by MODNET. (You need to run
MODNET even though you are installing this as a standard drive because
of a quirk of the Adaptec card). You should then proceed with INSTALL
and selection option #1 from the install menu. When asked if you wish
to init this drive answer Y. Also make sure your volume sizes are
limited to 128 Meg maximum (you'll have to cut this drive into two
volumes). This is a known GO installation procedure for this card with
this drive under this version of Netware. If you still have problems
after this procedure, at least we'll know that we are on stable ground
and can proceed with troubleshooting from there.
-------------------------------------------------------------------------
QUESTION:

I have a customer who has ran prepare from a Novell ELS version on a
ESDI drive with dmn used on it. The drive is a CDC MN94166-182e, Novell
Advance netware v2.0a, ARCNET. I also beleive that they ran Vrepair on
the drive. Any hope of recovering the data files? Can your program
Netutils recover the data files, or can you and what is your turn around
time? What are your charges for data recovery?

ANSWER:

You best bet here is to contact our data recovery department
directly at (800)872-2599. They could tell you what their turnaround
would be and what their rates are. I don't think Netutils would have
much of a chance of undoing this much damage.
-------------------------------------------------------------------------
QUESTION:

trying to find information on setting up a microscience 2160 esdi drive
with ontrack DMN ver 3.01 on sanyo 286
alwawayb get problem with drive 00: duplicate copies of disk redirection
data do not match
any reason this should happen in installing netware 286 ver 2.15

ANSWER:

You didn't tell me what brand and model of ESDI controller you
are using for this install. On a ESDI install this is the single most
important piece of information that I need. Without knowing which
controller you are using, I can only say that you should PROBABLY be
low level formatting (initializing) this drive thru your controller's
BIOS using debug, and then envoking it's 63 sector per track mode to
make this drive appear to have less than 1024 cylinders. You would then
use DMN to partition the drive only as a standard drive. If it's an
Adaptec controller, run MODDRVR anyway as if you were installing a
nonstandard drive. Using the translation mode will allow you to boot
the server directly from the hard disk after it has Netware on it. If
you would leave me another message with the brand and model of
controller you are using, I could recommend a specific install
procedure. (Or call the tech support line at (612)937-2121).

QUESTION:

the controller i'm using is a dtc 6180 with the micro science drive
2160 with 160 mg.

ANSWER:

With the DTC 6180, the recommended install procedure is as
follows:
1. Boot with DOS and run debug
2. At the debug prompt enter g=c800:5 and hit return
3. From the menu that appears on the screen, enter your bad tracks and
initialize the drive.
4. Exit the low level formatting program and let the machine reboot.
5. Run DMN/M
6. Once in the DMN program, confirm that the parameters shown in the box
on the left side of the screen are STANDARD and those parameters are
1274 cyls by 7 heads by 34, 35, or 36 sectors per track (depending on
how the drive is jumpered).
7. If the parameters are correct, press F6 to create the Netware
partition.
8. Once the partition has been created, press F7 to prepare the
partition.
9. Let DMN get thru at least 1 pass of the preparation, then press ESC
to terminate the scan.
10.Exit DMN and proceed with a standard netgen (no MODDRVR needed). Be
careful not to change the size of the partition from within netgen.
11.You will have to boot the server from a floppy since the drive is
over 1024 cyls. Just boot DOS and run NET$OS (no DMN/I needed).
If you have further questions about this install, leave them for me here
on the board or call the tech support line at (612)937-2121.
-------------------------------------------------------------------------
QUESTION:

my dmn version is 3.01. I am upgrading the store's
server with a 94196-766 hard drive from impiris. my novell is 2.15.
and controller is a 1007-se2. i set the drive up as 16 x 1024 x 64
sectors in the bios. when i use dmn the drive comes out as ~525MB.
is there a way to get the full 660mb plus. the hard drive is not
listed in the cdc drive tables.

ANSWER:

With this controller/drive combination, you will need to make
sure that the drive is jumpered for 53 sector per track, low level
format the drive with the controller's BIOS thru debug enabeling
alternate sectoring, make sure the parameters are 1630 x 15 x 52, and
exit debug. You then run DMN/M and make sure the parameters are showing
as 1628 x 15 x 52 STANDARD. You should then press F6 to create the
Netware partition, and F7 to prepare it. You may then proceed with a
standard Netgen. (No MODDRVR or DMN/I needed). When you run Netgen to
install Netware, make sure you select a custom configuration and when
Netgen asks you if you wish to change the partition table, answer NO.
After the installation is complete, you must boot this server from a
floppy because it's over 1024 cylinders. Again, no DMN/I is needed,
just boot from a DOS floppy and run NET$OS.
-------------------------------------------------------------------------
QUESTION:

What is the most current version of Netutils? I have version 1.2.
It will not recognize ESDI drives with more than 1024
cylinders, with or without translation. The program does work, but only
up to cylinder 1024. Using WD1007SEV2, with ROM enabled, 150MB Maxtor.
BTW I have used it to recover files from a downed server, tedious
because they must be copied one at a time, but is DOES WORK!


ANSWER:

The most current version of Netutils is version 1.2. You are
correct in saying that Netutils does not work with drives greater that
1024 cylinders... kind of...
Here's the specific limitations of Netutils at this point:
Netutils is limited to a drive that is CONFIGURED to be 1024 cylinders
or less, and 255M or less. In other words, you could have a Miniscribe
3180 that is 1250 cyls by 7 heads by 35 sectors per track, and if the
drive had been initialized thru the controller BIOS and set into 63
sector per track translation mode therefore appearing as 303 x 16 x 63,
you could use Netutils with no problem. This is because the drive is
appearing as having less than 1024 cylinders to the system, and also
having a capacity less than 255 Megabytes. (The 3180 is aprox. 150Meg).
On the other hand, if you had a Miniscribe 9380, even if you inited thru
the BIOS and used 63spt translation therefore making it appear to have
only 637 cyls by 16 heads by 63 spt (thus getting under the 1024 cyl
limit of Netutils), the drive's capacity would still be greater than 255
Megabytes, and you would have hit the second limitation of Netutils.
Other current limitations of Netutils are:
1. No SCSI or DCB VADD support
2. No work will be done on redirection tables.
3. No support for v2.15 if a MAC VAPP is present.
4. And actually, there are no guarantees on Netware 2.1x at all.
(although most functions actually will work fine under 2.1x, Netutils
was written for v2.0a, so nothing is guaranteed under 2.1x)
Now... The good news is that we are currently working on a new version
of Netutils that will not have any of the above limitations, and will
have quite a few additional features as well. Projected release date is
third quarter 1990. If you have any other questions, leave another
message here, or call us on the tech line at (612)937-2121.
-------------------------------------------------------------------------
QUESTION:

Has anyone come across any problems with installing a Maxtor 8760e 600mb
ESDI drive into an IBM m 80 and tried to use DMN to compsurf for
Netware. This drive continues to crash after serveral weeks of
operation, and we haven't found anything physically wrong with it.
The tech specs are as follows:
IBM m 80, WD 1007v MC1 mca esdi controller card, DMN v. 3.04
and a Maxtor 8760e ESDI 600 mb drive.
Urgent Help Needed.

ANSWER:

DMN v3.04 was not designed to operate on a PS/2. You should
upgrade your DMN to v3.10 to do this installation. Once you install
this drive with version 3.10 of DMN, I see no reason for problems.
You can upgrade your DMN by contacting our sales department at
(800)752-1333.
-------------------------------------------------------------------------
QUESTION:

I am using ONTRACK to install a new disk on a WD1007 controller and read
the intructions in HELP saying I needed to use SWBIOS. SWBIOS is not on
my ONTRACK disk. So I called this bulletin board hoping to download it.
I can't find it. Is it in the download area? If it is, how can I find
it and download it? THANX

ANSWER:

You didn't leave me the version number of your Disk Manager, but
evidently it is an OEM version that doesn't have SWBIOS on it (such as
the Seagate versions). Sounds like you should contact sales and
purchase the generic version of Disk Manager that has SWBIOS. You also
may have the option with your 1007 card to low level format the drive
thru debug and use it's 63 sector per track translation mode to make
this drive appear as having less than 1024 cylinders, thus preventing
the need for SWBIOS at all.
-------------------------------------------------------------------------
QUESTION:

I am installing a MAXTOR XT-8760E disk (650MB) with a WD controller (I
think the controller is a 1007 (?) using Disk Manager to partition the
disk into a small DOS partition and a 640 r/w partition. Everything
seemed to be working fine until I attempted to run an application
program and got an "Insufficient memory" message. I removed everything
but DM from the config.sys and autoexec.bat files, and the available
memory is still about 150K less than the 512K installed in this system.
If COMMAND.COM is using about 40-50K, DM appears to be using 100K of
memory, which seems unreasonable. I have not been able to get through on
the tech support voice line, so I would appreciate your help.

ANSWER:

With that size of partition, you have gone to a 16k sector
size, therefore if your BUFFERS statement in your CONFIG.SYS file is:
BUFFERS=20, you would be allocating 20x16k or a 320k buffer area. If
you don't have a buffer statement in your CONFIG.SYS file at all, it
defaults to 8 buffers, resulting in 8x16k or 128k buffer area. The
solution is to back down your buffer statement to BUFFERS=1, thus
creating a 16k buffer which should be more than sufficient for any
application. (When an application program dictates that it needs a
BUFFERS=20 in your CONFIG.SYS, the only thing they are really concerned
with is that you have a 10k buffer area created. They are ASSUMING that
nobody has anything other than a 512 byte sector size.)

QUESTION:

Thanks for the info on the large buffer size. I'll give that a try.
However, another problem has arisen: I was copying a number of files to
the disk (D:, the large 630MB partition). At about file number 255 or
256, the system locked up and I had to reboot. I understood that the
maximum number of files in a root directory was 512, so I don't know
where this 255/256 limitation comes from. But the most bizarre symptom
was a file that was created on the disk (named the same as one of the
files I was copying) with a length of 0 bytes. I cannot delete or rename
that file, even with Norton's WIPEFILE. All attempts result in a "divide
Overflow" message!!! Any suggestions??

ANSWER:

You have come across an anomaly with DOS 3.x. When you have a
partition greater than 512Meg under DOS 3.x, you will be limited to 256
directory entries in the root directory. You wil also be limited to 256
entries in any subdirectory as well, (and I thought subdirectories were
dynamic!). The solution is to keep your partitions down to below
512Meg, or to go to DOS 4.01 which doesn't have this problem.
-------------------------------------------------------------------------
QUESTION:

Hi. Wasn't sure where to put tech supp. question so am leaving it here.
Hardware: DTK 386/25 using CompuAdd Hardcache controller w/
Imprimis 94166.
Used controller card software for low level format. Using DM to
partition drive.
part #1 is DOS (32MB).
When preparing part # 2 (64MB) I get a mesage: "UNCORRECT ECC: . . ."
Question: Does this indicate a deffect in addition to what is in the
media deffect table? If so, does the prepare process lock out this bad
area? When you do a chkdsk and it shows bad sectors, are these bad
sectors the ones in the media defect list, or are they ones found while
formatting (in addition to the ones in the media deffect list.)
Question #2.
Do common defragmenting programs like Disk Optimizer and Compress(PC
Tools) work on the partitioned drives?

ANSWER:

When Disk Manager finds an Uncorrectable ECC I/O error when
preparing a partition, it will mark out the related entries in the FAT
table, thus effectively mapping out the bad spot. When CHKDSK shows you
the number bytes in bad sectors, it is reading the entries that have
been mapped out in the FAT table, which should include all defects in
the defect list, plus those found while preparing (formatting) the
partition. As for your second question, it depends on the utility you
are trying to use. If you have initialized the drive thru your
controller BIOS or controller software, the drive should be appearing as
a standard drive at this point, so that's not a problem. You do have
Disk Manager Write/Read partitions though, and there are programs that
don't understand our partitioning method. You will need to contact the
software company the makes the optimizer product and ask them
specifically if they work OK with Disk Manager partitions.
-------------------------------------------------------------------------
QUESTION:

Software: DM-N 3.01
Hardware: Everex ESDI controller EV-948
Joy Systems 286 with 2 MB RAM, AMI Bios
CDC 160 MB ESDI 94166-182 967 cylinders, 36 spt, 9 heads
Network: Advanced Netware 286 2.0a

Network Boards: Arcnet and NE1000 Ethernet
I have attempted several installs with unsuccessful results.
Background: The system has been successfully used as a file server with
MFM drives with no problems. The hard drive functions properly under
DOS.

Scenario 1: Used the bios on the controller to format the drive, then
used MODSPT 36. Installed two volumes, 120 mb and 33 mb. Since it
showed as standard drive on DMN, didn't use MODNET. Error message
Abend: DivideOverflow Interrupt

Scenario 2: Changed the drive parameters in DMN to 34 spt per track (to
make it a non-standard drive) and carried the changes through GENOS and
reinstalling. It seemed almost happy, showing the Mounting Volume SYS
and Mounting Volume Vol1 message. However, it hangs at that point.

Scenario 3: Disabled the Bios on the controller and formatted the disk
using DMN, MODSPT, and MODNET, encountering the same result as Scenario
2.

Scenario 4: Reenabled the bios on the controller, formatted once again.
Reinstalled once again, except that I used MODNET even though it was a
standard drive. Error: Abend: General Protection Interrupt

I believe that in all cases the appropriate sequence was followed. In
fact, there have been many more scenarios, some to silly to tell about.
I have been unable to get through on your help line, but I am running
out of ideas. Please give me some guidance on what to try next. Thanks.

ANSWER:

With the AMI BIOS you will probably have to disable the controller
BIOS and initialize this drive as nonstandard thru Disk Manager -N. The
only other problem that I see is that I am not familiar with the Everex
EV-948 controller. With some ESDI controllers, when you disable the
BIOS, you've got no way to disable the translation. Your best bet here
is to contact us on the tech support line at (612)937-2121 so we could
discuss this further and perhaps figure out a solution. Please be in
front of the machine when you call so we can try a few things.

FINAL OUTCOME:

I just wanted to let you know the final outcome on my problem
with installing a CDC 160 MB drive with an Everex ESDI
controller. I substituted a Western Digital w1007 in its place,
making sure that the bios was disabled on the board, and ran
through the process again with the same lock up problem that I
experienced with the Everex. I then looked further and replaced
the Arcnet board with another and everything worked. So it had
nothing to do with the Everex ESDI controller. So my final
conclusion: If one uses an AMI bios, then it is best to disable
the controller bios and use Disk Manager-N for the non-standard
driver option. Thank you for your help.
-------------------------------------------------------------------------
QUESTION:

Disk Manager-N Version 3.10
I am attempting to install Novell Advanced Netware 2.0a on a
Maxtor XT4380E Hard Disk ( 1222 Cyl 15 Heads 36 spt ) with a
WD 1007v SE2 Controler. Cylinders reduced by DMN to size limit of
Netware 2.0a.

I have to use a key card eliminator because the customers key card
and Novell disk serial numbers do not match.

I have run modspt, genos, modnet, & install. In order.

I boot the system, run DMN/I, then net$os. The server mounts
volumes, comes up to the : prompt, and workstations may link.

I down the server. Power off. Repeat the process and after
entering the net$os command the floppy appears to access,
then becomes quiet but the light stays on. No further action
on monitor.

I have repeated this several times on different computers,
286's and 386's. I have found no pattern as to when the
system will boot and when it won't. It won't more often than
it will.

I have sizeable investment in memory, hard disk, and
controler which I need to recover from customer. I've been
working on this for two weeks. Please Help!.

ANSWER:

It sounds like you are following the correct procedure here,
assuming that you first of all have disabled the BIOS and translation on
the 1007V by putting a jumper on W3 pins 1&2, and on W1 pins 5&6.
After disableing both the BIOS & translation on the controller, you
should run DMN/M, press F3 and select the Maxtor 4380E, let DMN truncate
the drive to either 967x15x36, 994x15x35, or 1024x15x34 (all are 255M),
press F4 to enter the defect list, F5 to low level initialize the drive,
F6 to create the Novell partition, and F7 to prepare the partition.
After at least 1 pass of the preparation, press ESC to terminate the
preparation, and exit DMN. Run MODSPT with whatever sector per track
value you ended up with in the configuration of this drive, followed by
GENOS, then MODNET. At this point, run DMN/I just for good measure, and
then run INSTALL. Select option number 1 from the install menu and
answer Y when prompted to initialize the drive. Proceed with a normal
install from here. If you follow this procedure, the server will be
using as much of the drive as Netware 2.0a can handle (255Meg), and you
will be able to either boot directly from the hard drive or from a
floppy with DMN/I & NET$OS if your key card defeater program dictates.
If the above process doesn't work, then you may have flakey hardware,
(the intermittent nature of the problem point that way anyway), or I
guess I'd be suspicious of your key card defeater program. If you start
from scratch with this procedure and it still doesn't work, then give us
a call on the tech support line at (612)937-2121 and maybe we can help
you figure it out.

P.S. Make sure you are keeping your volume sizes down to below 128M.
-------------------------------------------------------------------------
QUESTION:

I recently purchased a Micropolis 1355 ESDI PC PAK which came
with your DM software. I have two MFM drives on a Data Technology
controller. I would like to install the ESDI controller and drive
as drive # 3. Can I do that using DMDRVR and DM software ?
The ESDI controller is a WD 1007A-WA2. The CPU is a Micronics 386/25
using PHOENIX BIOS. Any help would be greatly appreciated.
Thanks.

ANSWER:

While it is true that both controllers have the ability to be
changed to a primary or secondary address, you computer BIOS never scans
for a secondary controller at boot time. Therefore the secondary
address was a nice idea, but the computer BIOS people have fallen short
in their support of it. Disk Manager, unfortunatly, can do nothing
about the situation. You will have to choose between installing just
the MFM drives, or just the ESDI drive. You can't have both.
-------------------------------------------------------------------------
QUESTION:

Here's the lay of the land:
Got an Adaptec 2322b controller, a Maxtor 8760E 660 Meg disk, and
Disk Manager-N. Trying to install Netware v. 2.15 rev. A with this
happy family, and they ain't gettin' along. In your Q&A.TXT I saw
explanations of the 2322b with other drives, and that drive with other
controllers, but what if you got them all together? I've formatted the
disk using the spare-sector approach, thinking maybe the bad block table
of netware was getting overloaded, but it's no go. Tried using a patch
supplied by Adaptec (patched ATDISK.OBJ) and it would at least load
netware (from the cold-boot loader, no less!) then run for a couple of
days, slowly coming to pieces -- more and more hot-fix errors, lost
files, trashed directories, until it got so bad the whole system would
come crashing down. Then we got DM/N and tried
using a real McCoy IBM AT, thinking that if it was CPU incompatibility,
that at least shouldn't be the case with the AT. But now we get through
the format, partitioning, preparation, loading the netware system file
jest fine, but when we try to boot netware (from a floppy, after copying
the net$os files onto it) it loads everything the dies with ABEND:
general protection interrupt running process MUXPRC. MUXPRC? Can't
find diddly about that in any of novell's documents. You guys must have
a clue about this combination of hardwares, eh? Should I try a WD
controller? I'm ready to try anything! The drive's reformatting now
(using the adaptec bios routine) -- please give any help you can.
Gracias, as they say here in Texas. . .

ANSWER:

Procedure for installing a Maxtor 8760E with a ACB 2322B under 2.1x Ntwr

First make sure that your 2322B is a 2322B-8 or a 2322B-16.
Either one of these numbers means it's a 15Mbit controller. If it's not
a -8 or -16, then it's probably the 10Mbit version of this controller
and won't be able to handle this drive. (Contact Adaptec Tech Support at
(408)945-2550 to identify the card if you don't see a -8 or -16 on it).
Because of the high number of sectors per track, alternate sectoring is
probably in order here. You can only have up to 64 defects before you
go over the 512 bad block limit of NetWare. (Each track consists of 8
blocks on this drive.) Alternate sectoring will swap out the
manufacturers bad spots with alternates, thus making the drive appear
error free at system level, giving DMN a fresh start and the ability to
find up to 512 more bad blocks without overflowing the bad block table.

Other than that, the procedure should be as follows.
1. Set your CMOS drive type to type 1
2. Initialize (low level) the drive & enter bad tracks thru the
controller BIOS using DEBUG. (g=c800:5)
3. Enter DMN in manual mode (DMN/M) and confirm that the drive paramters
are STANDARD as aprox 1630x15x50. If this is not the case, then
press F3 for parameter selection, and then press F10 for standard
params and write the config to the disk.
4. Press F6 to create the NetWare partition
5. Press F7 followed by F2 to prepare the partition. (This will run
forever, unless you hit ESC after at least one pass).
6. Run MODDRVR even though the drive appears as standard. (I know, it's
contradictory, but trust me. It's because of the Adaptec controller.)
7. NETGEN next, making sure you get the ATDISK.OBJ that you modified in
the last step linked in.
8. Install NetWare, making sure you do NOT manipulate the partition
table in any way from within NETGEN. Select a CUSTOM configuration,
and then answer N to the question about changing the size of your
NetWare partition. (NETGEN can't understand > 1024 cyls, so will show
a wrapped geometry. Only aprox. 600 cylinders).
9. Boot from a DOS floppy and run NET$OS to bring up the server.
(Because the drive is > 1024 cyls).

This is the same basic procedure that you would use with any 50-54
sectors per track drive on a ACB2322B under 2.1x NetWare.

If you follow this procedure, and are still having problems with your
install, then it would be time to give us a call on the tech support
line at (612)937-2121 so we could talk with you directly.
-------------------------------------------------------------------------
QUESTION:

Hello! I'm having a problem with the following configuration: DMN,
Everex 1800C 12 mHz 286, OMTI 8620 ESDI controller, Fujitsu
M2246E ~150 meg ESDI disk drive, Advanced NetWare V2.15C. Bios is AMI
(c) 1987, Rev F3-21 ..... S286-2098 4/26/89 K8 are some other numbers
on the screen during boot.
Problem goes like this: System has problems seeing hard disk when
any drive type is in CMOS. DMN works great when CMOS says no hard disks
installed. Run DMN, run MODDRVR, do the Netware thing. WHen it comes
time to install Netware to hard disk, it says no hard disks installed,
probably because CMOS says no hard disks are there. From reading a
previous message here, I've got to assume that I must disable the BIOS
on the OMTI 8620 controller. (Just a guess.) I've tried it three or four
different ways. The closest I ever got was ABEND error resetting at hard
disk controller. That was when I had a drive type in CMOS. Can you
help???

ANSWER:

According to our information, the OMTI 8620 is NOT a NetWare
compatible controller card. You will have to choose a different
controller to use if you wish to install NetWare on this machine.
-------------------------------------------------------------------------
QUESTION:

I have a WD1007V-SE2 ESDI controller and a CDC/Imprimis 94166-182 drive.
The Western Digital Tech Support department gave me instructions on how
to get this thing going under vanilla DOS 3.3. However, the lady there
also told me that I would be able to get "a lot" more capacity from the
drive if I used Disk Managerand she recommended that I call you. Will
Disk Manager in fact increase the capacity of that drive when attached
to the WD1007V-SE2 controller? If it does, I'll be glad to buy your
product. Thank you for any help you can provide.

ANSWER:

Disk Manager would not give you any more total capacity from this
drive at all. What it COULD do is allow you to have a greater than
32Meg partition under DOS 3.x (you could have a 1 meg C: to boot from,
and a 150 Meg drive D:).
-------------------------------------------------------------------------
QUESTION:

I am using Disk Manager - N in conjunction with a Compuadd Hardcache
card, and an Imprimis 630MB hard disk model 94196-766 in an AST premium
386. formerly it contained an ESDI card and a Miniscribe 300MB disk
(non-standard). I ran the patch to restore the driver to it's original
state. I regenerated the network. when I go into the Netware
installation, I get....

Abend: Error resetting the PC/AT hard disk controller.

I called Compuadd and they recommended formatting the disk at the
computer's slowest speed with all shadow RAM disabled. I tried this to
no avail. Also, I am using Novell SFT ver 2.11. Any suggestions?????

ANSWER:

I see that you spoke to tech support a couple of times last
week about this matter. The one thing I see that wasn't mentioned in
our record of your tech support call was that you will need to run
MODDRVR when using this card, even though the card is making the drive
appear as a standard drive. (the same method has to be taken with an
Adaptec controller). Your complete procedure should be as follows:
1. Init the drive thru Compuadd's low level routine & enter bad tracks.
(sector sparing is probably in order here)
2. Enter DMN with the /M switch (DMN/M)
3. The drive should be appearing as STANDARD 1628 x 15 x 53 (aprox)
If it's not, press F3 then F10 to install the standard params, write
them to the disk, exit DMN, reboot, re-enter DMN/M, and see if it
now shows the drive as STANDARD with the correct parameters.
4. Press F6 and create your NetWare partition
5. Press F7 and prepare that partition
6. Run MODDRVR as if you were installing a non-standard drive, then link
in the modified driver with NETGEN.
7. Install NetWare, being very careful not to manipulate the partition
table. Select a CUSTOM install, then answer NO when asked if you
wish to change the size of your partition.
8. Make a boot floppy to boot the server with. No DMN/I is needed, just
boot with DOS and run NET$OS.

If you are still having problems after attempting this procedure, then
you may have a faulty controller card.
-------------------------------------------------------------------------
QUESTION:

I HAVE A SPERRY AT 286 MOD. 3126-00 NOVELL 286 2.0A
I AM USING AN ESDI CONTROLLER DATA TECHNOLOGY CORP.#DTC6180
THE HARD IS A FUJITSU 2249E 330MB
I AM ALSO USING ONTRACK (DMN) V.3.03
I AM ABLE TO SELECT THIS DRIVE EXACTLY AND I ENTER THE BAD BLOCKS
MY PROBLEM OCCURS WHEN I TRY TO PARTITION/INITIALIZE/VERIFY
ALL HEADS ARE BAD OR ALL SECTORS NOT FOUND OR IT JUST KEEPS MARKING
ALL AS BAD. WHAT COULD I BE DOING INCORRECTLY ?
I HAD ONTRACK WORKING FINE WITH DMN V.2.41
HOWEVER, I WAS NOT ABLE TO SELECT FUJITSU 2249E LIKE W/V.3.03
W/ 2.41 ALL SEEMED TO WORK FINE AND I DID'NT GET ALL THESE ERRORS
LIKE W/ V.3.03. NOVELL WOULD LOAD AFTER USING V.2.41 DMN
BUT WHEN BOOTED IT GAVE AN ERROR MSG. WE DESCIDED TO GO FROM SCRATCH
W/ DMN 3.03 BECAUSE I NEW WE COULD SELECT FUJITSU 2249E EXACTLY.

ANSWER:

With the DTC6180, you'll have to initialize the drive thru debug,
then partition & prepare the partition thru DMN as a STANDARD drive and
run MODSPT with the sector per track value that you ended up with in
debug.
-------------------------------------------------------------------------
QUESTION:

I left a message that was answered 2 days ago, but some new issues have
cropped up. First, DMN is V3.04, WD1007-VSE2 controller,
Fujitsu M2246E hard disk, Everex/AGI 1800C 286 12 mHz. I spoke to your
excellent tech support people, but something new came up. Here goes:
Trying to install Netware 2.15C. First problem: I cannot place a drive
type in CMOS without getting a hard disk controller error. Here is the
most successful results I have had (this morning): Disable WD1007 BIOS
and translation; put in drive type 1, pass up the error message on boot;
run the Western Digital utility WDFMT to low-level the drive; run DMN
successfully to prepare the drive; actuall got Netware installed; but I
cannot reboot the system, either with a drive type in CMOS or not. Spoke
to Everex, they say that with the machine and system BIOS I have, they
run ESDI drives all the time. (Not sure if with Netware tho). I used
this hard disk for a year prior to attempting to use it for Netware
under DOS 4.01 with no problems. Do you or anybody on earth have any
idea waht is wrong? BIOS revision of system is F3-21A. Thanks for the
assistance! PS I'm under a strict time limititation to get this bad boy
running!

ANSWER:

With the WD1007 & the Fujitsu 2249, the recommended install
procedure would be as follows:
1. Enable both the BIOS & translation on the controller (remove jumpers
from both W-3 pins 1&2 and W1 pins 5&6).
2. Set your CMOS to drive type 1
3. Boot with DOS and run debug, at the hyphen prompt type G=CC00:5 after
which you will receive a menu on the screen.
4. Select the option from that menu to low level format the drive.
5. Select the option to enter your bad track list, and enter it.
6. Select the option to set drive type and hit the + key until you see
an option that has aprox 650 cyls, 16 heads, and 63 sectors/track.
7. Hit return to accept that option.
8. Exit from the low level format routine.
9. Reboot with DOS and make a fresh working copy of DMN.
10.Run DMN in manual mode. (DMN/M)
11.At the main screen of DMN, confirm the drive parameters are STANDARD,
and those paramters are about 650x16x63.
12.Press F6 to create a NetWare partition.
13.Press F7 to prepare that partition.
14.After at least 1 pass of the preparation, press ESC to exit testing.
15.Exit DMN
16.Proceed with a completly standard install. (No MODDRVR or DMN/I or
any need to boot from a floppy. The drive appears as a STANDARD drive
with less than 1024 cylinders using this method.)

If this method doesn't work, then you can try disabling the BIOS &
translation on the controller, and doing the entire job thru DMN as a
nonstandard drive. In this case you would have to run MODDRVR for a
nonstandard drive, DMN/I everytime you rebooted during the process, and
you would have to boot the server from a floppy because the drive would
be appearing as having greater than 1024 cylinders. But if you have a
BIOS that doesn't release control to the controller BIOS at boot time so
that it can supply the parameters of the drive to the system, then this
second method is your only other alternative.

P.S. Make sure that the drive is jumpered to either 35 or 36 sectors per
track. I have heard that there were some Fujitsu drives that came out
of the factory jumpered at a different SPT value.

QUESTION:

Maybe I typed in the drive type wrong.....its an M2246E and it is 823 X
10, with jumpers selected between 16 and 65 SPT, 35 is available

ANSWER:

Sorry about that, you didn't mistype it, I misread it. The
procedure would be the same though, except forget about setting the

controller to it's 63 spt translation mode, and when you get into Disk
Manager after initing the drive, it should show as STANDARD 821x10x35.

QUESTION:

I finally got the thing working with the help of the Fujitsu tech
support people. Anybody who has a Fujitsu M2246E (and maybe other of
their drives) there is a jumper on it to select Radial or Daisy
operation. Although I had only one drive in the system, I had it set to
radial (it worked fine under DOS that way) and it should have been set
to Daisy. After that, everything went smooth. I *DID* select 63 SPT
mode, though. Netware installed fine, but was there some reason I should
not have selected it??

ANSWER:

The 63 spt translation mode would not be necessary with this drive
because it is already under 1024 cylinders. That's the only reason the
63 spt translation mode exists; to translate a drive that has for
example 1224 cyls down to aprox 600 cyls by 16 heads by 63 spt for
operating systems that don't like a drive greater than 1024 cyls. If
you used the 63spt translation on this drive it would appear as aprox
285 cyls by 16 heads by 63 spt. It's not going to hurt anything, it's
just unnecessary.
-------------------------------------------------------------------------
QUESTION:

imprimis 94246-383
wd1007a-wa2 esdi ctrl
ontrack 4.02 dm
386-25 clone w/ami bios
using wd bios for low level format,
selecting 16 * 63 option,
using swbios and dm, drive will only format to 212mb rather than 383.

ANSWER:

The Imprimis 94246-383 is a 53 sector per track drive. (1747 cyls
by 7 heads by 53 sectors per track). You will have to have a 15 Mbit
controller to handle this drive. The WD1007A is a 10Mbit controller, it
can't handle the high transfer rate of this drive. What I'm saying is
don't even use this drive with this controller at the reduced capacity
that you're getting. It's not going to work at all. Replace the
controller with a 15Mbit controller and you'll solve both problems.
-------------------------------------------------------------------------
QUESTION:

I am using DMN v2.43
I am about to add a second CDC94166-182 ESDI drive to my 286ANW v2.15a
network. Intellectually, it seems like a straight forward process.
However, my palms keep sweating.
I plan to physically install the 2nd drive and make the appropriate
entry into the CMOS via the on-line "setup" facility of the Phoenix BIOS
3.10 Plus. Then I plan to initiate DMN, specify drive #2 and initialize
as a non-bootable netware disk. After exiting DMN, I will go
directly into NETGEN and configure Netware to recognize the 2nd
drive/volume.
Are there gremlins lurking about that I should know about? Any tips
for the unweary? Do you recommend talc for the palms?

ANSWER:

You didn't really leave enough info about your system for me to
make an appropriate responce. What brand and model of ESDI controller
are you using? Is the BIOS on the controller presently enabled or
disabled? What is your first drive? Was it initialized thru debug or
thru COMPSURF or thru DMN? If it was initialized thru DMN, is it
installed as STANDARD or NONSTANDARD? If STANDARD, is it using physical
or translated parameters? When you say you are planning to "make the
appropriate entry into the CMOS via the on-line setup facility", are you
referring to a user defined drive type? What kind of machine are you
using? Who's BIOS is it running? When you said this was v2.15a
netware, did you mean v2.15 advanced, or did you mean v2.15 advanced
revision A ? You'll need to be alot more specific for me to provide a
correct answer to your question. There is one suggestion that I can
make though... Back it up first.

QUESTION:

DMN v2.43
WRT my request for information contained in message 55 and your reply
contained in message 57, the following additional information is
provided:

ESDI Controller: DTC 6280 1:1 HD/FD
Controller BIOS: enabled
First Drive: CDC 94166-182 (969cyl/9hds)
How Initialized: I can't remember, but I think by DMN (Is there a way to
tell?)
Std/Non-Std: Non-Standard (the CMOS drive type is set at 1)
Machine: 286AT Clone (ELT-286), 16mz, CHIPS NEAT chipset, 4mb
BIOS: Award 286 Modular BIOS ver. N3.03 NFS
ANW: Novell 286ANW ver. 2.15 rev a
The 2nd ESDI drive is another CDC 94166-182. I could do a low level
format using DEBUG or DMN. I would perfer using DMN.

ANSWER:

Thanks for the further info about your system. As for your
question about whether the drive was inited thru DMN or debug, if
within the configuration menu of DMN, it says that the disk paramters
are REPLACED, it was formatted thru DMN (Not good in this case, I'll
explain in a moment). If it says that the disk parameters are STANDARD,
then it was formatted thru debug. With this controller you need to
leave the BIOS enabled when using NetWare. Whenever the BIOS is
enabled, you should be formatting thru debug. If you format thru debug,
the drive appears as STANDARD to the system. Also, this version of DMN
would not be able to handle a nonstandard drive under 2.15 Rev A. If
you actually have 2.15 (no rev letter) then a nonstandard drive would be
OK, but 2.15 Rev A needs at least v3.01 DMN.
So... taking all of the above into consideration, I would suggest the
following procedure:

1. Backup your first drive.
2. Down the server and make the hardware connections for the second
drive.
3. Boot the machine and run your setup program, setting your CMOS drive
type for the second drive to drive type 1, as you said the first
drive was already set to. (This is also another indicator that the
first drive was formatted thru debug, and the controller BIOS is
doing a parameter replacement at boot time, making this drive appear
as a STANDARD drive.)
4. Reboot the machine. (It'll take a while to time-out on the second
drive and give you a drive 2 error. This is normal because the
drive hasn't been formatted yet.)
5. Run debug. At the hyphen prompt type in G=C800:5 and you should
then get a low-level formatting menu on the screen. At this point
you will need to select the second drive, low-level format it and
enter the bad tracks as per the controller card's instructions.
6. Boot with DOS and run DMN in manual mode. (Just type in DMN.)
7. Go to the configuration menu and make sure both drives are appearing
as STANDARD, and with the correct parameters. If they are not, then
something didn't go right from within debug. You will have to go
back and fix whatever is wrong. Under no circumstance should you
attempt to select this drive from within DMN. In doing so, you
would be telling DMN that you were doing a nonstandard install.
This step has to be satisfied before proceeding.
8. Return to the Main Menu and then go to the Partitioning Menu. When
it asks you "which drive?" tell it drive number 2.
9. Create your NetWare partition on drive number 2.
10. Return to the preparation menu and SCAN the partition that you have
just created in the last step.
11. After at least one pass of the scan, exit DMN and proceed to your
NETGEN to tell NetWare of the existance of the second drive. At
this point, it will appear to NETGEN as though the drive has been
COMPSURFed, and you should follow the normal Novell instructions on
installing a second drive (except for compsurfing it).
-------------------------------------------------------------------------
QUESTION:

I have a 150 meg Miniscribe ESDI hd w/ a Western Digital Card that I
installed with your software... using the diskmanager, I recently
re-initialized my logical drive F:, but am now unable to format it. I am
currently using DOS 4.01. Am I screwed, or is there a way to format just
this partition (I thought I could before starting this)... Thanks!

ANSWER:

Do you mean that you actually re-initialized (low level formatted)
just your third partition on this disk? If so, how was this drive
formmated in the first place? Thru debug using the controller BIOS low
level format routines, or thru DIsk Manager? At what sector per track
value? If you can't just come into Disk Manager in manual mode, (DM/M),
go to the partitioning menu, answer N to "Modify Partition?" question,
select P to prepare a partition, answer N to "all partitions?" question,
and select partition #3 and get it prepared (formatted), then you may
have messed something up to the point where you'll have to reinitialize
the whole drive. Either leave me more detailed information, or call us
on the tech support line so we can talk in real-time.
-------------------------------------------------------------------------
QUESTION:

I will have literally thousands of small files (ie, less than 1k) on my
drive. What would be the best way to format/partition the drive so that
there is as little wasted space as possible? I am willing to trade some
performance for some capacity.

ANSWER:

A partition greater than 32 Meg will have a larger cluster size. A
cluster is your minimum allocation unit of space on the drive. So for
example, if you have a 30 Meg partition on the drive, your cluster size
is 2k. Therefore, any file on the drive takes up at least 2k. If it's
a 3k file, it would take up 4k of space. If, on the other hand, you
have a 150 Meg partition on the drive, you would end up with a 16k
cluster size. This could leave you with alot of slack space on the
drive. (Each file would take up at least 16k of space.) If you wish to
keep your cluster size down to 2k (that's the minimum), then keep your
partition sizes between 17 and 32 Meg. (If you go below 17 Meg in size,
you'll get a 4k partition.) The break points are as follows (under DOS
3.x):

1 - 16 Meg = 4k cluster size
16 - 32 Meg = 2k cluster size
32 - 64 Meg = 4k cluster size
64 -128 Meg = 8k cluster size
128-256 Meg = 16k cluster size
256-512 Meg = 32k cluster size

Disk Manager also has a /V switch available that gives you some manual
controller over your cluster sizes. (You can make them smaller than the
defaults shown above.) But if you do, beware that there are some
programs that won't deal with a nonstandard cluster size; starting with
CHKDSK.
-------------------------------------------------------------------------
QUESTION:

USING DMN VERSION 3.10
PROBLEM: WE HAVE TRIED TO INSTALL 8760E MAXTOR 676MEG DRIVE TO RUN
UNDER NOVELL SFT 2.15 RUNNING ON A ARC TURBO 12 AT COMPTER.
WE HAVE DONE THE FOLLOWING:
(1) INITIALIZED THE DRIVE WITH DEBUG - COMPLETED
(2) MAPPED OUT BAD SECTORS
(3) THEN RAN DMN IN MANUAL MODE AND RAN (F6) & (F7)
(4) THEN RAN NOVELL INSTALLATION COMPLETELY
(5) UPON BOOTING UP THE NOVELL NETWORK FROM FLOPPY WE GET THE
ERROR "NOT ENOUGH MEMORY FOR DISK FAT TABLES"
PLEASE LEAVE MESSAGE AS TO ANY TIPS THAT CAN HELP ME WITH MY PROBLEM

ANSWER:


Sounds like your install procedure was correct, (as long as you
didn't modify the partition table in any way with NETGEN). You should
be able to boot with a DOS floppy and run NET$OS and have this thing
run. One thing you didn't tell me is how much memory you really have.
The following is an excerpt from the Novell manual:

NOT ENOUGH MEMORY FOR DISK FAT TABLES.
Occurs on: 86 and 286. This is an Abend error.
Meaning: This error indicates that the file server required more
memory to store the File Allocation Tables (FAT) for all
mounted disk volumes than was available. NetWare requires 1
kilobyte of memory for each megabyte of mounted disk space
for storing the FAT.

Cause: The file server does not have enough dynamic memory
available to it or the server is attempting to mount more disk
space (volumes) than it can support with the memory that it
has. This error could also be generated as a result of a
failure in high memory which would cause the file server to
think that it has less memory installed.

Solution: Make sure that the file server has plenty of memory
available for FAT storage in addition to the memory used by
other portions of NetWare. Check that the file server's
memory is good. If no problems are found, then add more
memory to the file server.


Does your server have as much memory as suggested above? If not, then
that's obviously the problem. If it does have more than enough memory,
then let me know how much and we can go from there...
-------------------------------------------------------------------------
QUESTION:

I am installing two Imprimis WREN VI ESDI drives: a model 94196 766 Meg
drive and a model 94246-383 383 Meg drive. With the 383 meg drive I am
using a WD1007V-SE1 controller; with the 766 Meg drive I am using an
Adaptec ACB-2322B-8 controller. Both drives came jumpered for 54 spt.
The Disk Manager shipped with the drives showed the drives as 53 spt
devices. I used the WD1007 BIOS to format the 383 meg drive. It was
After formatting the drive there were five choices for typing: I
selected the last choice, "Special non-translation." Other choices
were: No drive present; Non-translation; Translation - 17 SPT; and
Translation - 63 SPT.
I then used dm /m to modify and write the drive configuration for 54
spt. The drive drive does partition and boot.
I have not yet formatted the 766 Meg drive with the Adaptec controller.
.
My questions are:
1. Did I use the best procedure?
2. Why are the WREN VI drives listed as 53 SPT devices rather than 54?
3. Are there advantages or disadvantages to using the 17 SPT or 63 SPT
translation modes on the controller? My understanding is that if
translation mode is used, some Int 13 calls may not work. If so, do I
get around this by typing the drive as special non-translation and using
dmdrvr.bin?
.
Thanks for you help. I find the topics of translation and INT 13 calls
with ESDI drives confusing and will welcome ANY insight.

ANSWER:

First of all, it has been confirmed by Western Digital that
the 1007V series will NOT support a 54 sector per track drive at a 1:1
interleave. Therefore I would suggest the following procedure:

1. Jumper the 94246-383 for 53 spt (a jumper on the drive it's self)
2. Low level format the drive thru debug and then enter the bad tracks
3. Select 63 spt translation mode. (I have no knowledge of certain
INT13 calls not working under this translation mode, and it makes the
drive appear to have less than 1024 cyls, thus negating the need for
SWBIOS which could have a conflict with some programs.)
4. Run DM in manual mode (DM/M) and go DIRECTLY to the partitioning
menu. DO NOT even enter into the configuration or initialization
menus at all.
5. Create your partitions and prepare them.
6. Alternatively, you could just use your DOS FDISK program to partition
the drive, instead of using DM in steps 4 & 5 above.

The ACB2322B-8 & 94196-766 procedure would be a little different. With
this drive you will have to use SWBIOS, because even at 63 spt mode, the
drive still has greater than 1024 cyls. The ACB2322B-8 doesn't have a
problem with 54spt drives at a 1:1 interleave that I know of, so the
procedure should be as follows:

1. Low level format the drive thru debug and enter the bad tracks.
2. Leave it set to physical parameters of 1630 x 15 x 54
3. Boot with DOS and run SWBIOS.
4. Run DM in manual mode and go to the configuration menu.
5. DO NOT select a drive or change CMOS, simply confirm that the
parameters of the drive are showing as 1628 x 15 x 54, and then
press W to write the configuration information to the disk.
6. Press R to return to the main menu, then press P to go to the
partitioning menu.
7. Create your partitions and prepare them. No matter what version of
DOS you are using, the only thing that can reside above cylinder
1024 is a write/read partition.
-------------------------------------------------------------------------
QUESTION:

I have a WELLS AMERICAN AT compatible, DOS 3.1, ONTRAK for IMPRIMIS, a
CDC 150MB hard drive, and a WD1007V-SE2 ESDI controller. I've disabled
the controller BIOS and Sector translation on the controller card by
jumpering pins 1-2 on W3 and pins 5-6 on W1. I'm trying to create a 2MB
and a 148MB partition. Everything looks fine when I'm all done, but as
I use the system things start falling apart. Yesterday the same cluster
got used both for a directory listing and a file (or something like
that). I've got version 4.0 of the ONTRAK software. The only thing I
don't like about the installation is that the specifications on the
drive say there are 36 sectors per track but your software says 34
sectors per track. HELP!!! (3 days!) :->

ANSWER:

The reason DM says 34 spt and the drive specs say 36 spt, is that
34 will always work, no matter what the drive is jumpered for. If you
are certain that the drive is jumpered for 36 spt, then go ahead and
tell DM 36 spt instead of 34. As far as the corruption that you are
getting, it sounds like an application that you are using may not
like anything other than 512 byte sectors. (Your 148 Meg partition under
DOS 3.x would have a 4k sector size.) Actually, it could be the DOS
it's self. You could try using true MS-DOS v3.3 to rule out that
possiblity. If you are still having problems with corruption on the
drive, then you might try an alternative approach (as follows):

1. Re-enable both the BIOS & translation on the controller.
2. Low-level format the drive thru the controller BIOS using debug, and
then enter the bad tracks.
3. Run DM in manual mode (DM/M) and go DIRECTLY to the partitioning
menu to create and prepare your partitions.

Using this method, the drive would appear as a STANDARD drive to the
system, and could get rid of some potential problems. If you have tried
all of the above and are STILL having corruption problems, then try
partitioning it with DOS 3.3 or 4.01 alone, without DM and see what
happens. If once again the corruption occurs, then you either have a
hardware malfunction, or one of your applications just plain doesn't
work.
-------------------------------------------------------------------------
QUESTION:

Computer is IBM PS/2 model 80-111; am trying to install a CDC WREN III
DRIVE Model 94166-182 as second system drive: it has Drive Selector Pin
#2 enabled. Upon Boot-up receive error code 10491 and computer freezes.
I have Executed IBM PS/2 reference disk (IBM-RD), used its test
procedure and IBM-RD recognized 2nd drive as 147 MB ESDI, however,
could not complete the tests as it was not properly formatted. After
IBM-RD tests computer re-booted and froze again.
I then attached a SPST switch to drive selector pin 2 and disabled
it until after new boot-up. Computer worked properly booting on IBM's
drive C:. Then enabled 2nd drive and executed ONTRACK Disk Manger (O-DM)
- but it did not recognize 2nd drive. Then tried to change CMOS ROM
however O-DM said:"BIOS disk drive parameter table unlocatable when
setup for No hard disks". (Note that IBM-RD recognized two ESDI hard
drives as being present.) I have also disabled disk controller ROM but
then O-DM doesn't work on anything at all.
At any rate, even if I could change CMOS ROM, I have no idea of what
the "Drive Type Code" for an IBM 115 MB hard drive is anyway...
It seems that there is a conflict between Model 80-111 and Ontrack
Disk Manager. I can't boot with new drive enabled and I can't get
O-DM to work on it after enabling it or on CMOS ROM. SO how'm I
going to do a low-level format on this new drive so I can use it?

ANSWER:

When installing a ESDI drive in a PS/2, CMOS drive tables are
not used at all. Your first procedure was correct, except you need to
press F1 to get by the 10491 error (even though it doesn't say this on
the screen). Quite frankly, you are thinking about this too much! If
you simply remove the terminating resister pack from the drive, jumper
drive select #2, connect it to the middle of the cable, connect the
narrow data cable for drive #2, boot the machine, press F1 at the 14091
error message, run DM, select drive #2, initialize, partition and
prepare it, it will work. Messing around with SPST switches (!), CMOS
drive types and the reference diskette will only cause you problems.
It's not nearly that complicated.
-------------------------------------------------------------------------
QUESTION:

I am preparing to do an ELSII install over the coming weekend and
thought I might contact you prior to head off any problems I may
run into. Using DMN with Wren III and Adaptec 2322B
controller. I am aware of the need to do a low level format and
leaving the BIOS enabled. What questions might I expect during the
low level format with debug and what are the proper responses for
this setup. I also know that I still must run MODELSII. If you can
think of any other problems I might come across let me know.

ANSWER:

Sounds like you've got the right ideas here. As far as
information on specific questions and responces from within the
controller BIOS's low-level format routine, you'll need to contact
Adaptec at (408)945-2550 and let them support their own product. In
general, the procedure is as follows:

1. Leave the controller BIOS enabled.
2. Set CMOS to drive type 1.
3. Boot the machine and run debug.
4. At the hyphen prompt enter G=C800:5
5. Low-level the drive and enter the bad tracks.
6. Select the "special options" menu and enable 63 spt translation.
7. Exit the low-level routine and reboot with DOS.
8. Run DMN in manual mode (DMN/M) and confirm that the drive is
appearing as a STANDARD drive with 63 sectors per track. If it
isn't, then press F3 followed by F10 to select STANDARD parameters,
write them to the disk when asked, exit DMN, reboot, run DMN/M and
then the parameters should be showing as STANDARD with 63 spt.
9. Press F6 to create the NetWare partition.
10. Press F7 to prepare the partition. (ESC after at least 1 pass).
11. Exit DMN and run MODELSII anyway as if this were a nonstandard
install (because of the Adaptec board).
12. Proceed with your ELSGEN

QUESTION:

Hi, I did the install Sat. but all did not go as I expected. The
drive is a Wren model 94216-106 which has 1024 cyls. You directed
me to specify 63 spt translation which I couldn't because the drive
did not contain over 1024 cyls. I continued with the default
setting of 36 sectors and was able to initialize and install the
drive. However when I was checking out the drive in FCONSOLE I
noticed that the HOT FIX table size was 717 with 483 remaining.
I had entered 42 bad block flaws during the intialization. Does
483 remaining blocks for HOT FIX sound right? I'm not sure if
I feel confortable with this install, especially since the client
is 75 miles away and I don't need unexpected problems poping up.
Yeah, I know, then why am I playing with networks .
Let me know what you think. Thanks

ANSWER:

Sorry about the confusion about the 63spt mode. When you say
"Wren III", it's kind of like saying "a station wagon". I falsely
assumed that the drive was one of the Wren III's with greater than 1024
cylinders. As far as the questions about the hotfix area, here's what
happens:

When you enter a known defect at initialization time, (either thru the
controller's low-level routine or thru DMN's flaw map), every sector on
that track gets marked as defective in it's own individual sector
header, (unless you are using BFI mapping, in which case only one or
two sectors get marked as bad per defect). When you prepare the
NetWare partition either thru DMN or COMPSURF, and it hits either a
sector that is marked as bad in it's header, or one the fails the
write/read/compare test, the block that that sector belongs to gets
added to the NetWare bad block table on the drive. At NetWare
installation time, NetWare reads this bad block table and flags the
corresponding entries in the FAT table with a special code (FFFE). The
bad block table entries also get entered into the redirect table and
alternate blocks in the redirect area get assigned.

From that point on, the bad block table is never looked at again. When
NetWare writes to the disk, it first checks the redirect table to see
if the block that it wishes to write on has been redirected. If it
has, then it instead writes on the block that has been allocated as the
alternate. If NetWare encounters an error when trying to write to a
block that hasn't been previously redirected, it will place FFFE in the
corresponding FAT entry, add that block to the redirect table, and
assign an alternate block in the redirect area. Under current versions
of NetWare, if this redirect table ever gets completely full, then
NetWare will shut off hotfix and revert back to the method of handling
errors that it used before the days of hotfix, (just checking the FAT
table for the FFFE code, and if found, simply skipping to the next
block instead).

The formula for figuring out how many bad blocks you are actually
mapping out in a given install is as follows: Number of sectors per
track plus 7 (for the possibility of blocks at the beginning and end of
each track overlapping onto adjacent tracks) divided by 8 sectors per
block (round up the result) equals number of blocks mapped out per
flawed track, worst case.

You have a 35spt drive. (35spt+7)/8 = 5.25 (round up to 6). So in
your case, each flagged track could take out up to 6 blocks. You have
a total of 42 flagged tracks, so 42 x 6 blocks per track = 252 total
deallocated blocks (worst case). You said that FCONSOLE stated that
you have 717 total redirect entries with 483 remaining. This would
mean that you have mapped out 234 blocks as bad (717-483=234). 234
blocks being mapped out is under the worst case number of 252.

So, when all is said and done, having 234 redirect entries used right
out of the starting gate makes absolute sense. Sorry about the
confusion at first, and I hope this clears up your questions and
concerns.
-------------------------------------------------------------------------
QUESTION:

DMN v2.43
First, thank you for your rather complete installation instructions
concerning setting up a 2nd ESDI drive.
I formated the CDC 94166-182 with the DTC 6280 controller under debug.
However upon entering DMN I (contrary to your very clear instructions)
selected the non-standard disk parms. I completed the DMN scan and
partiting processes and sucessfully NETGEN'd. Everything seems to be
working well.
When I re-enter DMN and select C)onfigure, the screen displays the
follwing:

Disk Parameters ( ... ) are now REPLACED

Drive 1 is set up as 967 by 9 by 34 and is actually 967 by 9 by 34
Drive 2 is set up as 967 by 9 by 34 and is actually 967 by 9 by 34

Obviously I have set up the drives as non-standard. My question is that
given that your instructions were very specific to set up the drives as
STANDARD and somehow I set them up as NON-STANDARD and everything seems
to be running well (very well in fact) what can be (or should be) done
at
this point? If it works, should I fix it? I look forward to your
comments.

ANSWER:

I guess my motto has always been, "If it works, don't fix it".
The fact that your nonstandard parameters are exactly the same as the
standard parameters, (both 967 x 9 x 34), is allowing this to work. In
the future, if initing thru the controller BIOS, you should make sure
you don't then select non-standard parameters.
-------------------------------------------------------------------------
QUESTION:

I am attempting to install a Micropolis 338mb model 1558 ESDI into a NEC
PowerMate 386/20 with a Western Digital WD-1007 WA2 controller w/ no
bios. My Disk Manager-N is version 3.10. What
steps must I take to install this drive for Novell Netware ELS Level I
version 2.12?

ANSWER:

The BIOSless version of the WD1007A that NEC (and others) use
presents a problem. The controller automatically locks into 63spt
translation mode when using a drive with greater than 1024 cylinders,
but doesn't have a BIOS to inform the system that it is using these
parameters. Fortunately, I believe the NEC has a CMOS drive type that
supports the translation mode directly with aprox 600 cyls by 16 heads
by 63 sectors per track. You will need to set your CMOS drive type to
this entry, low-level format the drive with the NEC supplied low-level
format utility, then use DMN to create and prepare the NetWare
partition only (as a STANDARD drive with the above 63spt parameters).
No patches to NetWare would be needed as you are installing a standard
drive.
-------------------------------------------------------------------------
QUESTION:

I am having trouble formating a 668 MB Imprimis hard drive on a Everex
step 25 pc
386 pc with AMI bios. The controller is a WD1007WA2 with translation
on. Am using debug for low level format, with the following parameters,
1630 cyl, 15hd, 35 secs. Low level goes ok. With using DM (4.02) I
invoke swbios, then choose non standard pars. I change secs from 34 to
35. At partioning menu 1st partion test ok (2MB) on the start of 2nd
partion (rest of disk space), I get "bad track near start of partion"
error msg with drive just clicking away. Have tried 34 secs instead of
35 and differant size 1st partions, no change. Any help on the correct
way to format this drive would be appricated.

ANSWER:

Your main problem here is that the WD1007A is a 10 Mbit
controller, and the Imprimis 94196-766 drive will require a 15 Mbit
controller such as the WD1007V. The 1007A won't handle this drive.
The drive has 53 sectors per track, not 34 or 35 as you are trying to
set it up as.
-------------------------------------------------------------------------
QUESTION:

PLEASE REFER TO OUR DM-N QUESTIONAIRE TEXT FOR ENVIRONMENT ETC.
WE RECEIVE THE ERROR "INVALID FILE A:ESDI.OBJ" WHEN RUN AGAINST THE
FILE THAT CAME WITH THE LATEST 286 2.15 ESDI.OBJ 5229 8-30-89 11:06a
WEDO NOT RECEIVE THE ERROR ON OLDER ESDI.OBJ 4994 11-11-88 3:45p
WHICH CAME ON ANOTHER 2.15 SYSTEM I INSTALLED LAST DECEMBER. WE ALSO
ARE UNCLEAR ON WHAT MIGHT PRODUCE THE ERROR IN THE NOVELL WHEN WE TRY
TO SELECT THE DISK DRIVERS. WE DO NOT HAVE A VADD AND DISK-MANAGER N
SEEMS TO HAVE ENDED WITHOUT ERROR FOLLOWING THE 10 MINUTE SCAN. WHEN WE
DO THE NETGEN AND SELECT DRIVE, WE GET AN ERROR FOLLOWING THE INSERTION
OF DSK_DRV_001 THAT STATES: "NO DRIVERS MATCHING "*.DSK" COULD BE FOUND."
SEVERITY "THE CURRENT OPERATION CANNOT BE COMPLETED" NETGEN WILL NOT
CONTINUE WITHOUT THE SPECIFICATION OF A DISK DRIVER. CAN YOU PLEASE
ADVISE ME ON HOW TO PROCEED? I HAVE BEEN INSTALLING NOVELL FOR 5+ YEARS
FROM 1.02a UP TO 386 AND HAVE NEVER EXPERIENCED ANYTHING LIKE THE PROBLEM
ABOVE. AS A BRIEF OUTLINE THE SYSTEM WAS PREPARED BY SELECTING DRIVE TYPE 1,
DEBUG G=C800:5 WD LOW LEVEL FORMAT, DMN PARTITION DEFINITION AND
PREPARATION, MODDRVR PRODUCING AN ERROR ON FILE SHIPPED, BUT NOT ON
EARLIER ESDI.OBJ FILE. THEN NETGEN. THE ERROR OCCURED THERE.

ANSWER:

By the information in your DMN registration questionnaire, I
see that you are installing a 1558 with a 1007A on a 286 under 2.15D.
When you initialize the drive thru debug with this combination of
hardware, you have the ability to invoke a 63spt translation mode (so
you will be able to boot the server directly from the hard disk instead
of a boot floppy), and the drive will appear as a STANDARD drive to the
system, (therefore no MODDRVR is needed at all). The procedure should
be as follows:

1. Enter DMN in manual mode, (DMN/M).
2. If the parameters appear as STANDARD 6XX x 16 x 63, then goto step 5
3. If not, then exit DMN, run DEBUG, enter G=C800:5, select the 63spt
translation mode, and exit.
4. Enter DMN/M and press F3 for parameter selection, then F10 for
standard parameters.
5. Press F6 to create the NetWare partition, and F7 to prepare that
partition.
6. Exit DMN and proceed with a completely standard, UNPATCHED NetWare
installation.

The driver that you mentioned (ESDI.OBJ) is for PS/2 installations
ONLY, therefore patching it becomes a moot point. You should be
linking in ISADISK.OBJ if you are setting up a 286 based server, and if
you follow the above procedure, you won't be running MODDRVR at all.
-------------------------------------------------------------------------
QUESTION:

HELP. I AM A DEALER WHO WAS INSTALLING A 2ND ESDI MICROPOLIS HD ON A
SERVER WITH A SIMILAR DRIVE. NOT IDENTICAL. BOTH MICROPOLIS. FIRST IS
FULL HEIGHT MODEL, SECOND IS THE HALF HEIGHT 161 MEG MODEL. WHEN WE RAN
THE DM-N THAT CAME WITH THE FIRST MICROPOLIS, AND WHICH WE BELIEVE WE
USED TO SET UP THE SERVER LAST YEAR, THE DRIVE TYPE CAM UP WITH THE
TRANSLATED SCHEME AND CLAIMED TO BE BAD. WE RERAN DMN IN MANUAL MODE
AND SELECTED THE DRIVE FROM THE LIST AND ALL SEEMED TO GO O.K. BUT WHEN
WE WENT INTO NETWARE TO PROCEED WITH INWSTALLATION WE RECEIVED AN
ABEND! THE ABEND IS THAT THE ROM TABLE IS INCOMPATIBLE WITH AN AT TYPE.
MAJOR PROBLEM: IS THAT EVEN AFTER DISCONNECTING THE SECOND DRIVE, THE
SERVER WILL NO LONGER BOOT BUT RETURNS THE SAME MESSAGE! WE DO NOT
KNOW HOW TO GET AROUND THIS OR WHY IT HAPPENED. DID DMN RESET THE BPB
WRITTEN ON THE NOVELL 1ST DRIVE. BUT IF SO, HOW DO WE RECOVER THE
DRIVE?
THIS IS USING THE DMN SUPPLIED WITH THE MICROPOLIS. IT IS VERSION 2.44
AND WAS USED TO SET UP DRIVE 1 LAST YEAR.

ANSWER:

You stated that you are using v2.44 DMN. This is a scaled
down version of DMN that was manufactured for Micropolis and cannot do
a nonstandard drive install under 2.1x NetWare, (hence the "invalid ROM
parameter table" abend as soon as you selected nonstandard parameters).
The version of DMN also cannot install a drive with greater than 1024
cylinders, so you will have to use the translated mode of the
controller if the drive is over 1024 cylinders. The only way to
install ESDI drives under this version of DMN is to initialize them
thru the controller BIOS using debug in order to make the drives appear
as standard and with less than 1024 cyls. You made the statement, "The
drive type came up with the translated scheme and claimed to be bad".
Which drive are you speaking of, and what do you mean by "claimed to be
bad"? If it was the first drive, and you changed the parameters, it
would indeed stop working. The reason you can't go back to just the
first drive is that you have now changed the parameters of that drive
to nonstandard. If you now go into DMN and select standard parameters
for that drive and then write the configuration to disk, it should come
back for you.

In general, there are two basic methods of installing ESDI drives. The
first is to leave the BIOS and translation on the controller enabled
and to initialize (low level format) the drive thru the controller BIOS
using debug. You must use the 63 spt translation mode if the drive has
more than 1024 cylinders. You would then use DMN to partition and
prepare (scan) the partitions only, (as STANDARD drives). Method
number two would be to first of all disable both the BIOS and
translation on the controller and install both drives as nonstandard
drives with DMN. Using this method, you would have to run the MODDRVR
program to modify your NetWare to accept a nonstandard drive. Your
version of DMN doesn't have the MODDRVR program, so the second method
is out of the question for you.

Installing a second drive under DMN and NetWare is not a problem, if
you follow the above rules, and not try to change the parameters of a
drive that is already partitioned and running. With this scenario, you
should have physically cabled up the second drive, inited it thru the
controller BIOS using debug, (envoking 63spt translation mode if over
1024 cyliders), and then run DMN SETUP5 (to install a second standard
drive). Or if you rather, run DMN in manual mode, go DIRECTLY to the
partitioning menu for the second drive, and create and scan the
partition only.
-------------------------------------------------------------------------
QUESTION:

I'm a fairly savvy Network administrator who has to bring up a new
Novell server with big drives in it. I'm using a PS-2 model 80. I have
two Imprimis Wren VI half height ESDI drives, unformatted capacity 383
MB, formatted capacity 330 MB. I have physically installed one of the
drives and disconnected the original IBM 80 MB drive but I can't get the
PS-2 to boot from a floppy now and I don't know how to proceed. If I
can't come up in DOS it doesn't seem that I'll be able to run DM-N. I'm
unsure of what drive select position I should use with both drives
present. I'd appreciate your holding my hand with as much detail as you
can spare the time to present in how I go about physically, electrically
and logically bringing these drives up in this @#$%^& MCA machine under
Netware 2.15c. Thanks a million.

ANSWER:

The first problem that you will have is that the IBM ESDI
controller is a 10Mbit controller. For the 94246-383, (1747x7x53), you
will need a 15Mbit controller, such as the WD1007V-MC1. It's not going
to work until you swap out the controller. Beyond that, you should
have both drives set for 53spt and set to drive select position 2. The
first time that you attempt to boot with DOS after installing these
drives, it will take the PS/2 a LOOOOONG time to boot. It will
eventually error out with a 10490 or 10491 error, (because the drives
haven't been initialized yet). After getting the error, you must press
the F1 key to go ahead and boot, (even though it doesn't say that on
the screen). It should then boot from your DOS floppy, and you will be
able to run DMN to format, partition, and prepare the drives for
NetWare. If you get this far and have more questions, feel free to
leave them here and I'll do my best to answer them.
-------------------------------------------------------------------------
QUESTION:

On the MAXSTOR DRIVE THE GEOMETRY IS 1221 X 7 WHICH THE WD1007
(TRANSLATION ENABLED) MAKES 949 X 9. DMN WILL NOT ALLOW FLAW ENTRIES
BEYOND 949 WHICH, IT SEEMS TO ME, MEANS THAT THE FLAW TABLE IS NOT BEING
SET CORRECTLY FOR FACTORY FLAWS. THE ATT 6386 HAS THE DRIVE TABLE SET
TO 949 X 9 ALTHOUGH IT KNOWS ABOUT 1221 X 7 FOR FLAW PURPOSES.
WHAT SHOULD I DO FOR SETTING FLAWS?
IS THIS A "NON-STANDARD" DRIVE REQUIRING FLOPPY BOOT?
DO I NEED TO RUN ANY OF THE MODxxx ROUTINES ON THE NOVELL SW?
YOU SHOULD BE AWARE THAT THE NOVELL DISK DRIVER SELECTED IS
THE ONE NAMED ISA OR AT COMPATIBLE.

ANSWER:

The AT&T version of the WD1007 doesn't have a BIOS on it.
Translation is always enabled if you are using a drive with greater
than 1024 cylinders. For this reason, you must set your CMOS drive
type in the AT&T to the one that directly supports the translated
parameters of this drive. Then enter the bad tracks and initialize the
drive thru the AT&T low level formatting utility (which understands the
fact that these are translated parameters), not DMN. After that, the
drive will appear in DMN as a STANDARD drive with the translated
parameters and you will use DMN to create and prepare the NetWare
partition only. You will not be selecting a drive or entering the bad
tracks thru DMN. Following this procedure, you are installing a
"standard" drive that appears to the system as having less than 1024
cylinders, so no patches should be needed and you don't have to boot
from a floppy.

QUESTION:

Your advice is sound. The only remaining problem is that the exact
geometry entry is 48 in the table. This results in an abend in Netware:
Improper ROM table Entry since #48 is beyond C800. Novell suggested
using MODDRVR to remove the test for this bounds condition in the
Novell ISA driver, then using everything standard from that point.
I will be trying their method and yours. Will let you know.

ANSWER:

If you are getting the "Invalid ROM parameter table..."
message, then MODDRVR would take care of that. Let me know what
happens.
-------------------------------------------------------------------------
QUESTION:

We have a new CDC drive that came with DiskManager v4.02 IMPRIMIS.
We're running an 80386-25c with AMI BIOS and VLSI (not NEAT) chipset
and DOS v4.01 - we have three controllers to play with:

(1) CompuAdd HardCache /4Mb
(2) Western Digital WD1007V-SE2 with feature F301R SuperBIOS
(3) DTC 6280-15T .

The HardCache formats out to 504 Mb without DiskManager. We haven't
got the foggiest idea how to run DM with this board, and if we CAN, we
would like to get 630 Mb out of the big CDC. CompuAdd would probably
want to bundle DM with their HardCache's, because people who buy boards
like this are likely to be running GONZO drives.
The WD "translation" scheme calculates out to 504 Mb, too. (we
haven't tried it yet). Can we get full, physical capacity of the big
CDC by using DM ? How about a little sketch of that procedure?
The CDC "translation" uses 60 SPT and calculates out to 480 Mb (again,
we haven't tried it yet). How 'bout it? Can we get more than 480 Mb
out of the drive by using DM ?

ANSWER:

If the Compuadd card has the ability to pass the physical
parameters of this drive to the system, (1630x15x54 more than likely),
then you can install this drive as a standard drive with DM using
SWBIOS for access beyond the 1024 cylinder area. I know the 1007V has
this ability, but you would have to jumper the drive to 53 sectors per
track rather than 54 sectors per track (the 1007V can't handle a 54
sectors per track drive at a 1:1 interleave).

In general, whichever card you use, you will need to enable it's BIOS,
initialize (low-level format) the drive thru debug or it's own software
utility, and set it into it's physical mode, (no translation). Then
run SWBIOS followed by DM/M and go into the configuration menu. If the
parameters are appearing as STANDARD, 1630x15x54 (or 53), then you've
got it. Just press W to write the parameters to the disk, return to
the main menu, go to the partitioning menu and partition & prepare the
drive.

There is one other option with the WD1007V. You could disable both the
BIOS and translation on the card and set it up as a nonstandard drive
thru DM (using SWBIOS). You could either format it at 2:1 interleave
with 54 spt, or 1:1 interleave at 53 spt.

As you see, there are lots of ifs, ands, or buts with this type of
large drive ESDI installation. So if this information doesn't do it
for you, the best solution would be to call us directly on the tech
support line so we can talk face to face. (You should be in front of
your computer and be ready to experiment when you do this).
-------------------------------------------------------------------------
QUESTION:

Disk Manager is not reading the Flaw map prepared by the AT&T low level
format routine. It reports 0 Flaws on the Surface Analysis screen. Most
of the flaws are probably invisible due to alternate track formatting
which can not be avoided. There are flaws in the ECC and some where
two sectors on a track are bad and it is these that are not being read.
AT&T tech support told me that they thought that factory flaw table
read by DMN had to be invoked manually. I can not find this screen, if
it even exists.
DMN has fixed the problem of ABED: Improper ROM .. by using MODDRVR
on the IASDISK.OBJ. This has permitted complete loading of NetWare
and checkout. DMN surface test found 7 bad blocks but I would feel
more confidant if I coud get the flaws to be recognized. Thanks

ANSWER:

The fact that DMN found 7 bad blocks leads me to believe that
these are the extra flaws that you were speaking of. Alternate
sectoring does map out the known bad sectors and hide them from the
system. These will not show up in DMN as flaws, but are effectively
mapped out by the controller card. DMN does not read any factory flaw
table on the drive. What it does do though, is to perform an extremely
intense test on the media. If there were additional bad spots on the
disk, DMN would find them. Also, your drive is using a translated
geometry. When DMN does find a bad spot, the head and cylinder that it
displays will relate to the translated geometry, not the physical
geometry, and therefore could be confusing if you are looking to see if
specific bad spots have been mapped out.
There IS no flaw table either created by the AT&T initialization program,
or read by DMN in this procedure. When you enter known flaws at
initialization time, the header of each sector that is involved with that
flaw is marked with a code that flags that sector as bad. Later, when you
try to read that sector, (with any program), it will return a read error.
When you prepare a NetWare partition thru DMN, and it gets either a read
error, or a write/read/compare test failure, it will add the flaw to an
actual flaw table that it builds for NetWare. If you are getting reports of
different or more bad blocks on each successive pass of the preparation
of the partition thru DMN, this is a sign of an intermittent hardware
failure somewhere. Any number of things could be causing it, such as a
bad HD, controller, cables, motherboard, or even a flaky power supply.
It is normal for DMN to pick up a few bad blocks on the first or maybe
even second pass of the preparation of the NetWare partition, but if
you keep getting more and more bad blocks on each pass, you've got a
hardware problem of some kind.
-------------------------------------------------------------------------
QUESTION:

NEED HELP DETERMINING WHY NOVELL'S (VER. 2.15) HOT FIX REDIRECTION TABLE
KEEPS GETTING TURNED OFF. WE FORMATED THE SERVER'S HARD DISK WITH DMN
(VER 3.04) AND IT RAN FINE FOR ABOUT 3 MONTHS. BUT LATELY THE HOT FIX
REDIRECTION TABLE FEATURE IS GETTING TURNED OFF FREQUENTLY. WE ARE
WONDERING IF IT HAS ANYTHING TO DO WITH DMN.

ANSWER:

As far as I know, the only time that the hotfix feature turns
itself off, is if the hotfix table gets full. Once it turns itself
off, it will stay off, unless you reformat the drive and reinstall
NetWare. The only exception to this that I can think of is if you had
the situation where you had 511 entries in the table and you hit two
errors at once. I think in this case that NetWare may be seeing that
it can't make two more entries into the table, (limit is 512), so it
turns off hotfix. The next time you boot the server, you've still got
the one entry left, so hotfix comes back on until you hit more errors.
This theory is kind of a wild shot in the dark, but the fact remains
that the only time hotfix gets turned off is if the table is full. So
the next questions is why is the table full. The source of the problem
could be either flaky hardware, or an improper installation procedure.
As far as the proper installation procedure goes, you should download
the Q&A.EXE file from this BBS, (read bulletin #10 for more info).
Within the Q&A file, you will find the correct procedure for installing
this combination of hardware. I think the procedure is for installing
v2.0a NetWare, but the hardware initialization method is the same as it
would be for v2.15.
-------------------------------------------------------------------------
QUESTION:

I am currently running 3 partitions with a Maxtor 760MB ESDI drive.
DOS partition = 32mb
2 extended partitions approx. 200MB each, it seems the total I can
get is around only 525MB on the 760mb drive. Do you have a product that
will allow the remaining 100 or so MB to be accessed. I have set it up
with DM and its set for 16 heads, 1023 cyl. and 63 spt. On the wd1007v
the bios is enabled as is alternate sector functionality. I would like
to install Unix SYS. 5.3 on the remaining part of the drive if I could
get access to it.

ANSWER:

If you disable the BIOS and Translation (by placing jumpers on
W1 pins 5&6, and W3 pins 1&2) you should be able to set this up as a
nonstandard drive of 1630x15x53 through DM and get about 633 Meg (663
million) bytes of storage. This of course would rule out alternate
sectoring, because the BIOS is disabled. With the BIOS enabled, unless
you have the option from within the controller's low-level format
routine to set the parameters to 1630x15x52 (one less spt because of
alternate sectoring), you won't be able to get the full capacity of
this drive. Some versions of the 1007V were able to pass the physical
parameters of the drive directly to the system, while it seems that
later versions translate and truncate the parameters to 1024x16x63,
giving you only 503 Megabytes (528 million bytes) of storage.
-------------------------------------------------------------------------
QUESTION:

I've run into several problems while trying to upgrade a novell network.
I started with a IBM Model 80 with 110 meg drive using an ESDI
controller. The operating system was 2.12 SFT. The planned upgrade
included 4Meg on IBM extended memory board, 2.12 to 2.15 SFT, and the
addition of a CMS 320Meg Drive, plus a dual async adapter for more
printers. After completing the hardware and software installation, I
booted the server and the message "ERROR LOADING USER DRIVER" was
displayed followed by the machine entering basic. After consulting
with our vendor's support staff, I was informed there was a bug in
the Novell software that created a conflict on track zero of the hard
disk where both the boot information and the memory and fixed disk
information is kept. The vendor also said that if the reference
diskette was used after the boot track was installed by novell, that
the boot track would be corrupted. He suggested reformatting the
hard disk and running the installation in a specific order.
Enter the second problem.
The software sent with the CMS drive will not correctly format the
320 meg drive. The technical support from Ontrack informs me that
I need DM-N. Two questions. Will this software allow me to re- enter
the manufactures defect list and/or ready the drive to Compsurf specs?
Will this software resolve the conflicts with Novell's boot track or
is there a conflict? Is there a conflict with 2.15 Advanced Netware?

ANSWER:

Disk Manager-N would allow you to complete this installation. I am not
aware of any basic conflicts, but your vendor may know something about
the method CMS uses to format drives in a PS/2 that I don't. What Disk
Manager-N would allow you to do is to reformat the second disk and
integrate it into the PS/2 as if it were a standard drive to the
system. One recommendation though, I would suggest that you plan on
backing up the first drive and reinstalling them both with the new
NetWare. This is because DMN gives you access to one more head worth
of storage on the drive than NetWare alone does. When you are adding a
second ESDI drive to an existing first one on a PS/2, it is possible
that you could have a conflict because of this. If you backup the
first drive (it's silly not to in the first place), reinstall both with
DMN, then restore the first one from your backup, everything will work
fine.
-------------------------------------------------------------------------
QUESTION:

I have the following equipment:

a DTK 80286 10 Mhz computer with 1 MB on board
a BOCARAM AT with 1.5 MB of extended memory
MICROPOLIS 1355 ESDI drive with 150 MB, 1024 cyl, 8 head 34 spt
WD 1007A-WAH ESDI controller using jumper W3 (BIOS enabled)
Thomas Conrad 16bit turbo ARCNET adapter
Monocrome graphics adapter

I am using ONTRACK Novell version 3.10

My problem is:
I am installing NOVELL version 2.15 Rev. C and after preparing the drive
with ONTRACK, setting the computer BIOS to a drive type 1, I load the
operating systems without a flaw. When I reboot to load NOVELL, I get
the ONTRACK message and then this errror message:

ABEND: Invalid process ID passed by interrupt procedure to kernel.

So far I have done everything to check the memory, motherboard,
controller etc. and I still get the same error message. I even got
another similar machine to boot using another disk without a flaw !!!
Do I still need to modified the ISADISK.OBJ with MODDRVR ?
Last few attemps I had it modified, and still did not work.... I wonder
if it should be unmodified ?

ANSWER:

In general, with this combination of equipment, your best bet is to
low-level format the drive thru the controller BIOS using debug. Run
debug, and at the hyphen prompt type in G=C800:5 and press enter. This
should bring up a menu on the screen that allows you to low-level
format the drive and enter the bad tracks. After doing this, run DMN/M
and the drive should appear on the main menu screen of DMN as being a
STANDARD drive with 1022 cylinders, 8 heads, and 35 sectors per track.
At this point, all you do is press F6 to create the NetWare partition,
and the F7 to prepare that partition for NetWare. Under no
circumstance will you select the Micropolis 1355 from the configuration
menu. This will be a standard drive installation not a nonstandard
installation. Since you are using the Western Digital WD1007A, and are
formatting thru the controller BIOS, you won't have to run MODDRVR to
patch ISADISK.OBJ. After you prepare the partition, proceed with your
Netgen as normal.

An alternate method would be to disable both the BIOS and translation
on the WD1007A (by moving the jumper from W3 to W14) and low-level
format the drive as a non-standard drive thru Disk Manager-N. In this
case you would select the Micropolis 1355 from the configuration menu,
enter the bad tracks, low-level format (initialize) the drive, create
the partition, and prepare that partition, all through Disk Manager-N.
Using this method, you WOULD have to run MODDRVR prior to running
NETGEN, because you are installing a non-standard drive.

Also, another possible cause of the NetWare error message "ABEND -
INVALID PROCESS ID PASSED BY INTERRUPT PROCEDURE TO KERNAL" is
specifying ISADISK.OBJ twice during Netgen.
-------------------------------------------------------------------------
QUESTION:

I have a 676 Mb har disk. (Micopolis 1568-15) I need a driver for
OS/2 version 1.2 in order to access cylinders past 1024. Are you
working on a product???

ANSWER:

At this time we don't have a driver for OS/2, and there have been no
formal announcements as to whether or not we will be developing one in
the future.
-------------------------------------------------------------------------
QUESTION:

I am currently running 3 partitions with a Maxtor 760MB ESDI drive.
DOS partition = 32mb
2 extended partitions approx. 200MB each, it seems the total I can
get is around only 525MB on the 760mb drive. Do you have a product that
will allow the remaining 100 or so MB to be accessed. I have set it up
with DM and its set for 16 heads, 1023 cyl. and 63 spt. On the wd1007v
the bios is enabled as is alternate sector functionality. I would like
to install Unix SYS. 5.3 on the remaining part of the drive if I could
get access to it.

ANSWER:

If you disable the BIOS and Translation (by placing jumpers on
W1 pins 5&6, and W3 pins 1&2) you should be able to set this up as a
nonstandard drive of 1630x15x53 through DM and get about 633 Meg (663
million) bytes of storage. This of course would rule out alternate
sectoring, because the BIOS is disabled. With the BIOS enabled, unless
you have the option from within the controller's low-level format
routine to set the parameters to 1630x15x52 (one less spt because of
alternate sectoring), you won't be able to get the full capacity of
this drive. Some versions of the 1007V were able to pass the physical
parameters of the drive directly to the system, while it seems that
later versions translate and truncate the parameters to 1024x16x63,
giving you only 503 Meg (528 million) bytes of storage.

QUESTION:

Recently you answered a question regarding using the maximum capacity
of a Maxtor 8760E and the WD1007V-SE2. I appreciated the quick responce
to the question and the accuracy of the information. You stated that
with the BIOS disabled I would not be forced to use their translation.
I could set up this drive as a non-standard drive with DM. But I would
lose the alternate sector feature of the controller. I found that with
the use of WD's WDFMT utility I could disable the BIOS and still format
with a alternate sector per track. So my last question is the following:
If I do these steps listed below will DM be able to handle the
partitioning, 1) Disable WD on-board BIOS
2) Set my drive type (soft BIOS entry, type 48,49
to the actual physical parameters,
1632x54x15.
3) Format with WDFMT Utility (low-level)
So I can get my alternate sector feature.
4) Invoke DM to simply custom partion the drive.
Do you see any issues with this proceedure, all though I am currently
running in the translated mode and only using 528MB, I would like to get
the 678MB capacity of this drive AND use the alternate sector feature.

ANSWER:

Here are several problems/considerations that I see:

1. The WD1007V-SE2 can not handle a drive with 54 sectors per track at
a 1:1 interleave. (This has been confirmed by Western Digital).
If you plan on using a 1:1 interleave, you will need to jumper the
drive for 53 sectors per track, and enter the corresponding spt
value in your user defined drive type. Either that, or low-level
format the drive at 54 spt and an interleave of 2:1, if WDFMT gives
you that option.

2. If the drive parameters appear as STANDARD and greater than 1024
cylinders, SWBIOS will not automatically be invoked at boot time
unless you perform the following procedure. After formatting the
drive, run DM/M, go to the configuration menu, press N for
nonstandard parameters, and then ESC for standard parameters,
followed by W to write the parameters to the disk. This is known as
a nonstandard/standard configuration. After doing this, Disk
Manager will state (in the config menu) that the drive is setup as
1628x15x53, and is actually 1628x15x53 (both sets of parameters are
the same). After doing this, DMDRVR will automatically invoke
SWBIOS at boot time.

3. Some BIOS's have their own method of trying to handle a drive
greater than 1024 cylinders. If your BIOS is one of these and you
set your user defined drive type to 1630 cyls, the BIOS's method of
handling this may interfere with Disk Manager's SWBIOS.

Why, may I ask, is alternate sectoring so important to you? Unless
your operating system has a low bad block limit (such as NetWare), then
you'd be better off not using alternate sectoring, and in the process
would pick up an extra sector per track of storage capacity. If you
are installing DOS, my best/safest recommendation would still have to
be to disable the BIOS & translation on the controller and install the
8760E as a nonstandard drive with Disk Manager and forget about
alternate sectoring.

CLOSING COMMENTS:

The alternate sectoring is not that important, but does present a very
clean looking drive to DOS. The 8760E has approx. 125 bad tracks and are
spread throughout the disk. I felt that with this many I would be better
off using alternate sectoring, the space savings is not that much
considering, I'am forced to install Unix Sys. V root partition 30MB
before cyl. 1024. So the DOS piece of this drive is around 384MB
anyway. Also I have had issues in the past, especially with Word Perfect
when by chance I gets split over a bad sector after being installed on
the hard drive. All though I dont know for certain, I would assume the
alternate sectoring would increase performance, by what amount is a
guess. Seems I will just do it the way you suggested and bag the
creative alternatives, who knows what other problems could arise when
you have 4 O.S.'s installed and bootable on one hard drive. OS/2 is
also in this discussion, Thanks again.
-------------------------------------------------------------------------
QUESTION:

I will be installing a new ESDI drive on a server this weekend and
I wonder if you could give me the steps. The hardware is: Maxtor
XT 4380E, WD 1007 SE1, EMPAC 12MHZ 286 with AMI bios and 6 megs of RAM.
The LAN is Novell Advanced Netware 286 V2.12, I just bought Diskmanager
- N. I have done many large DOS hard disk installs but this is my first
install on Novell. I am taking out a Miniscribe 146 MEG ESDI and its
controller. (A tech at WD told me to use the new version of the 1007,
that's why I am installing a new controller also.) Should I use the
standard drive setup? The bios on the controller? Any help and
possible coveats would be greatly appreciated. Thanks in advance!

ANSWER:

The procedure for installing a Maxtor XT4380E & WD1007V-SE2 under
Advanced NetWare v2.12 is as follows:

1. After physically installing the drive and controller into the
machine, boot with DOS and run debug.

2. At the debug prompt, enter g=cc00:5 and hit return.

3. You will get a menu, and from the menu select the option to low-
level format the drive.

4. After low-level formatting the drive, select the option to enter the
bad tracks.

5. After entering the bad tracks, select the option from the menu that
says "SET DRIVE TYPE". After you select this option, you will be
given the opportunity to hit the + (plus) key and step through
several different drive geometries. When you reach the option with
approximately 600 cylinders, 16 heads, and 63 sectors per track, hit
return to accept it. This is a translation mode and makes the drive
appear to have less than 1024 cylinders, but a higher number of
heads and sectors per track.

6. Exit the debug routine, and let the machine reboot with DOS.

7. Run DMN/M and as soon as you're in, confirm that within the box on
the left side of the screen that the drive parameters are appearing
as STANDARD and that those parameters are approximately 600
cylinders, by 16 heads, by 63 sectors per track. If this is the
case, then so-far, so-good.

8. Select option F6 to create the NetWare partition, followed by option
F7 to prepare that partition. After one pass of the preparation, hit
ESC to stop the scanning.

9. Proceed with your NetWare installation as normal, except don't run
COMPSURF. Using this method, the drive is appearing as STANDARD to
the system, and as having aprox 600 cyls x 16 hds x 63 spt. Since
the controller is making the drive appear as a STANDARD drive, you
don't need to run the modification program MODDRVR, as you would if
you were installing a standard drive. Remember that your maximum
volume size under 2.12 adv NetWare is 255 Meg. You will have to
split this drive into 2 volumes.
-------------------------------------------------------------------------
QUESTION:

Hi. I'm running an AST 386-33 w/Imprimis 383H at work w/o DM and a
MBS 286-16 w/Micropolis 1355 at home. Both are running WD 1007A's.
I'm using DM at home that was shipped with the Micropolis PCpak, and
should be registered if you included form w/software. Let me first say
that no problems thus far have been encountered that could be attributed
to DM. Until now. See desc in Windows Survey. Many late nights on
this one. More background: a couple of weeks ago, I started getting
"Packed file is corrupt" when trying to invoke COMPRESS from PCTools 5.5
-blew a fresh copy in & no change. Then HOG (piechart of partition
hogs) did the same thing. All of this was on C:. OK, reformat C:, run
barebones & everything was OK, except that I couldn't access the rest
of the partitions without DM. Blew the root directory back in w/DM
& everything was OK. Great, I said to myself, everything's cool, back
to messing w/Windows 3.0. Surprise, surprise - COMPRESS, HOG, and now
WOLFPACK won't work. Hey, I can get alternative utilities, but GAMES??
... anyway the discussion over the past few nights was that if it was
a virus (despite all the anti-viral stuff I have) got into my system,
it has to be attached to DM. This after much messing around
reformatting C: and playing about. Never suspected Windows 3.0
until reading a UL of your Bulletin 12. Jeez, I guess I just don't
understand how it can mark executables like that. Anyway, I don't
need this. And I have a better monitor at home too...
.... Any suggestions as to restoring executables would be MUCH
appreciated ... or updated drivers or whatever. Running out of
time, will dial back in & browse for a bit.

ANSWER:

We have had other experiences of the message "Packed file is corrupt"
come up when executing DM.EXE. This was traced to a bug in the
MicroSoft C compiler EXEPAK program that we were using. It is
basically a load point problem. The work around was to load an
additional copy of the command interpreter by typing in COMMAND at the
C:> prompt, and then run DM.EXE afterwards. This would effectively
shift the load point of DM.EXE in memory, and alleviate the false error
message. Your problem is a little different though... You are not
trying to execute DM.EXE, but other programs. It could be the same
type of load point problem if these other programs are also packed with
EXEPAK. The possibility also exists that these other programs are
actually corrupt. What are you using for backup and restoring these
files in question? Is this backup/restore program compatible with
partitions greater than 32Meg and with a greater than 512 byte logical
sector size?

You mentioned two separate machines; one at home using DMDRVR, and one
at the office without DMDRVR. Are you having this problem on both
machines or just on the one at home? Did these problems not exist
before, and start immediately after attempting to install Windows?

Concerning a DMDRVR/Windows conflict; the 1355 has only 1024 cylinders,
so SWBIOS is not in the picture, therefore there would be no chance for
a SWBIOS/Windows conflict. Even if the drive had greater than 1024
cylinders, and there were a SWBIOS/Windows conflict, it would surface
as the machine hanging when trying to execute Windows in 386 enhanced
mode, and would have nothing to do with data corruption. The only other
conflict you will have is that you won't be able to create a permanent
swap file for Windows if DMDRVR.BIN appears in your CONFIG.SYS file.
The swapfile program simply refuses to proceed if it sees our driver.
Again, this is simply the inability to do something and in no way has
anything to do with data corruption. Doesn't sound like either of
these conflicts could be related to your problem.

If you would like to try a newer version of DMDRVR, you should be able
to contact our sales department at (800) 752-1333 and upgrade to the
most current version, but I don't think it's likely to be the cause of
your problems, (since everything used to work OK on your system).

I would suggest that for a test, you strip out everything in your
CONFIG.SYS file except DMDRVR, boot cleanly with no AUTOEXEC, and then
try to execute fresh copies of the programs in question. If that
works, then work backwards to find out where it trips. At that point,
try the load point workaround and see if it helps. If you can nail
this down a little more, we'd stand a better chance of figuring out
what's happening.

QUESTION:

Thanx much for the prompt reply. I guess I
wasn't really prepared to leave much of a lucid message last nite.
Anyway, in response to your concerns: Only the home machine (286) is
acting up. The info on the C compiler bug is interesting, no doubt
that may have something to do with the seemingly random nature of this
nasty bug. Here's a little quickie. Hard boot the machine from floppy
bare bones - no config.sys or autoexec.bat. Programs in question runs
just fine. Boot from C: bare bones - everything seems ok except no
access to partitions D thru G. Boot from C: with just DMDRVR installed,
got "Packed file is corrupt" again. I do not use unreadable or packed
-proprietary backup software. All backups are kept in ready-to-use
form, with duplicates both here at home & at work. Renamed the
config.sys, booted bare bone again, & programs work ok. With DMDRVR
installed, even overwriting the executables w/fresh copies does nothing,
still results in "Packed file..." error, moreover, it seems to be
affecting more executables.
The reason I think Windows 3.0 may be involved somehow is that this
whole phenomena started around the time I brought Windows home to
learn more about it, and appears to be getting worse.
Anyway, I am satisfied (for now) that the problem seems to be DMDRVR.
Will try the trick of another COMMAND shell, plus blow in a fresh
copy of DMDRVER - will try to log back in tomorrow & keep you posted,
maybe spend more time browsing thru the BBS.
For now, it's back to dinking around with the 286. Again, thanx for
you interest. If you think of anything else regarding this bug, please
post.

QUESTION:

The DMDRVR is definitely responsible for executables refusing to run.
Whether the controller, Windows 3.0, or whatever is causing it, I can't
accept continuing contamination (?!), as more and more EXE files are
being marked as "Packed file is corrupt". Perhaps a somewhat nasty
synergy between the MS C EXEPAK, DM 4.01, and Windows 386?? Why does
it proliferate? I have tried every virus scanner I have access to, and
none reports a virus. A combination of the above + hardware? Aaaargh.
Listen, what do think about using the 63 SPT and FDISK even though
the Micropolis 1355 has 1024 cylinders? All references have been for
drives > 1024. Any problems? I've done a byte by byte compare on
the DMDRVR BIN files and have not found any differences.
Recommendations?

ANSWER:

I have taken your concerns to our engineering department manager and he
has provided me with the following information about the EXEPAK load
point problem that causes the FALSE message "Packed file is corrupt".

1. Your .EXE files are NOT being corrupted or modified in any way. You
should be able to prove this by copying them from your hard disk to
a floppy and then over to another machine and have them work on that
machine.

2. The EXEPAK load point problem is universal. The companies that
write the programs that you are having problems with are also
evidently using the EXEPAK program. It does NOT have to do with
DMDRVR.BIN, Windows, your DOS version or hardware combination
specifically, (except possibly your BIOS), but is in fact a LOAD
POINT PROBLEM. When you load DMDRVR.BIN, you are shifting the load
point of all subsequent programs, the same as if you loaded any
other TSR or driver of the same size. You should be able to prove
this by booting with a clean system (no CONFIG or AUTOEXEC) and
playing with the buffers statement until you recreate the same
scenario.

3. We have not yet been able to absolutely pin down or reliably
recreate the scenario in which you will have the EXEPAK load point
problem. It has been extremely elusive so far. We had a machine
that seemed to reliably produce this message in our lab once, until
one day it just quit doing it. We have no idea why.

4. MicroSoft admits that there is a problem with EXEPAK, but their
stance is that EXEPAK is no longer being produced, updated, or
supported. There will be no fix from MicroSoft.

5. From what we have seen, the EXEPAK load point problem seems to be
linked with the Phoenix BIOS, although Phoenix states that they have
no knowledge of the problem.

6. Even if you take out DMDRVR.BIN and use FDISK instead, you will
probably have the same problem again eventually, depending on what
combination of DOS version, TSRs, and drivers that you use, (all of
which of course change the load point of subsequent programs).

7. The best chance you have for combating this problem is to fiddle
with your FILES and BUFFERS statements (to change the load point of
applications), or perhaps change your DOS or BIOS versions.

QUESTION:

Thanks for taking the time to pursue this slippery phenomena, and
especially the well-written recommendations. I should add that I
am using Quarterdeck's QRAM on the machine as well (instead of Windows
3.0's HIMEM). Since all this started happening, I have been thinking
of updating the Phoenix BIOS to a version that supports user-supplied
drive types & the C&T chipset. It's hard to believe that this problem
is not more prevalent... Thanks for your efforts... will have to
wrestle with this some more armed with your concise diagnosis of what
the probable cause is. ONTRACK has my highest regards. If I come up
with a workable solutiion, will post. Thanks again.
-------------------------------------------------------------------------
QUESTION:

I have run DMN on my drive CDC 94166-182 and it shows different
parameters than when I actually choose the drive in manual mode. The
test works fine with the initial parameters but with the CDC parameters
chosen DMN reports Track 0 bad disk unusable. When the test and
preparation process is complete, and I try to install Novell Adv286
V2.15C I get an Abend: Invalid ? sent to Kernel. I believe the problem
is with the SMS Omti Controller, in which the drive parameters seem to
be hard coded. Also of interest, is the fact that the CMOS must be set
to no hard disk in order for the Controller to work. I thought that DMN
would solve this problem and hope that it still might, but I seem to
need alittle guidance. Also I have no documentation for the controller
and would greatly appreciate any information on possibly finding this
company SMS/Omti. Thank you.

ANSWER:

According to my information, the OMTI 8620 is not NetWare compatible.
You will have to obtain a different controller before you'll be able to
run NetWare. DMN can't make a controller NetWare compatible.
-------------------------------------------------------------------------
QUESTION:

I have DM 4.02 generic. I also have a DTK 386/20, a Maxtor
XT-8760 ESDI drive, and a Konan FTL Times 10 ESDI controller. My BIOS id
Phoenix 386. I can sucessfully format the drive, but after it makes it
bootable and finishes, it won't boot up. I have tried using SWBIOS and
even making a special disk table entry with the correct parameters. It
simply refuses to boot. I know this drive is good as is the controller
and cables. Anything else you might suggest? Thanks.

ANSWER:

I am not familiar with the "Konan FTL times 10 ESDI controller". With
an ESDI controller, when in doubt low-level format thru the controller
BIOS using debug. Try running debug, and at the hyphen prompt type in
g=c800:5 to get to the controller BIOS low-level formatting routine.
(If this doesn't give you a menu to low-level format the drive, then
you'll have to call Konan for more explicit help on using their card).
If you can accomplish a low-level format thru the controller BIOS, the
drive should appear as a standard drive to the system after that. You
would then perform the following procedure. Run SWBIOS, then DM/M.
One inside DM, press C and hit enter for the configuration menu. The
parameters should be appearing as STANDARD and drive 1 should show as
being actually 1629 x 15 x 53. At this point press W to write these
standard parameters to the drive, (don't attempt to select the 8760
from within Disk Manager, we are doing a standard drive install here).
Return to the main menu, and then go to the partitioning menu to create
and prepare your partitions. If it still doesn't boot after that, then
try a FORMAT C:/S. If you have to do a FORMAT on it, you will then
have to copy the CONFIG.SYS file and the DMDRVR.BIN file back onto
drive C: afterwards. As I said before, this is a unknown controller to
me, so if you have further problems with this install, you would be
better off calling our tech support line at (612)937-2121 when you are
in front of the machine so we could do some experimenting.
-------------------------------------------------------------------------
QUESTION:

Hi - I'm just now installing this new computer, and it came with your
driver. The drive is a Maxtor XT-4170E (157 megs formatted, 1224 cyls,
8568 tracks, 36 sectors per track), and the controller is the Ultrastor
Ultra 12(F) cacheing ESDI controller (32k). It's all new to me.

I had already run into the invalid media problem, and I found your
fix from the bulletin. I just read about the Windows problems in
the MSWIN forum on CompuServe - your solution seems acceptable for now.

Anyway, here's some simple questions. 1) Why the limit of 132 megs
in a partition (I'm using DOS 4.01)? Why can't I have one 157 meg
partition? 2) I put dmdrvr.bin and xbios.ovl in a subdirectory, and
changed config.sys to find dmdrvr in the subdirectory. Will dmdrvr
still be able to find xbios.ovl? 3) Is this drive/controller
combination using sector translation? I don't pretend to really
understand translation, but I know that Spinrite II says it can't work
with translated drives. I suspect it will also have a problem with a
caching controller. I haven't tried yet, but do you think I'll be able
to run Spinrite II on this machine?

ANSWER:

The 132 Meg limit would appear to be when you bump up against the
1024th cylinder. No DOS partition, under any version of DOS, (that's
right, not even 4.01), can go beyond cylinder 1023. End of story.
The only thing that can reside out there is a Disk Manager write/read
partition. If you do the low-level format of this drive thru the
controller BIOS using debug, (at the debug hyphen prompt type in
g=c800:5), you should be able to invoke the controller's translation
mode to make this drive appear to have less than 1024 cylinders, but a
higher number of heads and sectors per track. This would negate the
need for XBIOS.OVL and the VIRTUALHDIRQ=OFF line in your SYSTEM.INI
file. You could have the whole drive as drive C:, and wouldn't need to
use DMDRVR.BIN at all. If you do leave it as it is, and use DMDRVR &
XBIOS, you must leave XBIOS.OVL in the root directory. DMDRVR.BIN
won't look for it in a subdirectory. As far as SPINRITE goes, it's my
understanding that they don't function correctly on a drive with
greater than 1024 cylinders, and if they can't function on a drive with
a translating controller either, then it looks like they just can't
function on this drive at all. But I don't absolutely know these
statements to be facts. You should really ask them about it directly.
-------------------------------------------------------------------------
QUESTION:

Hi.....I have a Miniscribe #3180E with a DTC6280 ESDI Controller Card
from Data Technology. I use (happily so!) your Disk Manager software
which was included with my harddrive - Serial #MIN02615326. The
Hardrive is 156Meg partitioned into 7 drives & I've formatted the drive
utilizing the Controller Card's re-mapping feature. At present I have 2
concerns - I'm awaiting delivery of my Windows V3 update & have noted
the apparent problems & wondering what problems (if any) I will be
faced with.... Secondly, I am going to purchase a 2nd ESDI drive of
approximately 170 Meg (this one has to be a 3 1/2") and would like to
know if I can use my OEM version of Disk Manager to partition that drive
or - is my driver going to have to be updated? My DMDRVR.BIN file is
dated 10-6-88. My dealer is presently searching
out a drive for me which will probably be a Seagate / Control Data model.
Any help you can provide would be very much appreciated..... I will call
back in a few days for a reply. Thanks...

ANSWER:

Since you have used the controller's internal low-level formatting
routine and are using translated parameters on the first drive, I would
suggest following the same procedure for the second drive when you get
it. Since you are using the controller's translating feature, these
drives appear to the system as having less than 1024 cylinders. Since
the drives appear to have less than 1024 cylinders, SWBIOS is not
needed. Therefore, you will not have to place the VIRTUALHDIRQ=OFF
line in your SYSTEM.INI file for Windows. If you are using DMDRVR.BIN
in your CONFIG.SYS file, Windows will not allow you to create a
permanent swap file. You can use temporary swap files though and they
will work just fine. You should not have to upgrade your DM for the
second drive, since you will be low-leveling the drive thru the
controller BIOS, which makes it appear to the system as a standard
drive. You can use DM to partition the drive only, (as a standard
drive).
-------------------------------------------------------------------------
QUESTION:

Hello. I am trying to install Disk Manager V4.03 on a DTK 386 AT
(25mhz) system using an MAXTOR ESDI drive having 760Meg capacity. The
model number is XT-8760E. I am using a Western Digital WD1007V-SE2
Controller with Translation enabled and MS-DOS 4.01

The problem is that your listing says that MAXTOR XT8760 is a 650Meg
hard drive. The drive I am trying to use is an XT8760E 760Meg drive.
HELP!! I would like to be ablto get access to the other 110 meg.
How can I get full use of my drive? Thanks!!

ANSWER:

What you are seeing is the difference between formatted and unformatted
capacity. It is common procedure for a drive manufacturer to advertise
his drive with an unformatted capacity rating, (because it sounds
better). If you are formatting this drive for DOS (which has 512 byte
physical sectors), you would in fact only get 632 Meg (663 Milion
Bytes). This is how to figure out the formatted capacity of a drive:

cylinders x heads x sectors per track x 512 bytes per sector = Tot Bytes

In your case: 1630 x 15 x 53 x 512 = 663475200 bytes (632.7 Megabytes)
(remember that a megabyte is 1024 x 1024 bytes)

Hope this clears up your concerns...
-------------------------------------------------------------------------
QUESTION:

I have a compaq Despro 386/33 and am trying to install a Maxtor Xt4380E
disk drive with your Diskmanager-N (I have already registered via the
script). When trying to install this morning I ran into two problems.

When asked to enter the bad-block map (ie cylinder/head method) my disk
has 151 defects. This puts the number over 512 and therefore the disk
is useless. Diskmanager suggests an alternate method using the Offset
as well as the cylinder/head information. My defect map lists Head,
Cylinder, Byte and Length info. Is the "Byte" information really the
offset? The byte numbers range the mid hundreds to almost twenty-one
thousand. Not having to much experience with PC hardware I am not sure
if this is the case.

With limited time to ba able to take down the server (basically 9pm
until 7am) I thought I had nothing to lose by trying the head/cylinder/
byte method. This got me to the next phase of running the Netware
partition, but also ran into problems. Every single Head/Cylinder in
the sequential test failed with a SECTOR NOT FOUND error. Given the
explination in the trouble shooting area of the installation to reinit
the drive or trash it I am a little concerned.

Am I doomed to buy another $4200 drive or is there a chance that this
will work?? Signed, a tired and weary Systems Engineer.

ANSWER:

It looks like you have spoken to someone on our tech support line since
you left this message. Their recommendation to you of setting it up as
a standard drive of 611x16x63, initializing with the Compaq utility,
and only partitioning the drive thru DMN, (as a standard drive), is
correct. The reason you have to jump through so many hoops on this one
is that the Compaq version of the WD1007A doesn't have a BIOS, but
translation is always enabled. Therefore, when using a drive with more
than 1024 cylinders, the controller will automatically assume that you
will be speaking to it with translated parameters. This works fine if
you are using it in a Compaq that has a CMOS entry that supports the
translated parameters, and has a utility that understands that these
ARE translated parameters, and not physical parameters.
-------------------------------------------------------------------------
QUESTION:

Last weekend I installed a second ESDI into a Northgate 386-20
mhz a year old. It had come with a Miniscribe 3180E and an
Adaptec 2322B controller, which says Rev A, but not 8 or 16 (chip
stickers are C. 1987). The Miniscribe was initialized (last year)
using DEBUG and set up to map at 63 stp, with the drive setup in
CMOS as "type 1" a meaningless small type. The board does not
have 53-55 stp mapping in it, according to the DEBUG invocation
of its special features menu. To make things worse, vendor sent
wrong manual (for a WREN IV) with this drive, which I'm
unfamiliar with.
The new drive is WREN VI, 94196-766, 16 heads x 1630 cyls, and
I think jumpered for 54 spt, but I had difficulty figuring how to
jumper it even for drive 2, don't have any info on jumpering. It
came with DM 4.02 for imprimis with this drive on the tables and a
little bit about Adaptec in the horrid on-disk help file. I used
SWBIOS and DM/S/M to initialize it, since I could not use DEBUG.
Seemed to initialize OK, partitioned OK, boots (as drive 2) OK,
although tak3es rather long to boot up because partition 1 wqas
setup as boot partion (MSDOS 3.3) since the imprimis will
eventually replace the miniscribe and be reterminated as the only
drive on this sys. DM recognizes it as 1630 x 165 x 532.
I read up on your ESDI Q&A file & apparently this Adaptec will
not handle this particular drive. Has SWBIOS + DM done so
correctly, is it safe to use the drive? Can I rejumper the drive to
63 spt so as to use the drive's BIOS? Should I disable the BIOS
(there is a CMOS setting for 16 heads x 1630 cyls)? Is the drive
safe to use as is? What are performance penalties of leaving it as
is? If I have to get a new controller, probably will get Western
Digital, not Adaptec, anything to watch out for to make sure the
WD1007 will handle this drive? I would rather not get new
controller if it is OK as is, but much of your answers seemed to
say this Adaptec cannot handle this particular Imprimis.
What problems will there be with Windows 3.0 on this drive set
up as described? I take it I cannot use a permanent swapfile &
must put VIRTUALHDIRQ=OFF in SYSTEM.INI? Drive was
purchased for extensive grafix work involving Windows, if it later
goes back on network, it will be Artisoft Lantastic (NETBIOS, not
Novell). I can't seem to Leave a msg, so am leaving as SysOp
comment, hope you can make it public when you reply as this helps
others. There wasn't much about ESDI & that only small Seagates
in the public discussions, so I downloaded everything in your files
area, where apparently ESDI is `causing much madness & pain. Did
I recall correctly that Ontrack doesn't support SCSI or something
like that?
Note: apparent error in my file, Ontrack recognized the drive
as 1630x16x53, not 165 heads (giant jukebox).

More info....

I have a new Imprimis 94196-766 ESDI installed with the DM 4.2 that came
with the drive. Previous drive is an ESDI Miniscribe 3180. Controller
is Adaptec 2322, there is an "A" stucker pasted on the board next to
model number. This drive only has 63 spt in its table. The older drive
was set up with Ontrack v. 4.0 (DMDRVR.BIN). The current one, since it
has 1630 cyls, can't escape using SWBIOS. It is jumpered for 54 spt, I
did not try to change it to 63 so I conceivably could have initialized
thru DEBUG as the first drive was. With SWBIOS, it was initialized
through DM/S/M, shows up as 1630 x 15 x 52. Computer is Northgate
386-20, Anyway, the 2 drives are partioned etc. OK now.
Drive 1 is partitioned 32 (boot partition), 60, 60. Drive 2 is
partitioned 32, 150, 150, 150, 190 which is probably kind of stupid in
relation to bufering. Drive j" is the last drive. The first partition
is no longer bootable as everyone suggested this would cause trouble.
The DMDRVR.BIN which is on Drive 1 is from the older (version 4.0
general-purpose) DM, although I used 4.2 to set up the new drive.
Right now, it is taking these 2 drives more than 30 secondws--almost
infinite time--to boot. This horrid time comes about through frequent
need to switch CONFIG.SYS and AUTOEXEC to use windows 3, or 386-max and
my normal DOS programs. Is this caused by the old Adaptec controller?
Would it be worthwhile to get a WD 1007V-Se2?
Computer is 20 mzh Northgate 386 w/4 megs RAM on motherboard, 1 5
1/4" floppy.
Windows has been set up on drive g (i.e. second partition on the new
drive) and seems to work OK, though very slow. I put virtualhdirq=off in
its sys.ini file, and also emmexclude=c400-c7fff to disable vidmem
problem that was occurring. I did not use SMARTDRV, and did not set up
a permanent swap file. The result is it's very slow, loading, loading
any other windows, etc.
In yesterday's (July 9) InfoWorld pg 3 "At Deadline" column, there is
a story from MicrSoft, fllowup of week previous where they had a news
story about problems of Windows 3 and DM. This story has some garbled
info from MicroSoft about "what to do with DM" as regards large hard
drives. It says "more than 1024 cylinder drives are at risk if you are
using DOS 3.30 or greater." I called MS's new tech support-for-windows
and asked "does that mean if you are using PC DOS 3.1 you are not at
risk?" I spoke with someone there who said:
"We have tried a large HD (a maxtor) with 1664 cyls using DOS 3.1 and it
was OK (presumably with the permanent swap file, because nothing is ever
OK with Smartdrv in my experience, only PC Qwik-Cache works without
trashing fATS on large drives), but he said Microsoft guarantees nothing
. Well, I've had problems with 3.3 before and it takes up more memory
anyway. I don't have a 3 1/2" drive, but if I did there are device
drivers I can use w/3.1, which I would rather use anyway as it is the
only DOS I actually trust. MicroSoft 3.3 has caused network problems,
especially w/Northgate Slimlines which took down a net I was supporting
(Lantastic, NETBIOS) when 2 new Slimline VGA stations were added, the
problems went away when the Slimlines were set up with IBMPCDOS 3.1
instead of Northgate DOS 3.3.
So my question now about Ontrack + 1630 cyl 54 spt Imprimis is would
DOS 3.1 make a difference either in the long boot time or in ability to
use a cache (Qwik, not Smartdrive) and a permanent disk swap file?
I wish I could post this, in case any users have experience with the
Imprimis 94196 using smaller, reliable DOS 3.1 as opposed to later, big
buggy DOS'es.
I am willing to try an experiment w/my new hard drive, if you have
any suggestions as to what I should test to find out if the drive is
safge with some Windows speedup when using dos 3.1, I mean I have about
a week when I can do this, then the drive has to get to work on a big
project that will soon have it full.

ANSWER:

Whew... That was some message you left! Now, where do I start...
If the Adaptec 2322 doesn't have a -8 or a -16 after the model number,
it is more than likely a 10 Mbit controller. The Imprimis 94196-766
would require a 15 Mbit controller. My best suggestion is to contact
Adaptec tech support directly (at 408-945-2550) in order to absolutely
identify this controller, and to find out if it is a 15 Mbit controller
or not. If it is not, then DO NOT use it with the big Imprimis. Speed
degradation would be the least of your problems in that case.

If you choose to use a WD1007V instead, you will have to physically
jumper the drive for 53 spt instead of 54. It has been confirmed
through Western Digital that the 1007V is not able to handle a 54 spt
drive at a 1:1 interleave. If you jumper the drive to 53spt, it'll
work just fine with the 1007V.

The 63 spt translation mode that both controllers offer (it is a
function of the controller, not the drive) is not sufficient to make
this drive appear to have less than 1024 cylinders (which is the whole
point of a translation mode). There would be no point in invoking the
63 spt translation mode with this drive. The only controller that I
know of that has a translation mode capable of making this drive appear
to have less than 1024 cylinders is the Ultrastor controller. It has a
translation mode that would make this drive appear to the system as
having 644 cylinders by 64 heads by 32 sectors per track.

In any event, unless you end up with the Ultrastor controller, this
drive will end up appearing to the system as having more than 1024
cylinders, thus the need for SWBIOS, the VIRTUALHDIRQ=OFF line in your
SYSTEM.INI file, DMDRVR.BIN in your CONFIG.SYS (therefore the inability
to create a permanent swap file), and the need to stay away from
SMARTDRV.SYS (which doesn't work correctly on drives with greater than
1024 cylinders). Your DOS version has nothing to do with any of these
facts. Where the DOS version comes in is if the drive were under 1024
cylinders, but was a nonstandard drive with a head or sector per track
mismatch between what it was "setup" as and what it "actually" was,
then there would be problems when using SMARTDRV.SYS only if your DOS
version were 3.3 or higher. If your drive has over 1024 cylinders,
then you've got a problem with SMARTDRV.SYS, no matter what version of
DOS you are using.

Incidently, Disk Manager does indeed work just fine with SCSI drives.
-------------------------------------------------------------------------
QUESTION:

We used DMN-N to format a Maxtor XT8760E drive with an Adaptec 2322B
controller. We installed Netware 2.12 on the system and ran it in
an IBM AT 6MHz computer. We later tried to transfer the disk, cables,
and controller into a Northgate 386 computer running at 33 MHz. When
we tried to boot up the system, we get the message "abend: stack
overflow detected by kernel". The configuration runs fine in the AT,
though. I'd like to know if anyone is using a similar configuration
to ours. If so, I'd like to hear from them.

ANSWER:

The earlier 2322 adaptec controllers have been known to have timing
problems in a fast machine such as yours. 2322 boards with a -8 or -16
after the model # solved some fo the problems, I suggest you try slow-
ing the system down to see if it solves your problem if it does you
might want to up-grade to a newer rev d 2322 controller
-------------------------------------------------------------------------
QUESTION:

we are trying to intergrate a micropolis esdi drive into a ibm model
80-111 using a real ibm esdi controller and an existing real ibm esdi
drive with dm for ps2 however we cannot get the second drive to primary
format. is there an on board formatter on these controllers? we have
some information to start format using debug but it is not very clear.
the ibm diagnostic rejects the disk because the error map is in the
wrong location do you have any suggestions as to how we can get the
drive working with dm ps2 or any other proceedure kind regards dl

ANSWER:

The IBM ESDI controller is a ten megabit transfer rate card and
cannot handle a drive with greater than 35 sectors per track. You did
notsay what the model # of your drive was so I cant comment on wether
or not this controller will handle it... also there is no debug routine
to LL on an IBM ESDI other than standard "IBM" for which you use the
refrence disk to install the drive... The only known utility on the
market that can properly command this card and LL anonstandard drive
is the Disk Manager PS/2 version for micro channel
-------------------------------------------------------------------------
QUESTION:

We are installing a pair of Wren ESDI drives in a PS/2 Model 80 using
the Western Digital 1007-MC controller. When controller and drives are
initially installed in the 80, the system gives a 164 error message and
indicates improper C: and D: drives. Seagate has indicated that DMN-3.1
will allow proper installation of these drives. The 3.1 has been
ordered for overnight delivery. Would appreciate any guidance on
the use of DMN for this purpose (for use with Adv SFT Netware).

ANSWER:

If the "Wrens'" that you speak of are 54 sector per track drives, you
will have to physically jumper them to 53 sectors per track. Other
than that, there should be no problem. V3.10 of DMN has the ability to
handle ESDI drives correctly in the PS/2. If you have problems with
your install, (I don't think you will), then you'll need to fill out
the DMN registration form on this BBS and leave a specific question
concerning the problems you are having.
-------------------------------------------------------------------------
QUESTION:

Configuration:
IBM PS/2 Model 80-311 - IBM 300mb ESDI Drive, attempting to add
Imprimis WREN VI model 94196-766 drive using DiskManager-N Version 3.01.

Questions:

1) What selection should we be making on tracks/sectors - currently
jumpered for 'undefined'
2) What selection should we be making for drive # - currently 2

Problems:

Cannot get DMN to recognize existance of WREN VI drive. PS/2 powering
up with errors - F1 will continue and boot DOS.

Planning to use Novell Netware 2.15C on this system. This is the first
of three identically configured systems. Our vendor provided the WREN
VI drives & ontrack as a bundled set.

ANSWER:

The IBM PS/2 ESDI controller is only a 10Mbit controller. The Imprimis
94196-766 will require a 15Mbit controller. You will need to get a
WD1007V-MC1 or equivalent before this will work. Also, v3.01 of DMN
was not designed to work on the PS/2. You will need to upgrade your
DMN to the microchannel version of DMN, at least v3.10 as well.
-------------------------------------------------------------------------
QUESTION:

We have a 9380E 330M drive with a 1007 Western Digital Controller
During a Word Perfect Session, the computer locked up and after
rebooting we discovered that it would no longer boot from the C drive
and it wouldn't recognize the partitions d,e,f, or g. After performing a
few functions with DM and Norton utilities, we discovered very little.
I wrote a copy of a config.sys and DMDRVR.BIN on a 1.2 meg dos diskette.
At this point we could access the other partitions, but most of the
files we tried to access or run, gave a data error reading drive "x". I
eventually got the computer to boot from C:, but we kept having the same
problems.
At this point we're not sure what to do. After performing disk tests we
found that head 10 on every cylinder has a read defect. So, do you
think the drive is configured wrong, controller is bad, harddisk shot,
or what? We would appreciate a quick response.
While I'm here, we would also like to inquire as to the cost of the disk
recovery service as we think we will probably have to take advantage of
it.The last backup was performed in March! Thanks for your time. I
look forward to hearing from you.

ANSWER:

If you are getting read errors on head 10, every cylinder, it sounds
like you do indeed have a hardware problem of some kind. That problem
could be caused by the controller, cables, or hard disk itself. There
is really no way to isolate the cause of the problem without doing some
swapping out of parts. If you would like to double check the
configuration of the drive, call our tech support voice line, we could
lead you through that process. (Be in front of the machine when you
call). If all else fails, the Data Recovery Department can be reached
at (800)872-2599. They can fill you in on the cost of their services.
-------------------------------------------------------------------------
QUESTION:

I did a quick check of the message base and I see that #188 is just
about where I am. You didn'nt seem to give a solution, just a work around.
I am having problems booting a Novell server.
1) Boot DOS, 2)DEBUG and low level format, 3) DMN /M to set disk
parameters, 4) DMN automatic to setup partitions, 5) NETGEN to install
Advanced Netware.

Even attempting to boot from a floppy meets with failure. Actually, the
CMOS got screwed up and the hard disk was unavailable until I reset the
parameters with v4.0 IMPRIMIS of DM.

Am I reading the manual wrong? It looks pretty straight forward in
there, why does this thing refuse to boot under Netware? This is my
first Novell installation but I am a veteran of other networks. Any help
would be appreciated.

ANSWER:

The procedure you saw outlined in message number 188 was not a
"workaround". It was the correct procedure for working with this
particular ESDI controller. An ESDI installation is a non-trivial
process to be sure. You need to know the intricities of the ESDI
controller that you are using, and how it is going to react to any
given scenario. With an ESDI installation, you can't expect to just
slap in any conbination of controller and drive, then just type in DMN
and expect everybody to be happy. There can be much more to it than
that. What follows is a document that I put together at one time to
attempt to explain ESDI installations:

The ESDI interface allows you to chose between two different install
methods. Either one of these two methods will work, as long as you
pick one method and stick with it all the way through. The two methods
are briefly described below and each have their own advantages.
Remember that we are speaking about ESDI interfaces IN GENERAL. There
are many different brands and models out there with different features
and different requirements, and space doesn't allow us to go into
detail about each and every different controller's own little quirks.
(Even if we did try to put together such a document, it would be huge,
and outdated as soon as it was written!) If these methods don't work
with your particular controller, call our tech support line at
(612)937-2121, and we may be able to tell you how to handle that
specific controller.

METHOD 1:
The first method is to lowlevel format the drive through the ESDI
controller's BIOS using DEBUG and then use Disk Manager to partition
the drive only. The advantages of using this route are:

1. When you lowlevel the drive through the controller's BIOS, the drive
appears to be a standard drive to your system from that point on.
This means that under DOS 4.01 for example, you can have the entire
disk as drive C:, (as long as it doesn't have more than 1024
cylinders.) Which brings us to the next benefit...

2. Within the controller's lowlevel program, you usually have the
ability to invoke a 63 sector per track translation mode. What this
is designed to do is to translate a drive with more that 1024
cylinders to appear to have less that 1024 cylinders, but a higher
number of heads and sectors per track (16 hds & 63 spt). The reason
for this is that DOS has a 1024 cylinder limitation and cannot
access any cylinder above that. Also, Netware cannot boot directly
from a hard disk with more that 1024 cylinders. The translation
mode makes the drive APPEAR to the system to have less than 1024
cylinders. Remember that this is a translation of the drive
geometry only, you still have the full capacity of the drive
available.

The basic procedure for this method is as follows:
1. Boot with DOS
2. Run your SETUP program and set your CMOS drive type to type 1.
3. Run DEBUG
4. At the hyphen prompt enter the entry address of the controller's
lowlevel format program. (usually either G=C800:5 or G=CC00:5).
5. From within that program, lowlevel format the drive and enter your
defect list.
6. If you are using a drive with more that 1024 cyls, invoke the
translation mode.
7. Exit the lowlevel routine, at which point the machine will reboot.
8. Run DM or DMN in manual mode (with the /M parameter)
9. If running DM, go to the config menu and press W to write the
standard parameters to disk.
10.Go to the Partitioning menu, create your partitions and prepare
them.
11.If you are using DMN, proceed with your Netware install.

METHOD 2:
The advantage to this second method is that it is an almost fully
automatic install using Disk Manager to do the entire job. However, in
some cases, the advantages obtained from the first method outweigh the
ease of the second method.

1. First of all and most importantly, DISABLE THE BIOS AND TRANSLATION
on your controller board. If you leave them enabled, they will be
fighting with DM over what the parameters of the drive are, and no
one will win. Be aware that some ESDI controllers have no way to
manually disable the translation mode. For example: If you disable
the BIOS on the Adaptec 2322, it will automatically go into a
translation mode that you have no way of disabling. Controllers
such as this will not function correctly using this second method.
2. Boot with DOS.
3. If your drive has over 1024 cylinders and you are using DM, run
SWBIOS.
4. Run DM or DMN in automatic mode and let it do the whole job.
5. You will have to keep your DOS boot partition very small because of
the sector per track mismatch that you will have with your ESDI
drive.
6. If you are using DMN, you will have to run the appropriate
modification utility because you are installing this as a
nonstandard drive.

Remember, if you use this second method, you are installing a drive
that is appearing as nonstandard to your system. The DOS boot
partition, (your drive C:), will be limited in size to the standard
portion of your disk, (whatever the parameters are for the CMOS drive
type that you are set to). Anything above and beyond that area of the
drive will have to be a Disk Manager write/read partition.
-------------------------------------------------------------------------
QUESTION:

We've just replaced an existing file server's Seagate ST4096 mirrored-
pair drives for a pair of ST4384E drives. The installation went fine,
and the server booted normally from the cold boot loader. After we
restored the files we had backed up from the original drives, we
couldn't boot the server from the cold boot loader any more. Instead,
we can boot it by first booting DOS, and then running NET$OS from the
diskettes. If we let it try to boot from the loader, the message from
the loader appears very briefly, and then we get a screenful of raw data
from one of the data files in the system, and the system hangs. We have
tried many things, including redoing the low-level format on the drives,
and starting the installtion over again. Oh, I forgot. The first thing
I tried was to re-load just the system files (NET$OS and the cold boot
loader) from within NETGEN. Didn't help, same problem. Then I tried
re-initializing the drives, redefined volumes, reload the whole system ,
etc. No help there, either. I finally had to re-do the low-level
format on the drives, and do the entire installation procedure from
square one. Then I could boot it off ot the hard drive. I thought
maybe it was just a glitch, and I restored my files again. To my
delight (argh!), the problem came back.
So, my client is currently booting from a floppy disk. It's not
catastrophic, but I'd sure like to know what's going on. I'm not even
certain that your software is at fault, But this is the only
installation I've ever had that I couldn't restore files on normally.
Thnx for your help.

ANSWER:

I have a theory as to what may be happening with your install. We have
seen this in the lab, and it took us a while to figure out what was
going on, but it turned out to be as follows: NetWare's cold boot
loaders cannot boot directly from a hard disk with more than 1024
cylinders. What happens is, at boot time the cold boot loaders are
presented with a geometry of 1224 cylinders, what the cold boot loaders
see at this point (because they don't understand anything over 1024) is
a "wrapped" geometry of only the excess of over 1024 cylinders, (in
this case about 200 cylinders). NetWare then assumes, (since the
geometry is being seen as only 200 cylinders), that it's redirect area
must be at about cylinder 190 or so. When the system is clean, (no
data on the drive except for NetWare itself), the area around cylinder
190 is blank, so therefore NetWare sees no problems with what it
interprets as it's redirect area, and boots up without any error
messages. As soon as you put data on the drive though, and the area
beyond cylinder 190 gets written on, the next time you boot NetWare it
sees "junk" in what it interprets as it's redirect area and the whole
thing tips over rather ungracefully, (with garbage on the screen).

The solution for this is to take advantage of the 1007V's 63 sector per
track translation mode. With a WD1007V controller, you have the
ability to low-level format the drive thru the controller's BIOS (using
debug, g=cc00:5), and invoke a 63 sector per track translation mode.
This translation mode that is provided by the controller will make the
drive appear to have a higher number of heads and sectors per track,
but a lower number of cylinders, (aprox. 635 cyls x 16 hds x 63 spt in
this case). Since the number of cylinders are appearing as being less
than 1024, NetWare will be able to boot directly from the hard drive.
In this scenario, you will be using DMN to partition and prepare the
drive only, as a STANDARD drive. The procedure for doing this is as
follows: After low-leveling the drive, entering the defect list, and
setting the 63spt translation mode thru the controller BIOS's low-
level routine, run DMN/M (manual mode), confirm that the parameters on
the left side of the screen are appearing as STANDARD and that they are
aprox. 635x16x63. If this is the case, then so far so good. Then all
you do is press F6 to create the NetWare partition, and then F7 to
prepare the partition. After this, you proceed with your Netgen.
Using this method, you should be able to restore your data onto the
drive with no ill effects.
-------------------------------------------------------------------------
QUESTION:

I'm trying to figure out a way to get the entire 336mb
on a Micropolis 1558 recognized in non-translation mode on the ACB-2322B
with about a 100-mb, bootable ISC 386/ix SVR3.2.02. I've got the later
AMI bios which allows user enterable parameters on type 47, but the ISC
fdisk still hits the 1024 cyl. limitation, and the ISC install process
croaks when I try sector translation. Guess I should call Adaptec and
find out about their latest ESDI controller (2322D?). Thanks for your
prompt response to prev. posting...
dlp (WOOPS - just reread my msg - line 8 should read: "100mb bootable
partition" - right now I'm booting DOS from a floppy and hitting reset
and flipping up the FD latch to boot UNIX from the UNIX partition).
THANKS AGAIN!!

ANSWER:

I need a little more specific information about what it is you are
trying to do... What follows are separate parts of your question
followed by my initial answers each of your questions. As you see, I
am a bit confused. Please elaborate, and I might be able to be of more
help.

> "I'm trying to figure out a way to get the entire 336mb on a
> Micropolis 1558 recognized in non-translation mode on the ACB-2322B
> with about a 100-mb, bootable partition."

You can do this. Go into the Adaptec's BIOS with debug (g=c800:5),
select "Special options menu" and turn translation off. Then boot with
DOS, run SWBIOS, run DM in manual mode (DM/M), press C to go to the
config menu, press N to select nonstandard parameters, when it asks you
to select your drive press ESC for standard parameters instead. (This
is known as a "nonstandard/standard" configuration and insures that
SWBIOS will load at boot time.) At this time, the parameters should be
appearing as being "setup" as 1222x15x35 and "actually" 1222x15x35.
Press W at the configuration menu to write the parameters to disk,
return to the main menu, go to the partitioning menu, create a 100M DOS
partition, and then split the rest of the disk up into one or more Disk
Manager write/read partitions (which can reside over cylinder 1024).
Prepare all partitions, and you should be ready to go.

> "I've got the later AMI bios which allows user enterable parameters
> on type 47, but the ISC fdisk still hits the 1024 cyl. limitation."

It will. That's why Disk Manager and SWBIOS exist. But don't enter
values of greater than 1024 cylinders into a user defined drive type,
as it will prevent SWBIOS from loading at boot time. SWBIOS assumes
that since the parameters are appearing as "STANDARD" the BIOS must
know how to handle it by itself. This is not the case in your
scenario. You should zero out your user defined drive type, and set
your drive type to type 1, and use the low-level routine on the Adaptec
to low-level the drive. After this, the BIOS on the Adaptec will make
this drive appear as standard to the system without the aid of a user
defined drive type. Then follow the above procedure for the rest of
the install.

> "The ISC install process croaks when I try sector translation."

What is "ISC" and the "ISC install process"?

> "Right now I'm booting DOS from a floppy and hitting reset and
> flipping up the FD latch to boot UNIX from the UNIX partition)."

Whoa! Wait a second! This is the first time you've mentioned UNIX...
How is it involved?
-------------------------------------------------------------------------
QUESTION:

I am using a Seagate/CDC/Imprimis 94186-383H (320 meg) hard drive with
1,224 cylinders. I am not currently using Disk Manager on this drive,
since CompuAdd's MSDOS 4.01 took care of setting it up as one huge drive
C: Am I correct in assuming that I will have no problem with Windows 3,
or will the 1,224 cylinders still give me Windows problems?

I *am* using Disk Manager on eleven other computers with large hard
drives (DM supplied with each drive). I know a few of them are over
1,024 cylinders, but I'll check the others using your helpful
instructions in Bulletin # 12.

Do you know yet whether Ontrack will change Disk Manager to work with
Windows 3, or will Microsoft change Windows to work with Disk Manager
(and other partitioning products)? Your product is GREAT, and I have
very much appreciated it over the years. If I have to make a choice,
I'll keep Disk Manager and my large hard drives and ignore Windows...
but I hope there will be a solution.

Thanks for providing this BBS. I manage a number of PC's at a
State University. I'll check back to keep up to date on your efforts to
resolve the Windows compatibility issue. I also Sysop a multiline BBS
system. May I put Bulletin 12 and some of your Q&A files on my system
for my users?

ANSWER:

If you are indeed getting all 320 meg of this drive as drive C: with
DOS 4.01, your controller (you didn't mention which controller you were
using) must be doing a translation of the drive geometry. Most ESDI
controllers have a 63 sector per track translation mode that will make
the drive appear to have a greater number of heads and sectors per
track, (16 & 63 respectively), and a lower number of cylinders (about
637 in this case). Using this translation mode, your drive which is
actually 1224x15x35 will appear to the system as being 637x16x63. The
reason for this translation mode is that DOS has a 1024 cylinder
limitation. Using this capability of your ESDI controller gives you
the best of all worlds. The main benefits are listed below:

1. When using the controller BIOS to low-level format the drive, (as
was done in this case), the controller BIOS makes this drive appear
as a "standard" drive to the system. Therefore, you are not setting
up a "nonstandard" drive in this case, and have no chance for a head
or sector per track mismatch between what the BIOS of the computer
thinks it has for a drive, and what Disk Manager thinks it has for a
drive. This fact allows you to use the full sector per track value
of the drive in the DOS boot partition, (instead of only 17 spt
which the BIOS would normally see).

2. The drive appears to the system as having less than 1024 cylinders,
while retaining the full capacity of the drive. Because of this
fact, you aren't using SWBIOS (or XBIOS) to get beyond cylinder
1023. Therefore, you obviously won't have a conflict between
SWBIOS/XBIOS & Windows, (so you don't need the VIRTUALHDIRQ=OFF line
in your SYSTEM.INI file in order to run Windows in 386 enhanced
mode).

3. The two above facts combined, (the drive appears as "standard" and
less than 1024 cylinders), means that DOS can see the whole drive
all by itself without the help of Disk Manager. You can have the
whole drive as drive C: with DOS 4.01. This means that you don't
have to use DMDRVR.BIN, therefore the Windows swapfile program won't
object to it's presence and will allow you to create a permanent
swapfile for Windows. It also means that SMARTDRV.SYS would be able
to function correctly without a chance of it causing data
corruption, (or at least not because of these two reasons).

The next version of Disk Manager/XBIOS that is released will be able to
function correctly with Windows in 386 enhanced mode without the need
for the VIRTUALHDIRQ=OFF line in the SYSTEM.INI file. It is my
understanding that MicroSoft is in the process of modifying
SMARTDRV.SYS to understand nonstandard drives and drives with greater
than 1024 cylinders. We also hope that they will modify the swapfile
program to allow it to function correctly on a nonstandard drive, (and
DMDRVR.BIN is present in the CONFIG.SYS file).

As far as bulletin number 12 goes; Please, by all means place it on
your BBS, (and any other BBS you get a chance to). Spread it all over.
It is important that anybody who is using Windows sees it.

The Q&A files on the other hand, I would rather you didn't place on
your BBS. I ask this because they are not official statements from
Ontrack Computer Systems, and should not be presented as such. They
are simply reprints from the message base of this BBS. The questions
are answered by myself, to the best of my ability and understanding.
But, the fact remains that I am a human, (yes there is actually a lowly
tech support person at the keyboard writing this), and have been wrong
before. (NO!... yes, I'm afraid so...). Also, the Q&A files are
constantly changing. When I find a situation in which I have
misinterpreted a person's scenario and/or question, or they have not
told me the whole story (an amazing percentage of the time), or they
have actually told me untruths, or the controller manufacturer has
changed it's BIOS and the board no longer works the way I said it does,
or Disk Manager changes, or new problems with specific combinations of
hardware and software come to light, or my answer was simply wrong for
any number of other reasons including ignorance on my part, I try to
update the Q&A files to correct my answers. If the Q&A files were
being published on other BBS's, they would then be out of my hands and
I wouldn't have the chance to update them with this new information.
You are welcome to inform your users that the Q&A files are available
on this BBS, and encourage them to call this BBS and download them, but
I do ask that you don't post them on other BBS's. Thanks, SysOp.
-------------------------------------------------------------------------
QUESTION:

I've got a PS/2 model 80-041 and I'm trying to install a CDC Wren V
(94186-383H). So far I've been able to get the machine to recognize
that the drive is present (I checked by using the reference disk test
programs) but the bios and DM seem convinced that the drive has
299 cyl, 64 heads and 32 sec...
this is far from the truth - the drive is configured for 34 spt and has
1224 cyl with 15 heads...
nothing I do seems to change this...
I even tried to just go blindly ahead and do a low level format but I
can't enter in the bad sector table...
help - what am I doing wrong??????
(so far, seagate and ibm have only pointed fingers at each other as to
the problem - I fear that my ESDI card - the origional card that came
with the model 80-041 - may be defective... incidentally: if I swap
the Wren V I'm attempting to install with the origional 70 meg ibm
drive, everything works fine... so I'm really not sure if it's the card)
this whole thing really leaves my cheese in wind...

ANSWER:

When installing a 94186-383H on a PS/2, the parameters of 299x64x32 are
absolutely correct. The PS/2 uses a 64 head x 32 spt translation for
ESDI drives. The ESDI controller is not defective. This is the way it
is designed to work, and both IBM and Seagate should have been able to
tell you this. If you are using Disk Manager for the PS/2, version
p4.0, this install should go without a hitch. It knows that 299x64x32
are translated parameters and the drive is really 1224x15x34, and will
allow you to enter the bad track list correctly. If you are using any
other version of Disk Manager than version p4.0 (the only one designed
for use with the PS/2), it won't work.
-------------------------------------------------------------------------
QUESTION:

I am having a problem using disk manager (manufactured for Imprimis) to
format a 300mb ESDI hard drive in a compaq 386/25 computer with a Western
Digital 1007a Controller. It seems to format OK once I've set the
configuration information correctly; however when I partition it into
logical drives C:-L: (10 partitions, approx. 33mb each), it works until I
get to K: and half way though K: every sector returns a RECORD NOT FOUND
error. It is driving me up a tree so any help would be appreciated.

ANSWER:

Is this the Compaq BIOSless version of the WD1007A, or is it a real
WD1007A with a BIOS?

If it is a real WD1007A with a BIOS on it, I would suggest that you
enable both the BIOS and translation on the card (W3 on, W14 off), boot
DOS, run DEBUG, enter g=c800:5 to run the controller's low-level
routine, low-level format the drive, enter the bad tracks, and set the
card into it's 63 sector per track translation mode, (all thru the
controller's low-level routine accessed thru debug). After the drive
has been formatted, you should be able to run Disk Manager in manual
mode, (DM/M), and have the drive appear as a standard drive of aprox.
600 cylinders by 16 heads by 63 sectors per track. You would then use
Disk Manager to partition the drive ONLY. Since you are using the
63spt translation mode of the controller, the drive appears to the
system as having less than 1024 cylinders so you won't have to use
SWBIOS. If you have Compaq DOS 3.31, make sure you use the /4 switch
in Disk Manager, or you can just use the FDISK program instead of Disk
Manager since the drive will be appearing as standard to the system and
as having less than 1024 cylinders.

If this is the BIOSless Compaq version of the WD1007A, then you must
set your CMOS drive type to the entry of 600x16x63, use the Compaq
utility to low-level format the drive, (or use DM with the /7 switch to
low-level format), then partition it the same as in the above example,
(either thru Disk Manager, (with the /4 switch if you are using Compaq
DOS 3.31), or thru FDISK since the drive is appearing as standard).

With either method listed above, if you are using Compaq DOS 3.31 or
DOS 4.x, you can have the whole drive as drive C: if you wish.

More explicit instructions for this procedure can be found in the
ESDI_Q&A.EXE file, in the files section of this BBS. Download it and
take a look.
-------------------------------------------------------------------------
QUESTION:

I AM INSTALLING THE DISK AS THE SECOND DISK IN THE SYSTEM. THE FIRST
DISK IS THE STANDARD IBM ESDI 115 MB DRIVE. THE FIRST ATTEMPT, I USED
DISK MANAGER N TO COMPLETELY PREPARE THE NEW DISK. WHEN I SELECT THE
NOVELL INSTALL UTILITY IT TELLS ME THERE ARE ONLY 56MB IN THE NOVELL
PARTITION. i CALLED ON TRACK AND WAS TOLD TO DO THE FOLLOWING..START DMN
ENTER THE MANUAL MODE, SELECT THE SECOND DRIVE] THEN CTRL F10 TO ERASE
ALL 32 SECTORS. THEN TO GO INTO COMPSURF BYPASS FORMAT, SELECT 0 PASS
AND LET IT INITIALIZE THE DISK. COMPSURF RECOGNIZES THE DISK,
INITIALIZES ALL 14095 THEN STARTS TESTING. i TERMINATED THE TESTING AND
STOPPED COMPSURF THEN STARTED THE NOVELL INSTALL UTILITY AGAIN IT
RECOGNIZES ONLY 56MB AS AVAILABLE FOR NETWARE. WHERE DID I SCREW UP?
THANKS WILL CHECK FOR YOU REPLY ON MONDAY.

ANSWER:

Unfortunately, installing a 300+ Meg drive in a PS/2 under NetWare 2.0a
is a no-go situation. NetWare 2.0a has a 255 Meg physical drive size
limitation. With the PS/2 ESDI controller, you cannot do a "non-
standard" install, therefore is no way to truncate the drive to that
255 Meg limitation. If NetWare 2.0a wakes up to find a drive greater
than 255 Meg in the system, it will only see the excess of 255 Meg, (56
Meg in this case). This is not going to work, no matter what you do.
You will either need to upgrade your NetWare to a 2.1x version, or get
a drive with a total capacity of less than 255 Meg.
-------------------------------------------------------------------------
QUESTION:

I am running DMDRVR.BIN (V 4.02 I think) on a 9380 Miniscribe ESDI.
However, even as fast as the ESDI drive is (with a WD1007 WA2
controller), I would still like to run a cache program (like either the
one with PC-TOOLS V6 or PC-KWIK (which I came to find are one and the
same, at least for V 5.5 it was), since I have some large directories
and processing can be slow. Is there any problem with this combination ?
I have my 320mb drive partitioned as C=32mb, D=250mb and E the rest ?
How about a list of what other disk cache programs that dont work/should
not be used with DM !
I eventually want to go to DOS 4.01 to not have to use DM, since due to
the large cluster size, I waste a lot of space in directories (currently
over 500K on my D: drive).

ANSWER:

The only caching program that we are aware of that works correctly on a
drive with greater than 1024 cylinders is "Super PC-Kwik" by Multisoft,
(v3.251 or later). Since you have a WD1007A, you have another option
though. The WD1007A has a 63 sector per track translation mode that
would make this drive appear to the system as being 637 cyls by 16 hds
by 63 spt. If you are using this ability of the card, most any caching
program should work. Also, if you are thinking about going to DOS
4.01, using this 63spt mode would allow you to have the whole drive as
drive C:, (without the help of our driver).
-------------------------------------------------------------------------
QUESTION:

HELP! I have just purchaced a AMI 386-33Mhz motherboard, a DPT cache
controller with 4.5 Meg Ram, a Seagate WREN (actually an Imprimis)
hard disk with 766Meg unformatted (1632 cyl, 15 heads)
It is model 94196-766. I have Disk Manager Ver 4.03 (says on the
package) but the software comes up as Ver 4.02
My problem is that I don't seem to access the entire capacity of the
drive. ( I should get 663.5 Meg formatted....I get around 400 Meg)
My Dealer called you and got some info, I followed it with not much
better luck. I also have purchased DOS ver. 4.01 to see if that helps,
again...not really.
I have tried making the LAST partition large so that NO partition
starts after cyl 1024...again no luck. DM gives the following message
"NO RECORD FOUND I/O ERROR DRIVE 1 CYLINDER xxxx HEAD yy"
where xxxx seems to be > 1024, and head is 1-15.

One of my concerns, and reasons for DOS 4.0 was also the cluster size
which I would like to keep as small as possible.
After looking at the on-line help for cluster size, I think a
partition structure like this makes sense.
C: 1-2 Megs
D: 255 Megs
E: 127 Megs
F: 33 Megs
G: 244 Megs or a bunch of 127 Meg Drives.

Please help me get access to all my drive. I am a software developer
and this "down time" is costing me quite a bit.

ANSWER:

According to my information, with the DPT ESDI controllers you are best
off low-level formatting the drive through the controller's onboard
BIOS using debug, (G=C800:5 more than likely), or their DPTFMT software.
Once you have accomplished this, you should boot with DOS, run SWBIOS,
then DM/M (manual mode). Once in DM, go to the configuration menu and
make sure the parameters show as "standard", and are aprox 1630x15x54.
If this is the case, then so far so good. You next want to W)rite these
"standard" parameters to the disk, (don't actually attempt to select
the 94196-766 from within DM), then go to the partitioning menu to
create and prepare your partitions. Using this method, you should be
able to get 644.7 Meg of capacity from this drive.

(1630 cyls x 15 heads x 54 spt x 512 bytes/sector = 675,993,600 bytes)
(675,993,600 bytes / 1024 / 1024 = 644.7 Megabytes)

Of course every partition that even partially resides above cylinder
1024 will have to be a Disk Manager write/read partition, (even with
DOS 4.01).

If you have further problems, it would probably be best for you to call
our Tech Support Voice line at (612)937-2121, so we can walk through it
with you and make sure each step along the way is working correctly.
Make sure you are in front of the computer and ready to experiment when
you call.
-------------------------------------------------------------------------
QUESTION:

Any problems using Disk Manager with Novell Netware 386 version 3.0
or version 3.1. I am duplexing two Miniscribes 9380E's as Type 1
in standard bios. I plann to use SWBIOS to emulate the correct hard
disk types. Any comments on easy installation of the duplexing using
disk manager.

ANSWER:

Afraid I've got bad news for you... Disk Manager-N is not compatible
with NetWare 386 at this time. Even if it was, SWBIOS would not be
involved as it only comes with the Disk Manager DOS version, not the
NetWare version. You should be able to initialize these drives with
the Compaq low-level format utility, after which they will appear as
"standard" drives to the system, and NetWare 386 will be able to see
them all by itself.
-------------------------------------------------------------------------
QUESTION:

We've been through the following steps 4 times and each time after
restoring the hard disc..it seems to work for a few hours...and then we
get a bad Allocation Table reading on the C: drive. When we go to
reboot lo and behold, the boot portion is gone along with command.com
and the system.

The following hardware is being used:

Micronics 486-25 Motherboards
Fujitsu M2263E Hard Disc
6280-15TX Controller
MultiSync 3D Monitor
ATI VGA Wonder Card w/ 512K Version 4.1


We are in the process of upgrading over 425 "older" 386 machines that
were using these drives along with the CompuAdd Cache Controller card.
Due to the size of sectors 1658, we are not able to utilize their
controller, because you can not turn their bios translation off...which
is a prerequisite for DM.

(Would really like to know if there is a cache controller card that
would work in it's place though! THANKX!)

The following steps are being made for setup:

1: Run Debug with the G=c800:5 setting {No Problems}
2: Run Debug again and perform the low level format. {No Problems}
3: Enter DM with the following command DM/4/M/C {Sorry..we're running
Compaque 3.31 Dos}
4: DM recognizes our drive as it should be: 1658 cylinders, 53 sectors
and 15 heads
5: We Proceed to Partition menu and set Partition 1 = 416 MB and
Partition 2 = 276 {the balance of the drive}
6: We write these changes to disk
Everything A.O.K. thru here...then......
7: We run Prepare and DM goes through and verifies the disk and when it
begins formatting and preparing we encounter the following error at the
bottom of the screen:

Verifying Data Area
No Record Found I/O Error Drive Number 1 Cyl. 99 Head 7

We can let DM finish up its preparation, load our tape backup onto the
drive and everything seems to be OK. System boots, can do file copies
across the C & D drive etc. but then out of nowhere we get an error
returning to the DOS prompt FILE ALLOCATION TABLE {CORRUPT} something
to that effect. We try to reboot and naturally the boot sector etc. is
gone.

We did in fact have the SMARTDRIVE.SYS in our Config.Sys file all three
times when the system crashed.

Would that have caused our problems, or do we have another internal
problem {referring to the No Record Found I/O Error we get during DM
preparation}?

*Would appreciate your suggestion on another Cache Controller card that
would in fact work with these drives and DM.

**What Disk Technician {Spinrite ?} would you recommend for use on
these systems?

***Is it safe to use Norton's Optimize on this system?

****Is it safe to use PC-CACHE.COM version 6.0 on this system?

If SmartDrive.SYS is the culprit, would it be possible to have two
Config.Sys files set up, one for Windows Enhanced with Smartdrive in
and without the DMDrvr.Bin and have another without Smartdrive but with
DMDRVR.BIN to access the data on the D drive? Or will this still
corrupt data and allocation tables?

I really appreciate any help you can provide me, as I am receiving the
418 drives and controllers later in the week for installation over the
weekend and would like to have "at least one" system up and going
before we begin tearing the other systems down!!!!

ANSWER:

In the procedure that you outlined for me, I see that you may have
missed one step... When you run DM/4/m/c and go to the configuration
menu, you need to press W to write the parameters to the disk, prior to
going to the partitioning menu. This is done to insure that XBIOS gets
invoked at boot time. Also, SMARTDRV.SYS is DEFINITELY a problem in
this situation. Anytime you are using a drive with more than 1024
cylinders, you MUST remove SMARTDRV.SYS from your system. It will
cause data corruption similar to what you are seeing. The dual
CONFIG.SYS file scenario that you spoke of would work, but it sounds a
little messy... Removing SMARTDRV.SYS altogether and using a caching
program that can function on a drive with greater than 1024 cylinders,
(such as Super PC Kwik), sounds like a better solution to me.

As far as other utilities go, I am not certain that any of the ones
that you mentioned will function on a drive with greater than 1024
cylinders. We, (Ontrack), do have one that will function with a drive
with greater than 1024 cylinders though. It is called Dosutils, and is
available through our sales department (800)752-1333, (or through most
any software distributor). It includes a comprehensive set of disk
diagnostics, and the "nondestructive" low-level format that it seems
you are looking for.
-------------------------------------------------------------------------
QUESTION:

We have a little situation with a WD1007-se1 card and a Seagate 4383E
drive. While we know we may not boot novell from it, when writing the
netware partiton the system seems to work ok but when we load Novell
ANW 2.15 Rev C we get an abend:Impropper rom parameter table for disk
controller. - the book says run moddrvr on the ATDISK.obj file. OK but
this version of ANW has NO atdisk.obj file onthe DSK_DRV_001 disk?
When we skip to the ISADISK.obj we get an error message saying INVALID
file, Program aborted. (who invented this word abort:) so where to now?
if we cannot load novell onto the drive oops!
we tried the generated Compsurf file which just terminated and returned
to DOS.
ANY ideas? DiskManager Program is Ver3.10, about 3month old.


ANSWER:

With the WD1007, you have the ability to initialize the drive through
the controller BIOS and make the drive appear as being "standard" to
the system and having less than 1024 cylinders, (so you don't have to
patch NetWare at all, and you can boot the server directly from the
HD). The procedure for taking advantage of this ability of the WD1007
is contained in the ESDI_Q&A.EXE file on this BBS. Download it and
take a look. Otherwise, if you want to install this as a "nonstandard"
drive, you will need to contact our sales department and upgrade your DMN
to the most current version, (v3.11). This will cure the "invalid file"
message when trying to patch ISADISK.OBJ on your version of NetWare.
-------------------------------------------------------------------------
QUESTION:

I AM INSTALLING A MAXTOR 8760E DRIVE ON MY FILE SERVER. THE SERVER
ALREADY HAS 2 155MEG DRIVES , EACH RUNNING OFF SEPARATE NOVELL DISK
COPROCESSOR BOARDS. THE COLD BOOT LOADER ACCESSES THE FIRST DRIVE TO
BOOT THE SYSTEM. MY QUESTION IS THIS...IF I INSTALL THIS NEW DRIVE AS
WORKING VOLUMES STARTING AT THE 3RD DISK POSITION, DO I STILL HAVE TO DO
THAT FUNNY BOOT OFF THE FLOPPY THAT'S MENTIONED IN THE DISKMANAGER-N
MANUAL???

ANSWER:

I you can actually get this drive addressed and showing up under DMN
and Netgen as drive 3, get it partitioned and prepared for NetWare and
get volumes defined and initialized on it, (I'm glad you didn't ask me
how to actually do that...), then the answer to your question is no.
You won't have to boot from a floppy. Also, the Ultrastor controller
has a translation mode that is able to make this drive appear as having
less than 1024 cylinders. So even if this were the first drive, you
could take advantage of this translation mode and not have to boot from
a floppy.
-------------------------------------------------------------------------
QUESTION:

I HAVE A 320 MB MINISCRIBE ESDI DISK SET UP WITH DISK MANAGER 4.02.
WHEN I AM USING THE DRIVE, I OCCASIONALLY GET STRANGE NOISES FROM IT,
KIND OF A RRR,RRR,RRR,THUNK! THE DRIVE DOESNT SEEM TO BE WRITING OR
RETRIEVING ANY DATA AT THAT TIME. I HAVE CHECKED BOTH THE SURFACE OF
THE DISK AND THE CONTROLLER WITH DM DIAGNOSTICS AND CHEDKED THE DISK
MEDIA WITH NORTON AND PC TOOLS DISKFIX. I REGULARLY COMPRESS THE DISK
AND RUN CHECKDSK. SO FAR, THERE ARE NO PROBLEMS BESIDES THE NOISE AND
THE ACCOMPANYING LAG IN RESPONSE TIMES. I WOULD APPRECIATE ANY INSIGHT
INTO THIS PROBLEM. ALSO, WHEN I TRY TO ALLOW CE WRITES IN DM DIAG, IT
GIVES ME AN ERROR STATING THAT IT COULD NOT VERIFY THE TRACK. IT ALSO
TELLS ME THAT THERE IS A 10.6 MEG DRIVE #2 WHEN THERE SHOULD (I THINK)
NOT BE ONE PRESENT. THANKS

ANSWER:

The sound you are hearing from your drive is probably caused by it
doing a "retry" on a sector it is having problems reading. But did I
understand you correctly when you said that it sometimes does this when
you are not attempting to access the drive at all, (like when it is
sitting idle at the DOS prompt)? If so, then should really call
Miniscribe tech support to see if they think something else might be
wrong with it. As for your questions about DIAG; Did you select
"allow CE cylinder writes" and then select "format the CE cylinder"
before attempting any write/read test on it? Also, some ESDI
controllers protect what we consider to be the CE cylinder from any
access at all. The second drive that DIAG shows is a phantom drive.
If you look at that screen again, you will see that the CMOS drive type
is set to 0, (no drive installed), but the other parameters shown are
those for a 10 meg drive. This is because those are the values that
are actually placed in memory when you select drive type 0 (not
installed). We probably should have blanked out all other parameters
if the drive type was 0, to avoid confusing people, instead of showing
what was actually in memory.
-------------------------------------------------------------------------
QUESTION:

What to do for: INVALID FILE A:ESDI.OBJ
received from MODDRVR on ESDI.OBJ (or any .OBJ file).
My DM-N version is 3.04. 3.10 does the same thing.
My *.OBJ files are from ANW 2.15c .
The goal is to install Netware on a 760 Meg HD.

Server: Arche Rival 386-SX .
Cntrlr: UltraStor-12F .
Disk : Maxtor XT-8760E

I've never had this problem before. I've used DM-N five times,
but never with this release of ANW.
DMN.EXE itself runs fine. Preparing the .OBJ for NETGEN is my next
step, tomorrow morning. Tonight is dedicated to LLF thru DMN.
I have not read Q & A on this BBS. If one is specific to this problem
please give me the name of this file.
If another version is required or an upgrade, please be specific about
the version. It might be faster to buy another one than getting the
upgrade. As usual, time is priority #1.
Thank you again for this BBS. Keep on the good work.

continued...

I just downloaded ESDI_Q&A.EXE and found a msg about INVALID FILE.
Ok now for ISADISK.OBJ vs ESDI.OBJ .
I now make the relation between ISADISK.OBJ, the driver I pick in
NETGEN and ESDI drives I use.

Now, about 63 SPT translation. With this 760 Meg (675.5 according to
DM-N) I am still over 1024 cyl. Should I use translation or not ?
Is the only drawback of NOT using it that it cannot boot directly
from the hard disk (for a smaller drive) ?

Also, msgs about UltraStor, always tell to LLF thru the BIOS on
the controler. UltraStor's doc tells it too. But DM-N worked
fine on my Maxtor XT-8760E directly out from the box. No LLF was
done on it, neither from me or the manufacturer. What is the real
story about this ?

continued...

Back to square #1. The INVALID FILE A:ISADISK.OBJ is still stopping
the installation. No such thing in ESDI_Q&A.TXT.
What's next ?

ANSWER:

When using the Ultrastor controller, you shouldn't need to run MODDRVR
at all, (if you follow the correct procedure). What I would suggest
you do is to low-level format this drive thru the Ultrastor
on-board BIOS (run debug, G=C800:5), set it into 64 head x 32 spt
translation mode, (not the 16 head by 63 spt mode), exit to DOS, reboot
the machine, and then run DMN in manual mode (DMN/M). As soon as you
get into DMN, the drive parameters should appear as "standard", a
little over 630 cylinders, by 64 heads, by 32 sectors per track.
Remember, this is a translation mode, not the physical parameters.
Don't worry. Do not attempt to select the Maxtor 8760, (or any other
drive), from within DMN! This would really confuse things when using
this method. All you need to do at this point is to press F6 to create
the partition, and then F7 to prepare it. Once the preparation is
complete, run Netgen and install NetWare. Using this method, you don't
have to run MODDRVR at all, (which is good, because your version of DMN
does not support a "nonstandard" drive under this version of NetWare),
and you don't have to boot from a floppy, because the drive appears to
the system as having less than 1024 cylinders.

NOTE:

Since the time this message was written, questions have come up about the
compatibility of the Ultrastor's 64x32 translation mode with NetWare.
Per Ultrastor's manual the 64x32 mode is not compatible with NetWare, and
testing in our lab has supported this statement. If you have the chance,
try the 64x32 mode and let me know what you find out. If it doesn't work,
then you'll have to go back to the 16x63 mode, or physical parameters.
-------------------------------------------------------------------------
QUESTION:

I have a customer who wants to use their PS/2 model 80 and drives other
IBM drives. They want to know what all drives disk manager for the ps/2
supports. Also, they are not using DOS 3.3. They are using XENIX. Will
disk manager allow them to low level format their drives so the ps/2
can recognize what they are? Thank you for your time.

ANSWER:

Disk Manager for the PS/2 supports over 300 different drives.
(Too numerous to list here). In general, even if the drive that
you wish to install is not directly supported by Disk Manager,
you can manually enter the parameters of the drive into Disk
Manager, and it will then support it as well, (which means you
can install about any drive). The important thing to know is
that the PS/2 is not able to support a MFM drive that has more
than 1024 cylinders under any circumstance, (unlike AT style
machines). You can install one in it, but the drive will be
truncated at 1024 cylinders. On the other hand, ESDI drives with
more than 1024 cylinders are no problem, since the IBM ESDI
controller will go into a translation mode that makes the drive
appear to have more sectors per track and more heads, but less
than 1024 cylinders.

We do not have a driver to support a nonstandard drive under
XENIX, but since an ESDI drive installed in a PS/2 with Disk
Manager will appear as a standard drive to the system, XENIX
should have to problems with it. If you are installing a
nonstandard MFM drive, some versions of XENIX (SCO for one...)
will allow you to specify the parameters of the drive to it
directly at operating system installation time, therefore there
should be no need for a driver.
-------------------------------------------------------------------------
QUESTION:

I have a Micropolis 1355 ESDI drive, a DTC ESDI controller, and the Disk
Manager/Disk Manager-N package that came bundled with it. This drive
serrved us well for a year in a Novell server before being replaced with
a bigger drive today. It was my plan to install the old drive into my
workstation. No such luck. I don't know where I'm going wrong, but DOS
steadfastly insists the drive is 17spt and the DMDRVR steadfastly
insists it's 34spt. So if the boot partition is readable to DOS in
17spt, once the DMDRVR.BIN is loaded the file system goes away and, of
course, COMMAND.COM suddenly can't be found. If the boot partition is
set at 34spt, none of the MS-DOS files are findable in the first place.
I hate calling for help so I've spent the last three hours trying
various combinations of automatic and manual mode options, sequence of
events, and even reading the whole HELP package one topic at a time.
Still no luck. Documentation assures me that ESDI is no problem, and
that only a certain WDC controller is liable to cause grief. So what is
wrong with my procedure and/or equipment?
Again, remember that this drive and controller, in a different
AT/386 with essentially the same AMI BIOS, worked for a year in a Novell
network server. What the problem is in MS-DOS world, I just don't know.
I've also tried DOS 3.3 and 4.01 both. ...help!!!

ANSWER:

With the DTC 6180/6280, you should be able to enable the BIOS on this
controller, (place a jumper on SW1-4), low-level format the drive thru
debug (G=C800:5), and make the drive appear as a "standard" drive to
the system. This would circumvent the 17/34 spt problem that you are
now having, (DOS should see 34 spt also). After you low-level the
drive through debug, run Disk Manager in manual mode, (DM/M), and go
DIRECTLY to the partitioning menu, (don't even enter the configuration
or initialization menus). Create your partitions and prepare them.
Or... with DOS 4.01, after initializing thru debug, you can just run
FDISK and have the whole thing as drive C:.
-------------------------------------------------------------------------
QUESTION:

Hardware: Micropolis 1355 PC Pak and DM v1.3
Trying to install ELS I v2.12
In middl eof installation I get error
"Problem with Drive 00: Drive not setup for Hot Fix
Abend: Invalid drive passed to dsik process.
I have run DEBUG with and tryied partitioning the drive with DMN
but I dont know if it really writes the aprtition. Though it says
that I am ready to Install Netware the current partition table says
"No Partition"
I have also tried running DMN with the bios disabled. When I tried
this I got basically the same results: it initialized fine, but when I
tried to partition it is said "Ready to install Netware and No Partition
Could you help me with the Novell Message? Can this version of DMN handle
ELS Level I v2.12?

ANSWER:

The bad news is: The version of DMN that you have is not able to
install a "nonstandard" drive under ELS level 1 v2.12.

The good news is: With the WD1007 controller, you can make this drive
appear to the system as a "standard" drive. If you follow the general
instructions in method number 1 of bulletin number 5 on this BBS, (ESDI
considerations), you should be able to complete this install, (barring
actual equipment failure). An overview of this procedure is as
follows: Low-level format, and enter the bad track map through the
controller's BIOS using debug, (g=c800:5). Run DMN in manual mode and
go directly to the partitioning menu. Create your NetWare partition
and scan it. (If after you do this, you are not able to go back into
DMN's partitioning menu and show the partition as actually being there,
then you have a hardware problem with either the disk or controller).
Then proceed with your NetWare install as normal. Since the drive is
now appearing to the system as a "standard" drive, there are no patches
needed.
-------------------------------------------------------------------------
QUESTION:

I have a Minisribe model 3080 and do not have a disk Manager for this
drive type. I have the versoin of disk manager -n that supports it but
I need to create a Dos partition on the drive. Its running in a ALR
386 flexcash machine. WD1007 controller. I need to get a hold of this
manager as soon as possible for one of our clients. I would like to be
able to down load it if I can?

ANSWER:

You may order Disk Manager though our sales department (800)752-1333,
or though most any computer dealer or software distributor, but you can
not download it from this BBS. It is not a public domain or shareware
program.
-------------------------------------------------------------------------
QUESTION:

I am using the DPT Smart Cache ESDI controller card for 2 Micropolis
675MB Hard Drives. The drives were partitioned and prepare with DM
Ver.4.02. Customer wanted to use the drives for very large files, each
of them may exceed 400MB. When the file size was larger than 121MB he
used to get an error message "General failure error in writing drive ...
R)etry,I)gnore or F)ail ? When "F" was pressed a message "Fail INT - 24
" was displayed. It seems that the error was a software one and related
to Disk Manager's limitation.
I had tried to contact you by phone continuously for couple of days and
was unable to get thro' your very limited and busy phone lines.
Appreciate if you will advise if you have been notified of the above
problem and if it has been solved in your later version of DM.
Best regards.

ANSWER:

There is no inherent "limitation" of Disk Manager that would cause an
error message to occur when trying to create a large file. You stated
that the customer "used to get" an error message. By this do you mean
that he is no longer getting an error message?

In general, with the DPT smart cache controller, you should use the DPT
utilities that come with the controller to low-level format the drive.
Then run SWBIOS, followed by DM in automatic mode. When asked if you
wish to "initialize" the drive, answer NO. Disk Manager will then
allow you to partition the drive, and should work fine with any file
size.

QUESTION:

What I meant was the customer is constantly getting the error.
Now I tried DM Ver.4.20 and it seems to work better, it had gone beyond
121MB barrier and worked fine even up to 350MB. However, the customer
called again telling me that the error appears on both drives when there
are 220MB left on the drives.

ANSWER:

The only thing that I can think of as to why you would be having this problem
creating large files is that it is not actually a "large file" problem, but is
in reality a "over 1024 cylinders" problem. When using the DPT controller with
a 600+ Meg drive, there is a very specific install method that you must follow.
The following is a reprint of a document that is being distributed by DPT, and
has to do with using this size of drive with their controller.


Installation Steps for Disk Manager and
DPT Controller with BE3011 to break
DOS 528MB barrier


DOS's maximum addressable parameters for a physical hard disk are 1024
cylinders by 16 heads by 63 sectors per track. This means that the
largest physical drive that DOS can address is 528MB (1024x16x63x512
bytes per sector). In order to get the full capacity of a drive that
is larger than 528MB, Disk Manager must be used to partition and high
level format (prepare) the disk. The low level, or initialization, is
still performed by DPTFMT. Follow these steps:


Equipment needed: PM3011/70 disk controller with Y7 enabled
BE3011 BIOS Expansion Chip
Disk Manager version 4.0 or later
DOS 3.3 or later
Large capacity ESDI disk drive such as
Micropolis 1568
Maxtor 8760
HP 97548E
CDC/Imprimis Wren 6


1. Run DPTFMT with the FORMAT AND CERTIFY option.
2. Select Novell 2.1x from the operating system list after the drive
has formatted. Selecting this OS will set the drive up for 2048
cylinders, as opposed to the drive being setup for 1024 cylinders
in preperation for DOS.
3. Note drive type selected on the next screen. The capacity should be
close to what the drive really is. Write the parameters down on a
piece of paper.
4. Hit return until the system reboots. Boot under DOS with your floppy.
5. Run SWBIOS from your Disk Manager diskette (not necessary with versions
4.2 or later).
6. Start Disk Manager by typing in DM/M (DM/M/4 for Compaq DOS 3.31 or
NEC DOS 3.3+)
7. Enter C for Configuration Menu.
8. Enter N for Non-Standard Parameters.
9. Hit ESCape (do not select drive).
10. Note parameters displayed in upper right corner of screen. These
parameters should match the parameters you wrote down earlier.
11. Enter W to write the parameters to the disk.
12. Enter R to Return to Main Menu.
13. Enter P for Partition Menu.
14. Enter Y to modify partition.
15. Enter N for New Partition Table.
16. Enter N to not use default table.
17. Enter A to Allocate Partition.
18. Enter D for DOS type partition.
19. Select partition size. First C: (bootable) partition must be 33MB or
less for DOS 3.3.
20. Enter A to Allocate another Partition.
21. Enter W for Write-Read Partition.
22. Select partition size. This partition may be as large as 512 Meg under
DOS 3.x, and the remainder of the drive as a third partition, (or split it
up into even more partitions if you wish). With DOS 4.x, you can have the
entire remainder of the drive as one single Write-Read partition.
23. Enter R to Return to Partition Menu.
24. Enter P to Prepare Partition.
25. Enter Y to Prepare all DOS/Write-Read Partitions.
26. Enter Y to place a system on the DOS partition.
27. Follow the prompts until all partitions are prepared. Upon completion
of the preperation (high level formatting), the system should now
boot from the hard disk.
-------------------------------------------------------------------------
QUESTION:

I am having trouble with my version of DMN 2.43 (which I know is old) in
prepping a Micropolis 1664-7 ESDI 300 meg drive for Netware. I would
like to do this tomorrow, and probably can't get a new version by then,
so could you give me the config info so I could create an entry for
that drive.

ANSWER:

If you want to upgrade your DMN to version 3.11, you will need to contact
the sales department at (800)752-1333 and order an upgrade.

I think you should be able to do this install without the upgrade
though. If you can low-level format the drive though the controller
BIOS, (or perhaps the controller card comes with it's own low-level
format routine on a disk), then set the card into 63 sector per track
translation mode, then the drive should appear to DMN, (and NetWare),
as being a "standard" drive with approximately 667 cylinders, 16 heads,
and 63 sectors per track. (The drive is physically 1779x7x54, but
using the translation mode would make it appear to the system as being
667x16x63). You will then be able to run DMN, go DIRECTLY to the
partitioning menu, create your NetWare partition, and SCAN it. Once
this is accomplished, you can proceed with a completely standard,
unpatched NetWare install, and since the drive appears to the system as
having less than 1024 cylinders, you can boot directly from the hard
drive, (instead of a floppy as you would with a drive that appeared to
have more than 1024 cylinders). This all assumes that the Everex ESDI
controller is emulating a WD1007V, and can make the drive appear as
"standard" to the system. If this is not the case, then you will in
fact have to upgrade your DMN to the current version.

Also, I am assuming that you are using the EV-348, which is a 20 Mbit
transfer rate controller. You didn't say "EV-348", you just said
"Everex". If this is not the EV-348, then you'll have to get a 15 or 20
Mbit controller in order to handle this 54 sector per track drive.
-------------------------------------------------------------------------
QUESTION:

OK, I GIVE UP - HERE IS ALL THE INFO - CAN ANYONE TELL ME *EXACTLY HOW
TO SET UP THIS STUFF SO IT WILL WORK!!!!!!
DRIVE - ST-4182E CONTROLLER WD1007V/SE2 CPU-INTEL 386-25
SOFTWARE - NOVELL NETWARE 2.0A [NO SFT] DMN 3.10
NO MATTER WHAT COMBINATION OF DEBUG, DMN AND BIOS I TRY, I STILL EITHER
GET A "DIVIDE OVERFLOW" ERROR OR THE SYSTEM HANGS UP ON THE COLD BOOT
LOADER MESSAGE.

THE DRIVE IS SUPPOSEDLY 969,9,34 BUT CAN BE SET TO 35 OR 36 SPT BY
JUMPERS ON THE BOARD AND/OR SWITCHES ON THE DRIVE.

THERE ARE TOO MANY COMBINATIONS OF VARIABLES, ESPECIALLY THE JUMPERS ON
THE WD BOARD. I NEED HELP!!!!!

WD'S TECH LINE SAID TO JUMPER W1 9-10 TO FORCE TO 35 SPT - THIS DID
NOTHING. ANY HELP WOULD BE GREATLY APPRECIATED.

ANSWER:

The drive should be jumpered at 35spt. The card should have jumper W3
pins 1&2 removed to enable the BIOS, jumper W1 pins 5&6 removed to
enable translation, jumper W1 pins 11&12 removed to disable alternate
sectoring, and jumper W1 pins 9&10 installed to strap it into hard
sector mode, (It then pays attention to what the drive is set to. In
this case 35 spt). Initialize the drive through the controller BIOS
using debug, (G=CC00:5), enter the bad tracks, and then exit the
formatting program.

Let the machine reboot and run DMN in manual mode, (DMN/M). The
parameters on the left of the screen should say "standard" and
967x9x35. Press F6 to create the NetWare partition, and then F7 to
prepare it. When it has completed one pass of the preparation, press
ESC to stop it.

Run MODSPT 35, GENOS, and then INSTALL to continue with your NetWare
installation, being careful not to manipulate the partition table.
Since the drive is being installed as a "standard" drive, you will not
be running MODNET or DMN/I at any time. When you run INSTALL, it will
ask you if you wish to "initialize" the disk. You must answer Y to
this question. Novell uses the word "initialize" to describe the
creation of the FAT tables and Directory area, as well as writing
information into the volume descriptor table. They do not mean "low-
level format" when they say "initialize".

Also, keep your volume sizes down to no more than 127 Meg. Some early
versions of 2.0a had a 128 Meg volume size limitation, and when they
saw a volume size greater than 127 Meg, they would tip over with a
"divide overflow" message. Keep the volume size down to 127 Meg for
the time being, just to be safe. Once you get it up and running, you
can experiment with larger volume sizes if you wish.

Also, make sure your LAN card is installed correctly, and you specified
the correct parameters for it when you ran GENOS. The server must be
physically attached to the LAN and the LAN must be properly terminated
before the server will come up.

If after following this procedure, you are still having problems, let
me know and I'll try to be of more help.

QUESTION:

well, i'm making progress..... i tried what the other bb user said, let
dmn do the whole job, using the imprimis 4182 choice. the system then
goes into what was described by another user in your esdi file - it
hangs up at the cold boot about 50-75% of the time (but it does come up
occasionally!!) my board settings were w1 1-2-on 3-4-on 5-6-on
7-8-off 9-10-off 11-12 off w3 1-2-on all others at default
I installed 2 volumes, 100 and 45 as the install program still only sees
145 of the 151.5 mb that dmn says i have. I will try your method also.
did the other guy ever find any reason for his intermittent cold boot
hangup?? What I was wondering regarding the ARCnet was if he had
replaced one brand with another or an 8-bit with a 16-bit or
verse-vicea.
Oh well, I have 3 days to play, thanks again for all your help.

ANSWER:

Since there is no such thing as "intermittent software", your problem
with the server hanging up at the cold boot about 50-75% of the time
must be hardware related in some way. Have you tried the install
method that I gave you yet? Let me know the results.

QUESTION:

TRIED YOUR METHOD WITH NO RESULTS - THE ONLY WAY IT WILL COME UP IS TO
DISABLE THE BIOS AND TRANSLATION, USE DMN, CHOOSE THE DEFAULT SETTINGS
FOR THE IMPRIMIS DRIVE, AND DO THE WHOLE THING WITH DMN. EVEN THEN IT
ONLY COMES UP ABOUT HALF THE TIME. AM IN THE PROCESS OF SWITCHING BOARD
LOCATIONS, BOARDS, ETC TO TRY TO PIN DOWN THE PROBLEM. 20 VS 25 MHZ HAS
NO EFFECT, WITH/WO CACHING ON THE BOARD HAS NO EFFECT, 7/4 ECC HAS NO
EFFECT, 1/0 WAIT STATE HAS NO EFFECT. ABOUT TO SEND THE WHOOLE THING
BACK AND TRADE FOR AN MFM. ANY IDEAS APPRECIATED.

ANSWER:

Again, the intermittent nature of this problem leads me to believe that
you are having a hardware malfunction or incompatibility. I would try
swapping out the controller, or try putting this controller and drive
into a machine with a different BIOS. Other than that, I'm out of
ideas.

CLOSING REMARKS:

our conclusions exactly!!! we put the wd board and drive, still
condigured the same, into a 386sx with a phoenix bios - came up first
time and worked every time! took the )(*&^%$#[email protected] 386 AMI back to the
distributor, but the only other board they had in stock also had a AMI -
they are going to get us either and award or a phoenix for it.
meanwhile we have another controller coming in tomorrow - will try that.
was also told that you can't use interrupt 2 or 9 for the arcnet card
on a 386, so switched to 7 with no different results. from now on i
stick to mfm!!! will let you know the outcome when and if. thanks for
all your help and info.
-------------------------------------------------------------------------
QUESTION:

WHEN WILL ONTRACK HAVE DOSUTILS SO IT WILL RUN ON IBM'S PS2'S SUCH AS A
MODEL 80 WHICH I HAVE? HOW MUCH WILL IT COST TO UPGRADE TO THIS VERSION
FROM 2.01? THANKS

ANSWER:

Dosutils is compatible with the PS/2 model 80, with the exception of
the "initialization" functions when using an ESDI drive. It is not
able to low-level format an ESDI drive in a PS/2 because of the PS/2's
ESDI translation scheme. Disk Manager for the PS/2 can do this
initialization, and I assume that eventually Dosutils will be upgraded
to have this ability as well, but the current version (v2.01) cannot.
-------------------------------------------------------------------------
QUESTION:

Recently, I got a CiompuAdd HardCache ESDI controller with 4 megs of
read-write cache RAM. I had previously been using an Adaptec 2322, and
although I was able to make it work with my new 15 MB Imprimis WREV VI
94196, together with an older 10 MB Miniscribe 3180, I thought it was a
good excuse to get the cache controller, especially because I'm in the
middle of a big database/graphing project.
After intalling the new controller with the 2 ESDI drives, using the
initializxation utility FMT.EXE that CompuAdd supplies, everything
worked fine. In particular, with the old Adaptec in place, booting
always took a long time, while there seemed to be some sort of hunting
contention going on for the big Imprimis drive. That went away with the
new installation.
However, I wasn't getting any "cache effect." CompuAdd supplies a
software utility for reading and tuning the cache, and this simply
terminated whenever I tried to run it.
A call to CompuAdd tech brought the info "You have to use OnTrack's
DM version 4.2." I called your sales office and ordered v4.20.
The disk that came does *not* have any SWBIOS on it (of course SWBIOS
is needed for drives that even with mapping to 63 spt show more than
19024 cylinders, as the Imprimis with 1632 cylks does). Novertheloess,
I simply copied DMDRVR.BIN to my boot partition (which is on the older,
slower ESDI drive) and now find that HCU.EXE, the HardCache tuner
utility works fine. I have kept it in default until I olearn more about
performance tuning with database records of different sizes, but I now
find that in runs on the "preliminary" databases (about 3-5000 recordsd)
which are worked over about 25 times before merging to the big database,
there is a speed increase of about 600% after the first run-through
that reads them into the cache.
There is a speed increase of anbout 10x in merges of these small
databases into the big database that holds them all.
In short, for people who are using HardCache, which is a great
bargain as far as hardware cache controllers go, get DM version 4.2, or
you will have alol that RAM sitting there, it will get the records read
into it, but it will not give them back for subsequent operations. I
checked both w/Ontrack (this BBS) and CompuAdd before buying the
controller; no one mentioned this 4.2 bit. Also, I observe that in the
download text hints several people ask about the HardCache and asre not
told this incompatibility. One of them they are using it for Novell,
with DM/N, I am not sure of what the effects are there, but it can't be
good, i.e. why pay $1000 for a cache that is nonfunctional? So they
should upgrade, if DM/N has solved this driver incompatibility problem.

ANSWER:

Sorry about the confusion concerning the DM/Compuadd situation. I
simply was not aware, (and am still not convinced), that the caching
ability of the Compuadd card was not able to work correctly with
versions of DM prior to version 4.20. Actually, it may only be that
Compuadd's benchmark program doesn't work correctly with versions of DM
prior to version 4.20, and the actual caching is still occurring. Did
you try making your own benchmarks, or did you rely on Compuadd's
benchmark program? I am not sure how a software device driver could
possibly get in the way of a hardware cache...

It would be nice if every company in the industry would be interested
in working with other companies to find and correct possible
incompatibilities between their products, but unfortunately this is not
the case. We have tried repeatedly to talk with Compuadd and to obtain
sample controllers from them for testing. At this point, they don't
seem to be interested in working with us. Almost every other major
controller manufacturer has been overjoyed at the prospect of having us
test their controllers with our products. So, for now we have to rely
on rumors.

As you have noticed, v4.20 of DM does not use SWBIOS anymore. The
replacement for SWBIOS is XBIOS.OVL. When you upgrade from a version
of DM prior to v4.20, you MUST copy XBIOS.OVL into the root directory
of your boot disk, (as well as the new DMDRVR.BIN file). If XBIOS.OVL
is not in the root directory, then all access to cylinders beyond 1024
will result in a "general failure" error message. If you have not
copied XBIOS.OVL into the root directory of your boot disk, you should
do so immediately.

Since DMN for NetWare functions differently than DM for DOS, (there is
no "driver" as such in DMN for one thing), your concerns about getting
the caching benefit from the Compuadd card in a NetWare environment are
not applicable. (DM and DMN are like apples and oranges).
-------------------------------------------------------------------------
QUESTION:

I am interested in acquiring you new Modgen Utility. Please let me know
how this may be done.

ANSWER:

MODGEN comes with DMN v3.11. From your registration information, I see
that you are using the Micropolis ESDI PAK v2.44 OEM version of DMN. I
don't know if an OEM version of DMN is upgradable to a generic version.
You'll have to contact our sales department and inquire about it.

On the other hand, you probably don't need to upgrade at all. If you
follow the instructions that came with the ESDI PAK, (low-level format
the drive through debug, turn on the 63 sector per track translation
mode of the controller, address the drive as "standard" though DMN),
you would need to patch NetWare at all. (You didn't tell me what brand
and model of controller you have. If it is a WD1007 controller, the
above will work. Other controllers may not.)
-------------------------------------------------------------------------

QUESTION:

Help. I'm use elsii 2.15 on an arche 386 machine, I have run models
and the update of the files went smoothly. I ran dmn /i and then regened
the system. installation went great it updated the files, but when I go
to boot the server in dedicated mode i get an odd pattern on the screen.
so i use the dmn /x2e0 and booted the net$os from the floppy and i get
a abend: general protection int. error. what else can i do ? this is an
emergency, any info would be greatly appreciated
thanks

ANSWER:

Since the 3180 has more than 1024 cylinders, you will either have to
boot the server from a floppy, or low-level format the drive through
the controller's on-board formatting routine, enable it's 63 sector per
track translation mode, and install it as a "standard" drive. Both of
these procedures are outlined in bulletin number 5, and in the ESDI_Q&A
file. If after downloading and reading these files you still have
problems, then leave me another message, and I'll give you a step by
step install procedure.

I would not mess with the /X2E0 switch, as it was really only needed
with one old version of Advanced NetWare v2.1x.
-------------------------------------------------------------------------
QUESTION:

I am trying to install a Micropolis 1568 ESDI drive (760 meg) with
a WD 1007v-SE2 controller. I've read the tech bulletin on ESDI
considerations. I used the WD controller BIOS and set the drive to
1023x16x63 - as suggested in the first option in the bulletin. By the
way the actual geometry of the 1568 is 1630x15x34. Anyway, I used
DM version 4 in manual mode, created one partition of 1022 cylinders and
formatted it. (four hours). Disk Manager put DOS version 4.01 on it and
it booted up fine 🙂 hooray!

But I've only got 527,138,816 bytes of disk capacity. Here's why:
512 bytes per sector times
16 sectors per cluster times
64,348 clusters per disk equals 527,138,816 bytes per disk
From the translation: 1022 cylinders x 16 heads x 63 sectors
equals 1,030,176 sectors which matches 16 x 16,348 from above.

Am I missing something? How does the suggested 1023x16x63 ESDI
translation get access to the rest of the capacity of the drive?
Thanks for your help with this.

ANSWER:

With the WD1007V-SE2, if you low-level the drive through the controller
BIOS, it will truncate the drive at 1024 cylinders which when in 63 spt
translation mode amounts to about 504 Meg. Since the controller is
truncating the parameters, you have no access to the rest of the drive.

(1024 cyls * 16 hds * 63 spt * 512 bytes/sec = 528,482,304 total bytes)
(528,482,304 / 1024 / 1024 = 504 Megabytes)

In order to get the full capacity from this drive, you will have to
disable the 63 sector per track translation mode, which will make the
drive appear to the system as 1630 cyls x 15 hds x 53 spt (not 34 spt
as you stated). Then run DM in manual mode, (run SWBIOS prior to DM if
you have DM v4.03 or earlier), go to the configuration menu, press S
for standard parameters (which should be showing as 1628 x 15 x 53),
press W to write the configuration to disk, then go to the partitioning
menu to create and prepare your partitions. Anything above the 1023rd
cylinder will have to be a Disk Manager write/read partition.

Using this method will give you access to the full capacity of your
drive.

(1628 cyls * 15 hds * 53 spt * 512 bytes/sec = 662,661,120 total bytes)
(622,661,120 / 1024 / 1024 = 631.96 Megabytes)

(760 Meg is the UNFORMATTED capacity of this drive. After formatting
for DOS, you will get 631.96 meg of capacity. This difference between
unformatted and formatted capacity is caused by DOS, not Disk Manager.)

You could also disable both the BIOS and translation on the controller,
and do the whole job though DM as a nonstandard drive. Using this
method would give you the full capacity of the drive as well, but you
would want to restrict your DOS boot partition (drive C:) to about a
megabyte or so in order to avoid wasting space on the disk. When
installing a nonstandard drive, the DOS boot partition will only use
the "standard" portion of the drive. This "standard" portion of the
drive is whatever your CMOS drive type states the number of heads and
sectors per track are. You will probably have a head and sector per
track mismatch between the "setup as" and "is actually" parameters as
shown in the configuration menu of DM. The DOS boot partition will
only utilize the "standard" number of heads and sectors per track. So
if you use this method, create a 1 or 2 meg DOS boot partition, and
then you can have the entire rest of the drive in one big chunk as a
write/read partition, which you will access as drive D:.
-------------------------------------------------------------------------
QUESTION:

I'm setting up a Micropolis 1568-15 ESDI drive using the Ultrastor
controller and I'm clueless as to how to procees. I wanted to leave a
few questions. I know that I have to LL format using the Ultra's
onboard BIOS, but: How do I respond to SECTOR MAPPING, TRACK MAPPING,
and CYKINDER TRUNCATION? Once thats done and I go to DM (version 4.02)
whats next? Do I need to call up SWBIOS first? Do I just go the
partitioning menu and define and then PREPARE my partitions? Do I have
to write any configuration information? How should I set the Drive Type
in the computer's (AMI) BIOS? Finally, why does DM see my DOS 3.3 and
identify it as DOS 3.2? Does it matter?

ANSWER:

Here's what my database says about the Ultrastor controller:
-----
Ultrastor 12F ESDI controller, 20 Mbits, 16 Mhz bus speed, Hard sector
drives, JP10 all open = Bios disabled. 5&6 closed = g=C800:5 for LL
format. Has 16x17, 64x32, 16x63 & native SPT options available. Also
has track mapping option, (purpose unknown as of 6-26-90), which is
identified as a means of overcoming 528 Mbyte limit for DOS but not for
NOVELL, OS/2, or UNIX/XENIX. Drive parm setup menu will set CMOS, then
force a reboot, before entering the bios again to LL format the drive.
-----
I would set the sector mapping to 64 head by 32 sectors/track mode,
forget about track mapping, and set truncation to NO. Since you are
using the 64x32 translation mode, the number of cylinders will appear
to be about 600, so you won't be using SWBIOS. Then run DM in manual
mode, go directly to the partitioning menu, create and prepare the
partitions, making sure to keep your partition sizes below 512 meg
maximum.

As for the 3.2 vs 3.3 question, DM copies the DOS version information
onto the hard disk's master boot record (MBR) directly from the DOS
diskette's MBR. You DOS diskette must have incorrectly identified the
version as 3.2. All in all, it doesn't really matter. As long as it
responds v3.3 when you do a VER command.
-------------------------------------------------------------------------
QUESTION:

For my Novell install, should I use MODDRVR to change the ISADISK.OBJ
file or should I be using an ESDI disk driver?
I'm using debug to low level format the drive and allowing drive
geometry translation. (63 SPT x 16 HEADS x # cyclinders).

ANSWER:

With the WD1007V-SE2, if you are low-leveling through the controller
BIOS and are using the 63 spt translation mode, the drive appears to
the system as a "standard" drive and you don't need to run any NetWare
modification utilities at all. After low-leveling the drive through
debug, run DMN in manual mode, (DMN/M), and the drive parameters should
be appearing as "standard" and 6XX x 16 x 63. Don't try to pick the
94186-383H at this point, (which would really confuse it), just press
F6 to create the NetWare partition, followed by F7 to prepare it for
NetWare. As far as NetGen goes; Just link the straight, unmodified
ISADISK.OBJ into your NetWare, and it will see the drive. The ESDI.OBJ
driver is for PS/2 installs only. Since you are installing this as a
"standard drive", you won't be running any NetWare modification programs.
-------------------------------------------------------------------------
QUESTION:

I have an SCC Technology, 386sx with 1.2,1.44 floppies, a 2400 modem,
an IMSI mouse, a ESDI 6280-15T controller, and a Maxtor XT-8760E,
677mb (formatted) drive.
I am using DOS 4.01, and ran into a problem today.
I was doing a backup onto a BSM tape drive. It was about 80% of the
way through the backup of about 150MB of data, when it complained that
something was wrong with one of the files, a larege one of 62 MB. I
discontinued the process, and started to take a look at the disk and
that particular file when it became very obvious that the data on the C:
partition was gone. At one point the system told me that there was an
invalid or missing command.com, and it was missing, and I could not get
it to recover, so I reset, and then everything was gone.
We do CD-ROM applications, and I was working on the very final
stages of the release when all this happened. I have a previous backup
but it is two weeks old. If I can save two weeks of work fine. If I have
to reformat, I want to do it under DOS 3.3x versus DOS 4.01, but I want
to have access to the full 677mb available on the MAXTOR. The other
reason for DOS 3.3x is that I am having problems with some
applications under DOS 4.0x

ANSWER:

It sounds like the BSM tape software might not be compatible with a
drive with more than 1024 cylinders. If this is the case, you might
have had a 1024 cylinder "wrap" occur. What this means is that the
tape software may be bypassing our driver, and when the software tried
to write to the 1025th cylinder, it couldn't handle access to beyond
the 1024th cylinder by itself and actually wrote to cylinder 0 instead,
(it "wrapped" around). A call to BSM would be in order. Ask them
specifically if their software is compatible with a drive that has more
than 1024 cylinders.

Above and beyond that, your main interest is salvaging your data. The
absolute best bet would be to contact our Data Recovery Department at
(800)872-2599. Other than that, there is a remote chance, (assuming
that you did actually have a 1024 cyl "wrap" and not a actual hardware
failure), that you could get your second partition back. This
procedure is already detailed elsewhere on the board. Instead of
writing it all again, I will refer you to messages: 726, 736, 741, 751,
and 775. These messages are a conversation with another user that had
a "wrap" occur, and with some help he was able to get his second
partition back. If you type in R;726 from the main menu of this BBS,
you will be able to read message number 726. At the end of that
message, at the "End of message command?" prompt, type in a T to do a
"thread" read. This will take you to message 736 automatically. At
the next "End of message command?", just hit ENTER to proceed to the
next "threaded" message, which will be number 741, etc...

If after reading these messages you still have further questions, just
ask. (And if all else fails, don't forget about the Data Recovery
Department).
-------------------------------------------------------------------------
QUESTION:

I currently have an ACER 386 System running Netware 286 Ver. 2.12. It
has a Miniscribe 9380E disk attached to a Western Digital WD1007A-WA2
ESDI controller. This combination was installed using DMN Ver. 2.41. I
am planning to install another 9380E and WD1007A controller in the
server in the duplex mode. I am going to purchase a current copy of DMN
(Ver. 3.11) to perform this installation. Please let me know what the
correct procedure is to complete this installation, along with any
pit-falls I should be aware of. Thanks for your time and assistance!

ANSWER:

As was suggested by the tech that you spoke to yesterday, you will need
to purchase a copy of DMN v3.11 for this server. Also, duplexing
support is a function of the NetWare disk driver that you are using,
not DMN. In your case, with Advanced NetWare 286 v2.12, the disk
driver is called ATDISK.OBJ. I don't think that ATDISK.OBJ supported
duplexing, (a call to Novell would be in order to find out for sure).
I believe that the current NetWare driver (ISADISK.OBJ) is capable of
duplexing, (again, call Novell to be sure). What this all means is
that you may have to upgrade your version of NetWare in order to
accomplish duplexing. Once you have a driver that can duplex, and you
have that driver and your hardware configured correctly for duplexing,
(per Novell's instructions), and the driver linked into NetWare, then
DMN v3.11 when run from NetGen will be able to see both controllers and
drives since it is talking through the driver.
-------------------------------------------------------------------------
QUESTION:

I have an ACER 1100 20 mhz 386, WD 1007A with two MS3180Es attached,
parted out with DM 4.03 as C: - 6 megs D: - 144 megs, disk2 E: 150 megs
I have a mountain file safe tape drive. Backed up the E: drive instead
of the D: drive and formatted the D: drive using DM/M. Went to restore
D: drive and discovered mistake. I need to recover the data on D:
partition.

I have DOSUTILS, have tried DL but to no avail, I see nothing. Do you
have any suggestions? I need this information bad.

Read through the bulletins. Our Experience here has been to NOT RUN ANY
DISK OPTIMIZERS with ontrack, optune, Norton or any of that stuff. It
has also been our experience to have more than one copy of a backup, but
I kinda didn't keep to that policy, eh?

ANSWER:

When you said that you "formatted" the partition, did you mean you high
level formatted the partition, ("prepared" the partition), or did you
mean that you low-level formatted the partition, ("initialized" the
partition)?

If you have "prepared" the partition in question with DM, then the Data
Recovery department may have a chance of piecing things back together
again manually. If you have actually "initialized" (low-level
formatted) the partition in question with DM, then there is absolutely
no chance of getting any data back.

There is nothing that you or I can do with either DM or Dosutils in
either situation. If you have "prepared" the partition, the FAT tables
and the DIR tables have been zeroed out. The data is still physically
on the disk, but there is no way to know which sector goes with which
sector and belongs to which file, or even what that file name is. What
you have is a bunch of sectors on the disk that don't belong to
anything. It's like a jigsaw puzzle where all the pieces are square.
It would probably even take the DR department quite a bit of work to
get anything usable off the disk. The problem is made even worse if
you have not used any "optimizers" to keep the files unfragmented as
you said.

Your only chance is to contact the DR department at (800)872-2599, and
see what they have to say about it.
-------------------------------------------------------------------------
QUESTION:

we got an hitachi dk515-78 disk with an Adaptec 2328 contrller
it has 1360 cyls, 14 hds, 69 s/t. We try to get as much as possible
usable for Netware 386. We seem limited to 1024 cyl, 16 hds, 63 s/t
by using BIOS' of cntrller. Any clue?

ANSWER:

Since DMN does not support NetWare 386 at this time, and also has an
upper limit of 63 spt, it's not going to be able to help you. It
sounds like 1024 x 16 x 63 will indeed be your limit on this drive with
this controller. On the other hand, you may want to investigate the
Ultrastor 12F controller, which has a PS/2 style translation mode of 64
heads by 32 sectors per track, therefore it doesn't go over 1024
logical cylinders until you hit a gigabyte. It may be able to give you
the full capacity of this drive under NetWare. Check with them first
to be sure though...
-------------------------------------------------------------------------
QUESTION:

I have a Compaq 386/20E running on Compaq DOS 3.31(I have 4.01 also). I
have a WD 1007 ESDI controller with the floppy drive section jumpered
off on the card. I am using a Micropolis 1558(320-380Meg) hard drive. DM
4 sets up the drive for me(30 meg C-L drives) but I have to hit F1 when
I boot the system. The drive was set up by DM as a type 28. When I try a
straight DOS setup I get 250 meg from the drive but it boots normally. I
hope you have some ideas what I can do. The F1 is a problem for the BBS.
If I loose power and no one is home to hit F1, the BBS is down. Thanks.

ANSWER:

Concerning the controller that you are using: Is it the BIOSless
Compaq OEM version of the WD1007A, or is it a real WD1007A with a BIOS?

If it is the BIOSless Compaq version, you will have to manually run the
setup program that came with your machine and set it up for the drive
type with parameters of 618x16x63 (WD1007A translation mode), then run
DM/7/4.

If it is a real WD1007A with a BIOS, you could enable the BIOS and
translation by removing the jumper from W14 and placing it on W3 of the
controller. Then set your CMOS drive type to type 1, run debug, at
debug's hyphen prompt type in G=C800:5, (which will get you into the WD
low-level routine), low-level format the drive and enter the bad
tracks, select the option to set the drive type, hit the + key until
you see 618x16x63, hit return to accept these parameters, and then
either partition the drive with DOS alone or DM/4/M.

In either case, you will be using the WD1007A's 63 sector per track
translation mode. This allows the drive to appear to the system as
though it has a larger number of sectors per track and a larger number
of heads, but a smaller number of cylinders, (to get under the 1024
cylinder limit).

There are other possible procedures as well, (as always, this is a
multiple choice question with more than one correct answer), and if you
continue to have troubles I could review those with you as well.
-------------------------------------------------------------------------
QUESTION:

I thought that I would ask a question first rather than after I get
myself into trouble. We are bidding a network with a 600MB drive. We
used a MAXSTORE 8760E with DTC Controller on the last on without any
problems. Does you have any experience with the Fujitsu
M2263E and what controller would be appropriate for a ADV NTW 215C.

ANSWER:

The drive itself doesn't really make much difference. What you need to
be careful of is selecting the correct controller for your drive.

When using a 600+ meg drive, you have to make sure you are using a
controller that is capable of at least a 15 megabit transfer rate,
(because of the high number of sectors per track on the drive). You
also need a controller that can either pass the true parameters of the
drive to the system, (all 1600+ cylinders instead of truncating the
drive geometry at 1024 cylinders), or has a translation mode that can
handle this large of a drive, (such as a 64 head by 32 sector per track
translation mode). The old style 16 head by 63 spt translation mode
that most controllers offer breaks down at about 503 meg, (it goes over
1024 cylinders). Of course, the controller must also be NetWare
compatible.

The Ultrastor 12F is one example of a controller that fits the above
criteria, and I am sure that there are others as well.
-------------------------------------------------------------------------
QUESTION:

I am having a slight problem with my Micropolis 1664-7 hard drive, Ultra
store and DM ver 4.22.
I have set the drive parameters through the controller bios using debug.
I then ran dm/i/d set the drive to 1664-7 entered the defect list with
BFI included set the interleaf to 1 and the rest to the default values.
DM then initialized the disk and I set my partitions to 100 mb primary
and the rest secondary. The setup went fine. The problem comes when I
boot the computer. I get a bad boot on every other boot. The message I
get is " Disk boot failure, Insert system disk and press enter". I warm
boot after this message and the disk boots fine. Any ideas on how I
might a good boot on every boot? I am using msdos 4.01. Thanks.

ANSWER:

When you low-level formatted (initialized) the drive with DM after
setting the drive parameters through the controller BIOS, you undid
whatever you accomplished while in the BIOS routine. This is because
the controller writes information to the disk in order to keep track of
what it is supposed to do. When you initialized the disk through DM
afterwards, you zeroed out this information.

With the Ultrastor controller, you should low-level format the drive
through the controller BIOS, set the controller into it's 64 head by 32
spt translation mode, and then partition the drive only with Disk
Manager as a "standard" drive. Using this method, you could have the
whole drive as drive C: if you wish, since this drive (1664-7) would
appear to the system as a "standard" drive having only about 320
cylinders, but 64 heads and 32 spt.

QUESTION:

Thanks for your quick reply. I did as you advised except that I set the
ultrastore controller to 63 spt translation mode. My parameters are now
665x16x63. Anyway every thing is as before. The disk works fine but I
still have to boot twice to get it going. I think I may have a hardware

problem either with the disk or with my computer bios. What do you
think? Your thoughts on the matter would be most welcome. Thanks again.

ANSWER:

I agree. If you have followed the procedure that I suggested and you
are still having problems, it sounds like this may actually be a
hardware based rather than a procedural based problem. I would not
suspect the computer's BIOS, but would suspect either the controller
card or the drive itself. Do you have another drive or controller that
you could swap into the machine and try?

QUESTION:

Regarding the problem I wrote to you about a couple of weeks ago -
Micropolis 1664-7 hard drive and the Ultrastore controller ( having to
boot twice to get the hard disk recognised ).
I found the source of my problem and thought I would share my findings.
The problem is that the Ultrastore controller will not work properly
when I have two video cards in the system.( no matter what bios address
I set for the controller). If I were to remove one of the cards the
computer boots from the hard drive first time everytime! If I put the
second video card back in I have to boot twice. Unfortunately I need two
monitors to run Lumena ( a graphics program ). So, I am stuck with
having to boot twice. I guess I could change controllers, but will I
have the same problems? Sigh... such is life.

ANSWER:

Thanks for letting me know what was causing this. I'm glad you got it
figured out. Whether or not you would have the same problem with a
different controller is anybody's guess. Does Ultrastor or the two
video card manufacturers have anything to say about this?
-------------------------------------------------------------------------
QUESTION:

I am upgrading my server from ELSII NetWare to Adv. Netware 2.15C.
I successfuly used Disk Manager-n to install ELSII.
Disk Manager-N instructions for Advanced Netware advise to run Moddrvr
prior to running NETGEN. The documentation mentions inserting the disk
DSK_DRV_001 which should contain the file atdisk.obj. The Ver. 2.15C
disk does not contain that file. It does contain a file named
ISADISK.OBJ dated 2/1/90. Is this the replacement for that file? Do I
need a more recent version of Disk Manager-N.

ANSWER:

If you are low-level formatting this drive through the WD1007A's
controller BIOS using debug, (G=C800:5), then the drive appears to the
system as a "standard" drive, and you wouldn't have to run MODDRVR at
all. After low-level formatting the drive through the WD1007's BIOS,
you would run DMN, and then go DIRECTLY to the partitioning menu to
create and scan your NetWare partition. You would not select the drive
model number at all from within DMN.

Otherwise, if you wish to disable the BIOS and translation on the
WD1007A and install this drive as a "nonstandard" drive, then you will
have to upgrade your DMN though our sales department since DMN v3.04
was released prior to v2.15C NetWare, and therefore is not able to
patch it correctly for a "nonstandard" drive.
-------------------------------------------------------------------------
QUESTION:

In reading your messages (the big ESDI text file), I noticed
something about an Ultrastor controller being a controller
with "enough" translation to take my Miniscribe 9780E drive
below 1024 cylinders. Only one problem -- no one has heard
of it that I have talked to. Do you have a voice phone
number, fax phone number or address for them?
Also, do you know of any other controllers with similar
geometry reductions? (ie. with one factor (heads or sectors)
doubled which would cause the 1660 to be divided by 2 and
the cylinders would become approximately 830).

ANSWER:

There is at least one other controller manufacturer that makes a
controller capable of "BIG" translation schemes. DPT or Distributed
Processing Technology of Maitland Florida make a SmartCache ESDI
controller model # PM3011E/75 that will allow you to get your drive
down below the 1024 cylinder limit. Their phone # is (407) 830-5522
or FAX # (407) 260-5366. The Ultrastor Corporation is located in
Irvine, California their phone # is (714) 581-4100. The model # of
their controller is ULTRA 12(F) ESDI....
-------------------------------------------------------------------------
QUESTION:

We have recently purchased your DMN for help in installing Novell networks.
We have a problem with installing a Micropolis 1558 as the only disk in a
ELS 2.15 Level II file server. We keep getting garbage to the screen
when we try to boot the network. How can we avoid this??

ANSWER:

I understand that you have spoken to one of our techs on the voice line
since leaving your message on this BBS. When you install a drive with
greater than 1024 cylinders under NetWare, you must boot the server
from a floppy as stated in the HELP program on the DMN diskette.
Failing to do this will result in a screen full of garbage as you have
noticed. If you are still having problems with this installation, or
have other questions, just ask.

Also, you stated in your message that you purchased a copy of DMN to
help in installing Novell networks. I notice that you used the plural
form of the word "network". I feel that I should remind you that DMN
is licensed for one copy per server. Using the same copy of DMN to set
up more than one server is a violation of the license agreement.
-------------------------------------------------------------------------
QUESTION:

Reading the on-disk docs for DM-N and Netware/286 v.2.1x I'm
a bit concerned about what my end-product will be after I install
this Micropolis ESDI 1558-15 drive (337 Mb formatted capacity).
This will be my primary server drive (only one, in fact); does Netware
indeed limit me to a 255Mb volume? Must I then define a second volume
of (337-255) Mb? Will I then be able to boot from the 255 Mb vol,
or will the number of physical cylinders (1224, I think) disallow
me from booting from this physical disk?

ANSWER:

The Micropolis ESDI PAK usually comes with a WD1007A-WA2 controller, so
for purposes of discussion I'll have to assume that this is what you
actually have.

With the ESDI PAK version of DMN, you must follow the directions for
low-level formatting the drive through the controller's BIOS using
debug as outlined in the ESDI PAK written documentation. Once you have
accomplished this, the drive will appear to the system as a "standard"
drive with less than 1024 cylinders. The fact that it appears to the
system as having less than 1024 cylinders means that you will be able
to boot directly from the server's hard drive.

After the low-level formatting, you will proceed with the directions
found within the HELP.EXE program on the documentation diskette that
comes with ESDI PAK. Briefly, all you do is run "DMN SETUP4" and it is
an automatic process from there.

NetWare v2.1x does indeed limit volume sizes to 255 meg maximum, so you
will have to split this drive into at least two volumes. Be careful
not to confuse "partitions" with "volumes". The NetWare partition is
defined from within DMN. NetWare limits you to one NetWare partition
per disk, although that partition can cover the entire disk. The
NetWare partition is then split into multiple "volumes" from within
NetGen.

QUESTION:

We've installed successfully the Micropolis drive per your Ontrack
and BBS instructions. One anomaly: SYS (first volume) has 255 megs, as
expected; but "SYS2" (the leftovers) is only 40 Mb. Now, DM-N didn't
report any bad tracks, and the manufacturer's defect map had only one;
part of the discrepancy may be due to Hot Fix (?). Where did the rest
go?

ANSWER:

Actually, 255 meg & 40 meg volume sizes sounds about right.
The Micropolis 1558-15 is 1224 cylinders by 15 heads by 36
sectors per track, (spt). Two cyls are reserved by the
controller at the end of the drive for drive parameters, and bad
track list. This leaves 1222 cyls available for data storage.
The WD1007A formats the disk at 35 spt by default and reserves 1
spt for sector sparing leaving 34 spt for data storage.
1222x15x34 is then translated to 618x16x63 by the controller to
make the drive appear to the system as having less than 1024
cylinders. DMN vN2.44 will see the drive as 618x16x63 and will
also cut a couple of cylinders off for landing zone and CE
cylinder. Since DMN vN2.44 is not aware that these are
translated parameters, it will cut off two translated cylinders
leaving 616x16x63 of available disk. Hotfix will take another
2%, and the difference between millions of bytes and megabytes
accounts for the rest.

616 * 16 * 63 * 512 bytes/sector = 317,915,136 bytes.
317,915,136 - 2% (default) for hotfix area = 311,556,833 bytes.
311,556,833 bytes / 2^20 = 297 megabytes for data storage.
297 meg - 255 meg for first volume = 42 meg leftover for vol2.
-------------------------------------------------------------------------
QUESTION:

I am installing Advanced NetWare 286 v2.15C on a computer with a
Micropolis 1568-15 and an Ultrastor 12F ESDI controller. Can you
give me any pointers?

ANSWER:

When using the Ultrastor controller, you must low-level format the
drive through the Ultrastor's own low-level formatting routine built
into it's onboard BIOS. To access it, you have to run debug, and at
the debug hyphen prompt, type in: G=C800:5 and hit return. This will
bring up a menu and will allow you to low-level format the drive. Once
the drive has been low-level formatted, you will need to set the
controller into it's 64 head by 32 spt translation mode and save those
settings. This translation mode will make the drive appear to the
system as having less that 1024 cylinders, allowing you to boot the
server directly from the hard disk. This procedure also allows the
controller to make the drive appear to the system as a "standard"
drive, and therefor there is no need to patch your NetWare to accept a
"nonstandard" drive, (you won't be running the MODDRVR patch program at
all).

More detailed information concerning the Ultrastor's low-level format
routine accessed through debug will have to come from your Ultrastor
manual, or from Ultrastor tech support directly.

Once you have completed the above procedure, you will need to run DMN
in manual mode, (DMN/M). As soon as the main menu comes up, confirm
that the drive parameters appear as being "standard", and that those
parameters are showing as approximately 640 cylinders, by 64 heads by
32 sectors per track. If this is not the case, then you have missed a
step somewhere along the way and should go back to the debug routine to
check your work. Once the drive appears to DMN as "standard" and the
parameters are as shown above, all you need to do is press the F6 key
to create your NetWare partition, and then press the F7 key to prepare
that partition. At no time should you ever attempt to select the
Micropolis 1568-15 from within DMN. Since you are using the controller
to translate the parameters of the drive, telling DMN that this is a
1568-15 would confuse it greatly. Simply use DMN to create and prepare
the NetWare partition. Once this is accomplished, install NetWare as
you normally would.

NOTE:

Since the time this message was written, questions have come up about the
compatibility of the Ultrastor's 64x32 translation mode with NetWare.
Per Ultrastor's manual the 64x32 mode is not compatible with NetWare, and
testing in our lab has supported this statement. If you have the chance,
try the 64x32 mode and let me know what you find out. If it doesn't work,
then you'll have to go back to the 16x63 mode, or physical parameters.
-------------------------------------------------------------------------
QUESTION:

I'm hoping that you'll steer me in the right direction. I have an IBM
PS2/80 server running Novell SFT 2.15C. It has a standard 300Mb IBM
drive with the original IBM ESDI controller. We need to upgrade this
machine so we bought a Maxtor XT8760E 652Mb drive with a Western
Digital WD1007V-MC1 controller and DiskManager-N ver. 3.11. After
looking through the DMN help files and the Novell manuals, I'm a little
disoriented and not sure about where to start. I'd like to impact the
current server as little as possible because it is a production machine.
I'd also like to be able to tell my colleagues that we will be able to
make full use of the new drive without much problem. Thanks in advance.

ANSWER:

I'm going to recommend that you reformat the first drive when you
install the second in your server. This is for two reasons:

1. You are changing the controller board. Any time that you change a
controller board you should reinitialize the drive, even if the
controller board is the exact same brand and model as the original
one. Even identical boards can be a bit out of calibration in
opposite directions from each other, which could cause problems.
When changing the actual brand and model of the controller, it is
sometimes absolutely necessary to reinitialize, and always highly
recommended.

2. When using DMN to add a drive to a PS/2 server that already has a
drive that was originally prepared with COMPSURF, it is highly
recommended that you reprepare the first one at that time as well.
The IBM PS/2 ESDI translation mode makes the drives appear to the
system as having 64 logical heads. NetWare, when not patched by
DMN, will for some reason only use 63 out of these 64 heads in the
NetWare partition, which is fine until you introduce DMN. As part
of the installation process, DMN patches NetWare to use all 64
heads. After the install, the next time NetWare looks at the first
drive it will attempt to use all 64 heads, but the files that are
already on that disk were written as if it only had 63 heads. This
will more than likely cause problems.

Using DMN to reformat the first drive as well as formatting the second
will circumvent both of these potential problems. So my recommendation
is to backup the data on the first drive, and then redo both with DMN.

P.S. Make sure that the Maxtor 8760 is physically jumpered for 53
sectors per track, and not 54. The WD1007V-MC1 cannot handle a 54
sector per track drive at a 1:1 interleave.
-------------------------------------------------------------------------
QUESTION:

INSTALLING ADVANCED NETWARE 2.11
EQUIPMENT:
MODEL DFI-MINI286 12MHZ COMPUTER WITH 3MEG RAM
HITACHI DK-515 78 ESDI HARD DRIVE - 780MEG UNFORMATTED 671MEG FORMATTED
ADAPTEC ACB-2322D ESDI CONTROLLER
AWARD 3.03 BIOS

USING THIS MACHINE TO INSTALL SOFTWARE ONTO DRIVE ONLY. WILL
THEN PUT DRIVE AND CARD INTO 386 DFI COMPUTER SAME BIOS WITH 6 MEG

NO PROBLEM WITH LOW LEVEL FORMAT, MODGEN OR NETGEN AND CONFIGURING
SYSTEM. COLD BOOT LOADER INSTALLS OK. WHEN ASKED TO INSERT GENDATA AND
CLOSE DRIVE DOOR, THE FOLLOWING MESSAGE DISPLAYS:
ABEND: NOT ENOUGH MEMORY FOR DISK FAT TABLES

AT THIS POINT I RELOADED 3.3 PCDOS WITH A MEMORY DRIVER CALLED EMMA.SYS
TO USE THE EXTENDED MEMORY AS EXPANDED MEMORY JUST TO INSURE THAT I HAD
ENOUGH MEMORY. SAME EXACT RESULTS. I READ THE MESSAGE ABOUT NETWARE
NEEDING 1K FOR EACH MEG OF DISK SPACE IN THE ESDI_Q&A FILE. EITHER MY
286 WILL NOT ALLOW IT TO GO ANY FURTHER AND I NEED TO LOAD ALL OF THIS
ON THE 386 OR I AM MISSING SOME KEY STEP. DM-N SAVED ME A TON OF TIME.
GREAT PRODUCT. ANY THOUGHTS OM THIS CHALLENGE ?

ANSWER:

In general, NetWare uses extended memory exclusively, not expanded
memory. Your memory card must be physically able and jumpered to
provide true extended memory as it's default. Also, if your system
board only has 512k of RAM on it, the add-on board must also backfill
the remaining 128k to get base memory up to 640k. This alone will not
circumvent your error during NetGen though.

Versions of Advanced NetWare prior to v2.12 attempted to load the FATs
into memory during the install portion of NetGen for some reason.
Since Netgen is run under DOS, the only available memory is 640k. The
FATs, along with any other memory resident programs, (including NetGen
itself), can quickly use up the available memory and generate the error
"ABEND: NOT ENOUGH MEMORY FOR DISK FAT TABLES". Large drives aggravate
this condition.

Solution: Try an install without the NetGen utility, as described
below from a Network drive.

Prompt should be at W:\NETWARE> (or equivalent), with discreetly named
subdirectories containing the various Novell diskettes.
Then enter these three command lines:

Netgen\instovl 1 support
Netgen\ninstall pn o1 (o not zero)
Netgen\instovl 2 support

Or if you are doing a floppy based NetGen, with the NetGen diskette in
drive A: and at the A:\> prompt enter these three command lines:

instovl 1 support
ninstall pn o1
instovl 2 support

This procedure should get you around the error, (or you could upgrade
to Advanced NetWare v2.15...)


Now, I have a couple of questions for you...

Given: 1. The Hitachi DK515-78 is 1359 cyls x 14 hds x 69 spt.
2. Neither DMN nor NetWare support drives with > 63 spt.
3. The ACB2322D does not offer a 64x32 translation mode, but
does offer a 16x63 translation as well as a "logical
split" capability to make this drive appear as two
logical devices at the system level, (not to be confused
with dividing the drive into two partitions).

Therefore, the following is my concept of what would have to happen to
get the full capacity of this drive: The card would either have to be
able to present the drive geometry as 1300x16x63, (I don't know if it
will do this or if it will truncate to 1024x16x63), or use it's
"logical split" capability to make your DK515-78 appear to the system
as being two drives, each 650x16x63, (or perhaps the first one would be
1024x16x63, and the second one would be 275x16x63. I don't know).
Either method would enable you to utilize this entire drive, but would
require that you low-level the drive through the ACB2322D's on-board
BIOS and set the proper translation modes at that time as well. You
would use DMN to create and prepare the NetWare partition(s) only.
(When using the "split" option, DMN would then see this drive as being
two distinct devices, and you would therefore create a NetWare
partition on each of them.) NetWare would also see this as being two
distinct devices. One question remains concerning the "split" method,
and that is will NetWare like seeing two "drives" with CMOS set to only
one?

My questions are: How did you set this up? What are the drive
parameters showing as within DMN? (Please be specific with your
answers).

QUESTION:

THANK YOU FIRST OF ALL FOR THE TIMELY RESPONSE.
I WILL NEED TO DO A LITTLE MORE RESEARCH TO ANSWER ALL OF YOUR QUESTIONS
AND ABSORB YOU ANSWERS. I DID NOTICE HOWEVER, THE BUSINESS OF THE
2 LOGICAL DRIVES ON MY CONTROLLER. IT WOULD BE NICE TO HAVE ALL OF THE
EXTRA DRIVE SPACE. I WILL GET BACK TO YOU VERY SOON.
AGAIN, THANKS.

ANSWER:

Since your message said that "everything went fine" up to the point of
installing NetWare on the HD, I guess I assumed that you were indeed
getting full capacity on this drive already. Are you not?
-------------------------------------------------------------------------
QUESTION:

Having spent some time with the intended server this weekend,
and being frustrated as all get out, I have some questions I would
dearly like you to answer.
I have an EPSON IIe with 4 megs of RAM that I have beeying for
a month to install NOVELL 286 SFT Levell II, Rev 215d. Drives are
Seagate/IMPRIMIS 94246-383, controller WD1007V-S21,
Problems:
(1). DMN (Version 3.10) does not run any longer.
It used to, but now it tells me that the FLAWMAP files are invalid.
What are these, and how can I repair them? (BTW, I have no backup
of the original system disk, as the license is valid for only one
server, so what the heck?) Anyway, the FLAWMAP files on the
disk show 6-byte files. So what 6 bytes can I give it to let me
run the thing again?
(2). While I think the documentation for DMN-N stinks, it made
sense to me that I could install the drives as STANDARD without
a problem. Use the WD Bios to do the low-level format, enter bad
blocks, and then get and choice on exit for translation mode.
Now, on the WD controller there are no jumpers for W1, but I am
never offered a 63 SPT translation option. I have paged through
the offerings, but no 63 SPT offering. Why?
(3). I did jumper the drive for 53 SPT, and that is what the
BIOS format routine shows. But no translation mode is offered
other than 17 SPT.
(4). As I understand it, I went ahead anyway with DM-N to try
and install Netware. This went okay, but the Netware partition
editor would refuse all requests for a larger partition that 255
or thereabouts. Your sales literature says that I can use the
full drive under 2.1x. If so, how? On this board you say to
leave the partition table stuff alone in Netware. How else can
I get to use the full capacity of the drive?
Confused, bewildered, and thankfull for any help.

ANSWER:

1. As for the "FLAWPMAP files are invalid massage", it sounds like the
DMN diskette has been damaged somehow and you will need to contact
sales concerning that one. Their number is (800)752-1333. You
cannot repair the FLAWMAP files by hand. You should ALWAYS work
from a working copy of DMN, as you would with any other piece of
software. (I hope you aren't working from your NetWare master
diskettes!)

2. You should be able to install this drive as a "standard" drive by
formatting it though the WD1007V's BIOS. Since you will be
installing this as a "standard" drive, you won't be modifying your
NetWare at all. The only reason I can think of that you wouldn't have
the option of 63 spt mode was if you answered YES to the "Enable
alternate sectoring?" question during that process. This would
prevent you from being able to select 63 spt translation afterwards.
Try re-lowleveling it again and answer NO to the alternate sectoring
question. Otherwise, you'll have to contact WD and find out what's
wrong with your card.

3. If the drive is appearing to the system as having more than 1024
cylinders, the partition editor in NetGen will be confused and will
show only the number of cylinders over 1024, (723 or so in your
case). Are you sure that you aren't talking about volumes instead
of partitions? Under 2.1x NetWare, there is a 255 meg volume size
limitation, not a 255 meg partition limit. DMN will create a
NetWare partition that covers the entire drive, and then you will
have to split this up into two volumes from within NetGen. Since
DMN has already created the NetWare partition, there is no need to
modify it from within NetGen. If the drive is appearing as over
1024 cylinders, you will destroy the partition table if you attempt
to modify it with NetGen. Because of these two facts combined, we
state that you should never even look at the partition table from
within NetGen. If you just leave it alone everything will be fine.

QUESTION:

I am sorry about the sloppiness in referring to volumes/partitions.
It is the DOS mind set yonow.
I am answering no to the alternate (spare) sectoring questions, and
there is no jumper installed to allow it.
I have low-levelled with the WDIOS routine about 10 times now, and
still do not get a 63 SPT option. (I have paged through the choices
several million times).
Is it possible that this choice will not appear if the drive
geometry after 63 SPT will still exceed 1024?
As to using my original disk for DM-N, sorry. And yes I use the
Novell distribution disks to install for all but what is required
for Netgen and Shgen. Guess I am paying the price.
Talked to Western Digital. Problem is that unless on follows the

following routine, you will not be offered a choice of full translation
with the WD 1007V Bios routine:
i). Low level format
ii). Enter defect list (by hand if it cannot be read from the drive).
iii). Verify entire drive.
The key is the verify routine. Unless it is done the full
translation option will not be offered.
iv). Set drive type to 63 SPT.
.
Hope this helps someone else with ESDI and DM-N. The controller
in question is WD1007V-s21.

ANSWER:

Thanks for the information from WD. I don't really know why you would
be forced to "verify" the drive in order to get the translation mode
options to appear, and this is contrary to my previous understanding of
how the card worked, but maybe it's something for other people to try if
they are having trouble.

QUESTION:

Sorry guys...
I tried the low-level format; did the Verify routine; added manually
the bad sector (CYL HD BBFAI); and then chose option 7 to set the
drive type. Again, no translation option appeared.
This is the WD1007V Version 2.2 BIOS.
Am I an unhappy man? Yes.
I'll try again tomorrow.
I agree that the Verify option seems silly given all that
one has to do after the low-level in order to run NETWARE. But
sometimes there are little tricks (quirks) that get the darn
think to work.
I really thought we had this one nailed.
Anyway, I will re-check the drive jumpers to make sure we are not
in soft-sectored mode, re low-level, re verify, and call WD
in the morning.

ANSWER:

If the card and drive are jumpered properly, and you are not selecting
alternate sectoring when you init the drive, and you still aren't
getting the option of 63spt mode, then I would say that the card is
probably broken. You asked if the WD1007V disabled the 63spt mode if
it wouldn't get the drive to appear as less than 1024. It wouldn't
disable 63spt mode, it would still offer 63spt mode but would truncate
the drive at 1024x16x63, (about 503 meg). This wouldn't affect you
anyway since your drive would appear as approximately 642x16x63 in
63spt mode.

After checking the jumpers again, if this still doesn't work I would
try a different controller. You do always have the option of leaving
it at 1747x7x53, (you would need to use DMN to create and prepare the
partition since it would extend beyond the 1024 cylinder boundary), and
boot the server from a floppy diskette... But if the 63spt mode isn't
working correctly, I would be suspicious of the card in general and
would want to try a different one.
-------------------------------------------------------------------------


 January 12, 2018  Add comments

Leave a Reply