ASUS K8V SE Deluxe MoBo
AMD Athlon 64 processor
I have one SATA RAID drive with XP on it, one ancient WD 4GB and a
newer WD 40GB. The 4GB is already an OS/2 HPFS data drive (I only left
it in the box to get the info off it) and the 40GB has a partition set
up for booting, a DOS partition and a HPFS data partition. I am trying
to install OS/2 onto the 40GB into the partition set up for booting.
NOTE: I created this partition prior to the system upgrade and never
installed to it.
I am using a DVD/CD RW drive - I believe ASUS (my Daughter provided
all the hardware and installation into my box as a Christmas present
and I see no identifying marks on the drive).
I followed the instructions I found online at
http://www.os2voice.org/VNL/past_issues/VNL0404H/vnewsft.htm about
removing TESTCFG.SYS and APM.SYS from the CONFIG.SYS on the boot
floppies. Note: I used 4.51 to create the floppies.
The installation proceeds through the Installation Disk, and Disk 1,
but Disk 1 never seems to finish. The screen requests Disk 2, but the
floppy drive (not a new drive) is still acting like it's reading off
and on in a regular pattern.
If I put in Disk 2, a bunch of errors flash up the screen and then I
get:
" S/2 is unable to operate your harddisk or diskette drive. The system
is stopped. Correct the preceding error and restart the system."
I figured that it was possible that the Disk 1 floppy was bad and
attempted to create the floppies from the 4.52 CD, after booting into
XP. This fails when attempting to create Disk 1.
At this point, I'm stuck. I don't know much about what I'm doing here
and would appreciate any help offered. In particular, how to create
the boot floppies from the 4.52 CD and any drivers I need to
add/remove from the floppies. It seems I saw something some time back
about a problem creating the floppies under 4.52, but I don't recall
what the solution was.
In the information I read at the above mentioned website, it was also
suggested to get the latest OS2LDR and run the PatchLdr utility on it.
I dunno anything about this suggestion. The OS2LDR is what came on the
Software Choice CD. - I could try to get the one from the 4.52 CD
instead of the 4.51 CD if that would make a difference.
Thanks in advance for your help. Please CC replies to my email or
contact me directly via email as the system I'm using for posting here
is rather primitive and I don't want to miss out on any suggestions.
Stick the IDE drive into an existing machine that has OS/2 on it. Xcopy
the existing system on to that volume. Modify to remove APM and Testcfg
drivers from config.sys. Move the disk back to the new machine. It
should now boot. You'll have to go through the process of changing the
LAN NIC and audio drivers, but since you are using SNAP or SNAPse on the
old machine, video should work on the new machine.
This is the easiest way, believe it or not, of getting AMD64 working
with OS/2.
BTW, You cannot use boot floppies for 4.51 to boot 4.52 CD. You'll have
a version mismatch between kernels and other files. Use the 'create
boot floppies' on the 4.52 CD to make yourself a new set of boot
floppies. Modify the new floppies like you did as per the Voice article.
--
Bill
Thanks a Million!
I'm having problems quoting your post - only had new post, no reply to
link - so I copied it below:
Stick the IDE drive into an existing machine that has OS/2 on it.
Xcopy
the existing system on to that volume. Modify to remove APM and
Testcfg
drivers from config.sys. Move the disk back to the new machine. It
should now boot. You'll have to go through the process of changing
the
LAN NIC and audio drivers, but since you are using SNAP or SNAPse on
the
old machine, video should work on the new machine.
This is the easiest way, believe it or not, of getting AMD64 working
with OS/2.
BTW, You cannot use boot floppies for 4.51 to boot 4.52 CD. You'll
have
a version mismatch between kernels and other files. Use the 'create
boot floppies' on the 4.52 CD to make yourself a new set of boot
floppies. Modify the new floppies like you did as per the Voice
article.
MY REPLY:
The problem is that I no longer have machine with OS/2 on it. I have a
notebook with XP and the box that my daughter upgraded (which will
boot to XP). The only parts of the old machine that I have are the HDs
(the drive that used to boot is dead), the floppy drive, monitor,
keyboard, mouse, video card and box that holds it all. My daughter
added the RAID drive and installed XP on it - my intention is to have
a dual boot machine with OS/2 as the main boot partition and XP for my
wife's use.
I'm not using SNAP (that I know of), was using Matrox drivers and
transferred that card to the new MoBo.
I am unable to create floppies from the 4.52 CD - it fails when trying
to create the first diskette. I can create floppies from 4.51. IIRC, I
read something a while back on a newsgroup about problems creating
floppies from 4.52 CD, but I don't remember what the solution was.
> I am unable to create floppies from the 4.52 CD - it fails when trying
> to create the first diskette.
Try cdinst.bat with XP
I've done that. There is a failure when creating Disk 1 - never get the
opportunity to try to create Installation or Disk 2.
An update:
I managed to find a friend who had an old computer sitting around. He
let me install OS/2 4.51 on it and I managed to create the 4.52 install
disks. I removed the TESTCFG.SYS line from Config.SYS (and the file
from the floppy), but there are no lines referring to APM.SYS.
When I try to use these disks to install onto my box, I'm getting a
trap in OS2Kernl on Disk 1. I don't have all the information handy and
don't have the time at the moment to duplicate the error (I did
duplicate it 3 times earlier). Dunno why I could get all the way
through Disk 1 to the point it was asking for Disk 2 the other day...
Can I run the PatchLdr utility from my friend's computer that has OS/2
on it? Will it make a difference? I'm willing to drive over to his
house again if it will, otherwise, I'd just rather not waste the time.
Is it possible that the RAID controller is causing a problem? That's
where XP is loading from. IIRC, the boot sequence is set up A:, CD,
E:(RAID drive). There is a RAID driver on one of the floppies.
What are the latest versions of OS2LDR and OS2KERNL? Where can I obtain
them and what is the procedure to get them onto the floppies?
http://www.os2site.com/sw/upgrades/kernel/index.html
Believe you don´t need any special RAID drivers if you are using the
latest dani drivers. Don´t think the newest OS2LDR (with the new
kernels) require Dani´s patch. Anyway, you are not there yet.
According to what I read at the site mentioned in my original post,
Dani has managed to install on a box with AMD 64 and 1GB RAM - just
what I'm attempting... The steps she listed made it sound simple.
Will add the basedev line you suggest to Config.SYS if it isn't already
there. My original setup was pretty ancient also - about 7 years old.
Figure I probably need the patch, if I don't get a later OS2LDR, since
I now have 1GB RAM. Found out the other day that the patch won't run
from an XP DOS prompt, so I'll have to take the disk over to my
friend's house to run it in an OS/2 VDM.
Not using Dani's drivers on the floppies, but was using them (IDE and
ATAPI) on my original setup and plan to use them on this setup if I can
ever get an install. Is it just a matter of copying the ADD and FLT
files to the floppies and adding lines to Config.SYS (remming original
lines of course)?
Still don't understand how I got further along the other day. My
original problem (using floppies created from the 4.52 CD in my
original setup) occurred after Disk 1 had completed and the screen
requested Disk 2. The floppy drive kept turning on and off - putting
Disk 2 in anyway resulted in much scrolling of error messages and a
message about not being able to access the HD (maybe floppy and/or CD
too, don't recall off the top of my head). I figured that I had a bad
Disk 1 and attempted to re-create them from XP DOS prompt, which failed
on Disk 1 and apparently ruined it.
Exception in module OS2KRNL
TRAP 000e ERRCD=0000 ERACC=**** ERLIM=********
EAX=07027406 EBX=00000000 ECX=fefd081ff EDX=00000000
ESI=FFFFFFFF EDI=07027406 EBP=000055f0 FLG=00013246
CS:EIP=0168:fff3f4e9 CSACC=c09b CSLIM=ffffffff
SS:ESP=0030:000055bc SSACC=1097 SSLIM=0000487f
DS=0160 DSACC=c0f3 DSLIM=ffffffff CR0=8001001b
ES=0160 ESACC=c0f3 ESLIM=ffffffff CR2=0702765c
FS=0000 FSACC=**** FSLIM=********
GS=0000 GSACC=**** GSLIM=********
The system detected an internal processing error at location
##0168:fff1eb1c - 000e:cb1c.
60000, 9084
07861200
Internal Revision 14.089_W4
The system is stopped. Record all of the above information and contact
your service representative.
Any suggestions?
If you get PatchLDR.zip from Hobbes there is a file MEMTEST.EXE which
will show you which switch to apply to patchLDR.exe, in my case /E820.
It has to be applied to your new box. So you have to try to get an OS2
prompt, either with the maintenance diskettes or with the AltF1 blob.
You can edit the \os2\boot\config.x from XP if you install HPFSW2K.ZIP.
Like Daniela I did not do an installation, which you are attempting. I
cloned a functoning version to the new box, and am stuck, but got a bit
further than you. Hope Bill can help us to get yet a bit further.
Toast the APM.SYS, testcfg.sys (hardly used realy) and remove
ibmint13.sys should get your copied configuration running.
Roderick Klein
Thanks Roderick. Sure did not work for me! Have disk drives 2 x 200GB
so I need type 48 LBA support and suppose that ibmint13.i13 gives that
support.
Krister
> Seems our problems were different. My problem was getting disk
access.
> To get going with my maintenance floppies, I think I applied the
> latest versions of IBM1S506.ADD, OS2DASD.DMD and IBMATAPI.FLT (or
> IBMDECD.FLT), and that it was only later that I moved to the Dani
versions.
>
> If you get PatchLDR.zip from Hobbes there is a file MEMTEST.EXE which
> will show you which switch to apply to patchLDR.exe, in my case
/E820.
> It has to be applied to your new box. So you have to try to get an
OS2
> prompt, either with the maintenance diskettes or with the AltF1 blob.
> You can edit the \os2\boot\config.x from XP if you install
HPFSW2K.ZIP.
>
> Like Daniela I did not do an installation, which you are attempting.
I
> cloned a functoning version to the new box, and am stuck, but got a
bit
> further than you. Hope Bill can help us to get yet a bit further.
When you refer to the latest versions of IBM1S506.ADD, OS2DASD.DMD and
IBMATAPI.FLT, are you referring to the 4.52 version or is there a later
version? Can I just replace them with Dani's S506 and ATAPI files? If
so, what about OS2DASD.DMD?
I have been unable to get to a command prompt, I get the trap I posted
yesterday, but I haven't tried AltF1 at the OS/2 blob. Will give that a
shot later and see if it works. Not sure if that will work either since
MEMTEST requires DOS and I'm not sure if it will run under OS/2.
I already have the PatchLdr file and have unzipped it. I attempted to
run the patch program against the floppy version of OS2LDR, but it
wouldn't work so I ran the patch program against the install I made on
my friends box and copied it to my floppy - that got me the trap
yesterday, but I think it's the same trap I got before. If I can get
MEMTEST to run on the upgraded box (it won't run in a DOS window under
XP, tried already) and find the switch to apply, can I do the patch on
the version that I have on my friend's box (after copying the backup to
the original filename) and then copy that to the floppy?
Maybe I can just apply the patch with one switch, try it, and, if it
doesn't work, try the other switch? It will have to be done to my
friend's box though and I'll have to get the patched loader to
floppy...
Sure am glad he let me borrow his box until I can get things working...
> You can edit the \os2\boot\config.x from XP if you install
HPFSW2K.ZIP.
Will this file allow me to access HPFS paritions? Will it allow me to
copy files from HPFS to the XP parition (NTFS)?
George
"Mike Strickland" <m.stri...@att.net> schreef in bericht
news:1107633480.1...@g14g2000cwa.googlegroups.com...
Are you installing MCP 1 or 2 or eCS any version? Then the default IBM
or Dani drivers should work, since they had 48 bit LBA addressing for
about the past five years or more. I'd recommend the latest Dani
driver, renamed to IBM1s506.add, as she has the support for the latest
in chips, whereas the IBM drivers do not have that support.
Thanks Bill. I had a functioning MCP2 partition on the old box.
Cloning this to the new box (AMD64, 1GB RAM, 2x200GB disk drives)I had
problems with disk access, also from the AltF1 blob and from the
maintenance diskettes. The problem with the maitenance diskettes were
that they did not include ibmint.i13 as I had produced these diskettes
on the old box which apparently did not require ibmint.i13. Applying
the lastest IBM basedevice drivers or the latest Dani drivers solved the
disk access problem. For DANIS506.ADD had to use the switch /!BIOS.
So my problem is not disk access any more, it is the DOSCALL1.DLL
failure, which comes at the very end of the booting process. Have
changed to Daniela´s TESTCFG.SYS, remmed out all the APM related
drivers, tried Veit´s APMAA64.SYS, on your suggestion Bill I have
installed the EXPARTCP (except the newer basedevice drivers), applied
the Daniela´s PatchLDR and tried various newer kernels. But I am stuck
with the DOSCALL1.DLL failure.
Was hoping we could help Mike so that he might join me in trying to
solve the DOSCALL1.DLL problem
As to versions of the basedevice drivers I am not sure. I think the
4.52 drivers were not good enough for my 200GB drives. Anyway, I did
get the latest IBM drivers.
Yes, believe you can replace with the latest DANI drivers. Actually I
prefer DAMIDASD.DMD as I can boot without being in the first partition
on the first drive. I have had problems with the DANIS506.ADD, even
with applying the switch /!BIOS, but think this is because I have more
than one primary partition.
Correction, DANIS506.ADD does not work for me even with the /!BIOS
switch. Gives drive access failure. Have to use IBM1S506.ADD. Have
four primary partitions on this drive of which the second is OS2 4.52
: The system detected an internal processing error at location
: ##0168:fff1eb1c - 000e:cb1c.
: 60000, 9084
:
: 07861200
: Internal Revision 14.089_W4
Just for comparison,
The system detected an internal processing error at
location ##0168:fff1ea1f - 000e:ca1f
60000, 9084
08861200
Internal revision 14.100c_W4
is what I was getting when eCS1.2 was trying to do APM on my Thinkpad.
Close (addresses) but maybe no cigar.
Ian
--
: Are you installing MCP 1 or 2 or eCS any version? Then the default IBM
: or Dani drivers should work, since they had 48 bit LBA addressing for
: about the past five years or more.
For what it's worth, my Nice New Thinkpad falls over at install unless
I use Dani's drivers in compatibility mode.
Ian
--
FYI: Those addresses reported there is the address of the error
reporting routine in the kernel. That address will change between
builds of the kernel as the size of the kernel changes. The important
information as to the location of the fault and what it was doing is in
the information above this line in the trap screen. In other words, you
just reported that the error routine worked, and nothing more.
: FYI: Those addresses reported there is the address of the error
: reporting routine in the kernel. That address will change between
: builds of the kernel as the size of the kernel changes. The important
: information as to the location of the fault and what it was doing is in
: the information above this line in the trap screen. In other words, you
: just reported that the error routine worked, and nothing more.
Thanks! My bad.
Ian
--
Krister Hallergard wrote:
<snip>
>
> Correction, DANIS506.ADD does not work for me even with the /!BIOS
> switch. Gives drive access failure. Have to use IBM1S506.ADD. Have
> four primary partitions on this drive of which the second is OS2 4.52
>
Have you tried a CID install? Does the old machine still work? Do you
know which NIC driver is required by the new box? Then hook the LAN up
and try a CID install. Look in the /CID folder of the old box, and the
LAN CID Guide in the Assistance Center under /Information/tasks.
Suggesting this as a different view on the problem. It just requires
modifing the script used to install MCP2 on the old box to load the
system needed by the new box.
Or else create a new Boot CD. This just about the same task complexity
as the CID task.
Use from the start the newest kernel, Danis' latest hard drive & atapi
drivers, her TESTCFG.SYS, Veit's APMAA64.SYS, and redo the MBR and
partitions using LVM and not anything else (Microsoft or Linux). Use
LVM during the boot install to mark the chosen volume as installable.
My bet is that you used Windows XP (or Windows anything since 3.11) to
make the partitions, and now the partitions are not aline on cylinder
boundries, causing these failures? Be sure that in BIOS the IDE drive
is recognized as LBA drive of the proper size and the BIOS otherwise is
at the factory defaults (not over clocked). Disable the CPU caches,
which was a trick used with prior new technology and might be needed to
get an install.
Cutting edge technology makes one bleed a lot.
Ian Johnston wrote:
<snip>
>
> Thanks! My bad.
>
I made the same mistake eight years or so ago.
More correctly, the answer is "maybe", not catagorically "yes". The
hpfs ifs comes from NT 3.5 and will NOT handle large partitions. I'm
trying to recall the exact break points but there are 2 distinct ones
AIR from messing with this over the years. The first is something
like 4 gig +- where writes wrap (a BAD thing!). The other, best I
reacall was larger, maybe 8 gig or so, beyond which W2K/XP simply
refuse to accept the partition as being formatted. Use care and don't
count on those numbers without testing first as I haven't done more
than cursory looks since NT 4 and I can't be sure about the sizes but
the limitations still exist as I tried the W2k version just last week.
FWIW, I had some nasty corruption problems using the hpfs drivers
under both NT 4 and W2K on partitions where the drive was larger than
8 gig as well. It's been a crap shoot for me - now I treat the HPFS
partitions as readonly and use the ntfs ifs to go the other way back
to hpfs under OS/2.
--
Will Honea
Reading the thread quickly. Get the latest kernel from testcase from IBM
(don't have th link handy) and latest loader. MCP 2 has quite an "old"
kernel.
Roderick Klein
> To do the MEMTEST I used a Win98 boot diskette.
>
> As to versions of the basedevice drivers I am not sure. I think the
> 4.52 drivers were not good enough for my 200GB drives. Anyway, I did
> get the latest IBM drivers.
>
> Yes, believe you can replace with the latest DANI drivers. Actually I
> prefer DAMIDASD.DMD as I can boot without being in the first
partition
> on the first drive. I have had problems with the DANIS506.ADD, even
> with applying the switch /!BIOS, but think this is because I have
more
> than one primary partition.
Thanks for the tip about Win98 - I think I have one of those sitting
around somewhere...
I have not had problems in the past with DANIS506.ADD, had 4 (IIRC)
primary partitions on 2 different drives. This setup has 3 primary
partitions, I'm pretty sure.
Were you working on similar hardware?
IIRC, one if the partitions is 19 gig - do I understand correctly that
I can't even read that partition? Will there be hell to pay if I try?
Might it be worth my while to go ahead and invest in eCommStation's
latest version? I was reading on their site that they've released a
version that's supposed to install on AMD 64 "problem systems".
Thanks Bill. I am not convinced that my problems are due to buggy
partitions. I used a Ranish partion manager, a dos tool.
I'm trying to remember what NT did with a 10 gig HPFS partition -
without much luck. AIR, W2K just flat refused to recognize the
formatting on the partition when it got too big (wanted me to format
it - right!). If you do read-only access you should be OK. Just be
aware that Win will not handle EA's, regardless of format.
I don't know about where exactly I ran into repeatable write
corruption, but I have this vague recollection of it being on HPFS
partitions > 4 gig. I may be wrong there, but I never had any issues
with r/w from NT or W2K on HPFS partitions below the 2.1 gig size -
same limits as for FAT 16 so that's usually what I used for transfer
partitions.
My solution for this situation, BTW, was a network cable between two
computers. One ran OS/2, the other ran W2K. NETBEUI made short work
of the format and file sharing issue <g>.
--
Will Honea
IF DFsee says the partitions are okay, then they are. Please check.
That does not hurt. DFsee is on Hobbes.
-> I downloaded what I think is the latest testcase, but XP says I need a
-> password to unzip the package. Any place I can get a copy of the loader
-> and kernal that isn't password protected?
->
-> Might it be worth my while to go ahead and invest in eCommStation's
-> latest version? I was reading on their site that they've released a
-> version that's supposed to install on AMD 64 "problem systems".
->
Where did you get them? The kernels on testcase are not password
protected, but some private fix files are. See
http://www.os2site.com/sw/upgrades/kernel/index.html for an archive of
recent updates. The testcase site itself is cleared off after a few
days, so at present there aren't any on it.
Mark
--
From the eComStation of Mark Dodel
http://www.os2voice.org
Warpstock 2005, Where?/When? Stay tuned to - http://www.warpstock.org
Thanks, I think I'll give it a try since no harm will be done. If it
works, at least I can see the files and get them to a safer location...
I have seen this Config.X mentioned a couple of times. Is this a file
that I should look at editing on my floppies to aid installation?
I see from your sig that you're using eComStation, so I'll ask your
opinion on the question I posed in my earlier post about upgrading to
eComStation.
INT 12 returns: 626 KiBytes
INT 15 (88) returns: 64512 KiBytes above 1 MiB
INT 15 (C0) returns: PS/2 function INT 15 (C7) is not supported
INT 15 (DA88) returns: not supported
INT 15 (E801) returns: 1046784 KiBytes above 1 MiB available
1046784 KiBytes above 1 MiB
configured
INT 15 (E820) returns memory list:
639 KiBytes @ 0 K
1046720 KiBytes @ 1024 K
no patch required
I don't understand these results, since all that I have read indicates
that there should be a patch needed for newer hardware.
Update:
I copied the OS2LDR and OS2KRNL (renamed to OS2KRNLI) to my Install
floppy. I still get the trap - info below.
Exception in module OS2KRNL
TRAP 000e ERRCD=0000 ERACC=**** ERLIM=********
EAX=ffdfc000 EBX=00000000 ECX=0000f000 EDX=00000000
ESI=FFFFFFFF EDI=00000000 EBP=000055f0 FLG=00013246
CS:EIP=0168:fff3f6d7 CSACC=c09b CSLIM=ffffffff
SS:ESP=0030:000055bc SSACC=1097 SSLIM=0000487f
DS=0160 DSACC=c0f3 DSLIM=ffffffff CR0=8001001b
ES=0160 ESACC=c0f3 ESLIM=ffffffff CR2=ffe38000
FS=0000 FSACC=**** FSLIM=********
GS=0000 GSACC=**** GSLIM=********
The system detected an internal processing error at location
##0168:ff1e9fb - 000e:c9fb.
60000, 9084
07869090
Internal Revision 14.100o_W4
The system is stopped. Record all of the above information and contact
your service representative.
Does this mean anything to anyone?
Yes there is also config.x on your floppies that you might have to edit.
Believe config.x is invoked first and than config.sys, and I guess
this has to to with changing floppies. Don´t suppose they should be
allowed to contradict eachother.
->
-> Mike Strickland wrote:
-> > Have gone to the site you suggested and downloaded w420050130.zip
-> and
-> > one called half strict (don't have a clue what that is) called
-> > w420050130d.zip. Will try the w420050130.zip first.
-> >
-> > I see from your sig that you're using eComStation, so I'll ask your
-> > opinion on the question I posed in my earlier post about upgrading to
-> > eComStation.
->
-> Update:
->
-> I copied the OS2LDR and OS2KRNL (renamed to OS2KRNLI) to my Install
-> floppy. I still get the trap - info below.
->
-> Exception in module OS2KRNL
->
-> TRAP 000e ERRCD=0000 ERACC=**** ERLIM=********
-> EAX=ffdfc000 EBX=00000000 ECX=0000f000 EDX=00000000
-> ESI=FFFFFFFF EDI=00000000 EBP=000055f0 FLG=00013246
-> CS:EIP=0168:fff3f6d7 CSACC=c09b CSLIM=ffffffff
-> SS:ESP=0030:000055bc SSACC=1097 SSLIM=0000487f
-> DS=0160 DSACC=c0f3 DSLIM=ffffffff CR0=8001001b
-> ES=0160 ESACC=c0f3 ESLIM=ffffffff CR2=ffe38000
-> FS=0000 FSACC=**** FSLIM=********
-> GS=0000 GSACC=**** GSLIM=********
->
-> The system detected an internal processing error at location
-> ##0168:ff1e9fb - 000e:c9fb.
-> 60000, 9084
->
-> 07869090
-> Internal Revision 14.100o_W4
->
-> The system is stopped. Record all of the above information and contact
-> your service representative.
->
-> Does this mean anything to anyone?
->
Not to me, but if you post it in comp.os.os2.bugs you may get some
help. The kernel developer lurks there.
mark
->
-> Have gone to the site you suggested and downloaded w420050130.zip and
-> one called half strict (don't have a clue what that is) called
-> w420050130d.zip. Will try the w420050130.zip first.
->
-> I see from your sig that you're using eComStation, so I'll ask your
-> opinion on the question I posed in my earlier post about upgrading to
-> eComStation.
->
I just installed eCS 1.2 on a new Thinkpad T42p and the install was
flawless but that was a Pentium M CPU. Serenity has a beta release of
1.2 with fixes for installing on an Athlon64. I'm told it fixes a
lot of problems with installing on that hardware but I have no
experience since I don't have any.
Downloaded Dfsee and got four warning messages. First the disk layout:
ID Dr Type, description Format OS GB
Disk1
01> C: Prim 0C FAT32-EXT FAT32 XP 188 Total 188
Disk 2
02 Free Space Wasted 0
02 F: Prim 06 FAT16 FAT16 W95 2
03 G: Prim 07 Inst-FS HPFS OS2 2
04 Free Space Wasted (reserved for Linux) 8
04> D: Prim 0C FAT32-EXT FAT32 W98 33
05 E: Prim 07 Inst-FS NTFS W2k 144
07 Free Space Wasted (reserved for PartMgr) 2 Total 191
Warning messages as follows:
1. Pid01 Partition doesn´tend on the last head (cylinder boundary)
2. Pid01 Partition extends beyound end of disk
3. Pid04 Active (bootable) but not on 1st disk. Requires modern BIOS
or smart MBR code to boot
4. Pido4 Cylinder >1023 butMBR doesnot have I13x support!
Use the ´NEWMBR 1´ command to refresh the MBR code.
Warning 4 does also occur for Piod5 if that partition is the active one.
On disk 2 I am using the Ranish compact partition manager for booting
and that is what Dfsee finds in the MBR
Warnings 1 and 2: The two disks are identical but Disk 1 shows 324
fewer cylinders. The supplier did not include the XP CD in the delivery
but a rescue disk (called Reload), which would reinstall Disk 1 from
scratch using some 3 GB compressed and hidden files, XP and other
bundled programs. This space does not show up in the partition table.
The Disk 1 disk and partition structure is exactly as delivered. As per
Ranish:
HD1 188,242M [23,997cyls x255heads x63 sects =385,521,472 sects]
Starting Ending Partition
# Type Row System Type Cyl Head Sect Cyl Head Sect Size[KB]
0 MBR Master Boot Record 0 0 1 0 0 1 0
1 Pri Unused 0 0 2 0 0 63 31
2 >Pri 1 Windows FAT-32LBA 0 1 1 23,997 153 27 192,760,704
3 Pri Unused 23,997 153 28 23,997 153 28 0
Could this be the reason why OS2 gives a late DOSCALL1.DLL boot failure
when I use IBM1506.ADD or get early disk access boot failure when using
DANI506.ADD?? Tried hiding the Pid01 partition, but no change.
Disk 2 as per Ranish:
HD2 190,782M [24,321cyls x255heads x63 sects =390,721,968 sects]
Starting Ending Partition
# Type Row System Type Cyl Head Sect Cyl Head Sect Size[KB]
0 MBR Master Boot Record 0 0 1 0 0 1 0
1 Pri Unused 0 0 2 0 254 63 8,032
2 Pri 3 DOS FAT-16 1 0 1 260 254 63 2,088,450
3 Pri 4 Windows NT NTFS 261 0 1 520 254 63 2,088,450
4 Pri Unused 521 0 1 1,559 254 63 8,345,767
5 >Pri 1 WindowsFAT-32LBA 1,560 0 1 5,715 254 63 33,383,070
6 Pri 2 Windows NT NTFS 5,716 0 1 23,996 254 63146,842,132
7 Pri Unused 23,997 0 1 24,321 80 63 2,605,081
Any ideas?
You have two choices: redo all the partitions using LVM to create the
partitions and have to reinstall everything (backup and restore is a
good ideal), or you can install another hard drive and put OS/2 on it.
Is Rannish a boot sector boot manager? If so, you are going to need to
lose it. It does not put the X13 marker that OS/2 expects and
installing LVM/newmbr:0 would wipe it out. Now Martin's Airboot has the
marker and is a MBR boot manager and has been proven to work with all
operating systems. Airboot is on Hobbes. If you do put OS/2 on a
separate disk, make its partition a logical one, using LVM. This will
require a temporary install of OS/2 Boot Manager at the end of the disk,
which you can delete later after the install is complete.
Thank you Bill. Will have a look at Airboot. I have a spare harddisk
and could use that for OS2 and proceed as you say. But will the other
two buggy drives interfer with the boot process?
Or if I use my Disk 1, and shrink the XP partition to add an OS2
partition in the evacuated space (created by LVM), would that have a
chance of working? (I am very reluctant to redo the XP partiton from
scratch, especially since I have no way of putting the XP boot record back).
As to redoing Disk 2 from scratch using LVM I am a bit reluctant to put
in all that work without a guarantee of not getting stuck again with a
DOSCALL1.DLL failure. The boot process goes on for quite a while.
Believe the basedevice drivers all get installed, USB drivers,
FAT32.IFS, UPF filesysten, Fax modem, ODIN, USBPRINT.SYS, FAT32 cache,
and than there is a long wait ending in the DOSCALL1.DLL failure. Would
the boot process go on for so long if tnecause is buggy partition table??
Krister Hallergard wrote:
> William L. Hartzell wrote:
<snip>
I am not the expert on this. Believe it, the author of DFsee is. Buy
DFsee, I believe Jan, the author, can tell you how to fix this without
lost data.
There are two other partitions that it can't see.
I took the drive out of the computer it was in and moved it to my
friend's. Now that computer will not boot from HD - it appears that the
drive that was originally in it is set to master and single. I can't
find out how to get rid of the single part so I'm booting it with
floppies and CD.
I decided to try to install OS/2 to one of the partitions on the drive
I moved, so I pulled the cable from the original drive in my friend's
computer, attached my drive and changed the jumper to master.
Attempting to install shows that the partition table is possibly
corrupt - dunno where that came from since the partitons were created
with LVM originally - mebbe the XP install to that drive did something
(had originally intended to install both XP and OS/2 into partitons on
this drive, but the XP partition wasn't large enough when I attempted
to get service pack 2 and I just installed it on the RAID drive).
At this point I had a brainstorm and figured that I could add the
latest USB drivers to my boot floppies and possibly have access to my
USB portable HD.
Unpacking the drivers went just fine, they were unpacked to the I:
partiton (one of the ones I need to copy). At that point, I could no
longer perform any actions on I:. I can change to the drive letter, but
DIR and any other commands I've tried fail with an unknown error - OS/2
message file is unavailable.
I have started copying the data from H: (the other partition I need to
copy) using floppies and have all but given up after all day yesterday
swapping diskettes - moving the data to the external drive via my
wife's work notebook which has a 3.5 drive.
If I can get my hands on a copy of the latest USB drivers, can I
possibly access the USB HD? If so, could someone please unpack the file
and send me the files in the package?
What can I do to regain access to the partition that I have lost
access to? Will CHKDSK fix it? Do I need to take the drive to pros?
I've gotta have the data on the I: partiton at all costs.
Found out why I had problems with using DANIS506.ADD. I had used
BASEDEV=DANIDASD.DMD /MT:Ad /BD:C to boot the fourth partition on
Disk 2, which I had set as boot disk in BIOS. (With this setting the
W2k Boot.ini refers to the disk as "disk 0" which lead med to believe I
should use A for the first disk) But I should have used /MT:Bd!
Apart from being able to use DANIS506.ADD, nothing changed.
DOSCALL1.DLL failure towards the end of the boot process.
So I decided to add another 3GB disk drive with one only partition for
OS2 HPFS. Implemented all the patches I know that are recommended for
AMD Athlon 64, and made sure this worked on my old box. Without any
changes at all I moved this disk drive to my new box and booted:
DOSCALL1.DLL failure towards the end of the boot process. Is the
problem the other two 200GB disk drives with buggy MBR or is it the AMD
64 CPU?
Disconnected the two big buggy drives, but still got the DOSCALL1.DLL
failure. Seems as the usual AMD64 problem. Must have missed something.
Try it with at most 512MB Ram.
Hendrik
Thank you Hendrik. Tried it and am afraid it did not work, still
DOSCALL1.DLL failure
Krister
Does the error message give to you an ordinal number of the failing
call? Maybe someone will recognize and have a suggestion? I'm out of
ideals.
Thanks Bill. If it is the popupos2 you asking about, this is it:
02-24-2005 08:32:39 SYS3175 PID 0014 TID 0001 Slot 0009
C:\OS2\PMSHELL.EXE
c0000005
1c039fda
P1=00000000 P2=ffffffff P3=XXXXXXXX P4=XXXXXXXX
EAX=00009807 EBX=0000ef7e ECX=00000000 EDX=c4280000
ESI=00040000 EDI=f40f0327
DS=0000 DSACC=**** DSLIM=********
ES=9807 ESACC=00f3 ESLIM=0000113f
FS=9807 FSACC=00f3 FSLIM=0000113f
GS=0000 GSACC=**** GSLIM=********
CS:EIP=e01f:00009fda CSACC=00df CSLIM=0000dd58
SS:ESP=0047:0000ef5a SSACC=00f3 SSLIM=0000efff
EBP=0004ef82 FLG=00012246
DOSCALL1.DLL 0003:00009fda
Does anyone recognize this??