Boot from USB PRM (Partitioned Removable Media ) failed after loading OS2LVM.DMD error: "OS/2 is unable to operate your hard disk or diskette drive"
Rainer Stroebel
Software: OS/2 MCP2 with current fix pak and driver level
1. create a maintance partiton on my primary HDD with bootabel 4.3
minium features: only boot to cmd prompt, no wps support
USB Support basic and USBMSD.ADD
Test the system - system is in Boot Manager Menue
2. take a CF 1 GB Card and create a partiton on the medium
How to it: readme.tx from LVM fixpak / Update 14.105 (XR0L002)
3. the PRM partition has a drive letter of C: in my configuration
4. xcopy y:*.* c:*.* /H /O /T /S /E /R /V
y: is the drive letter of the mini maintance system
5. \os2\install\bootdisks\sysinstx c:
6. modify the config.sys
1. drive letter Y: --> C:
2. Change USB statments in sequence and parm according
booting from USB diskette
souce of instructions:
- usbbasic package: USBBASIC.txt 14.162
- mcp2: Technote.txt
3. create an alternative config.sys with the original usb from
primary min maint partiton config.sys
7. boot from PRM
7.1 Activate USB BIOS support
7.2 change the startup/ boot seq in bios
The PRM is an Harddrive for the bios
set the primary hdd to the T23 inacitce in boot menue
8. The system boots from the PRM
8.1 OS/2 blob
8.2 Warp logo
8.3 the loaded driver are recorded
8.4 .
.
os2dasd.dmd
os2lvm.dmd
8.5 full screen mode
meassages from usb driver
meassages form dani driver
9. Error "OS/2 is unable to operate your hard disk or diskette drive"
Ths error shows up with config.sys
- with modified usb sequence
- with "normal, standard usb" statements
- with or without dani driver preasent e.g no driver for the
primary HDD
any ideas ????????? need help
Rainer
===============================================
enclosure: the config.sys with mod. usb statements
rem
rem config.sys with usb stamtments arranged according boot from USB
deskette instructions
rem
BASEDEV=USBD.SYS /V /I13
BASEDEV=usbuhcd.sys /V
BASEDEV=usbuhcd.sys /V
BASEDEV=usbuhcd.sys /V
rem str BASEDEV=usbuhcd.sys /V
rem str BASEDEV=usbuhcd.sys /V
rem str BASEDEV=usbehcd.sys /V
rem BASEDEV=usbd.sys /REQ:USBUHCD$,USBOHCD$,USBEHCD$ /V
rem BASEDEV=usbd.sys /V
BASEDEV=usbhid.sys /V
DEVICE=C:\OS2\usbmouse.sys /V
DEVICE=C:\OS2\usbprt.sys /LPT
BASEDEV=usbmsd.add /V /FLOPPIES:0 /REMOVABLES:1 /CDS:0
rem BASEDEV=CWUSBMSD.ADD /V /FLOPPIES:0 /REMOVABLES:1 /CDS:0 /FIXED_DISKS:1
RESERVEDRIVELETTER=M
LIBPATH=.;C:\OS2\DLL;
SET PATH=C:\OS2;
SET DPATH=C:\OS2;
SET LANG=de_DE
SET ULSPATH=C:\language
DEVICE=C:\OS2\unicode.sys
rem IFS=C:\OS2\cdfs.ifs /Q /W
IFS=C:\OS2\hpfs.ifs /CACHE:1024 /AUTOCHECK:*
rem IFS=C:\OS2\udf.ifs /V
BASEDEV=ibmkbd.sys
BASEDEV=chkdsk.sys
BASEDEV=ibm1flpy.add /A:0 /FORCE:2 /U:0 /F:1.44MB /U:1 /F:1.44MB
BASEDEV=danis506.add /W
rem str BASEDEV=danis506.add /VLL /A:1 /BAY
rem BASEDEV=daniatap.flt /V /A:1 /U:0 /TYPE:W
BASEDEV=os2dasd.dmd /V
BASEDEV=os2lvm.dmd /V
COUNTRY=049,C:\OS2\country.sys
CODEPAGE=850
DEVINFO=KBD,GR,C:\OS2\keyboard.dcp
DEVINFO=SCR,VGA,C:\OS2\viotbl.dcp
rem DEVICE=C:\OS2\os2cdrom.dmd
rem BASEDEV=os2aspi.dmd /ALL
DEVICE=C:\OS2\testcfg.sys
DEVICE=C:\OS2\dos.sys
DEVICE=C:\OS2\pointdd.sys
DEVICE=C:\OS2\mouse.sys
SET SAVECONNECT=1
SET COPYFROMFLOPPY=1
rem DEVICE=C:\OS2\com.sys
BASEDEV=print01.sys /IRQ
PRINTMONBUFSIZE=134,134,134
rem DEVICE=C:\OS2\apm.sys
rem RUN=C:\OS2\apmdaemn.exe
SET TEMP=C:\Temp
SET TMP=C:\Temp
SUPPRESSPOPUPS=C
MEMMAN=SWAP,PROTECT
SWAPPATH=C: 512
IOPL=YES
PROTECTONLY=YES
AUTOFAIL=YES
THREADS=1024
DISKCACHE=D,LW,AC:C
MAXWAIT=1
SET VIDEO_DEVICES=VIO_VGA
SET VIO_VGA=DEVICE(BVHVGA)
SET TZ=CET-1CES,3,-1,0,7200,10,-1,0,10800,3600
SET PROMPT=$i[$p]
SET KEYS=ON
PROTSHELL=C:\OS2\cmd.exe
SET OS2_SHELL=C:\OS2\cmd.exe
SET COMSPEC=C:\OS2\cmd.exe
Doug Bissett
...snip...
Some things, that might help:
1) Use LVM, or DFSEE, to permanently attach a drive letter to your USB
drive. If you don't do that, the drive letter will, likely, change,
when the USB drivers do get loaded, and LVM takes control. The result
will be exactly what you describe.
2) Let bootAble build the whole system on the USB drive for you. I
have no idea why you would need to change what it builds. Doing a copy
of a system built on Y:, to C:, then expecting to simply change the
drive letter in CONFIG.SYS, misses about 90% of what needs to be
changed.
3) I remember, at one time, that OS/2 could NOT be booted from a
removable device (other than diskette). In the past, when I built
bootable systems on a removable device (an ORB disk), I used the /!RMV
parameter on the DANIS506.ADD line. That parameter tells the Dani
driver to report that the device is not removable. I don't know of any
similar parameter for the USB support drivers (if they need it).
4) You would, likely, want to make the USB drive startable, and put
the USB device boot before the hard disk boot, in the BIOS settings,
so that the USB device will boot, without needing the Boot Manager,
that may be on a non-working hard disk, at some point.
Hope this helps...
--
From the eComStation 1.2 of Doug Bissett
dougb007 at telus dot net
(Please make the obvious changes, to e-mail me)
Rainer Stroebel
Hello Doug,
thanks for your comments and ideas.
Your message does help me a lot to do big step
in analyzing the problem and bring it to a solution.
Here are the details:
>...snip...
>
>Some things, that might help:
>
>1) Use LVM, or DFSEE, to permanently attach a drive letter to your USB
>drive. If you don't do that, the drive letter will, likely, change,
>when the USB drivers do get loaded, and LVM takes control. The result
>will be exactly what you describe.
>
>
>
The media (compacted flash card) is installed as PRM with a drive
letter c: in the
LVM DLAT Table. I use a FAT file system to keep it simple.
>2) Let bootAble build the whole system on the USB drive for you. I
>have no idea why you would need to change what it builds. Doing a copy
>of a system built on Y:, to C:, then expecting to simply change the
>drive letter in CONFIG.SYS, misses about 90% of what needs to be
>changed.
>
>
>
I built a maintenance system with bootable on HDD drive Y: to do test boot.
Again to keep it simple, the system a stripped down to minimum e.g.
no wps, no cdrom support, no udf, no hpfs - just boot to command prompt.
So the change from y: to C: in config.sys for the system on the PRM
should do the
job.
Do I miss something?
>3) I remember, at one time, that OS/2 could NOT be booted from a
>removable device (other than diskette). In the past, when I built
>bootable systems on a removable device (an ORB disk), I used the /!RMV
>parameter on the DANIS506.ADD line. That parameter tells the Dani
>driver to report that the device is not removable. I don't know of any
>similar parameter for the USB support drivers (if they need it).
>
>
>
Right, I agree - this is the key problem !
os2dasd.dmd and os2lvm.dmd does not support removables as boot devices!
For cdroms this is solved by the DANIBOOT.FLT by changing the device type
from removable to fixed disk ( and do I13 support reading from cdrom
and loading the CDROM drivers
in protected mode.
This is very well discussed / documented at
http://www.ami.com.au/BootableOS2CD.htm
"Creating a Bootable CD-ROM *for OS/2" *Copyright 1998 by Allen Dermody
Conclusion:
Handle the PRM as fixed device and the system should boot to command prompt!
How to do it?
1.Add an parameter to usbmsd.add to handle removable device as fixed device
e,g, pass the "right" device type to os2dasd.dmd and os2lvm.dmd
2. Ask very polite Chis for a modification of CWUSBMSD.add :-)
3. Built a filter driver to switch the device type form removable to fixed
as done in CD_boot.FLT, DANIBoot.FLT and DANIS506.ADD with the
device parameter /!RMV
This is probably the most general solution and could enable the
booting from other removable devices
in the future e..g. firewire HDDs
I have start the process to the bounty system at os2word to promote
the building of the flt driver
I will ask Dani very polite about it :-)
4) You would, likely, want to make the USB drive startable, and put
>the USB device boot before the hard disk boot, in the BIOS settings,
>so that the USB device will boot, without needing the Boot Manager,
>that may be on a non-working hard disk, at some point.
>
>
>
See the bios mod in the problem description - already done
LVM GUI does not allow a bootmanager on a PMR and
does not reserve the 7,x MB at the beginning of the drive space
>Hope this helps...
>
>
Indeed it does help a lot :-)
kind regards
Rainer
Peter Brown
I saw a project on os2world that looks like it will be of interest to
you and may help with some of your questions:-
http://www.os2world.com/cgi-bin/news/viewnews.cgi?category=33&id=1140513347
Regards
Pete
Rainer Stroebel
> Hi Rainer
>
> I saw a project on os2world that looks like it will be of interest to
> you and may help with some of your questions:-
>
> http://www.os2world.com/cgi-bin/news/viewnews.cgi?category=33&id=1140513347
>
>
>
> Regards
>
> Pete
>
Hello Pete,
thanks for the URL - just read the project description and the forum
comments.
"Kif" uses an IDE - TO - CF Adapter at his project.
So he use the ibmS501.add or the DANIS501.add driver to access the boot
drive.
For additional HDD space he use a second CF Card via USBMSD access.
He has some trouble with this CF - TO - IDE Adapter.
It does not work 100 % link an IDE drive !!!
I just passed the procurement source info for an alternate CF - To -
IDE Adapter in the forum to "Kif".
If the Adapter works according to the IDE spec, I do see no need for an
USBMSB access to storage at the project.
My goal is to boot directly from USBMSD device on any computer with
the proper bios USB boot support.
kind regards
Rainer
Rainer Stroebel
Hello Doug,
just some additional info - new facts
http://www.warpupdates.mynetcologne.de/english/hard_largeata.html
Oliver Rick documented the /RF parameter for os2dasd.dmd - many
thanks to Oliver Rick for his excellent work at warpupdates!!!
Just done some tests.
The BASEDEV=OS2DASD.DMD /V /RF is accented as correct syntax at boot!!!
The boot still stops with the error message ---> back to the lab for
further research and ask for help / ideas from the community
Does the undocumented parameter are as expected ?
If yes - there is an additional problem / bug on the " way to
success".
kind regards
Rainer
Doug Bissett
<RainerS...@t-online.de> wrote:
I think you would still need to use LVM to fix the drive letter, so it
cannot change when the system switched from the INT13 support, to the
full USB driver support. It may also be possible to NOT install any
USB driver support, so the device keeps on using the INT13 support.
>
> >2) Let bootAble build the whole system on the USB drive for you. I
> >have no idea why you would need to change what it builds. Doing a copy
> >of a system built on Y:, to C:, then expecting to simply change the
> >drive letter in CONFIG.SYS, misses about 90% of what needs to be
> >changed.
> >
> >
> >
> I built a maintenance system with bootable on HDD drive Y: to do test boot.
> Again to keep it simple, the system a stripped down to minimum e.g.
> no wps, no cdrom support, no udf, no hpfs - just boot to command prompt.
>
> So the change from y: to C: in config.sys for the system on the PRM
> should do the
> job.
>
> Do I miss something?
I think so. There are still some INI entries, that could be pointing
to the wrong place (Y:).
> >3) I remember, at one time, that OS/2 could NOT be booted from a
> >removable device (other than diskette). In the past, when I built
> >bootable systems on a removable device (an ORB disk), I used the /!RMV
> >parameter on the DANIS506.ADD line. That parameter tells the Dani
> >driver to report that the device is not removable. I don't know of any
> >similar parameter for the USB support drivers (if they need it).
> >
> >
> >
> Right, I agree - this is the key problem !
>
> os2dasd.dmd and os2lvm.dmd does not support removables as boot devices!
>
> For cdroms this is solved by the DANIBOOT.FLT by changing the device type
> from removable to fixed disk ( and do I13 support reading from cdrom
> and loading the CDROM drivers
> in protected mode.
Daniboot is only used with the El-torito boot method, AFAIK. You
should be building a maintenance partition, on the device, which would
not use that, at all.
> This is very well discussed / documented at
>
> http://www.ami.com.au/BootableOS2CD.htm
>
> "Creating a Bootable CD-ROM *for OS/2" *Copyright 1998 by Allen Dermody
>
> Conclusion:
>
> Handle the PRM as fixed device and the system should boot to command prompt!
That is an ancient article, and many things have changed since then...
> How to do it?
>
> 1.Add an parameter to usbmsd.add to handle removable device as fixed device
> e,g, pass the "right" device type to os2dasd.dmd and os2lvm.dmd
Unless IBM releases the "latest" USB driver code, I doubt if this will
ever happen.
> 2. Ask very polite Chis for a modification of CWUSBMSD.add :-)
Possible, but Chris' driver is still based on old code, which might
just break a lot of newer devices.
> 3. Built a filter driver to switch the device type form removable to fixed
> as done in CD_boot.FLT, DANIBoot.FLT and DANIS506.ADD with the
> device parameter /!RMV
>
> This is probably the most general solution and could enable the
> booting from other removable devices
> in the future e..g. firewire HDDs
This is the most likely to work properly.
> I have start the process to the bounty system at os2word to promote
> the building of the flt driver
>
> I will ask Dani very polite about it :-)
She may not have the tme, or the interest, to do it, but it can't hurt
to ask.
> 4) You would, likely, want to make the USB drive startable, and put
>
> >the USB device boot before the hard disk boot, in the BIOS settings,
> >so that the USB device will boot, without needing the Boot Manager,
> >that may be on a non-working hard disk, at some point.
> >
> >
> >
> See the bios mod in the problem description - already done
I saw that, after I posted...
> LVM GUI does not allow a bootmanager on a PMR and
> does not reserve the 7,x MB at the beginning of the drive space
You really shouldn't need boot manager on the device, but you would
need to make it startable (active, in Windows terms), so it would
immediately try to boot the code that is on the device. Unfortunately,
doing that with any of the current tools (except, possibly, DFSEE),
will make Boot Manager not startable on the hard disk (that can be
fixed, after you remove the USB device).
>
> >Hope this helps...
> >
> >
> Indeed it does help a lot :-)
>
>
> kind regards
>
> Rainer
Good luck...
Oliver Rick
> http://www.warpupdates.mynetcologne.de/english/hard_largeata.html
> Oliver Rick documented the /RF parameter for os2dasd.dmd - many
> thanks to Oliver Rick for his excellent work at warpupdates!!!
> Just done some tests.
> The BASEDEV=OS2DASD.DMD /V /RF is accented as correct syntax at boot!!!
> The boot still stops with the error message ---> back to the lab for
> further research and ask for help / ideas from the community
> Does the undocumented parameter are as expected ?
"The known parameters from the NEWDASD beta package /RF (removable as
fixed) - now activated by default - [...]"
So the parameter doesn't change anything.
The *BOOT.FLT drivers don't fake a fixed disk, but a floppy disk image as a
floppy drive. A physical floppy drive then becomes the second, while a
possible second drive becomes inaccessible.
I'm not sure if OS/2 recognizes any non-locked PRM until they're made
available via lvm /RediscoverPRM.
Have you tried to use the medium as a large floppy instead (see Readme of
latest LVM Update)?
You might ask Veit Kannegieser, the author of MemDisk ("El Torito" No
Emulation CD-ROM boot support and boot eCS from a memory hard disk), if he
has an idea.
/Olli/
--
WarpUpdates International/Deutschland
http://www.warpupdates.mynetcologne.de/
Doug Bissett
<RainerS...@t-online.de> wrote:
I think that "/RF" is no longer q documented parameter, and as such,
it may actuall do nothing, if used.
>
> Just done some tests.
>
> The BASEDEV=OS2DASD.DMD /V /RF is accented as correct syntax at boot!!!
>
> The boot still stops with the error message ---> back to the lab for
> further research and ask for help / ideas from the community
>
>
> Does the undocumented parameter are as expected ?
That was documented, at one time, but I don't think it applies to any
of the LVM based systems.
> If yes - there is an additional problem / bug on the " way to
> success".
Yes, you have some way to go, yet... I must give it a try, when I find
some time...
Rainer Stroebel
>On Sun, 13 Aug 2006 Rainer Stroebel wrote:
>
>
>
>>http://www.warpupdates.mynetcologne.de/english/hard_largeata.html
>>
>>
>
>
>
>>Oliver Rick documented the /RF parameter for os2dasd.dmd - many
>>thanks to Oliver Rick for his excellent work at warpupdates!!!
>>
>>
>
>
>
>>Just done some tests.
>>
>>
>
>
>
>>The BASEDEV=OS2DASD.DMD /V /RF is accented as correct syntax at boot!!!
>>
>>
>
>
>
>>The boot still stops with the error message ---> back to the lab for
>>further research and ask for help / ideas from the community
>>
>>
>
>
>
>>Does the undocumented parameter are as expected ?
>>
>>
>
>"The known parameters from the NEWDASD beta package /RF (removable as
>fixed) - now activated by default - [...]"
>
>So the parameter doesn't change anything.
>
>
>
It does change something at my eCs 1.2R test partition - code level
without any upgrade
The Icons for the removables are different in the drives folder with /RF
and without.
It does not change the icons in my MCP2 partition -with all current
fixes and
Kernel 14.104a
It does change the behavior of the eject command at Jan van Wijk System.
It does not change the DFSEE indicator
in col. ID in the output of the FDISK -r- command.
At boot from the USB the os2lvm.dmd driver is displayed ( ALT - F2
active )
a little longer with /RF before the error message comes up.
I have looked for a way to verify this :-) have not found anything.
>The *BOOT.FLT drivers don't fake a fixed disk, but a floppy disk image as a
>floppy drive. A physical floppy drive then becomes the second, while a
>possible second drive becomes inaccessible.
>
>I'm not sure if OS/2 recognizes any non-locked PRM until they're made
>available via lvm /RediscoverPRM.
>
>
>
This is one of my open problems.
The HDD maintenance mini system with boot to the command prompt shows a
problem:
the system is on drive Y: and the PRM is drive letter C:
1. First command after boot
copy c:\config.sys y:\test.xxx
produce the error sys0015 drive letter
unknown !!!!!!!!!!!!
2. the second command:
dir c: does work !!!! and than the
3. copy c:\config.sys y:\test.xxx than does work too.
This is through for a FAT or a HPFS maintenance partition
any ideas ??????
I can not find out the reason for access problem to the boot drive:
1. because if it is an removable drive ( in case /RF does not doe
anything )
and os2dasd.dmd and os2lvm.dmd does not support booting form this type
or
2. if /RF does the "right" thing
the drive is not read for reading the country command from config.sys
via the protected mode drivers - the basedev driver just load before
It should be ready for an read command !
May be there is an bug in the basedev drivers and an work around
is implemented via the usb monitor /
via lvm /RediscoverPRM as you questioned
>Have you tried to use the medium as a large floppy instead (see Readme of
>latest LVM Update)?
>
>
>
That was my first try -- run into bios startup prolbems
does not investigate it ---> just switch to PRM with the USB MSD Device.
Might go back to it and look more in detail for the bios startup
>You might ask Veit Kannegieser, the author of MemDisk ("El Torito" No
>Emulation CD-ROM boot support and boot eCS from a memory hard disk), if he
>has an idea.
>
>
>
good idea - I have already look at memdisk documentation to learn
more about the boot process.
Rainer Stroebel
I am a bit confused about your statements about the ini files/entries.
The mini maintenance system for command prompt built by bootable does not
built/generate any os2.ini or os2sys.ini files in the /os2 directory.
My understanding is:
The ini files are needed by the WPS and application programms
( and the USBprint.sys driver )
The BMR does not see the PRM device .
By booting from HDD the BMR gets control -
the BMR does not look / start the PRM via I13
even if there is an entry in the BMR tables for the system on the PRM.
( the USB drive LED is not green )
By booting from the PRM ( via bios start up pointing first to the USB in
start up boot list )
the BMR does not get control - the system is directly loaded from the
primary partition on the
PRM
>
>
>>>Hope this helps...
>>>
>>>
>>>
>>>
>>Indeed it does help a lot :-)
>>
>>
>>
>>
>>kind regards
>>
>>
>>
>> Rainer
>
>
>Good luck...
>
>
I truly need it :-)
Kind regard
Rainer
Rainer Stroebel
Looking forward to it -
the main/core problem by these test is
to find out the right case
- does it not work because I do not do the right thing
- it does not work because of bugs
-:)
Working in teams and do the testing on different configuration does help
a lot.
kind regards from Germany
Rainer
Steven Levine
at 12:25 AM, Rainer Stroebel <RainerS...@t-online.de> said:
> the main/core problem by these test is
>to find out the right case
> - does it not work because I do not do the right thing
> - it does not work because of bugs
It's neither of the above. I can not work because the code to implement
booting from USB removable media was never implemented for eCS/OS2.
You can test all you want, but this will not change.
Regards,
Steven
--
--------------------------------------------------------------------------------------------
Steven Levine <ste...@earthlink.bogus.net> MR2/ICE 2.67 #10183
Warp/eCS/DIY/14.103a_W4 www.scoug.com irc.fyrelizard.com #scoug (Wed 7pm PST)
--------------------------------------------------------------------------------------------
Rainer Stroebel
>In <ebqt8o$t3u$01$1...@news.t-online.com>, on 08/15/2006
> at 12:25 AM, Rainer Stroebel <RainerS...@t-online.de> said:
>
>
>
>>the main/core problem by these test is
>>
>>
>
>
>
>>to find out the right case
>>
>>
>
>
>
>> - does it not work because I do not do the right thing
>>
>>
>
>
>
>> - it does not work because of bugs
>>
>>
>
>It's neither of the above. I can not work because the code to implement
>booting from USB removable media was never implemented for eCS/OS2.
>
>You can test all you want, but this will not change.
>
>Regards,
>
>Steven
>
>
>
Hallo Steven,
OS/2 2.1 does not boot from CD-ROM or removable devices like the ORB.
How OS/2 and the follow up eCs has two different methods to boot
from CD-ROM.
Veit Kannegieser has done and doing a greate job with MEMDISK .
Daniela Engert implemented the boot from ORB devices in DANIS506.
So lets find out exactly why the current OS/2 eCs does not boot from USB.
1. Define the problem, identify the components and functions involved.
2. Develop a concept how to boot from USB with the know restrictions.
3. Implement the concept.
With your in deep knowledge of the OS/2 internals
you could certainly contribute to the project with concepts and code.
kind regards
Rainer
Rainer Stroebel
> Oliver Rick schrieb:
>
>
>> Have you tried to use the medium as a large floppy instead (see
>> Readme of
>> latest LVM Update)?
>>
>>
>>
> That was my first try -- run into bios startup prolbems does not
> investigate it ---> just switch to PRM with the USB MSD Device.
>
> Might go back to it and look more in detail for the bios startup
>
>> You might ask Veit Kannegieser, the author of MemDisk ("El Torito" No
>> Emulation CD-ROM boot support and boot eCS from a memory hard disk),
>> if he
>> has an idea.
>>
>>
>>
> good idea - I have already look at memdisk documentation to learn
> more about the boot process.
Hello Oliver,
i have done some additional testings with "Big Floppy" format.
The T23 BIOS still identifies the USB storages ( formated as big
floppy ) as Hard Disk Boot Device in the Startup List of the BIOS.
Consequently the boot step process fail in an very early stage.
The BIOS reads a few records ( LED is shortly blinking ) and than
stops. No OS/2 blob on screen.
Back to the PRM format and booting from the first primary partition :-)
kind regards
Rainer
PS: I do have run sysinstx C: for the big floppy :-)
Mike Luther
Rainer Stroebel wrote:
> Steven Levine schrieb:
>
>> It's neither of the above. I can not work because the code to implement
>> booting from USB removable media was never implemented for eCS/OS2.
>>
>> You can test all you want, but this will not change.
>>
>> Regards,
>>
>> Steven
>
> So lets find out exactly why the current OS/2 eCs does not boot from USB.
>
> kind regards
>
> Rainer
>
I'm not sure this is what you want, not exactly anyway. But in at least
one case you can boot at least the R40, R51 and T43P Thinkpads here from
USB Floppy Diskette devices. I do that all the time now for a specific
reason. Then, using Jan's DFSEE at least version 7.1.5, I can clone the
entire Thinkpad to a commonly available USB 2.0 Hard Disk unit. That
does require a specific LVM rediscover command prior to working with the
USB device so that you can access it to do this. Using the technique, I
can also reverse the process and clone the entire USB 2.0 disk back into
a Thinkpad as well. Or as needed, I can transfer data to or from a
working Thinkpad to the USB 2.0 device for archival and recovery. But
if I want to get and instantly working rebuilt or built Thinkpad from
this clone whole drive issue from DFSEE, it has to be done from a floppy
diskette boot run unless one wants to boot to a command line for the
rebuilt or built unit and do a complete CHKDSK on every partition post
construction.
That does work to produce a complete working unit if you want to go
through the clone process that way. But since you are going to use the
floppy diskettes to build the unit, as I see it, why not just use the
floppy boot for the creation process as well?
And as another vector toward this whole toolset, I've not also tried
booting to an OS/2 command line using the <ALT F1> way of using DFSEE to
produce this backup operation. But with LVM rediscover that might also
work.
Thus you can use the DFSEE toolset to clone from a completely
functioning Thinkpad here provided you want to go through the cleanup
process on rebuild or build.
Now the next step. I've tried this and it works fine.
I can take the entire USB 2.0 Hard Disk unit with the cloned total
Thinkpad drive in it. I can then go into the BIOS settings for either
the R40, R51, or T43P units here. I can switch the positions of the
boot devices in the BIOS setups so that the USB 2.0 device is in a
proper position so that it is bootable. I can then completely boot the
tested Thinkpads to the external plug in USB hard drive completely as if
it was the internal boot hard drive.
IBM cooperated late last year with I and Jan to enable this technique.
But as far as I know other than what is already present in the latest
IBM drivers for OS/2 available through Passport Advantage for MCP2, the
only issues involved for the floppy diskette utility diskette boot run
are careful selection of what is in the floppy diskette CONFIG.SYS, what
is not, and the order of things in it. That as well as being sure that
the correct and needed drivers are on those floppy diskettes for the job.
Thus, to me, at least using the commonly available removable media
floppy diskette USB drive units can be used this way. And this
'qualifies' as two ways to use USB removable media to boot OS/2.
I have not tested the creation of 2.88 MB extended density floppy disks
out of the available tools to do this and tried any of the several
different USB floppy drive units to see if a single such disk could be
created to boot to OS/2 this way. If these USB floppy units will
recognize the larger diskette, I sure suspect you could do this as with
none of the bulky SCSI drivers and so on needed for the IDE drives in
the Thinkpads, it would sure look like there would be enough room on the
extended diskette to pull this one off. I've got all the third party
utilities to try this, but at the moment I'm covered up with other stuff
and can't take time to investigate it.
If this will help, then I can post the entire CONFIG.SYS for the
required floppy diskette boot. I can also post the required driver
content for the disks and thus the level of the build needed to
accomplish this. But as I read this thread, I think this is more of a
thrust toward using a plug-in USB stick device to boot OS/2. And I've
not gone at all toward that concurrent with what all else I'm swamped
with to do at the moment.
--
--> Sleep well; OS2's still awake! ;)
Mike Luther
Rainer Stroebel
Hello Mike,
thanks for your comments and report of your usage of the USB boot and
USB storage devices.
At lot of useful information!!
Let see if I a do understand it -:)
Mike Luther schrieb:
> Maybe this is the place and time to comment ..
>
Certainly, the knowledge is stored and accessible/researchable by google.
So no bits of value information and experience got lost. -:)
> I'm not sure this is what you want, not exactly anyway. But in at
> least one case you can boot at least the R40, R51 and T43P Thinkpads
> here from USB Floppy Diskette devices. I do that all the time now for
> a specific reason. Then, using Jan's DFSEE at least version 7.1.5, I
> can clone the entire Thinkpad to a commonly available USB 2.0 Hard
> Disk unit. That does require a specific LVM rediscover command prior
> to working with the USB device so that you can access it to do this.
> Using the technique, I can also reverse the process and clone the
> entire USB 2.0 disk back into a Thinkpad as well. Or as needed, I can
> transfer data to or from a working Thinkpad to the USB 2.0 device for
> archival and recovery. But if I want to get and instantly working
> rebuilt or built Thinkpad from this clone whole drive issue from
> DFSEE, it has to be done from a floppy diskette boot run unless one
> wants to boot to a command line for the rebuilt or built unit and do a
> complete CHKDSK on every partition post construction.
>
So - you have a Thinkpad with attached USB Floppy Drive and an USB
External HDD ( formated as PRM Partitioned Removable Media and accessed
via Basdev=USBMSD.add )
You have OS/2 Start/Maintenance Diskettes with modification for booting
from USB Floppy by USB Bios support of the ThinkPad.
Main goal: boot a "stand alone system" to backup/restore the primary
HHD ( without OS activites on the source disk - the source disk is
treated as read only )
> That does work to produce a complete working unit if you want to go
> through the clone process that way. But since you are going to use
> the floppy diskettes to build the unit, as I see it, why not just use
> the floppy boot for the creation process as well?
No, I do not " use floppy diskettes to build the unit"
1. I just have built a mini maintenance system on my HDD ( via
Bootable 4.3.2 )
2. Copy the system on an USB PRM primary partition
2.1 format the PRM
2.2 xcopy HDD to PRM drive with hotserv parm
2.3 sysinstx c: ( the drive letter of my USB PRM
device - my HDD configuration start with drive letter F: )
3. Change the bios startup to boot form the USB Mass Storage Device
and try to boot it.
>
> And as another vector toward this whole toolset, I've not also tried
> booting to an OS/2 command line using the <ALT F1> way of using DFSEE
> to produce this backup operation. But with LVM rediscover that might
> also work.
>
I have tested it.
The config.sys used to boot at F3 - boot to command prompt - does not
include the USB driver on my system.
If you use a config.sys with a USB support - it is not a stand alone
system - the HDD is not "read only" used
This is discussed in the yahoo dfsee support forum in thread "Go to a
bigger HDD by clone disk"
>
> Thus you can use the DFSEE toolset to clone from a completely
> functioning Thinkpad here provided you want to go through the cleanup
> process on rebuild or build.
>
> Now the next step. I've tried this and it works fine.
>
>
> I can take the entire USB 2.0 Hard Disk unit with the cloned total
> Thinkpad drive in it. I can then go into the BIOS settings for either
> the R40, R51, or T43P units here. I can switch the positions of the
> boot devices in the BIOS setups so that the USB 2.0 device is in a
> proper position so that it is bootable. I can then completely boot
> the tested Thinkpads to the external plug in USB hard drive completely
> as if it was the internal boot hard drive.
This is exactly what I want to do!!!!! Boot am OS/2 System from USB
HDD ( e.g. CF Card )
My approach only a to keep is simple - so I built a mint test system
without WPS support -
the T23 only has USB 1.1 - very very slow for handling x MBs !!
So what do I miss - I use the current drivers - what is the
relevant difference between our systems?
An USB stick device ( e.g. a CF Card ) formated as PMR is the
same as an external USB HDD - both are handled by USBMSD.ADD as
removables ??
or does handle USBMSD an USB HDD as fixed device ?
Is this the "rosetta stone" to the problem :-)
Mike,
Can you publish ??
1. Your config.sys from the USN Starter Diskette -
Just to document a working example for the description "How to
Boot from USB Diskette" in the usbbasic.txt file from the USB Basic
Driver Package
2. the disk layout of your cloned system - which you boot from
the USB HDD
As a first step to find out why my CF Card System does not boot :-)
I will try to reproduce you configuration by doing a test with a "real
fixed HDD" at the USB port of my T23.
Looking forward your reply
kind regards
Rainer
Rainer Stroebel
>
> Hello Mike,
>
>
>
>
> I will try to reproduce you configuration by doing a test with a
> "real fixed HDD" at the USB port of my T23.
>
I have just done the test with a "real fixed HDD" at the USB port
Mod the bios start up list and boot from the USB Mass Storage
--> no success --> "OS/2 is unable to operate...........
What magic do you do ??????????? :-)
Doug Bissett
<RainerS...@t-online.de> wrote:
> Doug Bissett schrieb:
...snip...
> >>I built a maintenance system with bootable on HDD drive Y: to do test boot.
> >>Again to keep it simple, the system a stripped down to minimum e.g.
> >>no wps, no cdrom support, no udf, no hpfs - just boot to command prompt.
> >>
> >>So the change from y: to C: in config.sys for the system on the PRM
> >>should do the
> >>job.
> >>
> >>Do I miss something?
> >>
> >>
> >
> >I think so. There are still some INI entries, that could be pointing
> >to the wrong place (Y:).
> >
> >
> >
>
> I am a bit confused about your statements about the ini files/entries.
>
> The mini maintenance system for command prompt built by bootable does not
> built/generate any os2.ini or os2sys.ini files in the /os2 directory.
>
> My understanding is:
>
> The ini files are needed by the WPS and application programms
>
> ( and the USBprint.sys driver )
You are, probably, right about that.
...snip...
> >You really shouldn't need boot manager on the device, but you would
> >need to make it startable (active, in Windows terms), so it would
> >immediately try to boot the code that is on the device. Unfortunately,
> >doing that with any of the current tools (except, possibly, DFSEE),
> >will make Boot Manager not startable on the hard disk (that can be
> >fixed, after you remove the USB device).
> >
> >
> The BMR does not see the PRM device .
>
> By booting from HDD the BMR gets control -
> the BMR does not look / start the PRM via I13
> even if there is an entry in the BMR tables for the system on the PRM.
> ( the USB drive LED is not green )
>
> By booting from the PRM ( via bios start up pointing first to the USB in
> start up boot list )
> the BMR does not get control - the system is directly loaded from the
> primary partition on the
> PRM
That should be correct, but I am wondering if the master boot record,
on the device, doesn't have the startable (active) bit on, so it will
attempt to boot from it. If that isn't on, the code will, likely, just
ignore it. Then again, I may be wrong about that.
...snip...
Okay, I am off to try this, to see what i can figure out...
Hope this helps...
Rainer Stroebel
...snip
>>>
>>>
>>The BMR does not see the PRM device .
>>
>>By booting from HDD the BMR gets control -
>>the BMR does not look / start the PRM via I13
>> even if there is an entry in the BMR tables for the system on the PRM.
>>( the USB drive LED is not green )
>>
>>By booting from the PRM ( via bios start up pointing first to the USB in
>>start up boot list )
>>the BMR does not get control - the system is directly loaded from the
>>primary partition on the
>>PRM
>>
>>
>
>That should be correct, but I am wondering if the master boot record,
>on the device, doesn't have the startable (active) bit on, so it will
>attempt to boot from it. If that isn't on, the code will, likely, just
>ignore it. Then again, I may be wrong about that.
>
>
>
like many other times DFSEE proves its value :-)
Here is the output for my 1 GB CF Card
ID
+---+--<disk 4>---</dev/hdd>----+--------+-----------<[ D4 ]
|27}|C:|Prim 06 FAT16 | 1|FAT |IBM 4.50|VL_PRM_1_OS|VN_PRM_P1, |149.9|
|28r|Z:|Log 06 FAT16 | 5|FAT |IBM 4.50|VL_PRM_LOG_|VN_PMR_Log_|100.9|
|32 | |FreeSp Pri/Log | |--- --|------ ----|-
------------------------------ - - - -| 745.0|
+---+--+--------------+--+----+--------+-----------+-----------+-----+
ID = The selection-id used by DFSee for this partition
Postfix: '>' for active partition, firstdisk
'}' active partition, not firstdisk
'!' active partition, bad bootsector
'*' for partitions bootable by BMGR
'-' bootable by BMGR, bad bootsector
'r' Removable device, can beejected
>...snip...
>
>Okay, I am off to try this, to see what i can figure out...
>
>Hope this helps...
>
>
Looking forward to your discoveries -:)
kind regards
Rainer
Mike Luther
With a very great deal of thanks to Rodney Almodavar from IBM Austin for
his and IBM's support of my quest here, I suggest you compare your Thinkpad
Utility Diskette Set made from the MCP2 latest fixpack operations with
what IBM suggested below. And you will need the that latest fixpack and
kernel as XRC_05 so noted to me as well. You'll also need at least certain
later of the device driver changes as well. And note, this is for an
IBM Thinkpad with ONLY the OS/2 operating system on the hard disk in it.
No complicated stuff, OK? Grin!
Here is the CONFIG.SYS file on the Disk #01 of the utility floppy disk set
which boots the R40, R51, T42, and T43P units I've tested and enables all
the USB operations we need. Pay particular attention to the lines in the
CONFIG.SYS file for BASEDEV=USBD.SYS /I13 and SET COPYFROMFLOPPY=1. Yes,
that copy from floppy line is there for a reason!
******* Begin CONFIG.SYS file below *******
buffers=32
iopl=yes
memman=swap,delayswap
protshell=sysinst1.exe
set os2_shell=sysinst2.exe
diskcache=D2,LW
protectonly=yes
libpath=.;\;\os2\dll;\os2\install;
ifs=jfs.ifs
ifs=hpfs.ifs /c:64
pauseonerror=no
codepage=850
devinfo=kbd,us,keyboard.dcp
devinfo=scr,ega,vtbl850.dcp
device=\dos.sys
device=\mouse.sys
set path=\;\os2;\os2\system;\os2\install
set dpath=\;\os2;\os2\system;\os2\install
set keys=on
basedev=ibmkbd.sys
basedev=ibm1s506.add /W
basedev=os2dasd.dmd
basedev=os2lvm.dmd
device=\testcfg.sys
basedev=ibmatapi.flt /A:1 /U:0 /TYPES:CLZ
BASEDEV=USBD.SYS /I13
SET COPYFROMFLOPPY=1
BASEDEV=USBUHCD.SYS
BASEDEV=USBUHCD.SYS
BASEDEV=USBUHCD.SYS
BASEDEV=USBEHCD.SYS
BASEDEV=USBHID.SYS
BASEDEV=USBMSD.ADD /FLOPPIES:1 /REMOVABLES:2
EARLYMEMINIT=TRUE
basedev=xdfloppy.flt
set os2_shell=cdinst.exe
set saveconnect=1
set cdrominst=1
ifs=cdfs.ifs /q
device=\os2cdrom.dmd
device=\refpart.sys
set os2_shell=cmd.exe
******* End CONFIG.SYS file above *******
Now here is the disk directory for the Disk #01 for the Utility Boot disk
set which will boot these same IBM Thinkpads and operate the entire USB
operations as described. It's been alpha sorted for clarity. Please note
the dates of certain of the files below. Note the file dates for:
IBM1S506 ADD 70490 1-03-05 11:43a
IBMATAPI FLT 39568 3-12-04 11:28a
IBMIDECD FLT 39568 3-12-04 11:28a
OS2DASD DMD 40709 7-19-05 12:34p
OS2LVM DMD 32519 10-12-04 1:13p
SCREEN01 SYS 12212 11-18-04 5:39p
SCREEN02 SYS 12174 11-18-04 5:39p
SNOOP LST 603 10-03-05 7:45p
USBD SYS 22988 9-10-04 3:48p
USBEHCD SYS 42112 9-10-04 3:49p
USBHID SYS 13292 9-10-04 3:49p
USBMSD ADD 40612 9-10-04 3:49p
USBUHCD SYS 27814 9-10-04 3:48p
******* Begin DISK #01 DIRECTORY below *******
AHA6360 SNP 27383 5-21-02 4:21p
AIC7770 ADD 42574 5-21-02 3:49p
AIC7770 SNP 17456 4-15-02 3:07p
AIC7870 ADD 154140 4-24-02 2:30p
AIC7870 SNP 9104 5-21-02 4:22p
AIC78U2 ADD 156412 1-18-02 12:45p
CLOCK01 SYS 4552 4-15-02 3:16p
CLOCK02 SYS 4318 4-15-02 3:16p
CONFIG SYS 970 10-04-05 10:08p
CONFIG X 627 4-17-01 10:08a
DELIVERY SYS 3806 10-25-01 4:40p
DPT20XX ADD 24700 5-21-02 3:50p
DPT20XX SNP 6700 4-15-02 3:03p
IBM1FLPY ADD 35418 10-08-02 12:02p
IBM1FLPY SNP 2393 4-15-02 3:04p
IBM1S506 ADD 70490 1-03-05 11:43a
IBM1S506 SNP 9160 4-15-02 3:03p
IBM2FLPY ADD 14424 5-21-02 3:45p
IBM2SCSI ADD 33115 5-21-02 3:46p
IBMATAPI FLT 39568 3-12-04 11:28a
IBMIDECD FLT 39568 3-12-04 11:28a
IBMINT13 I13 11522 5-21-02 3:46p
IBMKBD SNP 1404 4-15-02 2:59p
IBMKBD SYS 8714 5-21-02 4:03p
IPSRAID ADD 22310 4-24-02 6:47p
IPSRAID SNP 2624 4-15-02 3:09p
ISAPNP SNP 21374 4-15-02 3:01p
KBDBASE SYS 24989 5-21-02 4:05p
MOUSE SNP 3744 4-15-02 3:00p
NETDET1 SNP 11270 4-15-02 3:05p
NETDET2 SNP 14410 4-15-02 3:05p
OS2DASD DMD 40709 7-19-05 12:34p
OS2LOGO 60979 4-19-00 4:17p
OS2LVM DMD 32519 10-12-04 1:13p
PARALLEL SNP 3032 10-25-01 5:50p
PCIBUS SNP 2496 4-15-02 3:00p
PCMCIA SNP 3912 4-15-02 3:09p
PNP SYS 2152 4-15-02 3:01p
QL10OS2 ADD 37420 9-21-01 3:46p
QL10OS2 SNP 6516 4-15-02 3:08p
RESERVE SYS 5704 5-21-02 3:37p
RESOURCE SYS 55402 5-21-02 3:36p
RESRV SNP 5257 4-15-02 3:05p
SCREEN01 SYS 12212 11-18-04 5:39p
SCREEN02 SYS 12174 11-18-04 5:39p
SERIAL SNP 3280 4-15-02 2:59p
SNOOP LST 603 10-03-05 7:45p
SYM8XX ADD 63580 1-18-02 12:59p
SYM_HI ADD 67140 1-18-02 12:59p
USBD SYS 22988 9-10-04 3:48p
USBEHCD SYS 42112 9-10-04 3:49p
USBHID SYS 13292 9-10-04 3:49p
USBMSD ADD 40612 9-10-04 3:49p
USBUHCD SYS 27814 9-10-04 3:48p
XDFLOPPY FLT 15060 4-15-02 2:27p
******* End DISK #01 DIRECTORY above *******
Next, here is what is in the SNOOP.LST above for checking;
resrv.snp
netdet1.snp
ibmkbd.snp
ibm1flpy.snp
ibm1s506.snp
; SCSI Snoopers
aha6360.snp
;aha154x.snp
;aha174x.snp
aic7870.snp
ql10os2.snp
;ql40os2.snp
;ql510.snp
;ipsraid.snp
;btscsi.snp
;fd16-700.snp
;fd8xx.snp
;fd7000ex.snp
;flashpt.snp
;dac960.snp
; Misc. Snoopers
pcmcia.snp
;ir.snp
netdet2.snp
; CDROM Snoopers
;mitfx001.snp
;sbcd2.snp
;sony31a.snp
; Audio Snoopers
;sndgalax.snp
;auddrive.snp
;jazz16.snp
;sndblast.snp
;pas16.snp
;bsaudio.snp
; Misc. Snoopers
parallel.snp
mouse.snp
serial.snp
; Note: Place additional snoopers above this line
pcibus.snp
Next you need to check your Disk #02 to compare it to the below. It's been
alpha sorted for clarity. Please note the dates of certain of the files
below. Note the file date for:
CDFS IFS 45159 1-10-05 1:42p
CMD EXE 100620 1-01-05 6:30a
HPFS IFS 141378 1-10-05 1:41p
OS2CDROM DMD 50098 5-26-05 8:57a
******* Begin DISK #02 DIRECTORY below *******
ANSICALL DLL 512 3-08-02 7:16p
BKSCALLS DLL 512 3-08-02 7:16p
BMSCALLS DLL 512 3-08-02 7:16p
BVHINIT DLL 11158 9-10-01 4:17p
BVSCALLS DLL 512 3-08-02 7:16p
CDBOOT EXE 8903 10-25-01 5:16p
CDFS IFS 45159 1-10-05 1:42p
CMD EXE 100620 1-01-05 6:30a
CONFIG ADD 122 9-05-00 11:56a
COUNTRY SYS 50947 9-24-01 3:46p
DEL LST 51081 5-30-01 8:45a
DIR2 DAT 1824 10-04-05 5:35p
DOS SYS 1142 3-11-02 10:05a
DOSCALL1 DLL 137601 12-29-04 11:15a
HARDERR EXE 14889 12-29-04 11:15a
HPFS IFS 141378 1-10-05 1:41p
INST32 DLL 6688 10-26-01 10:09a
JFS IFS 190349 10-12-04 1:16p
KBDCALLS DLL 1024 3-08-02 7:16p
KEYBOARD DCP 25507 10-03-01 6:35a
MOUCALLS DLL 1024 3-08-02 7:16p
MOUSE SYS 30819 5-21-02 4:02p
MSG DLL 512 3-08-02 7:16p
NAMPIPES DLL 1024 3-08-02 7:16p
NLS DLL 512 3-08-02 7:16p
NPXEMLTR DLL 21812 3-08-02 5:50p
OS2CDROM DMD 50098 5-26-05 8:57a
OS2CHAR DLL 512 3-08-02 7:16p
POINTDD SYS 3846 5-21-02 4:10p
QUECALLS DLL 1024 3-08-02 7:16p
REFPART SYS 6679 9-05-00 11:59a
RMINFO DLL 22847 4-03-02 2:31p
RMVIEW EXE 32885 5-21-02 3:36p
SESMGR DLL 1536 3-08-02 7:16p
SYSINST1 EXE 4240 10-26-01 10:09a
SYSLEVEL OS2 165 3-25-02 3:19p
TESTCFG SYS 9808 5-21-02 4:11p
VCU MSG 1760 10-16-01 1:26p
VIOCALLS DLL 2048 3-08-02 7:16p
VTBL850 DCP 10495 10-03-01 6:36a
******* End DISK #02 DIRECTORY above *******
Lastly, you'll need to check your utility diskette #03. It's been alpha
sorted for clarity. Please note the dates of certain of the files
below. Note the file date for:
LVM DLL 167366 10-12-04 1:12p
LVM EXE 93069 10-12-04 1:37p
UHPFS DLL 105008 1-10-05 1:42p
******* Begin DISK #03 DIRECTORY below *******
CHKDSK COM 70720 3-11-02 9:58a
CHKDSK32 DLL 147268 1-01-05 6:43a
FORMAT COM 72064 1-01-05 6:31a
JFS MSG 12619 10-16-01 1:28p
LVM DLL 167366 10-12-04 1:12p
LVM EXE 93069 10-12-04 1:37p
LVM MSG 18184 10-16-01 1:26p
LVMH MSG 21126 10-16-01 1:26p
NLS DLL 512 3-08-02 7:16p
ODPANS DLL 21968 10-26-01 12:29p
ODPRTDRV EXE 12023 3-11-02 10:51a
OSDELETE EXE 69124 3-25-02 3:23p
OSO001 MSG 175354 3-08-02 3:56p
SHPIINST DLL 37328 10-16-01 6:52p
TEDIT EXE 9390 9-24-01 3:57p
TEDIT HLP 14596 9-05-00 12:13p
UCONV DLL 23331 9-04-02 5:30p
UHPFS DLL 105008 1-10-05 1:42p
UJFS DLL 286488 10-12-04 1:30p
VIOCALLS DLL 2048 3-08-02 7:16p
******* End DISK #03 DIRECTORY above *******
To use Jan's DFSEE 7.1.5 or later for the clone and back clone operation,
you obviously prepare a floppy diskette with his OS/2 toolset on it. But
to get it to run in a jam in on floppy boot, you will also have to add
certain of the IBM DLL's to that floppy. That's covered in documentation
for his product.
What isn't quite as clear is that after you boot the Thinkpad to a command
line from the diskettes, you will then need to stuff in the utiity diskette
and do a manual command line 'LVM /PRMrediscover' command. If you have
the drivers and all correct in the above, that is when you'll get to see
the USB 2.0 Hard Disk drive show up in Jan's toolset. And a way you go!
Lastly. You simply CLONE your whole Thinkpad to the external USB hard
disk unit. Then, if you wish, you go into the BIOS on the Thinkpad, and
make sure you have enabled USB support for it. Then you move the USB boot
device up to the top position in the boot device list! Poof! At least
here the entire Thinkpad operation will actually boot from the USB 2.0
external hard disk just like it did with the native disk inside.
Provided you have set up the required USB device drivers and so on or later
that are shown in the above explanations for the unit, of course...
Hope this helps!
Steven Levine
at 01:18 PM, Mike Luther <mike....@ziplog.com> said:
>I'm not sure this is what you want, not exactly anyway. But in at least
>one case you can boot at least the R40, R51 and T43P Thinkpads here from
>USB Floppy Diskette devices.
This is all documented in usbbasic.txt. The /I13 switch was introduced
something over than 4 years ago.
The functionality is not Thinkpad specific. It should work for any system
that can be configured to boot from diskette. Of course, IBM will not
sign up to support it on anything but Thinkpads.
In general booting from diskette is simpler than booting from hard disk
because there are no drive letter issues.
>I can take the entire USB 2.0 Hard Disk unit with the cloned total
>Thinkpad drive in it. I can then go into the BIOS settings for either
>the R40, R51, or T43P units here. I can switch the positions of the
>boot devices in the BIOS setups so that the USB 2.0 device is in a
>proper position so that it is bootable. I can then completely boot the
>tested Thinkpads to the external plug in USB hard drive completely as if
>it was the internal boot hard drive.
Now you have me confused. Are you saying you can boot directly from the
external USB hard disk drive or not?
If so, are you using BootManager and what volume are you booting from and
is the boot volume a primary?
>But as I read this thread, I think this is more of a
>thrust toward using a plug-in USB stick device to boot OS/2.
This is true, but since the same drivers are involved what works for an
external USB hard drive should work for a stick, barring defects.
Mike Luther
Steven Levine wrote:
> In <12e3idc...@corp.supernews.com>, on 08/15/2006
> at 01:18 PM, Mike Luther <mike....@ziplog.com> said:
>
>
>> I'm not sure this is what you want, not exactly anyway. But in at least
>> one case you can boot at least the R40, R51 and T43P Thinkpads here from
>> USB Floppy Diskette devices.
>>
>
> This is all documented in usbbasic.txt. The /I13 switch was introduced
> something over than 4 years ago.
>
> The functionality is not Thinkpad specific. It should work for any system
> that can be configured to boot from diskette. Of course, IBM will not
> sign up to support it on anything but Thinkpads.
>
Haven't tried it as non-laptops have floppy drives here. Thanks for the
thought though.
> In general booting from diskette is simpler than booting from hard disk
> because there are no drive letter issues.
>
Yes.
> Now you have me confused. Are you saying you can boot directly from the
> external USB hard disk drive or not?
>
Yes I can boot it from the external drive.
> If so, are you using BootManager and what volume are you booting from and
> is the boot volume a primary?
>
Not BM. I should have said that as well as noted that this is a
straight partition C: install, though with a number of other partitions
which are non-bootable. Sorry I didn't mention that.
>> But as I read this thread, I think this is more of a
>> thrust toward using a plug-in USB stick device to boot OS/2.
>>
>
> This is true, but since the same drivers are involved what works for an
> external USB hard drive should work for a stick, barring defects.
>
Which is one of the reasons I went ahead and posted the whole thing. I
was hoping that someone else might be able to carry the mission to
that. As I noted I'm totally swamped now with other things to do and a
couple of earlier Email comments to me about all this were nudging me to
try to go toward that light as well!
> Regards,
>
> Steven
>
>
Thanks Steve for your help..
Rainer Stroebel
> Better get your blinders on horse, grin. Very long post!
....snip
> With a very great deal of thanks to Rodney Almodavar from IBM Austin for
> his and IBM's support of my quest here, I suggest you compare your
> Thinkpad
> Utility Diskette Set made from the MCP2 latest fixpack operations with
> what IBM suggested below.
>
Thanks for your excellent instructions "How to boot/install from USB Floppy"
I will get hold of an USB floppy drive to test it.
The ThinkPad T23 can use a normal floppy drive by
- Ultra Bay 2000
- with special cable Floppy to Parallel Port
- via Ultra Bay 2000 in the DOCK II Station
So - there has been no need for an USB floppy in the past.
Even with the T41 I have done the BIOS update via boot diskette
with the Dock II and an diskette drive in the ultra bay 2000.
Two questions to the config.sys for the USB Install diskettes:
1. Device=refpart.sys
I found the following info in the ConfigTool Database:
Driver for accessing the Reference Partiton on IBM PS/2
Microchannel Machines
( Information by: Raphl-Macrus Boenicke )
IBM PS/2 MCA Machines does not have USB support.
==> obsolete on the disketts or do we need to add info to the
ConfigTool Database?
2. EARLYMEMINIT=TRUE
the need for this kernel instruction is a new fact?
ConfigTool DataBase:
=TRUE
Setting used with kernel 14.062e or higher
will allow device drivers, etc.,
access to the memory above 16mb early in boot.
Previously, this was only available after DD and IFS init
was completed.
This has various implications when enabled:
a) large VDISKs are possible. I tried
DEVICE=\os2\vdisk.sys 16000
b) AHA154X.ADD may do bad things to your system.
Don't even ask.
c) There may be some settings of HPFS386 cache that
are incompatible.
<<=NOTE=>> This feature is experimental and may not
work with later kernels.
====> Is it real required ???
> To use Jan's DFSEE 7.1.5 or later for the clone and back clone operation,
> you obviously prepare a floppy diskette with his OS/2 toolset on it. But
> to get it to run in a jam in on floppy boot, you will also have to add
> certain of the IBM DLL's to that floppy. That's covered in documentation
> for his product.
>
> What isn't quite as clear is that after you boot the Thinkpad to a
> command
> line from the diskettes, you will then need to stuff in the utiity
> diskette
> and do a manual command line 'LVM /PRMrediscover' command. If you have
> the drivers and all correct in the above, that is when you'll get to see
> the USB 2.0 Hard Disk drive show up in Jan's toolset. And a way you go!
>
> Lastly. You simply CLONE your whole Thinkpad to the external USB hard
> disk unit. Then, if you wish, you go into the BIOS on the Thinkpad, and
> make sure you have enabled USB support for it. Then you move the USB
> boot
> device up to the top position in the boot device list! Poof! At least
> here the entire Thinkpad operation will actually boot from the USB 2.0
> external hard disk just like it did with the native disk inside.
>
> Provided you have set up the required USB device drivers and so on or
> later
> that are shown in the above explanations for the unit, of course...
>
Adding the additional info from you last post/reply to Steven:
> Not BM. I should have said that as well as noted that this is a
straight partition C: install,
> though with a number of other partitions which are non-bootable.
Sorry I didn't mention that.
Well my test system boots from the USB CF Card. So the Bios Startup is
okay.
I have replaced the dani driver with the current IBM driver
and add the EARLYMEMINIT=TRUE instruction
Still getting the message
"OS/2 is unable to operate your hard disk or diskette drive"
after the kernel and all the basedev drivers are loaded.
The last driver loaded is the os2lvm.dmd at the boot screen and
all the output line for the USB and IBM1s506 drivers are displayed.
From your post to steve:
> Which is one of the reasons I went ahead and posted the whole thing.
> I was hoping that someone else might be able to carry the mission to
that.
> As I noted I'm totally swamped now with other things to do and
> a couple of earlier Email comments to me about all this were nudging me
> to try to go toward that light as well!
>
> Hope this helps!
>
>
Well I am clearly on the mission -:)
Your posting does help a lot!!!
It proofs the concept/implementation of booting from USB HDD work at
your configuration!!!
So what are the relevant diffences betten our to installations:
1. Hardware
ThinkPad R40, R51 and T43P versus T23
USB Floppy verus nornal floppy
Question: Is tthe USB floppey attached, wth you boot from USB
Mass Stroage??
2. Software
We are both on MCP 2 wtih the current fix level and driver level.
Both configurations boot from c: without an BMR
Diffences in the Config.sys:
This is the config.sys of my last test - boot form USB CF 16 MB
Card PRM Fat formatted :
==============================================================================0
rem
rem config.sys with usb stamtments arranged according boot from USB
deskette instructions
rem
EARLYMEMINIT=TRUE
BASEDEV=USBD.SYS /V /I13
BASEDEV=usbuhcd.sys /V
BASEDEV=usbuhcd.sys /V
BASEDEV=usbuhcd.sys /V
BASEDEV=usbhid.sys /V
BASEDEV=usbmsd.add /V /FLOPPIES:1 /REMOVABLES:1 /CDS:1
LIBPATH=.;C:\OS2\DLL;
SET PATH=C:\OS2;
SET DPATH=C:\OS2;
SET LANG=de_DE
SET ULSPATH=C:\language
DEVICE=C:\OS2\unicode.sys
IFS=C:\OS2\cdfs.ifs /Q /W
rem IFS=C:\OS2\hpfs.ifs /CACHE:1024 /AUTOCHECK:*
rem IFS=C:\OS2\udf.ifs /V
BASEDEV=ibmkbd.sys
BASEDEV=chkdsk.sys
BASEDEV=ibm1flpy.add /A:0 /FORCE:2 /U:0 /F:1.44MB /U:1 /F:1.44MB
BASEDEV=IBM1S506.add /W
BASEDEV=IBMIDECD.flt
BASEDEV=os2dasd.dmd
BASEDEV=os2lvm.dmd
COUNTRY=049,C:\OS2\country.sys
CODEPAGE=850
DEVINFO=KBD,GR,C:\OS2\keyboard.dcp
DEVINFO=SCR,VGA,C:\OS2\viotbl.dcp
rem DEVICE=C:\OS2\os2cdrom.dmd
rem BASEDEV=os2aspi.dmd /ALL
DEVICE=C:\OS2\testcfg.sys
DEVICE=C:\OS2\dos.sys
DEVICE=C:\OS2\pointdd.sys
DEVICE=C:\OS2\mouse.sys
rem str SET SAVECONNECT=1
rem str SET COPYFROMFLOPPY=1
SET TEMP=C:\Temp
SET TMP=C:\Temp
SUPPRESSPOPUPS=C
MEMMAN=SWAP,PROTECT
SWAPPATH=C: 512
IOPL=YES
PROTECTONLY=YES
AUTOFAIL=YES
THREADS=1024
DISKCACHE=D,LW,AC:+C
MAXWAIT=1
SET VIDEO_DEVICES=VIO_VGA
SET VIO_VGA=DEVICE(BVHVGA)
SET TZ=CET-1CES,3,-1,0,7200,10,-1,0,10800,3600
SET PROMPT=$i[$p]
SET KEYS=ON
PROTSHELL=C:\OS2\cmd.exe
SET OS2_SHELL=C:\OS2\cmd.exe
SET COMSPEC=C:\OS2\cmd.exe
========================================================================
arranging the usb in the "normal sequence" without the /I13 parm does
not make any difference in
the past, but I will test it again after this posting - Just to be sure.
Do you see any faults in this config.sys???
How does your config.sys in the USB HDD boot System looks ??
kind regards
Rainer
Oliver Rick
> i have done some additional testings with "Big Floppy" format.
> The T23 BIOS still identifies the USB storages ( formated as big
> floppy ) as Hard Disk Boot Device in the Startup List of the BIOS.
It must fit one of the floppy sizes. :-(
Do I remember correctly that you're trying to use a CF type II card in an
external x-in-1 USB 2.0 card reader attached to an USB 1.0 host controller?
> PS: I do have run sysinstx C: for the big floppy :-)
Copy the contents of the required sectors, maybe edit them afterwards?
BTW, do you run SYSINSTX.COM first before you copy anything else to the
medium?
Oliver Rick
BTW, Du bist einfacher zu lesen, wenn Du nur das nötigste zitierst.
>> "The known parameters from the NEWDASD beta package /RF (removable as
>> fixed) - now activated by default - [...]"
> It does change something at my eCs 1.2R test partition - code level
> without any upgrade
Odd, it shouldn't. The feature stems from pre-LVM times (1997), so maybe
the code is still there in the LVM-aware OS2DASD.DMD and causes unexpect
effects.
> The Icons for the removables are different in the drives folder with /RF
> and without. It does not change the icons in my MCP2 partition -with all
> current fixes and Kernel 14.104a
Probably the old icon wasn't replaced in eCS, because its appearance wasn't
expected.
> At boot from the USB the os2lvm.dmd driver is displayed ( ALT - F2
> active ) a little longer with /RF before the error message comes up.
From the Readme of the NewDASD Beta 2:
-+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+-
Drive letters for removable media, regardless of partition type, will
be placed AFTER the normal fixed disks IN DISK ORDER (ie., all primary
AND extended partitions, if any, from the first removable disk,
followed by next disk, etc)
If no media is detected at boot time, one drive letter will be
allocated. If, subsequently, media is inserted that has multiple
partitions, only the first partition will be accessible. This will
not cause any problems with the unaccessed partition.
If MULTIPLE partitions are detected on the media at boot time, the
media will NOT be removable, except at reboot. All partitions will
then be accessible.
-+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+-
So it takes some time until all checks are done. With LVM this is done when
called with /RediscoverPRM after boot.
> I have looked for a way to verify this :-) have not found anything.
Maybe Chuck McKinnis can write something regarding the changed icon
resources in eCS?
> The HDD maintenance mini system with boot to the command prompt shows a
> problem:
> [first access of PRM produces SYS0015, subsequent accesses succeed]
> This is through for a FAT or a HPFS maintenance partition
> any ideas ??????
The drive wasn't "discovered" yet. But the access initiated the discovery
process?
> the drive is not read for reading the country command from config.sys
> via the protected mode drivers - the basedev driver just load before
> It should be ready for an read command !
> May be there is an bug in the basedev drivers and an work around
> is implemented via the usb monitor /
> via lvm /RediscoverPRM as you questioned
OS/2 fails to pass access via INT13h to the driver. Did you try to load
USBMSD before all other ADDs (or even without the other *.ADDs)? Do the
DEVICE statements include drive letters or just a backslash?
Doug Bissett
Well, bootAble seems to build a valid boot system on the stick, but I
had to do it to the default (F:) drive. Then, I discovered that none
of my systems seem to have an option to boot from a removable USB
device. Boot Manager *might* do it, but I don't think you can count on
having Boot Manager, when you will need to boot from the stick. I need
to spend more time with this, and time, at the moment, is in short
supply.
I expect that by playing with LVM to remove the actual C: drive,
temporarily, that I could get the stick mounted as drive C:, then use
bootAble to build a maintenance partition on it, as drive C:. Then, it
would be necessary to change some of the CONFIG.SYS entries, to add
extra things ("/I13", then see if I can figure out what needs to be
done to try to boot from the device. I expected to find something in
my new Athlon64 system to be able to do that, but it isn't obvious, if
it is there (perhaps, it needs "Legacy USB" enabled, which prevents
the USB drivers from loading properly). Overall, I think it is MUCH
easier to simply use bootAble to create a boot CD, or DVD, since
almost all machines can now boot from one of them.
Rainer Stroebel
>On Tue, 15 Aug 2006 Rainer Stroebel wrote:
>
>
>
>>i have done some additional testings with "Big Floppy" format.
>>
>>
>
>
>
>>The T23 BIOS still identifies the USB storages ( formated as big
>>floppy ) as Hard Disk Boot Device in the Startup List of the BIOS.
>>
>>
>
>It must fit one of the floppy sizes. :-(
>
>Do I remember correctly that you're trying to use a CF type II card in an
>external x-in-1 USB 2.0 card reader attached to an USB 1.0 host controller?
>
>
Yes and No - it is an 1 in 1 USB 1.0 card reader attached to an USB 1.0
host controller ( T23)
This device was in use before the x-in-1 support has been implemented
the USB drivers. -:)
>
>
>>PS: I do have run sysinstx C: for the big floppy :-)
>>
>>
>
>Copy the contents of the required sectors, maybe edit them afterwards?
>
>BTW, do you run SYSINSTX.COM first before you copy anything else to the
>medium?
>
>
>
I do not remember the sequence for the "Big Floppy" -
for the PRM I have first do the xcopy and than the SYSINSTX C:
For the PRM it does work - the OS/2 kernel is loaded -
May be there is a difference for the Floppy format
good point !! - Have to test it :-)
kind regards
Rainer
eTroll
Rainer Stroebel wrote:
> Oliver Rick schrieb:
>
> >On Tue, 15 Aug 2006 Rainer Stroebel wrote:
> >
> >
> >
> >>i have done some additional testings with "Big Floppy" format.
> >>
Please look at http://forum.ecomstation.ru/viewforum.php?f=26
May be it can help partially.
I can make jfs bootable volume on usb drive and copy working
installation form hard disk to usb disk. After some manipulations with
SEDIT I can start boot prosess from usb drive while there is no any
other hdds (seem there is some extra check for boot fom only 1st hdd
in kernel)
Next I see white blobe, logo and "unable opereate your hard disk drive"
SY,
EK
PS: you can ask me at forum.ecomstation.ru, I rarely watch to
os2.bugs
Rainer Stroebel
>OS/2 fails to pass access via INT13h to the driver. Did you try to load
>USBMSD before all other ADDs (or even without the other *.ADDs)? Do the
>DEVICE statements include drive letters or just a backslash?
>
>
>
>
Hallo Oliver,
the USBMSD.Add driver is the first add driver in the config.sys.
I have just on point to verify:
"OS/2 is unable to operate your hard disk or diskette drive"
It is the " or diskette drive" in the error message -
I do have a diskette drive in the Ultra Bay 2000 - but no USB diskette
drive attached !!!
Dani wrote something about diskette access in Daniboot.doc
"ENHANCEMENTS
DaniBoot.FLT is enhanced over cd_boot.flt in many ways:
.
.
.
7) The presence of the floppy driver IBM*FLPY.ADD is no longer
required. "
===> So the floppy drive has to be there when CD_BOOT.ftl has been
used!!!!!
Do the drivers need an USB diskette device in the hardware configuration?
Mike Luther can probably answer it --> the question has been out in my
last reply to him.
Rainer Stroebel
>Please look at http://forum.ecomstation.ru/viewforum.php?f=26
>
>May be it can help partially.
>
>I can make jfs bootable volume on usb drive and copy working
>installation form hard disk to usb disk. After some manipulations with
>SEDIT I can start boot prosess from usb drive while there is no any
>other hdds (seem there is some extra check for boot fom only 1st hdd
>in kernel)
>Next I see white blobe, logo and "unable opereate your hard disk drive"
>
>SY,
>EK
>
>PS: you can ask me at forum.ecomstation.ru, I rarely watch to
>os2.bugs
>
>
>
Hello eTrool ( Papa Karlo ?? ),
thank for you additional facts to the subject -
we should keep the discussion on this forum. It is monitored, archived
and researchable by google.
I found the following posting on your link:
Post <http://forum.ecomstation.ru/viewtopic.php?p=591#591>Posted: Mon
Aug 07, 2006 10:50 Post subject: Reply with quote
<http://forum.ecomstation.ru/posting.php?mode=quote&p=591>
------------------------------------------------------------------------
Some info about problems with booting from USB and/or non-first hdd:
* LVM can't made bootable partition on USB disk, ALVM - can
o one of the reason is that
DosDevIOCtl(FileHandle,IOCTL_OEMHLP,OEMHLP_QUERYDISKINFO,..) return
zeros for USB disks (bug in usbd drivers ?)
* eCS's installer can't install system on USB drive - it even doesn't
show any usb drives with "show all volumes and partitions" option (This
is not mini-lvm window) May be this is result of OEMHLP_QUERYDISKINFO bug
* kernel want not to boot from second (non first) disk in system. This
is definitly some extra checks in kernel and it can be easily fixed
* if there is only one disk in system and this disk is usb than boot
starts, shows logo and than hangup with "the system can't operate your
hard disk" message. This is possibly os2dasd and usbd - related
* some extra bug with genegal LVM - it can't add to bootmanager (on hard
disk) botable volume from USB drive (the reason will be found in ALVM)
==================================================================================
==================================================================================
Well - Mike Luther write in his posting
he has been able to boot his system completely form an USB HDD.
No message "OS2 is unable to operate ........
So there is a way to work with the current os2dasd.dmd, os2lvm.dmd and
usbmsd.add .
We just have to find How to do it "more generally" :-)
kind regards
Rainer
Mike Luther
Rainer Stroebel wrote:
>
> Two questions to the config.sys for the USB Install diskettes:
>
>
> 1. Device=refpart.sys
>
> I found the following info in the ConfigTool Database:
>
> Driver for accessing the Reference Partiton on IBM PS/2
> Microchannel Machines
> ( Information by: Raphl-Macrus Boenicke )
>
> IBM PS/2 MCA Machines does not have USB support.
>
>
> ==> obsolete on the disketts or do we need to add info to the
> ConfigTool Database?
>
>
I have no idea ..
> 2. EARLYMEMINIT=TRUE
>
> the need for this kernel instruction is a new fact?
> ConfigTool DataBase: =TRUE
>
> Setting used with kernel 14.062e or higher
>
> will allow device drivers, etc.,
> access to the memory above 16mb early in boot.
> Previously, this was only available after DD and IFS
> init was completed.
> This has various implications when enabled:
>
> a) large VDISKs are possible. I tried
> DEVICE=\os2\vdisk.sys 16000
> b) AHA154X.ADD may do bad things to your system.
> Don't even ask.
> c) There may be some settings of HPFS386 cache that
> are incompatible.
>
> <<=NOTE=>> This feature is experimental and may not
> work with later kernels.
> ====> Is it real required ???
>
It's not involved with the R40 CONFIG.SYS I'll post below..
You commented later:
>
> Well my test system boots from the USB CF Card. So the Bios Startup
> is okay.
>
> I have replaced the dani driver with the current IBM driver
> and add the EARLYMEMINIT=TRUE instruction
>
> Still getting the message
>
> "OS/2 is unable to operate your hard disk or diskette drive"
>
> after the kernel and all the basedev drivers are loaded.
> The last driver loaded is the os2lvm.dmd at the boot screen and
> all the output line for the USB and IBM1s506 drivers are displayed.
>
>
> From your post to steve:
>
> > Which is one of the reasons I went ahead and posted the whole thing.
> > I was hoping that someone else might be able to carry the mission
> to that.
> > As I noted I'm totally swamped now with other things to do and
> > a couple of earlier Email comments to me about all this were
> nudging me
> > to try to go toward that light as well!
>
>>
>> Hope this helps!
>>
>>
> Well I am clearly on the mission -:)
>
> Your posting does help a lot!!!
> It proofs the concept/implementation of booting from USB HDD work
> at your configuration!!!
>
> So what are the relevant diffences betten our to installations:
>
> Question: Is tthe USB floppey attached, wth you boot from USB Mass
> Stroage??
>
No! To get this to work, after we are all done CLONING the complete
drive to the USB 2.0 Hard Disk unit, then we do not attach the USB
Floppy diskette drive to the Thinkpad we want to boot from the external
plugged in USB 2.0 Hard Disk unit. And further, we still enable booting
from USB device in the Thinkpad BIOS, and we place the boot device for
that USB hard drive at the top of the boot device parade from the F12
BIOS setting choices.
>
> 2. Software
>
> We are both on MCP 2 wtih the current fix level and driver level.
>
> Both configurations boot from c: without an BMR
>
> Diffences in the Config.sys:
>
Your CONFIG.SYS snipped ..
> How does your config.sys in the USB HDD boot System looks ??
OK what I have below is the actual CONFIG.SYS from an R40 that does this
but I've clipped away the PATH designations both to allow a decent
display since some of them are way up in the 250+ length. That and for
security reasons of not wanting to show everything else. You'll notice
the device partition check at the top has lots of drives on 'checkmate',
grin!
SET SCKILLFEATUREENABLED=1
BASEDEV=NVCHOOK.SYS /EX:60000
BASEDEV=XDFLOPPY.FLT
IFS=C:\OS2\JFS.IFS /AUTOCHECK:*
IFS=C:\OS2\HPFS.IFS /CACHE:2048 /CRECL:4 /AUTOCHECK:CDEFGIJK
PROTSHELL=C:\OS2\PMSHELL.EXE
SET USER_INI=C:\OS2\OS2.INI
SET SYSTEM_INI=C:\OS2\OS2SYS.INI
SET OS2_SHELL=C:\OS2\CMD.EXE
SET AUTOSTART=PROGRAMS,TASKLIST,FOLDERS,CONNECTIONS,WARPCENTER
SET RESTARTOBJECTS=STARTUPFOLDERSONLY
SET RUNWORKPLACE=C:\OS2\PMSHELL.EXE
SET COMSPEC=C:\OS2\CMD.EXE
REM ** Tells OS/2 where to locate Dynamic Link Libraries or DLLs.
LIBPATH= (Path not shown)
REM ** Path statement to programs just like DOS.
SET PATH= (Path not shown)
REM ** Tells OS/2 progs where look for data files for progs to use it.
SET DPATH= (Path not shown)
SET PROMPT=$i[$p]
REM ** Where the help files are located.
SET HELP= (Path not shown)
SET GLOSSARY=C:\OS2\HELP\GLOSS;
SET IPF_KEYS=SBCS
REM ** Comm apps need this to NO.
PRIORITY_DISK_IO=NO
FILES=50
BASEDEV=IBMKBD.SYS
DEVICE=C:\IBMCOM\PROTOCOL\LANPDD.OS2
DEVICE=C:\IBMCOM\PROTOCOL\LANVDD.OS2
DEVICE=C:\IBMCOM\LANMSGDD.OS2 /I:C:\IBMCOM /S
DEVICE=C:\IBMCOM\PROTMAN.OS2 /I:C:\IBMCOM
REM ** Normally this is on but on a Thinkpad disaster
REM BASEDEV=IBMKBD.SYS /NUMON
DEVICE=C:\OS2\BOOT\TESTCFG.SYS
DEVICE=C:\OS2\BOOT\DOS.SYS
DEVICE=C:\OS2\BOOT\PMDD.SYS
COUNTRY=001,C:\OS2\SYSTEM\COUNTRY.SYS
CODEPAGE=437,850
DEVINFO=KBD,US,C:\OS2\KEYBOARD.DCP
DEVICE=C:\OS2\MDOS\ANSI.SYS
SET LANG=en_US
SET TZ=UTC0
SET ULSPATH=C:\LANGUAGE;
DEVICE=C:\OS2\BOOT\UNICODE.SYS
BUFFERS=90
IOPL=YES
DISKCACHE=64,LW
REM ** Max seconds program may hog system, Use 1 with COMM progs **
REM ** MAXWAIT=3
MAXWAIT=1
MEMMAN=SWAP,PROTECT
SWAPPATH=D:\OS2\SWAP 2048 16383
BREAK=OFF
THREADS=1024
PRINTMONBUFSIZE=134,134,134
SET KEYS=ON
REM ** Points to on-line documentation for OS/2 and other OS/2 programs.
SET BOOKSHELF= (Path not shown)
REM ** Editor path call
SET SOMIR= (Path not shown)
SET SOMDDIR=C:\OS2\ETC\DSOM
BASEDEV=TIMER0.SYS
run=c:\norman\bin\zanda.exe /load
RUN=C:\OS2\EPW.EXE
RUN=C:\OS2\SYSTEM\LVMALERT.EXE
RUN=C:\OS2\EXTENDFS.EXE *
BASEDEV=PRINT01.SYS /IRQ
BASEDEV=IBM1FLPY.ADD
BASEDEV=IBM2FLPY.ADD
BASEDEV=IBM1S506.ADD /V
BASEDEV=I2OXPORT.SYS
BASEDEV=I2OSOSM.ADD
BASEDEV=OS2DASD.DMD
BASEDEV=OS2LVM.DMD
BASEDEV=CHKDSK.SYS
SET EPMPATH=C:\OS2\APPS;
PROTECTONLY=NO
SHELL=C:\OS2\MDOS\COMMAND.COM C:\OS2\MDOS /e:768 /p
FCBS=16,8
RMSIZE=640
DEVICE=C:\OS2\MDOS\VEMM.SYS
DOS=LOW,NOUMB
DEVICE=C:\OS2\MDOS\VXMS.SYS /UMB
DEVICE=C:\OS2\MDOS\VDPMI.SYS
DEVICE=C:\OS2\MDOS\VDPX.SYS
DEVICE=C:\OS2\MDOS\VWIN.SYS
DEVICE=C:\OS2\MDOS\VW32S.SYS
DEVICE=C:\OS2\BOOT\APM.SYS
DEVICE=C:\OS2\MDOS\VAPM.SYS
RUN=C:\OS2\APMDAEMN.EXE
DEVICE=C:\OS2\BOOT\OS2CDROM.DMD /Q
IFS=C:\OS2\BOOT\UDF.IFS
IFS=C:\OS2\BOOT\CDFS.IFS /Q /W
DEVICE=C:\OS2\MDOS\VCDROM.SYS
BASEDEV=IBMIDECD.FLT
BASEDEV=IBMATAPI.FLT
DEVICE=C:\OS2\MDOS\VMOUSE.SYS
DEVICE=C:\OS2\BOOT\POINTDD.SYS
DEVICE=C:\OS2\BOOT\MOUSE.SYS
BASEDEV=PCMCIA.SYS
BASEDEV=IBM2SS14.SYS
BASEDEV=AUTODRV2.SYS
DEVICE=C:\THINKPAD\VPCMCIA.SYS
DEVICE=C:\THINKPAD\PCMSSDIF.SYS
DEVICE=C:\THINKPAD\PCM2SRAM.SYS
DEVICE=C:\THINKPAD\FLSH2MTD.SYS
DEVICE=C:\THINKPAD\PCM2FLSH.SYS
BASEDEV=PCM2ATA.ADD /S:2 /!DM /NOBEEP
DEVICE=C:\OS2\BOOT\APM.SYS
DEVICE=C:\OS2\MDOS\VAPM.SYS
RUN=C:\OS2\APMDAEMN.EXE
DEVICE=C:\OS2\BOOT\OS2CDROM.DMD /Q
IFS=C:\OS2\BOOT\UDF.IFS
IFS=C:\OS2\BOOT\CDFS.IFS /Q /W
DEVICE=C:\OS2\MDOS\VCDROM.SYS
BASEDEV=IBMIDECD.FLT
BASEDEV=IBMATAPI.FLT
DEVICE=C:\OS2\MDOS\VMOUSE.SYS
DEVICE=C:\OS2\BOOT\POINTDD.SYS
DEVICE=C:\OS2\BOOT\MOUSE.SYS
BASEDEV=PCMCIA.SYS
BASEDEV=IBM2SS14.SYS
BASEDEV=AUTODRV2.SYS
DEVICE=C:\THINKPAD\VPCMCIA.SYS
DEVICE=C:\THINKPAD\PCMSSDIF.SYS
DEVICE=C:\THINKPAD\PCM2SRAM.SYS
DEVICE=C:\THINKPAD\FLSH2MTD.SYS
DEVICE=C:\THINKPAD\PCM2FLSH.SYS
BASEDEV=PCM2ATA.ADD /S:2 /!DM /NOBEEP
REM ** STANDARD DOS COMM DRIVER ROUTINE
REM ** DEVICE=C:\OS2\BOOT\COM.SYS
REM DEVICE=C:\OS2\BOOT\COM.SYS (1,3F8,4) (2,2F8,3) (3,3E8,15)
REM ** GWINN REPLACEMENT FOR IT IN ROOT DIRECTORY OF BOOT DRIVE.
DEVICE=C:\SIO\SIO.SYS (COM1,03F8,4,-) (COM2,02F8,3)
(COM4,INTERNET:2E8,NONE,-)
REM ** STANDARD DOS COMM DRIVER ROUTINE
REM ** DEVICE=C:\OS2\MDOS\VCOM.SYS
REM ** GWINN REPLACEMENT FOR IT IN ROOT DIRECTORY OF BOOT DRIVE.
DEVICE=C:\SIO\VSIO.SYS
SET DEVICEFONTDISABLED=NO
BASEDEV=USBUHCD.SYS
BASEDEV=USBUHCD.SYS
BASEDEV=USBUHCD.SYS
BASEDEV=USBEHCD.SYS
REM ** BASEDEV=USBD.SYS /REQ:USBUHCD$
BASEDEV=USBD.SYS /REQ:USBUHCD$,USBOHCD$,USBEHCD$
BASEDEV=USBHID.SYS
REM ** BASEDEV=USBMSD.ADD /REMOVABLES:2 /FLOPPIES:0 /A_USAGE:2 /V
REM ** BASEDEV=USBMSD.ADD /REMOVABLES:1 /FLOPPIES:0
BASEDEV=USBMSD.ADD
REM BASEDEV=USBCDROM.ADD
REM ** STANDARD DOS COMM DRIVER ROUTINE
REM ** DEVICE=C:\OS2\BOOT\COM.SYS
REM DEVICE=C:\OS2\BOOT\COM.SYS (1,3F8,4) (2,2F8,3) (3,3E8,15)
REM ** GWINN REPLACEMENT FOR IT IN ROOT DIRECTORY OF BOOT DRIVE.
DEVICE=C:\SIO\SIO.SYS (COM1,03F8,4,-) (COM2,02F8,3)
(COM4,INTERNET:2E8,NONE,-)
REM ** STANDARD DOS COMM DRIVER ROUTINE
REM ** DEVICE=C:\OS2\MDOS\VCOM.SYS
REM ** GWINN REPLACEMENT FOR IT IN ROOT DIRECTORY OF BOOT DRIVE.
DEVICE=C:\SIO\VSIO.SYS
REM DEVICE=C:\OS2\BOOT\USBKBD.SYS /V
REM DEVICE=C:\OS2\BOOT\USBCOM.SYS /V
REM DEVICE=C:\OS2\BOOT\USBMOUSE.SYS /V
REM DEVICE=C:\OS2\BOOT\USBQSER.SYS
DEVICE=C:\OS2\BOOT\USBSER.SYS /V
SET DISPLAYTYPE=VGA
SET DMIPATH=C:\DMISL\BIN
SET CLASSPATH= (Path not shown)
SET SWING_HOME=C:\JAVA11\Swing
SET MMBASE=C:\MMOS2;
SET DSPPATH=C:\MMOS2\DSP;
SET NCDEBUG=4000
RUN=C:\MMOS2\MIDIDMON.EXE
DEVICE=C:\MMOS2\SSMDD.SYS
DEVICE=C:\MMOS2\R0STUB.SYS
DEVICE=C:\MMOS2\MIDI.SYS
DEVICE=C:\MMOS2\VCSHDD.SYS
RUN=C:\MMOS2\QRYMMCD.EXE
DEVICE=C:\MMOS2\UNIAUD32.SYS /V
DEVICE=C:\MMOS2\UNIAUD16.SYS /V /C
DEVICE=C:\MMOS2\GWDTA.SYS
DEVICE=C:\MMOS2\VGENAUD.SYS
RUN=C:\MMOS2\GENAUDD.EXE
CALL=C:\IBMCOM\PROTOCOL\NETBIND.EXE
RUN=C:\IBMCOM\LANMSGEX.EXE
SET NLSPATH=C:\MPTN\MSG\NLS\%N;D:\os2tk45\msg\%N;C:\TCPIP\msg\ENUS850\%N;
SET ETC=C:\MPTN\ETC
DEVICE=C:\MPTN\PROTOCOL\SOCKETSK.SYS
DEVICE=C:\MPTN\PROTOCOL\AFOS2.SYS
DEVICE=C:\MPTN\PROTOCOL\AFINETK.SYS
DEVICE=C:\MPTN\BIN\VDOSTCP.VDD
DEVICE=C:\MPTN\PROTOCOL\AFNB.SYS /S:50 /C:80 /N:21
RUN=C:\MPTN\BIN\AFNBINI.EXE
RUN=C:\MPTN\BIN\CNTRL.EXE
CALL=C:\OS2\CMD.EXE /Q /C C:\MPTN\BIN\MPTSTART.CMD >NUL
RUN=C:\MPTN\BIN\VDOSCTL.EXE
DEVICE=C:\IBMCOM\PROTOCOL\NETBEUI.OS2
DEVICE=C:\IBMLAN\NETPROG\RDRHELP.200
IFS=C:\IBMLAN\NETPROG\NETWKSTA.200 /I:C:\IBMLAN /N
DEVICE=C:\IBMCOM\PROTOCOL\NETBIOS.OS2
SET IKEYMAN_HOME=C:\IBMGSK
SET IKEYMAN40_HOME=C:\IBMGSK40
SET IKEYMAN50_HOME=C:\IBMGSK50
RUN=C:\IBMLAN\NETPROG\LSDAEMON.EXE
RUN=C:\OS2\SYSTEM\LOGDAEM.EXE
RUN=C:\OS2\EPWROUT.EXE 1
DEVICE=C:\OS2\LOG.SYS
SET NWDBPATH=C:\IBMLAN\NETPROG
SET DLSINI=C:\IBMLAN\NETPROG\NETGUI.INI
SET INIT_FILE_NAMES=netgui
SET INIT_FILE_RANGES=200
SET WPS_COMMUNICATION=YES
SET LANINSTEP=D:\CID\SERVER\ibmls
SET TCPLANG=en_US
DEVICE=C:\TCPIP\bin\vdostcp.sys
SET INCLUDE= (Paths not shown)
SET LIB= (Paths not shown)
SET LOCPATH=C:\IBMI18N\LOCALE;C:\LANGUAGE\LOCALE
SET IPFC=D:\os2tk45\ipfc;
SET SOMBASE=D:\os2tk45\som;C:\OS2
SET SOMRUNTIME=D:\os2tk45\som\common;C:\OS2\DLL
SET MAXIMUS=D:\MAX\MAX.PRM
SET CPREF=CP1.INF+CP2.INF+CP3.INF
SET GPIREF=GPI1.INF+GPI2.INF+GPI3.INF+GPI4.INF
SET MMREF=MMREF1.INF+MMREF2.INF+MMREF3.INF
SET PMREF=PM1.INF+PM2.INF+PM3.INF+PM4.INF+PM5.INF
SET HELPNDX=epmkwhlp.ndx
SET WPSREF=WPS1.INF+WPS2.INF+WPS3.INF
SET SMADDSTAR=1
SET SMEMIT=h;ih;c
SET SMTMP=C:\
SET SMINCLUDE=D:\os2tk45\h;D:\os2tk45\idl;.;D:\os2tk45\som\include;
SET SMCLASSES=wptypes.idl
DEVICE=C:\IBMCOM\MACS\E100B.OS2
REM ** This line relates to IBM Thinkpad to allow display and monitor
REM ** choices for the DOCKSTATION monitor. Without this line on
REM ** Thinkpad installations that choice is missing.
SET VCFG_NO_DDC=TRUE
REM ** Another violation instanced by Lotus Smart Suite install.
REM ** needs to be removed if SSuite is installed.
REM SET TMP=C:\TCPIP\TMP
REM ** Original choice on install. Lotus Smart Suite changes this to
REM ** #:\LOTUS4\TEMP. Failure to hand rem this out seems a problem
REM ** on SSuite boxes.
REM SET TEMP=C:\
BASEDEV=PMVDMCC.SYS /F /V
REM ** DO NOT KNOW WHAT THIS IS YET HERE ON R40
SET HPP_BASE=C:\HOMEPAGE
SET SOMDTHREADPEER=1
SET TEMP=D:\LOTUSW4\TEMP
SET TMP=D:\LOTUSW4\TEMP
SET LOTUS_CLS=D:\LOTUSW4\compnent
SET DSQCOMMTRACE=OFF
SET DSSDIR=D:\LOTUSW4\WORK\APPROACH
SET DSSPATH=D:\LOTUSW4\APPROACH\
SET HOSTNAME=ZL3016
REM ** Below are SNAP/SDD items.
DEVINFO=SCR,VGA,C:\OS2\BOOT\VIOTBL.DCP
SET VIDEO_DEVICES=VIO_SVGA
SET VIO_SVGA=DEVICE(BVHVGA,BVHSVGA)
DEVICE=C:\OS2\MDOS\VSVGA.SYS
SET GRADD_CHAINS=C1
DEVICE=C:\OS2\GRADD.SYS
DEVICE=C:\OS2\MDOS\VMANWIN.SYS
SET GREEXT=SDDGREXT
SET C1=SDDGRADD
DEVICE=C:\OS2\SDDHELP.SYS /Q
REM ** Below is SNAP/SDD special option for comm port intensive boxes.
SET SNAP_INTCURSOR_DISABLE=Y
REM ** This produces plug-in USB card reader options in Connections
REM ** If $ICPMOS2 isn't in CONFIG.SYS OS2PCAARD has to be last line
REM ** in CONFIG.SYS and if it is then $ICPMOS2 has to be the last line
REM BASEDEV=OS2PCARD.DMD
DEVICE=C:\OS2\$ICPMOS2.SYS /G
*************** End of CONFIG.SYS file *****************
Notice Viet's little patch tool for the 'proper' handling of the last 2K
little piece of code in the DOS memory deal? It is really needed to be
able to gain the 'normal' 100K extra lower DOS memory in DOS-VDM sessions.
Notice also that comment at the end of the CONFIG.SYS file relative to
the last line needed in the CONFIG.SYS for Thinkpads. Apparently it is
on some; not on others. It *IS* here for things to work right. And at
this point each upgrade to the SNAP video driver installs moves the SNAP
code below it. As well some other IBM install stuff does too. I
frequently have to go in and hand correct that last item order in these
Thinkpad files to keep things working correctly.
As well, again, recall that I am not trying to 'install' an OS/2
operation to the USB device! What I am doing is to take a completely
functional clean hard disk image in total for just a PLAIN OS/2 all HPFS
partitioned and LVM's MCP2 operation - clone it completely to the USB
2.0 Hard Disk external device. Then, to boot from it, I am simply
placing that fully cloned external disk as the chosen disk in the BIOS
boot order.
Rainer Stroebel
> Sorry it took all day to get to this but .. more blinders .. long post!
>
> Rainer Stroebel wrote:
>
>>
>> Question: Is tthe USB floppey attached, wth you boot from USB Mass
>> Stroage??
>>
> No! To get this to work, after we are all done CLONING the complete
> drive to the USB 2.0 Hard Disk unit, then we do not attach the USB
> Floppy diskette drive to the Thinkpad we want to boot from the
> external plugged in USB 2.0 Hard Disk unit. And further, we still
> enable booting from USB device in the Thinkpad BIOS, and we place the
> boot device for that USB hard drive at the top of the boot device
> parade from the F12 BIOS setting choices.
>
===> So the " or diskette drive" in the error message does not
apply :-)
>> 2. Software
>>
>> We are both on MCP 2 wtih the current fix level and driver level.
>>
>> Both configurations boot from c: without an BMR
>>
>> Diffences in the Config.sys:
>>
> Your CONFIG.SYS snipped ..
>
>> How does your config.sys in the USB HDD boot System looks ??
>
>
> OK what I have below is the actual CONFIG.SYS from an R40 that does
> this but I've clipped away the PATH designations both to allow a
> decent display since some of them are way up in the 250+ length. That
> and for security reasons of not wanting to show everything else.
> You'll notice the device partition check at the top has lots of drives
> on 'checkmate', grin!
>
>
----Snip
>
> As well, again, recall that I am not trying to 'install' an OS/2
> operation to the USB device! What I am doing is to take a completely
> functional clean hard disk image in total for just a PLAIN OS/2 all
> HPFS partitioned and LVM's MCP2 operation - clone it completely to the
> USB 2.0 Hard Disk external device. Then, to boot from it, I am simply
> placing that fully cloned external disk as the chosen disk in the BIOS
> boot order.
>
>
Currently I have two test suits:
1. bottom up approach
Minimal system to boot to the command level ( build with bootable )
Systems with drive boot drive letter C:
2. bottom down approach
Copy of an clean MCP2 plus latest fixlevel installation
Systems with drive boot drive letter G:
no progress up till no: "OS/ 2 is unable ..." is reproduce able on
every test, no intermitting error :-)
to compare the your R40 config.sys with my config.sys from G: the
config tool to sort both config:
here is the results:
Your config.sys form R40:
REM >>> Section: GENERAL <<<
PROTSHELL=C:\OS2\PMSHELL.EXE
PRIORITY_DISK_IO=NO
COUNTRY=001,C:\OS2\SYSTEM\COUNTRY.SYS
CODEPAGE=437,850
DEVINFO=KBD,US,C:\OS2\KEYBOARD.DCP
BUFFERS=90
IOPL=YES
DISKCACHE=64,LW
MAXWAIT=1
MEMMAN=SWAP,PROTECT
SWAPPATH=D:\OS2\SWAP 2048 16383
THREADS=1024
PRINTMONBUFSIZE=134,134,134
PROTECTONLY=NO
DEVINFO=SCR,VGA,C:\OS2\BOOT\VIOTBL.DCP
REM >>> Section: BASEDEVICES <<<
BASEDEV=NVCHOOK.SYS /EX:60000
BASEDEV=XDFLOPPY.FLT
BASEDEV=IBMKBD.SYS
BASEDEV=TIMER0.SYS
BASEDEV=PRINT01.SYS /IRQ
BASEDEV=IBM1FLPY.ADD
BASEDEV=IBM2FLPY.ADD
BASEDEV=IBM1S506.ADD /V
BASEDEV=I2OXPORT.SYS
BASEDEV=I2OSOSM.ADD
BASEDEV=OS2DASD.DMD
BASEDEV=OS2LVM.DMD
BASEDEV=CHKDSK.SYS
BASEDEV=IBMIDECD.FLT
BASEDEV=IBMATAPI.FLT
BASEDEV=PCMCIA.SYS
BASEDEV=IBM2SS14.SYS
BASEDEV=AUTODRV2.SYS
BASEDEV=PCM2ATA.ADD /S:2 /!DM /NOBEEP
BASEDEV=IBMIDECD.FLT
BASEDEV=IBMATAPI.FLT
BASEDEV=PCMCIA.SYS
BASEDEV=IBM2SS14.SYS
BASEDEV=AUTODRV2.SYS
BASEDEV=PCM2ATA.ADD /S:2 /!DM /NOBEEP
BASEDEV=USBUHCD.SYS
BASEDEV=USBUHCD.SYS
BASEDEV=USBUHCD.SYS
BASEDEV=USBEHCD.SYS
BASEDEV=USBD.SYS /REQ:USBUHCD$,USBOHCD$,USBEHCD$
BASEDEV=USBHID.SYS
BASEDEV=USBMSD.ADD
REM >>> Section: IFS <<<
IFS=C:\OS2\JFS.IFS /AUTOCHECK:*
IFS=C:\OS2\HPFS.IFS /CACHE:2048 /CRECL:4 /AUTOCHECK:CDEFGIJK
IFS=C:\OS2\BOOT\UDF.IFS
IFS=C:\OS2\BOOT\CDFS.IFS /Q /W
IFS=C:\OS2\BOOT\UDF.IFS
IFS=C:\OS2\BOOT\CDFS.IFS /Q /W
IFS=C:\IBMLAN\NETPROG\NETWKSTA.200 /I:C:\IBMLAN /N
REM >>> Section: SET STATEMENTS <<<
SET SCKILLFEATUREENABLED=1
SET USER_INI=C:\OS2\OS2.INI
SET SYSTEM_INI=C:\OS2\OS2SYS.INI
SET OS2_SHELL=C:\OS2\CMD.EXE
SET AUTOSTART=PROGRAMS,TASKLIST,FOLDERS,CONNECTIONS,WARPCENTER
SET RESTARTOBJECTS=STARTUPFOLDERSONLY
SET RUNWORKPLACE=C:\OS2\PMSHELL.EXE
SET COMSPEC=C:\OS2\CMD.EXE
SET PATH= (Path not shown)
SET DPATH= (Path not shown)
SET PROMPT=$i[$p]
SET HELP= (Path not shown)
SET GLOSSARY=C:\OS2\HELP\GLOSS;
SET IPF_KEYS=SBCS
SET LANG=en_US
SET TZ=UTC0
SET ULSPATH=C:\LANGUAGE;
SET KEYS=ON
SET BOOKSHELF= (Path not shown)
SET SOMIR= (Path not shown)
SET SOMDDIR=C:\OS2\ETC\DSOM
SET EPMPATH=C:\OS2\APPS;
SET DEVICEFONTDISABLED=NO
SET DISPLAYTYPE=VGA
SET DMIPATH=C:\DMISL\BIN
SET CLASSPATH= (Path not shown)
SET SWING_HOME=C:\JAVA11\Swing
SET MMBASE=C:\MMOS2;
SET DSPPATH=C:\MMOS2\DSP;
SET NCDEBUG=4000
SET NLSPATH=C:\MPTN\MSG\NLS\%N;D:\os2tk45\msg\%N;C:\TCPIP\msg\ENUS850\%N;
SET ETC=C:\MPTN\ETC
SET IKEYMAN_HOME=C:\IBMGSK
SET IKEYMAN40_HOME=C:\IBMGSK40
SET IKEYMAN50_HOME=C:\IBMGSK50
SET NWDBPATH=C:\IBMLAN\NETPROG
SET DLSINI=C:\IBMLAN\NETPROG\NETGUI.INI
SET INIT_FILE_NAMES=netgui
SET INIT_FILE_RANGES=200
SET WPS_COMMUNICATION=YES
SET LANINSTEP=D:\CID\SERVER\ibmls
SET TCPLANG=en_US
SET INCLUDE= (Paths not shown)
SET LIB= (Paths not shown)
SET LOCPATH=C:\IBMI18N\LOCALE;C:\LANGUAGE\LOCALE
SET IPFC=D:\os2tk45\ipfc;
SET SOMBASE=D:\os2tk45\som;C:\OS2
SET SOMRUNTIME=D:\os2tk45\som\common;C:\OS2\DLL
SET MAXIMUS=D:\MAX\MAX.PRM
SET CPREF=CP1.INF+CP2.INF+CP3.INF
SET GPIREF=GPI1.INF+GPI2.INF+GPI3.INF+GPI4.INF
SET MMREF=MMREF1.INF+MMREF2.INF+MMREF3.INF
SET PMREF=PM1.INF+PM2.INF+PM3.INF+PM4.INF+PM5.INF
SET HELPNDX=epmkwhlp.ndx
SET WPSREF=WPS1.INF+WPS2.INF+WPS3.INF
SET SMADDSTAR=1
SET SMEMIT=h;ih;c
SET SMTMP=C:\
SET SMINCLUDE=D:\os2tk45\h;D:\os2tk45\idl;.;D:\os2tk45\som\include;
SET SMCLASSES=wptypes.idl
SET VCFG_NO_DDC=TRUE
SET HPP_BASE=C:\HOMEPAGE
SET SOMDTHREADPEER=1
SET TEMP=D:\LOTUSW4\TEMP
SET TMP=D:\LOTUSW4\TEMP
SET LOTUS_CLS=D:\LOTUSW4\compnent
SET DSQCOMMTRACE=OFF
SET DSSDIR=D:\LOTUSW4\WORK\APPROACH
SET DSSPATH=D:\LOTUSW4\APPROACH\
SET HOSTNAME=ZL3016
SET VIDEO_DEVICES=VIO_SVGA
SET VIO_SVGA=DEVICE(BVHVGA,BVHSVGA)
SET GRADD_CHAINS=C1
SET GREEXT=SDDGREXT
SET C1=SDDGRADD
SET SNAP_INTCURSOR_DISABLE=Y
REM >>> Section: LIBPATH <<<
LIBPATH= (Path not shown)
REM >>> Section: DEVICES <<<
DEVICE=C:\IBMCOM\PROTOCOL\LANPDD.OS2
DEVICE=C:\IBMCOM\LANMSGDD.OS2 /I:C:\IBMCOM /S
DEVICE=C:\IBMCOM\PROTMAN.OS2 /I:C:\IBMCOM
DEVICE=C:\OS2\BOOT\TESTCFG.SYS
DEVICE=C:\OS2\BOOT\DOS.SYS
DEVICE=C:\OS2\BOOT\PMDD.SYS
DEVICE=C:\OS2\BOOT\UNICODE.SYS
DEVICE=C:\OS2\BOOT\APM.SYS
DEVICE=C:\OS2\BOOT\OS2CDROM.DMD /Q
DEVICE=C:\OS2\BOOT\POINTDD.SYS
DEVICE=C:\OS2\BOOT\MOUSE.SYS
DEVICE=C:\THINKPAD\PCMSSDIF.SYS
DEVICE=C:\THINKPAD\PCM2SRAM.SYS
DEVICE=C:\THINKPAD\FLSH2MTD.SYS
DEVICE=C:\THINKPAD\PCM2FLSH.SYS
DEVICE=C:\OS2\BOOT\APM.SYS
DEVICE=C:\OS2\BOOT\OS2CDROM.DMD /Q
DEVICE=C:\OS2\BOOT\POINTDD.SYS
DEVICE=C:\OS2\BOOT\MOUSE.SYS
DEVICE=C:\THINKPAD\PCMSSDIF.SYS
DEVICE=C:\THINKPAD\PCM2SRAM.SYS
DEVICE=C:\THINKPAD\FLSH2MTD.SYS
DEVICE=C:\THINKPAD\PCM2FLSH.SYS
DEVICE=C:\SIO\SIO.SYS (COM1,03F8,4,-) (COM2,02F8,3)
(COM4,INTERNET:2E8,NONE,-)
DEVICE=C:\SIO\SIO.SYS (COM1,03F8,4,-) (COM2,02F8,3)
(COM4,INTERNET:2E8,NONE,-)
DEVICE=C:\OS2\BOOT\USBSER.SYS /V
DEVICE=C:\MMOS2\SSMDD.SYS
DEVICE=C:\MMOS2\R0STUB.SYS
DEVICE=C:\MMOS2\MIDI.SYS
DEVICE=C:\MMOS2\VCSHDD.SYS
DEVICE=C:\MMOS2\UNIAUD32.SYS /V
DEVICE=C:\MMOS2\UNIAUD16.SYS /V /C
DEVICE=C:\MPTN\PROTOCOL\SOCKETSK.SYS
DEVICE=C:\MPTN\PROTOCOL\AFOS2.SYS
DEVICE=C:\MPTN\PROTOCOL\AFINETK.SYS
DEVICE=C:\MPTN\PROTOCOL\AFNB.SYS /S:50 /C:80 /N:21
DEVICE=C:\IBMCOM\PROTOCOL\NETBEUI.OS2
DEVICE=C:\IBMLAN\NETPROG\RDRHELP.200
DEVICE=C:\IBMCOM\PROTOCOL\NETBIOS.OS2
DEVICE=C:\OS2\LOG.SYS
DEVICE=C:\IBMCOM\MACS\E100B.OS2
DEVICE=C:\OS2\GRADD.SYS
DEVICE=C:\OS2\SDDHELP.SYS /Q
DEVICE=C:\OS2\$ICPMOS2.SYS /G
REM >>> Section: DOS SPECIFICS <<<
FILES=50
BREAK=OFF
SHELL=C:\OS2\MDOS\COMMAND.COM C:\OS2\MDOS /e:768 /p
FCBS=16,8
RMSIZE=640
DOS=LOW,NOUMB
REM >>> Section: DOS DRIVERS <<<
DEVICE=C:\IBMCOM\PROTOCOL\LANVDD.OS2
DEVICE=C:\OS2\MDOS\ANSI.SYS
DEVICE=C:\OS2\MDOS\VEMM.SYS
DEVICE=C:\OS2\MDOS\VXMS.SYS /UMB
DEVICE=C:\OS2\MDOS\VDPMI.SYS
DEVICE=C:\OS2\MDOS\VDPX.SYS
DEVICE=C:\OS2\MDOS\VWIN.SYS
DEVICE=C:\OS2\MDOS\VW32S.SYS
DEVICE=C:\OS2\MDOS\VAPM.SYS
DEVICE=C:\OS2\MDOS\VCDROM.SYS
DEVICE=C:\OS2\MDOS\VMOUSE.SYS
DEVICE=C:\THINKPAD\VPCMCIA.SYS
DEVICE=C:\OS2\MDOS\VAPM.SYS
DEVICE=C:\OS2\MDOS\VCDROM.SYS
DEVICE=C:\OS2\MDOS\VMOUSE.SYS
DEVICE=C:\THINKPAD\VPCMCIA.SYS
DEVICE=C:\SIO\VSIO.SYS
DEVICE=C:\SIO\VSIO.SYS
DEVICE=C:\MMOS2\GWDTA.SYS
DEVICE=C:\MMOS2\VGENAUD.SYS
DEVICE=C:\MPTN\BIN\VDOSTCP.VDD
DEVICE=C:\TCPIP\bin\vdostcp.sys
DEVICE=C:\OS2\MDOS\VSVGA.SYS
DEVICE=C:\OS2\MDOS\VMANWIN.SYS
REM >>> Section: RUN, START AND CALL <<<
run=c:\norman\bin\zanda.exe /load
RUN=C:\OS2\EPW.EXE
RUN=C:\OS2\SYSTEM\LVMALERT.EXE
RUN=C:\OS2\EXTENDFS.EXE *
RUN=C:\OS2\APMDAEMN.EXE
RUN=C:\OS2\APMDAEMN.EXE
RUN=C:\MMOS2\MIDIDMON.EXE
RUN=C:\MMOS2\QRYMMCD.EXE
RUN=C:\MMOS2\GENAUDD.EXE
CALL=C:\IBMCOM\PROTOCOL\NETBIND.EXE
RUN=C:\IBMCOM\LANMSGEX.EXE
RUN=C:\MPTN\BIN\AFNBINI.EXE
RUN=C:\MPTN\BIN\CNTRL.EXE
CALL=C:\OS2\CMD.EXE /Q /C C:\MPTN\BIN\MPTSTART.CMD >NUL
RUN=C:\MPTN\BIN\VDOSCTL.EXE
RUN=C:\IBMLAN\NETPROG\LSDAEMON.EXE
RUN=C:\OS2\SYSTEM\LOGDAEM.EXE
RUN=C:\OS2\EPWROUT.EXE 1
REM >>> Section: Unknown <<<
BASEDEV=PMVDMCC.SYS /F /V
REM >>> Section: REM <<<
REM ** Tells OS/2 where to locate Dynamic Link Libraries or DLLs.
REM ** Path statement to programs just like DOS.
REM ** Tells OS/2 progs where look for data files for progs to use it.
REM ** Where the help files are located.
REM ** Comm apps need this to NO.
REM ** Normally this is on but on a Thinkpad disaster
REM BASEDEV=IBMKBD.SYS /NUMON
REM ** Max seconds program may hog system, Use 1 with COMM progs **
REM ** MAXWAIT=3
REM ** Points to on-line documentation for OS/2 and other OS/2 programs.
REM ** Editor path call
REM ** STANDARD DOS COMM DRIVER ROUTINE
REM ** DEVICE=C:\OS2\BOOT\COM.SYS
REM DEVICE=C:\OS2\BOOT\COM.SYS (1,3F8,4) (2,2F8,3) (3,3E8,15)
REM ** GWINN REPLACEMENT FOR IT IN ROOT DIRECTORY OF BOOT DRIVE.
REM ** STANDARD DOS COMM DRIVER ROUTINE
REM ** DEVICE=C:\OS2\MDOS\VCOM.SYS
REM ** GWINN REPLACEMENT FOR IT IN ROOT DIRECTORY OF BOOT DRIVE.
REM ** BASEDEV=USBD.SYS /REQ:USBUHCD$
REM ** BASEDEV=USBMSD.ADD /REMOVABLES:2 /FLOPPIES:0 /A_USAGE:2 /V
REM ** BASEDEV=USBMSD.ADD /REMOVABLES:1 /FLOPPIES:0
REM BASEDEV=USBCDROM.ADD
REM ** STANDARD DOS COMM DRIVER ROUTINE
REM ** DEVICE=C:\OS2\BOOT\COM.SYS
REM DEVICE=C:\OS2\BOOT\COM.SYS (1,3F8,4) (2,2F8,3) (3,3E8,15)
REM ** GWINN REPLACEMENT FOR IT IN ROOT DIRECTORY OF BOOT DRIVE.
REM ** STANDARD DOS COMM DRIVER ROUTINE
REM ** DEVICE=C:\OS2\MDOS\VCOM.SYS
REM ** GWINN REPLACEMENT FOR IT IN ROOT DIRECTORY OF BOOT DRIVE.
REM DEVICE=C:\OS2\BOOT\USBKBD.SYS /V
REM DEVICE=C:\OS2\BOOT\USBCOM.SYS /V
REM DEVICE=C:\OS2\BOOT\USBMOUSE.SYS /V
REM DEVICE=C:\OS2\BOOT\USBQSER.SYS
REM ** This line relates to IBM Thinkpad to allow display and monitor
REM ** choices for the DOCKSTATION monitor. Without this line on
REM ** Thinkpad installations that choice is missing.
REM ** Another violation instanced by Lotus Smart Suite install.
REM ** needs to be removed if SSuite is installed.
REM SET TMP=C:\TCPIP\TMP
REM ** Original choice on install. Lotus Smart Suite changes this to
REM ** #:\LOTUS4\TEMP. Failure to hand rem this out seems a problem
REM ** on SSuite boxes.
REM SET TEMP=C:\
REM ** DO NOT KNOW WHAT THIS IS YET HERE ON R40
REM ** Below are SNAP/SDD items.
REM ** Below is SNAP/SDD special option for comm port intensive boxes.
REM ** This produces plug-in USB card reader options in Connections
REM ** If $ICPMOS2 isn't in CONFIG.SYS OS2PCAARD has to be last line
REM ** in CONFIG.SYS and if it is then $ICPMOS2 has to be the last line
REM BASEDEV=OS2PCARD.DMD
REM >>> Section: <<<
============================================================================================
comments:
Are the following lines duplicate in your config.sys?
IFS=C:\OS2\BOOT\UDF.IFS
IFS=C:\OS2\BOOT\CDFS.IFS /Q /W
DEVICE=C:\SIO\SIO.SYS (COM1,03F8,4,-) (COM2,02F8,3)
(COM4,INTERNET:2E8,NONE,-)
==============================================================================================
Now my config.sys from the g: system
REM >>> Section: GENERAL <<<
ClockScale=4
ReserveDriveLetter=m
PROTSHELL=G:\OS2\PMSHELL.EXE
PRIORITY_DISK_IO=YES
COUNTRY=049,G:\OS2\SYSTEM\COUNTRY.SYS
CODEPAGE=850,437
DEVINFO=KBD,GR,G:\OS2\KEYBOARD.DCP
BUFFERS=90
IOPL=YES
DISKCACHE=D,LW
MAXWAIT=3
MEMMAN=SWAP,PROTECT
SWAPPATH=G:\OS2\SYSTEM 2048 2048
THREADS=1024
PRINTMONBUFSIZE=134,134,134
PROTECTONLY=NO
DEVINFO=SCR,VGA,G:\OS2\BOOT\VIOTBL.DCP
REM >>> Section: BASEDEVICES <<<
BASEDEV=USBUHCD.SYS /V
BASEDEV=USBUHCD.SYS /V
BASEDEV=USBUHCD.SYS /V
BASEDEV=USBD.SYS /V
BASEDEV=USBHID.SYS /V
BASEDEV=USBMSD.ADD /V
BASEDEV=XDFLOPPY.FLT
BASEDEV=IBMKBD.SYS
BASEDEV=TIMER0.SYS
BASEDEV=PRINT01.SYS
BASEDEV=IBM1FLPY.ADD /A:0 /FORCE:2 /U:0 /F:1.44MB /U:1 /F:1.44MB
BASEDEV=IBM2FLPY.ADD
BASEDEV=IBM1S506.ADD /A:1 /U:0 /ATAPI /FORCE
BASEDEV=I2OXPORT.SYS
BASEDEV=I2OSOSM.ADD
BASEDEV=OS2DASD.DMD
BASEDEV=OS2LVM.DMD
BASEDEV=CHKDSK.SYS
BASEDEV=IBMIDECD.FLT
BASEDEV=IBMATAPI.FLT
REM >>> Section: IFS <<<
IFS=G:\OS2\JFS.IFS /AUTOCHECK:*
IFS=G:\OS2\HPFS.IFS /CACHE:2048 /CRECL:4 /AUTOCHECK:GHKLPSTVX
IFS=G:\OS2\BOOT\UDF.IFS
IFS=G:\OS2\BOOT\CDFS.IFS /Q
IFS=G:\IBMLAN\NETPROG\NETWKSTA.200 /I:G:\IBMLAN /N
REM >>> Section: SET STATEMENTS <<<
SET USER_INI=G:\OS2\OS2.INI
SET SYSTEM_INI=G:\OS2\OS2SYS.INI
SET OS2_SHELL=G:\OS2\CMD.EXE
SET AUTOSTART=PROGRAMS,TASKLIST,FOLDERS,CONNECTIONS,WARPCENTER
SET RUNWORKPLACE=G:\OS2\PMSHELL.EXE
SET COMSPEC=G:\OS2\CMD.EXE
SET PATH=<------- line trunced
SET DPATH=<------- line trunced
SET PROMPT=$i[$p]
SET
HELP=G:\MPTN;G:\MPTN\HELP;G:\OS2\HELP;G:\MMOS2\HELP;G:\IBMLAN\NETPROG;G:\TCPIP\HELP;
SET GLOSSARY=G:\OS2\HELP\GLOSS;
SET IPF_KEYS=SBCS
SET LANG=de_DE_EURO
SET TZ=mez-1
SET ULSPATH=G:\LANGUAGE
SET KEYS=ON
SET BOOKSHELF=G:\IBMLAN\NETPROG;G:\OS2\BOOK;G:\MMOS2;G:\BOOKS;G:\TCPIP\HELP;
SET
SOMIR=G:\OS2\ETC\SOM.IR;G:\OS2\ETC\WPSH.IR;G:\OS2\ETC\WPDSERV.IR;G:\OS2\ETC\REXX.IR
SET SOMDDIR=G:\OS2\ETC\DSOM
SET EPMPATH=G:\OS2\APPS;
SET DEVICEFONTDISABLED=NO
SET DISPLAYTYPE=VGA
SET DMIPATH=G:\DMISL\BIN
SET CLASSPATH=<------- line trunced
SET SWING_HOME=G:\java11\Swing
SET MMBASE=G:\MMOS2;
SET DSPPATH=G:\MMOS2\DSP;
SET NCDEBUG=4000
SET NLSPATH=G:\MPTN\MSG\NLS\%N;G:\TCPIP\msg\DEDE850\%N;
SET ETC=G:\MPTN\ETC
SET IKEYMAN_HOME=G:\IBMGSK
SET IKEYMAN40_HOME=G:\IBMGSK40
SET IKEYMAN50_HOME=G:\IBMGSK50
SET NWDBPATH=G:\IBMLAN\NETPROG
SET DLSINI=G:\IBMLAN\NETPROG\NETGUI.INI
SET INIT_FILE_NAMES=netgui
SET INIT_FILE_RANGES=200
SET WPS_COMMUNICATION=YES
SET LANINSTEP=C:\CID\SERVER\ibmls
SET TCPLANG=de_DE
SET TMP=G:\TCPIP\TMP
SET VIDEO_DEVICES=VIO_SVGA
SET VIO_SVGA=DEVICE(BVHVGA,BVHSVGA)
SET GRADD_CHAINS=C1
SET VCFG_NO_DDC=TRUE
SET GREEXT=SDDGREXT
SET C1=SDDGRADD
REM >>> Section: LIBPATH <<<
LIBPATH= <------- line trunced
REM >>> Section: DEVICES <<<
device=G:\OS2\BOOT\UNICODE.SYS
DEVICE=G:\IBMCOM\PROTOCOL\LANPDD.OS2
DEVICE=G:\IBMCOM\LANMSGDD.OS2 /I:G:\IBMCOM /S
DEVICE=G:\IBMCOM\PROTMAN.OS2 /I:G:\IBMCOM
DEVICE=G:\OS2\BOOT\TESTCFG.SYS
DEVICE=G:\OS2\BOOT\DOS.SYS
DEVICE=G:\OS2\BOOT\PMDD.SYS
DEVICE=G:\OS2\BOOT\APM.SYS
DEVICE=G:\OS2\BOOT\OS2CDROM.DMD /Q
DEVICE=G:\OS2\BOOT\POINTDD.SYS
DEVICE=G:\OS2\BOOT\MOUSE.SYS
DEVICE=G:\OS2\BOOT\COM.SYS
DEVICE=G:\MMOS2\SSMDD.SYS
DEVICE=G:\MMOS2\R0STUB.SYS
DEVICE=G:\MMOS2\MIDI.SYS
DEVICE=G:\MMOS2\VCSHDD.SYS
DEVICE=G:\MPTN\PROTOCOL\SOCKETSK.SYS
DEVICE=G:\MPTN\PROTOCOL\AFOS2.SYS
DEVICE=G:\MPTN\PROTOCOL\AFINETK.SYS
DEVICE=G:\IBMCOM\PROTOCOL\NETBEUI.OS2
DEVICE=G:\IBMLAN\NETPROG\RDRHELP.200
DEVICE=G:\IBMCOM\PROTOCOL\NETBIOS.OS2
DEVICE=G:\IBMCOM\MACS\NULLNDIS.OS2
DEVICE=G:\OS2\LOG.SYS
DEVICE=G:\OS2\GRADD.SYS
DEVICE=G:\OS2\SDDHELP.SYS /Q
device=delay.sys /K :20 **** hier delay.sys **** Ort: ende config.sys ****
REM >>> Section: DOS SPECIFICS <<<
FILES=40
BREAK=OFF
SHELL=G:\OS2\MDOS\COMMAND.COM G:\OS2\MDOS
FCBS=16,8
RMSIZE=640
DOS=LOW,NOUMB
REM >>> Section: DOS DRIVERS <<<
DEVICE=G:\IBMCOM\PROTOCOL\LANVDD.OS2
DEVICE=G:\OS2\MDOS\VEMM.SYS
DEVICE=G:\OS2\MDOS\VXMS.SYS /UMB
DEVICE=G:\OS2\MDOS\VDPMI.SYS
DEVICE=G:\OS2\MDOS\VDPX.SYS
DEVICE=G:\OS2\MDOS\VWIN.SYS
DEVICE=G:\OS2\MDOS\VW32S.SYS
DEVICE=G:\OS2\MDOS\VAPM.SYS
DEVICE=G:\OS2\MDOS\VCDROM.SYS
DEVICE=G:\OS2\MDOS\VMOUSE.SYS
DEVICE=G:\OS2\MDOS\VCOM.SYS
DEVICE=G:\MPTN\BIN\VDOSTCP.VDD
DEVICE=G:\TCPIP\bin\vdostcp.sys
DEVICE=G:\OS2\MDOS\VSVGA.SYS
DEVICE=G:\OS2\MDOS\VMANWIN.SYS
REM >>> Section: RUN, START AND CALL <<<
RUN=G:\OS2\EPW.EXE
RUN=G:\OS2\SYSTEM\LVMALERT.EXE
RUN=G:\OS2\EXTENDFS.EXE *
RUN=G:\OS2\APMDAEMN.EXE
RUN=G:\OS2\SMSTART.EXE
RUN=G:\MMOS2\MIDIDMON.EXE
RUN=G:\MMOS2\QRYMMCD.EXE
CALL=G:\IBMCOM\PROTOCOL\NETBIND.EXE
RUN=G:\IBMCOM\LANMSGEX.EXE
RUN=G:\MPTN\BIN\CNTRL.EXE
CALL=G:\OS2\CMD.EXE /Q /C G:\MPTN\BIN\MPTSTART.CMD >NUL
RUN=G:\MPTN\BIN\VDOSCTL.EXE
RUN=G:\IBMLAN\NETPROG\LSDAEMON.EXE
RUN=G:\OS2\SYSTEM\LOGDAEM.EXE
RUN=G:\OS2\EPWROUT.EXE 1
REM >>> Section: Unknown <<<
set HOSTNAME=IBMT23
set ScKillFeatureEnabled=on
set restartobjects=startupfoldersonly
set ENH_SUSPEND=2
REM >>> Section: REM <<<
rem str BASEDEV=USBD.SYS /REQ:USBUHCD$,USBOHCD$,USBEHCD$ /V
rem BASEDEV=USBMSD.ADD /V /FLOPPIES:0 /REMOVABLES:1 /CDS:0
rem str DEVICE=G:\MMOS2\USBAUDIO.SYS /V
REM str DEVICE=G:\OS2\BOOT\UNICODE.SYS
REM SET DELDIR=G:\DELETE,512;H:\DELETE,512;I:\DELETE,512;K:\DELETE,512;
<------- line trunced
REM BASEDEV=USBD.SYS /REQ:USBUHCD$
REM BASEDEV=USBHID.SYS
REM BASEDEV=USBMSD.ADD
REM BASEDEV=USBCDROM.ADD
REM DEVICE=G:\OS2\BOOT\USBKBD.SYS
REM DEVICE=G:\OS2\BOOT\USBMOUSE.SYS
REM DEVICE=G:\OS2\BOOT\USBCOM.SYS
REM >>> Section: <<<
============================================================================================
I have found to diffences for the boot problem:
ClockScale=4 in my config.sys ===> set to rem for my next test
PRIORITY_DISK_IO=NO ====> set from YES ( default) to NO for my
next test
Does you or any body see somthing more?????
Doing the test and no progress.
No hardware, no config.sys relevant differences --> let us look to
the bldlevel
I have prepared a command sequence for quick checking ( ro make a
command file and run )
========================================================================
bldlevel os2krnl
bldlevel os2\boot\IBM1FLPY.ADD
bldlevel os2\boot\IBM2FLPY.ADD
bldlevel os2\boot\IBM1S506.ADD
bldlevel os2\boot\USBUHCD.SYS
bldlevel os2\boot\USBOHCD.SYS
bldlevel os2\boot\USBEHCD.SYS
bldlevel os2\boot\USBD.SYS
bldlevel os2\boot\USBHID.SYS
bldlevel os2\boot\USBMSD.ADD
bldlevel os2\boot\IBMIDECD.FLT
bldlevel os2\boot\IBMIDECD.FLT
bldlevel os2\boot\os2dasd.dmd
bldlevel os2\boot\OS2LVM.DMD
===========================================================================
Are these all the relevant components used up to the error message?
Here is the output from the command file from system G:
[G:\]bldlevel os2krnl
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:14.103#@_W4 IBM OS/2 Kernel
Vendor: IBM
Revision: 14.103
File Version: 14.103
Description: _W4 IBM OS/2 Kernel
[G:\]
[G:\]bldlevel os2\boot\IBM1FLPY.ADD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.085#@ Adapter Driver for ISA Diskette Controllers
Vendor: IBM
Revision: 10.85
File Version: 10.85
Description: Adapter Driver for ISA Diskette Controllers
[G:\]bldlevel os2\boot\IBM2FLPY.ADD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.057#@ Adapter Driver for ABIOS Diskette
Vendor: IBM
Revision: 10.57
File Version: 10.57
Description: Adapter Driver for ABIOS Diskette
[G:\]bldlevel os2\boot\IBM1S506.ADD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.161#@ Adapter Driver for ST506/IDE DASD
Vendor: IBM
Revision: 10.161
File Version: 10.161
Description: Adapter Driver for ST506/IDE DASD
[G:\]
[G:\]bldlevel os2\boot\USBUHCD.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.162#@ OS/2 USB UHCI compliant Device Driver
Vendor: IBM
Revision: 10.162
File Version: 10.162
Description: OS/2 USB UHCI compliant Device Driver
[G:\]bldlevel os2\boot\USBOHCD.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.162#@ OS/2 USB OHCI compliant Device Driver
Vendor: IBM
Revision: 10.162
File Version: 10.162
Description: OS/2 USB OHCI compliant Device Driver
[G:\]bldlevel os2\boot\USBEHCD.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.162#@ OS/2 USB EHCI compliant Device Driver
Vendor: IBM
Revision: 10.162
File Version: 10.162
Description: OS/2 USB EHCI compliant Device Driver
[G:\]bldlevel os2\boot\USBD.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.162#@ OS/2 USB Device Driver
Vendor: IBM
Revision: 10.162
File Version: 10.162
Description: OS/2 USB Device Driver
[G:\]bldlevel os2\boot\USBHID.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.162#@ OS/2 USB HID Class Driver
Vendor: IBM
Revision: 10.162
File Version: 10.162
Description: OS/2 USB HID Class Driver
[G:\]bldlevel os2\boot\USBMSD.ADD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.162#@ OS/2 USB MSD Class Adapter Driver
Vendor: IBM
Revision: 10.162
File Version: 10.162
Description: OS/2 USB MSD Class Adapter Driver
[G:\]
[G:\]bldlevel os2\boot\IBMIDECD.FLT
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.134#@ Single Filter Driver for ATAPI Disks
Vendor: IBM
Revision: 10.134
File Version: 10.134
Description: Single Filter Driver for ATAPI Disks
[G:\]bldlevel os2\boot\IBMIDECD.FLT
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.134#@ Single Filter Driver for ATAPI Disks
Vendor: IBM
Revision: 10.134
File Version: 10.134
Description: Single Filter Driver for ATAPI Disks
[G:\]
[G:\]bldlevel os2\boot\os2dasd.dmd
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:14.104#@ IBM DASD Manager
Vendor: IBM
Revision: 14.104
File Version: 14.104
Description: IBM DASD Manager
[G:\]bldlevel os2\boot\OS2LVM.DMD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:14.105#@ IBM Logical Volume Manager
Vendor: IBM
Revision: 14.105
File Version: 14.105
Description: IBM Logical Volume Manager
[G:\]
========================================================================
I have done tests with kernel:
14.103 from fixpak
14.103a from eCs 1.2r EN
14.104a from Testcase
Any difference in the build level ?
Kind regards
Rainer
>
>
Mike Luther
Rainer Stroebel wrote:
> ===> So the " or diskette drive" in the error message does not
> apply :-)
As you note :-)
>>> 2. Software
>>>
>>> We are both on MCP 2 wtih the current fix level and driver level.
>>>
>>> Both configurations boot from c: without an BMR
>>>
>>> Diffences in the Config.sys:
>>>
OK, but I'm curious about something! I've pasted a line from your
CONFIG.SYS below.
IFS=G:\OS2\HPFS.IFS /CACHE:2048 /CRECL:4 /AUTOCHECK:GHKLPSTVX
Now here is my compare line:
IFS=C:\OS2\HPFS.IFS /CACHE:2048 /CRECL:4 /AUTOCHECK:CDEFGIJK
I think I understand that LVM and whatever can allow you to boot from
other than the 'first' partition. But I wonder. Recall that I said
this was a totally OS/2 created hard drive for the Thinkpad. It has
nothing at all left of the original WIN XP whatever or the IBM restore
partition in it. It is completely native to OS/2 as set up that way
totally from the install work. Thus there is nothing in the boot
process here except the original as-created boot operation on the
logical drive which is the "C:" drive for me. This hard disk for this
Thinkpad has nothing but the raw OS/2 operation.
Since you are showing that you wish to AUTOCHECK a "G:" drive, thence
onward from there, I feel like you have something else also installed on
this 'drive' or whatever from the LVM standpoint. And I thus wonder if
what may be happening is that the boot process from the 'USB' device
process cannot proceed unless the actual boot partition is in fact the
"C:" partition and that is the actual physical first partition?
You did post that you have two approaches for this in test, though.
>
> 1. bottom up approach
>
> Minimal system to boot to the command level ( build with bootable )
>
> Systems with drive boot drive letter C:
>
> 2. bottom down approach
>
> Copy of an clean MCP2 plus latest fixlevel installation
>
> Systems with drive boot drive letter G:
>
>
> no progress up till no: "OS/ 2 is unable ..." is reproduce able
> on every test, no intermitting error :-)
Which answers why the CONFIG.SYS you posted does have that "G:"+ form in
it. But, however this works, I again have only a completely
'functional' OS/2 drive which DFSEE has CLONED, exactly, byte for byte,
into that external drive. Whatever the USB requested BIOS operation
sees when I do this is exactly what the Thinkpad sees when it is asked
to boot normally from the internal IDE drive which is the exact same
image. And that is a "C:" drive with nothing other than that initial
first partition right at the leading edge of the hard disk.
For this message I'd like to skip the compare/contrast issue(s) with
CONFIG.SYS except to appologize for the duplicate lines. Scowl.
>
> comments:
>
> Are the following lines duplicate in your config.sys?
>
> IFS=C:\OS2\BOOT\UDF.IFS
> IFS=C:\OS2\BOOT\CDFS.IFS /Q /W
> DEVICE=C:\SIO\SIO.SYS (COM1,03F8,4,-) (COM2,02F8,3)
> (COM4,INTERNET:2E8,NONE,-)
>
I'm sure so. I've asked for the ability to use a file read into the
message reply postings for Mozilla/Seamonkey such as I have in Post Road
Mailer and so on. Not furnished. So the correct work in pure text mode
for that post, when I use a clipboard copy and paste from Semi-Pro for
OS/2 or DOS or WordStar for DOS to do the real work, can produce errors
when I make them in pasting the stuff into the message.
I appologize profusely for my errors above.
>
> ============================================================================================
>
>
> I have found to diffences for the boot problem:
>
> ClockScale=4 in my config.sys ===> set to rem for my next test
>
> PRIORITY_DISK_IO=NO ====> set from YES ( default) to NO for my
> next test
>
>
> Does you or any body see somthing more?????
>
> Doing the test and no progress.
>
> No hardware, no config.sys relevant differences --> let us look to
> the bldlevel
>
> I have prepared a command sequence for quick checking ( ro make a
> command file and run )
>
OK, my next step is to do the same thing you did here and to see what
that gets us as to BLDLEVEL. I'll do this sometime today as possible
and try to make a comparator file so we can see that. That to produce a
separate message in reply to this. I'll try very hard to get it pasted
into the reply so I don't duplicate anything this time Rainer.
Thanks for your patience with me.
Rainer Stroebel
> Oliver Rick schrieb:
>
>>
>>> PS: I do have run sysinstx C: for the big floppy :-)
>>>
>>
>>
>> Copy the contents of the required sectors, maybe edit them afterwards?
>>
>> BTW, do you run SYSINSTX.COM first before you copy anything else to the
>> medium?
>>
>>
>>
> I do not remember the sequence for the "Big Floppy" -
>
> for the PRM I have first do the xcopy and than the SYSINSTX C:
> For the PRM it does work - the OS/2 kernel is loaded -
>
> May be there is a difference for the Floppy format
> good point !! - Have to test it :-)
>
Well.
I have done an the test with the PRM,
run sysinstx c: directly after the format of the drive
no change in test result --> boot --> load basedev --> "OS/2 is
unable ...... )
The test with the "big floppy" format is scheduled. :-)
Rainer
Mike Luther
Rainer Stroebel wrote:
>
> Any difference in the build level ?
>
Probably not significantly, but I have older USB drivers than you do,
per what I get when I run a batch file to collect all this. You also
have one driver installed which I do not per the run. I started this a
year or so ago and in that it is 'working' on this R40, I haven't
updated some of the drivers:
bldlevel os2krnl
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:14.104a#@_W4 IBM OS/2 Kernel
Vendor: IBM
Revision: 14.104
File Version: 14.104
Description: _W4 IBM OS/2 Kernel
---------
bldlevel os2\boot\IBM1FLPY.ADD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.085#@ Adapter Driver for ISA Diskette Controllers
Vendor: IBM
Revision: 10.85
File Version: 10.85
Description: Adapter Driver for ISA Diskette Controllers
---------
bldlevel os2\boot\IBM2FLPY.ADD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.070#@ Adapter Driver for ABIOS Diskette
Vendor: IBM
Revision: 10.70
File Version: 10.70
Description: Adapter Driver for ABIOS Diskette
---------
bldlevel os2\boot\IBM1S506.ADD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.161#@ Adapter Driver for ST506/IDE DASD
Vendor: IBM
Revision: 10.161
File Version: 10.161
Description: Adapter Driver for ST506/IDE DASD
---------
bldlevel os2\boot\IBM2FLPY.ADD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.070#@ Adapter Driver for ABIOS Diskette
Vendor: IBM
Revision: 10.70
File Version: 10.70
Description: Adapter Driver for ABIOS Diskette
---------
bldlevel os2\boot\USBUHCD.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.123#@ OS/2 USB UHCI compliant Device Driver
Vendor: IBM
Revision: 10.123
File Version: 10.123
Description: OS/2 USB UHCI compliant Device Driver
---------
bldlevel os2\boot\USBOHCD.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
unknown return code 161 from FileVerGetVersionStringFromFile
---------
bldlevel os2\boot\USBEHCD.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.123#@ OS/2 USB EHCI compliant Device Driver
Vendor: IBM
Revision: 10.123
File Version: 10.123
Description: OS/2 USB EHCI compliant Device Driver
---------
bldlevel os2\boot\USBD.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.123#@ OS/2 USB Device Driver
Vendor: IBM
Revision: 10.123
File Version: 10.123
Description: OS/2 USB Device Driver
---------
bldlevel os2\boot\USBHID.SYS
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.123#@ OS/2 USB HID Class Driver
Vendor: IBM
Revision: 10.123
File Version: 10.123
Description: OS/2 USB HID Class Driver
---------
bldlevel os2\boot\USBMSD.ADD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.145#@ OS/2 USB MSD Class Adapter Driver
Vendor: IBM
Revision: 10.145
File Version: 10.145
Description: OS/2 USB MSD Class Adapter Driver
---------
bldlevel os2\boot\IBMIDECD.FLT
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:10.134#@ Single Filter Driver for ATAPI Disks
Vendor: IBM
Revision: 10.134
File Version: 10.134
Description: Single Filter Driver for ATAPI Disks
---------
bldlevel os2\boot\os2dasd.dmd
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:14.104#@ IBM DASD Manager
Vendor: IBM
Revision: 14.104
File Version: 14.104
Description: IBM DASD Manager
---------
bldlevel os2\boot\OS2LVM.DMD
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature: @#IBM:14.100#@ IBM Logical Volume Manager
Vendor: IBM
Revision: 14.100
File Version: 14.100
Description: IBM Logical Volume Manager
---------
I sort of doubt that the newer USB drivers that you have would be
involved in this, but dunno. So I still come back to how the media is
'created' in the USB 2.0 case. In that it is a pure clone of a correct
total running object in the original hard disk case, made by Jan's DFSEE
7.1.5 or later.
Another thought comes to mind. This R40 is an IBM Depot Repair unit
with a new motherboard from early 2005 or so. According to SYSINFO 8.2,
the BIOS in this Thinkpad is Vendor: IBM PC Co., Date:07/05/04. Since
this unit wound up working, I'm not sure, but wasn't there something
about a more recent release from Lenovo which was 'supposed' to be added
to IBM Thinkpads for some reason or other as to hard disk
'compatibility' with something? The R51 and T43P which I don't have at
this facility involved here are later than this date. Without tracing
things down my memory says they are about March of 2006 in brand new
form. Thus whatever was in the BIOS versions for them will, I think, be
'standard' as of that date.
In case this turns out to be significant....
Rainer Stroebel
>
> ---------
>
> I sort of doubt that the newer USB drivers that you have would be
> involved in this, but dunno. So I still come back to how the media
> is 'created' in the USB 2.0 case. In that it is a pure clone of a
> correct total running object in the original hard disk case, made by
> Jan's DFSEE 7.1.5 or later.
I agree with you, the newer version of the USB should not be involved
with it.
I have check the build levels with my archive. I do have all the
required driver levels.
A back level test of the driver could be scheduled - as the last
desperate step to do :-)
>
> Another thought comes to mind. This R40 is an IBM Depot Repair unit
> with a new motherboard from early 2005 or so. According to SYSINFO
> 8.2, the BIOS in this Thinkpad is Vendor: IBM PC Co., Date:07/05/04.
> Since this unit wound up working, I'm not sure, but wasn't there
> something about a more recent release from Lenovo which was 'supposed'
> to be added to IBM Thinkpads for some reason or other as to hard disk
> 'compatibility' with something? The R51 and T43P which I don't have
> at this facility involved here are later than this date. Without
> tracing things down my memory says they are about March of 2006 in
> brand new form. Thus whatever was in the BIOS versions for them will,
> I think, be 'standard' as of that date.
> In case this turns out to be significant....
The latest T23 Bios and Controller Bios is of the same date range.
Bios:
Utility Version BIOS Version BIOS Date
---------------- ---------------- -------------------
1.01 1.01b (1AET38WW) 2001-07-27
.
.
.
1.17 1.17 (1AET61WW) 2003-05-29
1.18 1.18 (1AET62WW) 2004-07-06
Controller:
Released Versions ( August 2, 2004 )
-----------------
Version 1.00 (1AHT15WW) Initial Release
.
.
.
Version 1.06a (1AHT23WW) A support for new feature
Your Bios and the controller Bios of the R40 does look a bit outdated :-)
Machine types:
2681, 2682,
2683, 2896,
2898, 2899
Utility Version BIOS Version BIOS Date
---------------- ---------------- -------------------
1.00 1.00 (1OET34WW) 2002-12-16
.
.
1.15 1.15 (1OET49WW) 2004-07-05
.
.
1.27 1.27 (1OET61WW) 2006-06-29
Controller Bios
1.02 (1OHT30WW) Initial release for ThinkPad R40
.
.
1.14 (1OHT42WW) A fix for a problem
None of the entries in the Bios Change log in the read me of the bios
does relevate to our problem!
I agree with you - the probability for an bios bug related to our
problem is very low.
=============================================================================
Well sitting back and reevaluate all the fact we accumulated:
The main difference I do see at the moment is:
The R40 has a built in USB 2.0 Chip and work trough the USBEHCD.SYS
The T 23 has only a USB 1.0 Chip and work trough the USBUHCD.SYS
driver
To examine this differences does look promising to me. -:)
So i do same testing with a ThinkPad Dock II Type 2877 P/N 62P4547
1. Test with an USB 2.0 / Firewire Combo Card ( based on VIA6212L
and VT6306 ) No Name Product
- Under W2K the firewire connection is lost by copying a big
file to the external HDD after about 2 min.
- Under W2K and OS/2 the USB shows only USB 1.1 performance values.
- OS/2 HCIMonIt shows 2 additional usbuhcd and 1 usbehcd
controller -
and the additional driver were loaded at boot time.
- there is no bios hockup of the card into the boot sequence
like SCSI controllers do.
===> the USB Mass Storage device does not show up in the bios
start up selection
2. Adaptect AUA-3100B ( NEC 0720101GJ ) are not detected in PCI
Slot of the ThinkPad Dock II
- by the W2K Hardware Manager
- by OS/2 HCIMonIt
3. The Belkin F5U220qde Card with the some chip ( NEC 0720101GJ )
does show the same behavior!!
With this card i have run the PCI Test (Version 1.04vkA )
from Craig Hart / Veit Kannegieser
By the way a well done job of the authors !!!!
He is the relevant output:
Bus 9 (PCI), Device Number 0, Device Function 0
Vendor 1033h NEC Electronics Hong Kong
Device 0035h uPD9210/72010xx USB Open Host Controller
Command 0006h (Memory Access, BusMaster)
Status 8210h (Has Capabilities List, Detected Parity
Error, Medium Timing)
Revision 3Bh, Header Type 10h, Bus Latency Timer 00h
Self test 82h (Self test supported: Completion code 2h -
Failed!!)
Cache line size 56 Bytes (14 DWords)
PCI Class Unknown, type Unknown!
System IRQ 72, INT# -
New Capabilities List Present:
No 'New Capabilities' Are Currently Enabled
Well there is the next problem to investigate -:(
================================================================================
Cuurent State:
go for better USB 2.0 test hardware -:)
kind regards
Rainer
Mike Luther
Rainer Stroebel wrote:
> Hello Mike,
>
> Your Bios and the controller Bios of the R40 does look a bit
> outdated :-)
> None of the entries in the Bios Change log in the read me of the bios
> does relevate to our problem!
> I agree with you - the probability for an bios bug related to our
> problem is very low.
But you have taught me that the R40 is behind in these areas. OK,
depending on what happens on all of this and what I can find from
research on the R51 and T43P when I may have them here over the next
month or two .. I may update this R40 in this respect.
> Well sitting back and reevaluate all the fact we accumulated:
>
> The main difference I do see at the moment is:
>
> The R40 has a built in USB 2.0 Chip and work trough the USBEHCD.SYS
>
> The T 23 has only a USB 1.0 Chip and work trough the
> USBUHCD.SYS driver
>
>
> To examine this differences does look promising to me. -:)
>
Yes! And to support your thought I pass on more. During the research
to try this I had a USB 1.0 level Hard Disk controller. But this issue
here was also in the time line with the IBM Austin help on getting the
R40 to boot with the Utility Disk setup. If my memory is correct, I
could not get the 1.0 version of the controller to work with, at that
time, the R51. And in that time line, my present ADS USB 2.0 controller
would let me work with the R51, but using DFSEE from a normally booted
Desktop command line, as well as from the Desktop version of DFSEE.
As well, we discovered that some USB Floppy Diskette units will work
here and some older ones will not. I am using the TDK units which do
work. In the later discussion here of R51 and T43P units, an older USB
Floppy Diskette unit that was part of an original old Sony laptop
collection would not work. But a new purchase Sony USB Floppy Diskette
Unit works fine. My bet is that USB differences even in this hardware
are responsible.
It was at this point with the IBM Austin help that the matrix of what
needed to be added and changed in the Utility Diskette pack made from
the R40 as you see my levels first enabled their use. Since I already
owned the ADS USB 2.0 unit, it suddenly became accessible with the
improved diskette boot run. And that yielded the actual first actual
command line R51 from R40 clone after the LVM /PRMrediscover technique.
Which I have not investigated whether can also be done from an Alt F1
command line boot run as well. But probably would work.
What I think important here is that I first reverse cloned the R51 from
the USB 2.0 Drive Image from the R40 using the USB 2.0 unit and the
Utility DIsk modified set. That customized R51 which resulted from the
research, per my recall, would *NOT* work with the R51 owner's USB 1.0
Hard Drive unit! So she went out and bought a new USB Hard Drive unit,
a 2.0 unit. But we never got to test it on her R51. As the very last
step in this clone/customize research we were to clone her finished R51
to it the next morning.
That night she dumped a cup of coffee into the brand new R51. And
didn't realize you'd better yank the battery as fast as possible or else.
\ /
(!) Rabbit running from fried carrot patch ....
@ @
Result; totally destroyed project on the new R51. Back it went to IBM
Depot for expensive return. And in the enraged response to clean up the
workbench and all the clutter that was 'responsible' for the accident
... bye bye USB 1.0 older hard drive unit!
The research was transferred to her T43P for her paid-for copy of OS/2
MCP2 for it as well, during the R51 sojourn to IBM Depotville. It was
reverse cloned from the working ADS USB 2.0 Hard Disk R40 image. That's
how I first learned that the reverse clone for the R40 worked on the
T43P as well. It booted immediately from the reverse clone. Then,
suitably modified for the different audio chipset and Broadcomm NIC in
the brand new T43P ThinkPad, it was ..
I guarantee .. IMMEDIATELY cloned to her new USB 2.0 unit for backup!
And I guarantee .. then IMMEDIATELY tested for reverse clone and boot!
;)
> So i do same testing with a ThinkPad Dock II Type 2877 P/N 62P4547
Aha! She had bought a ThinkPad Dock as well for this project as well.
Because we both first thought that it would be the only way we could
clone things and so on. But when the project completed without ever
having to use it, I was behind on time after the coffee dump and double
work. So I never tested any of this with her dock. Looking back on
that I made a mistake. I should have gone ahead and learned more when I
had the hardware here.
> 1. Test with an USB 2.0 / Firewire Combo Card ( based on VIA6212L
> and VT6306 ) No Name Product
>
> ===> the USB Mass Storage device does not show up in the bios
> start up selection
> Cuurent State:
>
> go for better USB 2.0 test hardware -:)
Please let us know if the issue is 'solved' with USB 2.0 level
hardware. The reason I want to know as much as I can learn about this
is more forward. I want to try, if possible, to do a USB 2.0 Hard Drive
external reverse clone, with the appropriate substitution of Dani's hard
disk drivers if needed;
Into a brand new Intel 915GEVL motherboard unit.
If this next research thrust is successful, it will eliminate the need
to use SATA drive mobile drive trays to do build work on these units.
At present I'm using Jan's DFSEE 7.1.5 for this work with the same
technique and DRU mobile drive trays. But this would eliminate the need
for the drive trays if we could simply jam the same USB 2.0 Hard Disk
IDE unit in those cases. Then clone it to the SATA drive in them.
I have found that I can simply take a 'standard' IDE hard disk drive on
the single remaining IDE controller on the Intel 915GAVL/GEVL
motherboards. I can clone it to the internal SATA drive. And away we
go! But the USB stunt would save taking apart the box and give us a
plug-in way to do this.
Thanks!
Oliver Rick
> I have just on point to verify:
> "OS/2 is unable to operate your hard disk or diskette drive"
> It is the " or diskette drive" in the error message -
> I do have a diskette drive in the Ultra Bay 2000 - but no USB diskette
> drive attached !!!
> Dani wrote something about diskette access in Daniboot.doc
> "ENHANCEMENTS
> DaniBoot.FLT is enhanced over cd_boot.flt in many ways:
> .
> .
> .
> 7) The presence of the floppy driver IBM*FLPY.ADD is no longer
> required. "
===>> So the floppy drive has to be there when CD_BOOT.ftl has been
> used!!!!!
> Do the drivers need an USB diskette device in the hardware
> configuration?
No, *BOOT.FLT is only required if you boot from a CD/DVD with the "Floppy
Emulation". CD_BOOT.FLT needs the IBM1FLPY.ADD to enforce a physical floppy
drive, while DANIBOOT.FLT does not.
Oliver Rick
> It's not involved with the R40 CONFIG.SYS I'll post below..
You might like to compare your CONFIG.SYS to the ConfigTool database. I
think you can remove some of the entries.
> REM ** This produces plug-in USB card reader options in Connections
> REM ** If $ICPMOS2 isn't in CONFIG.SYS OS2PCAARD has to be last line
> REM ** in CONFIG.SYS and if it is then $ICPMOS2 has to be the last line
> REM BASEDEV=OS2PCARD.DMD
> DEVICE=C:\OS2\$ICPMOS2.SYS /G
> Notice also that comment at the end of the CONFIG.SYS file relative to
> the last line needed in the CONFIG.SYS for Thinkpads. Apparently it is
> on some; not on others. It *IS* here for things to work right. And at
> this point each upgrade to the SNAP video driver installs moves the SNAP
> code below it. As well some other IBM install stuff does too. I
> frequently have to go in and hand correct that last item order in these
> Thinkpad files to keep things working correctly.
The requirement listed above is just to make things easy. BASEDEVs always
load before DEVICES. DMDs are loaded as last BASEDEVs. What is meant is to
load OS2PCARD.DMD after OS2DASD.DMD. Since OS2LVM.DMD has to be loaded
immediately after OS2DASD.DMD, OS2PCARD.DMD has to be loaded after this
then.
Similar applies to $ICPMOS2.SYS. It must be the last DEVICE statement,
since they are loaded in order of appearance (where IFS counts as DEVICE
too). Actually, $ICPMOS2.SYS must be loaded after (a) certain driver(s),
but until someone runs through a move up/reboot cycle to find it out, it
should be placed after all drivers for devices which are affected by power
management. Maybe it just has to be placed after APM.SYS to work correctly.
Rainer Stroebel
----snip
--- snip
Hello Mike
Here are "intermediate" test result from the way to test with USB 2.0
Hardware.
I have used the following hardware:
CF Card 16 MB Sanyo DSM-C16G
CF Card 1 GB Kingston Elite pro 50x CF/1GB-s 9930407-009.A00LF
and the USB 1.1 CF Reader
CR-T2-UC CompactFlash Pocket Reader ( Vendor: Microdia product:
0x1050 displayed by USB RESMRG )
This combination does work without problems.
Now fro the tests a USB 2.0 CF Card reader is required.
Looking it my hardware resources a
Kingston FCR-HS2/ATA USB 2.0 ATA Card Reader ( Vendor:
Microdia product: 0x1050 )
in combination with an PC Card Adapter for IBM Microdrive
should do the job - a nice multi usable device
using the CF cards build with the Microdia usb 1.1 Procket Reader
with the T23
The combination CF Card Kingston Elite pro 50 x and the Kingston
FCR-HS2/ATA
does not work reliable at the T23 USB 1.0 port.
Comparing the contents the CF card with the original file on the
HDD come up with bit errors
doing the same test with the USB 1.0 CF reader or the slower 16
MB card --> no errors.
Conclusion: The "older" Kingston FCR-HS2/ATA reader does have timing
problems
with the new fast CF memory.
It is reasonable: At time this device was
developed no fast CF Cards exist !!!
Now going to an machine with USB 2.0 ports.
Frist test hardware :
A consumer PC : model MD8800 from ADLI/Medion ( a
German local band )
Enter the bios and modified the start up sequence - place
the USB before the internal HHD
Doing the test --> no boot from USB - the boot device USB
is ingored still booting from HDD
doing a quick google search shows - it is a known problem
for the MD8800
===> have to look for an USB 2.0 CF Reader
and go to test with ThinkPads -:) - USB BIOS start up
will work on this machines!!!
kind regards from the "testlab"
Rainer
Oliver Rick
> kind regards from the "testlab"
I just found the following paragraph in DaniDASD.DOC:
-+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+-
- it can overcome boot problems which result in "can't operate hard
disk" messages. This is common if OS/2 is booted without proper help
of OS/2 Bootmanager which is required to notify OS/2 of the
*correct* boot drive letter. If there is no boot manager with this
capability available (e.g. when booting from removable media), or
the boot manager fails to figure out the correct boot drive letter,
you can tell DaniDASD.DMD the correct one by means of the option
/BD:<drive letter>
which will override the boot drive letter assignment from the first
boot stage.
-+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+-
So no luck with LVM and without Boot Manager?
Mike Luther
Rainer Stroebel wrote:
>
> Hello Mike
>
> Here are "intermediate" test result from the way to test with USB
> 2.0 Hardware.
>
> Now going to an machine with USB 2.0 ports.
>
> Frist test hardware :
>
> A consumer PC : model MD8800 from ADLI/Medion ( a
> German local band )
>
> Enter the bios and modified the start up sequence - place
> the USB before the internal HHD
>
> Doing the test --> no boot from USB - the boot device USB
> is ingored still booting from HDD
> doing a quick google search shows - it is a
> known problem for the MD8800
>
>
> ===> have to look for an USB 2.0 CF Reader
> and go to test with ThinkPads -:) - USB BIOS start
> up will work on this machines!!!
>
>
> kind regards from the "testlab"
>
> Rainer
Today another Intel 915GEVL new box arrived to be used for an update box
on another of the older ones at a client location. So I did some more
'research' hoping to help out here. The Intel 915GEVL can take a
working IDE hard disk, connected to the single IDE cable it has. You
can then take an OS/2 Utility Disk floppy set made from a functioning
OS/2 IDE box and then boot the Intel 915GEVL from the floppy. You can
then use Jan's DFSEE 7.1.5 or later to clone that completely functional
IDE drive into a new SATA hard disk in these boxes. At that point you
can then simply boot the Intel 915GEVL with the cloned SATA hard disk
just fine. That's how I'm updating older IDE boxes to the newer
technology and it works fine.
But .. to be honest, without looking as I post this, I can't recall if I
am using Dani's driver instead of the IBM 506 IDE driver. I sort of
think so. Which may well be an interesting part of the failure I'm at
first seeing here.
Anyway, just for foolish fun, I decided to try to use the IBM R40 cloned
USB 2.0 Hard Disk image from the R40. I decided to configure the Intel
915GEVL BIOS 'properly' to see the USB devices. I then jammed in the
R40 Utility Disk floppy run set which will clone and reverse clone the
ThinkPad's. Suprise! Everything works fine until we see the "V"
(erbose) report of the 506 IDE driver run, which works fine with the
Thinkpads. Oops!
"OS/2 cannot operate your hard disk ..."
Hmmmm. Well there is another stunt that also works on some of these
things. You can take the DFSEE for DOS part of Jan's toolset. You can
boot many of these boxes from that floppy driver to the DOS version.
You can actually clone from an IDE drive to and IDE drive with that, but
I had never tried to use it with a USB 2.0 Hard Disk device, as well as
with an SATA device as well.
Interesting result. I can boot the 915GEVL from the DOS floppy. I can
run DFSDOS then. It will see both the USB 2.0 Hard Disk plugged in
device as well as the SATA device, according to what is on the screen.
I can pick a device as a DISK with which to work. I can then go through
the DFSEE menu to select an operation for cloning the very visible
devices one would normally see with DFSEE in the report game. And when
I finish and the clone operation should start, then no go.
The USB 2.0 external device sits there VERY slowly and winks about once
a minute while something is happening with it. And no data is ever
moved to the SATA drive in the new 915GEVL box.
Now there are two more research steps to try. I can move the IBM 506
driver on another test set of the ThinkPad floppies to Dani's even brand
new release today. I can see if that will affect this long shot test.
If I can either see Dani's driver 'find' the whatever hard disk from the
floppy boot, and auto start things, or use, as Oliver Rick suggests in
his response, the command line code to force whatever, then maybe I can
then perform this long shot reverse clone test from the R40 image. Just
for whatever.
The second more complicated test next would be to do working SATA drive
clone using the as working Intel box utility floppy disk sets, but clone
the SATA drive to the external USB 2.0 Hard Disk unit. Then, using that
same set of floppy diskettes which will clone SATA to SATA or IDE to
SATA images, I'd test to see if a USB 2.0 Hard Disk image could be used
to do this. If it would, then that would eliminate the DRU mobile drive
tray game. It would make it far simpler to clone work stations by just
jamming in the USB 2.0 image into the new case. And away we go .. But
this test will take longer and will also mean I have to get another new
USB 2.0 Hard Drive unit for pure research. $$$ ;(
I'll consider testing both of these during the next couple days and post
the results here for people. But it may be a couple days before I can
do all this as things stand.
Thanks for your help and Oliver Rick's suggestions too.
Mike Luther
Oliver Rick wrote:
> On Thu, 17 Aug 2006 Mike Luther wrote
>
>> It's not involved with the R40 CONFIG.SYS I'll post below..
>>
>
> You might like to compare your CONFIG.SYS to the ConfigTool database. I
> think you can remove some of the entries.
>
Yes, you are correct. The remarking is force of habit.
> The requirement listed above is just to make things easy. BASEDEVs always
> load before DEVICES. DMDs are loaded as last BASEDEVs. What is meant is to
> load OS2PCARD.DMD after OS2DASD.DMD. Since OS2LVM.DMD has to be loaded
> immediately after OS2DASD.DMD, OS2PCARD.DMD has to be loaded after this
> then.
>
>
OK. But can you teach me some more about exactly what must happen in
relation to the sub-position order of the BASEDEV and DEVICE issues?
Yes, I think I understand the issue about BASEDEV order for SCSI vs. IDE
drives in the CONFIG.SYS. That when you are attempting to use SCSI and
IDE drives both at the same time on a given system.
But how does this work out with SATA drives when you attempt to mix them
with USB 2.0 Hard Disk external units?
The reason I'm asking you is that the I didn't think it would work to
just jam it in and try the R40 Utility Disk set the Intel 915GEVL
system. As reported that produced the error 'OS2 cannot operate your
hard drive ..' I've never even tried the USB external hard drive
operation on any of these 915 GAVL/GEVL systems. I then built a
completely new OS/2 Utility Disk test for the 915 systems. I did
exactly the same technique to modify those floppy diskettes that we
worked out with IBM Austin on the Thinkpads.
Now I have a complete as suggested and modified Utility Disk set that
will completely boot a functional 915GEVL unit to the command line.
And, incidentally, that is with Dani's replacement driver for
IBM1S506.ADD, the release just previous to her brand new level toolset
of a couple days ago. It has been working just perfectly with the SATA
operations and Seagate drives in in these 915 units. It is also capable
of seeing and working with both the SATA and the earlier pure IDE hard
disk drives on the single older IDE cable in these units. That is how I
first moved the older IDE technology to the SATA units for update
purposes. That with her driver and DFSEE 7.1.5 or higher plus the DRU
Mobile Drive trays was the mirror sort of technique that I expanded from
the research on thee ThinkPads for the clone project.
But now, with the same general toolset components which work with the
ThinkPad IDE drive and USB 2.0 Hard Disk external units, things are not
working yet on the 915 Inter units. I'm sure that the Intel 915GEVL in
research here does 'see' and can attempt to access the same USB 2.0 Hard
Drive unit that has an image of the R40 on it. That for test purposes
only, but we have to be able to see and work with both the SATA and the
USB 2.0 units to go forward.
The complete new Utility Disk test set now goes all the way through the
boot run. And that boot run does show the little activity light wink on
that external USB 2.0 unit during the boot. Dani's driver in verbose
mode shows the correct information for the SATA drive in the unit. It
also shows the information on the second SATA drive from that diskette
boot run if you have one in the DRU mobile drive tray.
But nothing I can do will let me see the USB 2.0 external unit which
does report as even assignable to a boot drive in the 915 BIOS. And
despite that it 'winks' at me twice during the boot run, nothing I can
do with the same LVM /rediscoverPRM technique that works on the
ThinkPads will gain me access to the USB 2.0 device. I have, in this
case, moved forward to the later driver levels for the IDE and USB code
that has been released by IBM on Passport Advantage. That although it
seems that the USB drivers which were moved from 09-10-2005 are exactly
the same byte length as the 12-21-2005 release that was later made
available for download.
Can you help teach me what I may have done wrong here?
1.) Will you explain to me exactly what driver tools are actually used
to enable the working USB 2.0 Hard Disk external plug in unit in
the ThinkPads. That don't work here in this case?
2.) Can you teach me what settings I might have to hand alter to get
the IBM and Dani's driver to 'see' the light here? Grin!
3.) Might this issue be one sort of like the SCSI vs. IDE loader order
in your CONFIG.SYS such that the USB drivers including whatever
is needed in a specific order for what you will post me in reply
to my #1 question above, mean since we are working with SATA and
USB IDE, that this is where we are failing now?
> Similar applies to $ICPMOS2.SYS. It must be the last DEVICE statement,
> since they are loaded in order of appearance (where IFS counts as DEVICE
> too). Actually, $ICPMOS2.SYS must be loaded after (a) certain driver(s),
> but until someone runs through a move up/reboot cycle to find it out, it
> should be placed after all drivers for devices which are affected by power
> management. Maybe it just has to be placed after APM.SYS to work correctly.
>
> /Olli/
>
At some point after this research thrust, I'll try to get a look at what
happens to this in relation to the ThinkPads.
Thank you for any help you may be able to give me. Or anyone else here
as well!
Rainer Stroebel
---> snip
> Today another Intel 915GEVL new box arrived to be used for an update
> box on another of the older ones at a client location. So I did some
> more 'research' hoping to help out here. The Intel 915GEVL can take a
> working IDE hard disk, connected to the single IDE cable it has. You
> can then take an OS/2 Utility Disk floppy set made from a functioning
> OS/2 IDE box and then boot the Intel 915GEVL from the floppy. You can
> then use Jan's DFSEE 7.1.5 or later to clone that completely
> functional IDE drive into a new SATA hard disk in these boxes. At
> that point you can then simply boot the Intel 915GEVL with the cloned
> SATA hard disk just fine. That's how I'm updating older IDE boxes to
> the newer technology and it works fine.
>
Hello Mike,
can you do the following test with the Intel board?
1. Try to boot from your USB 2.0 external disk -
the R40 clone should do for the test
The USB Floppy Drive Hardware has been arrived
so I can replicate some of your scenarios - the USB 2.0 CF reader is
in the pipe line.
kind regards
Rainer
Rainer Stroebel
Hallo Olli,
thanks for coming up with this fact.
my understanding is
DaniDASD.dmd gives the user more freedom with drive letter
assignment in the "pre LVM world".
It gives some of the LVM functions for non LVM systems by the user of
DaniDASD.dmd and a proper
parm usage.
" This is common if OS/2 is booted without proper help
of OS/2 Bootmanager which is required to notify OS/2 of the
*correct* boot drive letter."
Does this statement is valid for the current OS2LVM.DMD without the usage of BMR?
It cant be general
-- > Mike reported success in booting from USB 2.0 PRM with his ThinkPad
R40 cloned configuration.
Mike Luther
Rainer Stroebel wrote:
>
> my understanding is
> DaniDASD.dmd gives the user more freedom with drive letter
> assignment in the "pre LVM world".
>
> It gives some of the LVM functions for non LVM systems by the user of
> DaniDASD.dmd and a proper
> parm usage.
>
>
> " This is common if OS/2 is booted without proper help
> of OS/2 Bootmanager which is required to notify OS/2 of the
> *correct* boot drive letter."
>
> Does this statement is valid for the current OS2LVM.DMD without the
> usage of BMR?
>
> It cant be general
> -- > Mike reported success in booting from USB 2.0 PRM with his
> ThinkPad R40 cloned configuration.
>
Last night I learned enough to finish the job of reverse cloning a USB
2.0 Hard Disk to a brand new Intel 915GEVL. Remove the USB stuff, push
button, instant running 915GEVL totally up and on line. And .. grin ..
the R40 USB 2.0 clone image was the one reverse cloned into the
915GEVL. I have one remaining problem in this process which is
highlighted by the above. But that one 'problem' has lots of parts, so
the learning process will help understand what is above and what I
haven't quite solved yet!
First, as reported, the Utility Disk set for the R40 will not boot the
new virgin Intel 915GEVL unit. I do have a complete different set for a
finished Intel 915GEVL which will boot to command prompt. However that
set had none of the required USB drivers on Disk #1 that are needed to
work with USB devices after a floppy boot.
So I modified a duplicated set of the 915GEVL floppies to include the
same changes as were in the R40 working set. Surprise! Instant Trap
"D" when the modified set was run. You may add the little file
ALTF2ON.$$$ to the floppy diskette boot set to see the driver load at
the bottom of the screen as well. OK, the Trap "D" hits just after the
OS2DASD.DMD is run. And that trap is produced here by the LVM load
attempted just after that. Why?
Interesting answer. That same set of modified 915GEVL floppies would
boot a fully functional working unit just fine! No Trap "D" pure
success, And the functional 915GEVL also, with the USB drivers active
has the USB floppy disk present in the CONNECTIONS folder. And after
boot, you can punch in the R40 USB 2.0 unit, do the LVM /rediscoverPRM
and see it. So we know USB is working for all of this. But watch what
happens in the next research step!
However, I happened to have left the USB 2.0 R40 image plugged into the
box during the the next floppy diskette boot attempt with the modified
'correct' files for the clone or the reverse clone process.
BLAM! "OS2 is unable to operate your hard disk..." Took me a while to
realize I'd left the USB 2.0 unit jammed into the box. When I yanked it
prior to the boot run, away we go through the floopy boot just fine.
And there is more to this part of the problem.
If I leave the USB 2.0 unit jammed into the box during a normal boot to
the finished unit, another surprise. The USB floppy object for the
floppy requested in the USB load as per /floppies:1 works fine! But the
instant you attempt to click on the 'Refresh removable media' object in
the CONNECTIONS folder, the object loader for LVM /refreshPRM jams solid
on. You never even can close it down with a KILLFEATURE operation. The
only hope is to reboot the whole box and then the only way to even get
the reboot finished during the closedown which can't kill the jammed LVM
operation, is a second <CTRL ALT DEL> which will finally let the reboot
work!
In other words, at least for the 915GEVL units, there is a problem with
LVM and a USB 2.0 Disk Drive unit plugged into the box at the time of
the boot run! Further, as long as the USB unit is NOT plugged into the
box during the use of the floppy diskettes with the USB system items
added, the completed 915GEVL box will finish boot with them. So maybe
we do have individual hardware unit compatibility problems here, just
like this or that PRM unit which will or will not work? Shudder.
But wait! There's more! Recall that a non-USB modified set of floppy
diskettes will boot a virgin 915GEVL box even with a completely brand
new never touched SATA hard drive. And recall that with the floppy
diskettes that are modified with the USB system items added produce the
Trap "D" on attempting to load the LVM tool? Why?
Answer. With the USB system items added, that Trap "D" only happens if
the as-installed hard disk in the unit is completely virgin and has
never been prepared for use. If I simply use DFSOS2 or DFSDOS and
'prepare' it for use and format it, or alternatively, use the non-USB
enabled floppy disks and boot to an OS/2 command line, then use the IBM
LVM tool to create a bootable "C" partition and format it in HPFS ..
1.) The USB modified floppy disk set boots up fine!
2.) And without the USB 2.0 unit plugged into the box during the boot
run, there is no "OS2 is unable to operate your hard disk."
Which at that point, from the booted system on the almost virgin box,
grin, you jam in the R40 USB 2.0 image unit. You do the LVM
/rediscoverPRM. The players report for duty at the show that night. Up
comes the ringmaster DFSOS2. He whips all the performers into their
right places on that 'near virgin' working disk. Then;
3.) After almost TWO HOURS of whirling dervish later, we are ready
for the curtain call!
The audience fidgets nervously in their seats after the two hour wait
for the grand performance. Up goes the reboot curtain;
PRESTO! Whole working 915GEVL with audio LAN and everything!
Sorry folks but I'm emotional. Sorta like New Year's Eve in 1969 or so
when as a guest of a client in Newton, Iowa, at the Holiday Inn, a
WONDERFUL rising star singer Don McClain, at a quarter to Midnight, said
to us;
"Folks, I'd like to try a new song on you I'm thinking of recording.
I've never done it in public. Based on your reaction I'll make
the decision to record it or not."
Then for the first time ever in public we heard the words,
"Starry starry night! Amber fields of violet haze ..."
And the song "Vincent" was given to the public.
Now if someone will help me unravel how to keep the USB enabled floppy
diskette set from TRAP "D" with a completely barren virgin hard disk in
that box so we don't have to use a double dip boot, I can finish this
deal. As well, I can then maybe do a test as Reiner wants to boot the
Intel 915GEVL from a plugged in USB 2.0 unit.
It begins to look like my original question about load order and perhaps
parameters of things is really critical here. And I can tell you the
the whole show now works with either the latest IBM1S506 or Dani's
drivers, either one. And at this point there is no solution I've found
so far for what I see.
Help? Thanks!
Oliver Rick
>> You might like to compare your CONFIG.SYS to the ConfigTool database. I
>> think you can remove some of the entries.
> Yes, you are correct. The remarking is force of habit.
I didn't think of the remarked entries. :-)
Besides the other entries you don't need most likely (as far as I can
assume), the CONFIG.SYS looks like it's based on the one created by the
OS/2 installation before the first reboot.
> But how does this work out with SATA drives when you attempt to mix them
> with USB 2.0 Hard Disk external units?
The rule is all fixed drives before removables.
> The reason I'm asking you is that the I didn't think it would work to
> just jam it in and try the R40 Utility Disk set the Intel 915GEVL
> system. As reported that produced the error 'OS2 cannot operate your
> hard drive ..' I've never even tried the USB external hard drive
> operation on any of these 915 GAVL/GEVL systems. I then built a
> completely new OS/2 Utility Disk test for the 915 systems.
The drivers involved are the same. The PATA/SATA drivers both support the
chipsets' hostadapters (ICH4-M and ICH6) and the USB host controllers (same
ICHs) haven't changed: UHCI (1.0) + EHCI (2.0).
> But nothing I can do will let me see the USB 2.0 external unit which
> does report as even assignable to a boot drive in the 915 BIOS. And
> despite that it 'winks' at me twice during the boot run, nothing I can
> do with the same LVM /rediscoverPRM technique that works on the
> ThinkPads will gain me access to the USB 2.0 device. I have, in this
> case, moved forward to the later driver levels for the IDE and USB code
> that has been released by IBM on Passport Advantage. That although it
> seems that the USB drivers which were moved from 09-10-2005 are exactly
> the same byte length as the 12-21-2005 release that was later made
> available for download.
Can you access the USB HDD after you have booted from the SATA HDD?
Do I understand you correctly: You have one internal fixed SATA HDD plus
one internal SATA HDD in a removable bay plus one external USB HDD?
DRU = Drive Replacement Unit?
> 1.) Will you explain to me exactly what driver tools are actually used
> to enable the working USB 2.0 Hard Disk external plug in unit in
> the ThinkPads. That don't work here in this case?
BASEDEV=USBEHCD.SYS
for one USB EHCI (2.0) host controller
BASEDEV=USBD.SYS
for general USB support. The switch /REQ:USBUHCD$,USBOHCD$,USBEHCD$ makes
no difference, maybe /I13 does.
BASEDEV=USBMSD.ADD
for USB mass storage support.
If you use an USB HDD > 40 GB and use an OS2DASD.DMD < 14.100 or if you
have more than one partition on the USB HDD please take a look at
USBMSD.TXT.
Due to my experience with PCs, I don't like/trust automatic/default
settings, so you might add /FLOPPIES:0 /REMOVABLES:1 to USBMSD.ADD with the
external HDD being the only USB storage device attached.
> 2.) Can you teach me what settings I might have to hand alter to get
> the IBM and Dani's driver to 'see' the light here? Grin!
See above, nothing required for *S506.ADD, because they're not involved in
USB.
> 3.) Might this issue be one sort of like the SCSI vs. IDE loader order
> in your CONFIG.SYS such that the USB drivers including whatever
> is needed in a specific order for what you will post me in reply
> to my #1 question above, mean since we are working with SATA and
> USB IDE, that this is where we are failing now?
OS2DASD.DMD usually serves the fixed drives first, then the PRMs. It's the
BIOS which shifts the PRM before the fixed disks. So maybe *S506.ADD has
to be placed after USBMSD.ADD. I don't know what happens internally
regarding the drive numbering.
> At some point after this research thrust, I'll try to get a look at what
> happens to this in relation to the ThinkPads.
The TPs seem to have a better (cleaner) implementation of the BIOS. In
your case almost everything is the same: the external HDD in its case, the
ATA-to-USB bridge chip in the case, the USB cable from the case to the USB
plug, the USB Enhanced Host Controller Interface in the I/O Controller Hub,
the USB driver stack, the OS. Only the BIOS is a different one.
I looked up an older mainboard test in a German computer magazine. Intel
boards support booting from USB floppy, optical drive, stick.
Hardware Voodoo? >;->
Finally, I like to add a revised article. Ignore the SCSI entries (boot
sequence should be "USB,A,C" or similar in your case). I'm not sure if the
IRQ of the USB host controller might be involved here. A change is
possible by rearranging cards in the PCI slots (PCIe as well?) or limiting,
exchanging, extending the number of IRQs in the mainboard BIOS ("PCI/ISA
PnP" vs. "Legacy" or how it may be named in these days, maybe with PCIe
these options have been dropped).
System hangs (sometimes followed by a trap) during installation or
reboot with messages
"Loading, please wait. ..." or
"OS/2 is unable to operate your hard disk or diskette drive."
Usually a cold boot is required.
OS/2 shows this issue every now and then during installation or reboots
*since version 1.x*! There were several fixes in the past that are
included in later FixPaks and OS/2 versions. It is not gererally
reproduceable (usually limited to a certain PC model or *clone*) and is
related to the *hardware* in most cases. Basically, it's an
initialization problem involving hardware interrupt 1 (keyboard
controller) sometimes, interrupt 6 (floppy controller), interrupt x/14
(storage hostadapter, 14 standard for ATA, x for SCSI etc.), maybe
shared hardware interrupts or conflict with IRQ of storage hostadapter
as well, software interrupt 13 (direct access storage devices). When
experiencing the hang the LED on the floppy drive might stay on. Other
reasons are either incorrect drivers (loaded) or drive letters have
been remapped.
Workarounds for problem related to *hardware*:
- disable busmastering for ATA(PI) (add /A:x /!BM after IBM1S506.ADD
resp. DaniS506.ADD in ?:\CONFIG.SYS)
- disable synchronous mode for CD/DVD drive resp. change sync rate to
async (SCSI BIOS)
- disable sync negotiation for CD/DVD drive (SCSI BIOS)
- disable disconnect feature for CD/DVD drive (SCSI BIOS)
- flash other (newer) version of SCSI hostadapter BIOS
- flash other (newer) version of mainboard BIOS
- set boot sequence to "A,C,SCSI" (mainboard BIOS)
- set option "OS/2 > 64 MB" to "No" (mainboard BIOS)
- set option "PnP OS" to "No" (mainboard BIOS)
- disable Power Management resp. change it from ACPI to APM (mainboard
BIOS)
- disable APIC mode (mainboard BIOS)
- disable hardware detection (Alt-F1, F6 at "#### OS/2")
- comment out unused (other hostadapter) drivers in ?:\CONFIG.SYS
- comment out unused snoopers in ?:\OS2\BOOT\SNOOP.LST
- change hardware interrupt for the (SCSI etc.) storage hostadapter
- use other driver revision
- use other (newer) driver revision
- disable memory shadowing (mainboard BIOS)
- disable L2 caching (mainboard BIOS)
- disable asynchronous RAM access (mainboard BIOS)
- disable disk caching
- exchange RAM modules (faster ones for old machines)
- turn off turbo switch (for old machines)
Workarounds for problem related to *software*:
- use correct driver for hostadapter
- use other (newer) driver revision
- comment out unused (other hostadapter) drivers in ?:\CONFIG.SYS
- comment out unused snoopers in ?:\OS2\BOOT\SNOOP.LST
- uncompress a compressed DOS partition, if OS/2 should be installed
onto it
The disabled features can be reenabled after a successful installation.
The original article was posted on Aug 11, 2001 with the subject
SCSI/"Loading ..."/"OS/2 is unable ..." issue summary
to news://news.ecomstation.nl/ecomstation.support.install
It referred to eComStation 1.0x installation CD-ROM failing to boot.
(eCS 1.0x uses the "El Torito" "floppy emulation" mode to boot while
eCS 1.1 and higher use the "no emulation" mode like Windows.)
Revised on Mar 2, 2002
Revised on Aug 24, 2006
Rainer Stroebel
> Almost there folks and this really seems important.
>
> Now if someone will help me unravel how to keep the USB enabled floppy
> diskette set from TRAP "D" with a completely barren virgin hard disk
> in that box so we don't have to use a double dip boot, I can finish
> this deal. As well, I can then maybe do a test as Reiner wants to
> boot the Intel 915GEVL from a plugged in USB 2.0 unit.
>
>
Hello Mike,
can you reproduce the error with an blank, virgin disk in the ThinkPad
R40?
Have you ever used the usb boot diskettes with an virgin disk in the
thinkpad primay drive and an USB 2.0 HDD attached?
kind regards
Rainer
Mike Luther
Oliver Rick wrote
>
> The rule is all fixed drives before removables.
>
Yes..
> The drivers involved are the same. The PATA/SATA drivers both support the
> chipsets' hostadapters (ICH4-M and ICH6) and the USB host controllers (same
> ICHs) haven't changed: UHCI (1.0) + EHCI (2.0).
>
>
Yes.
>> Can you access the USB HDD after you have booted from the SATA HDD?
>>
>> Do I understand you correctly: You have one internal fixed SATA HDD plus
>> one internal SATA HDD in a removable bay plus one external USB HDD?
>> DRU = Drive Replacement Unit?
>>
No .. in the case of the NEW virgin 915GEVL I'm researching, all it has
is the as installed SATA Seagate 70GB drive. And, yes, things are working.
> BASEDEV=USBEHCD.SYS
> for one USB EHCI (2.0) host controller
>
> BASEDEV=USBD.SYS
> for general USB support. The switch /REQ:USBUHCD$,USBOHCD$,USBEHCD$ makes
> no difference, maybe /I13 does.
>
> BASEDEV=USBMSD.ADD
> for USB mass storage support.
> If you use an USB HDD > 40 GB and use an OS2DASD.DMD < 14.100 or if you
> have more than one partition on the USB HDD please take a look at
> USBMSD.TXT.
> Due to my experience with PCs, I don't like/trust automatic/default
> settings, so you might add /FLOPPIES:0 /REMOVABLES:1 to USBMSD.ADD with the
> external HDD being the only USB storage device attached.
>
>
And yes, right on target too, but the above is EXACTLY what was going
wrong. No, I wasn't trusting the automatic/default settings, as you
don't like/trust either. But I had it WRONG as to what was installed.
I was trying to use the settings for an as working box with /FLOPPIES:1
/REMOVABLES:2 in the floppy diskette run as well. And that in
relationship to the issue where the only way we can get a floppy
diskette boot on the ThinkPad is to use the USB floppy unit as well!
So in the case of the ThinkPad Utility Disk set, that works. But your
coaching is right on the mark.
Since the Intel 915GAVL/GEVL units have a floppy disk drive in them and
a controller for the floppy drive unit, if it is enabled, it is of
course used for the floppy boot run. And no, you cannot use another
floppy diskette boot object for the LVM operations, in that the hardware
Trap "D" issue is spawned by the two floppy disk devices! Aha! If you
don't use the normal internal floppy drive unit and disable it in the
BIOS, just like you would see with the ThinkPad operations, then use the
USB floppy .. it works. But if you have to modify your logic for the
single removable choice operation if the internal BIOS enabled floppy
drive is in use.
Problem solved.
> OS2DASD.DMD usually serves the fixed drives first, then the PRMs. It's the
> BIOS which shifts the PRM before the fixed disks. So maybe *S506.ADD has
> to be placed after USBMSD.ADD. I don't know what happens internally
> regarding the drive numbering.
>
No, in this case the *S506.ADD still goes first. And when you use the
/V option of the MSD operation, you'll see it pop in there after the
*S506.ADD report if it too is using the /V option.
> The TPs seem to have a better (cleaner) implementation of the BIOS. In
> your case almost everything is the same: the external HDD in its case, the
> ATA-to-USB bridge chip in the case, the USB cable from the case to the USB
> plug, the USB Enhanced Host Controller Interface in the I/O Controller Hub,
> the USB driver stack, the OS. Only the BIOS is a different one.
>
> I looked up an older mainboard test in a German computer magazine. Intel
> boards support booting from USB floppy, optical drive, stick.
>
> Hardware Voodoo? >;->
>
And now for the final part of the report. OK, now that we have this all
figured out yes I can at least START the full OS/2 boot to a completely
raw out of the box for the Intel 915GEVL research case. In this case, I
enable USB boot in the BIOS. I disable the totally useless SATA new
drive and call the USB 2.0 Hard Disk unit the boot choice. In this case
I jam it in the front panel USB slot and punch up the power. Away we go
.. up comes the OS/2 screen and we get all the way through it until just
after the /V response for the R40 clone image. And then here comes the
"OS/2 cannot operate your message.."
Think about it. Of course not. Why? Because we have the same conflict
in the as installed /FLOPPY:1 /REMOVABLES:2 logic in the CONFIG.SYS
operation. It has to have that to work with it's floppy boot. And with
your USB floppy jammed into the ThinkPad, you have properly conformed
your boot logic for the operation that way. Which is also the LVM Trap
"D" creation for the Intel 915GEVL where the normal floppy is enabled in
the case and .. you have no USB floppy jammed into it as well!
I won't say that stupidity has its own reward, but let's just say
uninformaty has its own reward.
> Finally, I like to add a revised article. Ignore the SCSI entries (boot
> sequence should be "USB,A,C" or similar in your case). I'm not sure if the
> IRQ of the USB host controller might be involved here. A change is
> possible by rearranging cards in the PCI slots (PCIe as well?) or limiting,
> exchanging, extending the number of IRQs in the mainboard BIOS ("PCI/ISA
> PnP" vs. "Legacy" or how it may be named in these days, maybe with PCIe
> these options have been dropped).
>
>
Not the same quite in the brand new Intel 915 motherboards. Same sort
of logic, but the boot devices, permissions, and order of access are all
set by their actual names and 'seen' device types in the BIOS setup.
For research purposes I'm setting the floppy as first choice, the
DVD/CD-ROM as second, and in the case of the pure USB 2.0 external
device boot, I'm telling the BIOS to only permit that device.
And away we go.
The remainder of your advice and training are snipped here not because
they are to be ignored. Simply to focus on the solution in this case as
Reiner has more issues with us all here. I'm trying to focus toward the
original thread PRM discussion as well.
Thank you so much for your time and effort to help.
Mike Luther
Rainer Stroebel wrote:
>
> can you do the following test with the Intel board?
>
> 1. Try to boot from your USB 2.0 external disk -
> the R40 clone should do for the test
>
Yes! Please also read the long post to Oliver for the details on how
this was gotten to work. Courtesy of his teaching and work I can now
actually at least start the USB 2.0 Hard Disk boot process for a brand
new Intel 915GEVL unit with a totally untouched new Seagate SATA hard
drive in it. Further, by using the right combination of the CONFIG.SYS
USB parameters in it, including *NOT* enabling a FLOPPY unit and only
one REMOVABLE unit, I can actually use the previously made OS/2 Utility
Disk set for the 915GAVL/GEVL diskettes.
I can now boot directly to a fully accesable 'new' SATA virgin 915 drive
box. I can then jam in the USB 2.0 Hard Disk unit into the front panel
USB slot. I can the do the LVM /rediscoverPRM on the unit. I can the
see the virgin never used SATA drive and the full USB 2.0 R40 image with
DFSEE. And although in this case not actually done, my bet is I can
either romp the R40 image to that virgin drive or at least LVM and
format it to something and then pop the image to it. That since I've
already proven I can do it on another 915GEVL now.
Two hours is a long time to create a box compared to 30 minutes with a
DRU SATA mobile drive tray image object.
>
> The USB Floppy Drive Hardware has been arrived
> so I can replicate some of your scenarios - the USB 2.0 CF reader is
> in the pipe line.
>
>
> kind regards
>
> Rainer
What I would really like to do is do as you really want me to do. To
move this whole operation from the R40 off into a current Thinkpad. The
T43 line, for example. So that we can use the same image for a
workstation, for example, to reverse clone either the Thinkpad or the
Intel 915 desktop units and pop them right up for I/D REXX customization
of the new box in a couple minutes flat for the building LAN.
However, I'm not in a financial position to go further with this at this
point, sadly. And further, the R40 is too mission critical to my work
to really mess around with it any more than really needed.
Next, I want to *VERY* carefully study this whole technique. I want to
carry the whole research project here to see if I can substitue a full
4GM jam in USB stick to pull off this stunt. More important, I want to
see if I can substitute the stick for the mobile drive tray stunt with
OS/2. Both for pure backup as well as clone reasons. Ah yes, I do
realize what we are doing for the ability to romp off with whatever.
But either you understand what you know and use it with protection, or
somebody else strangles you with what you don't understand.
Rainer Stroebel
---> snip
> And now for the final part of the report. OK, now that we have this
> all figured out yes I can at least START the full OS/2 boot to a
> completely raw out of the box for the Intel 915GEVL research case. In
> this case, I enable USB boot in the BIOS. I disable the totally
> useless SATA new drive and call the USB 2.0 Hard Disk unit the boot
> choice. In this case I jam it in the front panel USB slot and punch
> up the power. Away we go .. up comes the OS/2 screen and we get all
> the way through it until just after the /V response for the R40 clone
> image. And then here comes the "OS/2 cannot operate your message.."
>
>
> Think about it. Of course not. Why? Because we have the same
> conflict in the as installed /FLOPPY:1 /REMOVABLES:2 logic in the
> CONFIG.SYS operation. It has to have that to work with it's floppy
> boot. And with your USB floppy jammed into the ThinkPad, you have
> properly conformed your boot logic for the operation that way. Which
> is also the LVM Trap "D" creation for the Intel 915GEVL where the
> normal floppy is enabled in the case and .. you have no USB floppy
> jammed into it as well!
>
>
---> snip
> Not the same quite in the brand new Intel 915 motherboards. Same sort
> of logic, but the boot devices, permissions, and order of access are
> all set by their actual names and 'seen' device types in the BIOS
> setup. For research purposes I'm setting the floppy as first choice,
> the DVD/CD-ROM as second, and in the case of the pure USB 2.0 external
> device boot, I'm telling the BIOS to only permit that device.
>
> And away we go.
>
Hallo Mike.
it is not clear to me - probably my lack of understanding the fine
details of the English language :-)
Does you manage to boot from the USB 2.0 disk after alternations of the
config.sys for the Intel configuration?
kind regards
Rainer
Mike Luther
Rainer Stroebel wrote:
> Hallo Mike.
>
> it is not clear to me - probably my lack of understanding the fine
> details of the English language :-)
No problem. I know no German, or Dutch, grin!
>
>
> Does you manage to boot from the USB 2.0 disk after alternations of
> the config.sys for the Intel configuration?
>
> kind regards
>
> Rainer
>
Yes, I can boot from USB 2.0 disk after alterations of the FLOPPY DISK
file for CONFIG.SYS on Disk #1. I can also copy the USB 2.0 drive
image backwards into a new SATA hard drive in the Intel case as well.
Both are described below.
Here is the critical part of the new CONFIG.SYS below:
basedev=os2dasd.dmd
basedev=os2lvm.dmd
device=\testcfg.sys
basedev=ibmatapi.flt /A:1 /U:0 /TYPES:CLZ
SET COPYFROMFLOPPY=1
BASEDEV=USBUHCD.SYS
BASEDEV=USBUHCD.SYS
BASEDEV=USBUHCD.SYS
BASEDEV=USBUHCD.SYS
BASEDEV=USBEHCD.SYS
BASEDEV=USBD.SYS /REQ:USBUHCD$,USBOHCD$,USBEHCD$
BASEDEV=USBHID.SYS
BASEDEV=USBMSD.ADD /V /FLOPPIES:0 /REMOVABLES:1
basedev=dpt20xx.add
Notice the /FLOPPIES:0 above? If you don't make that a zero (0) then
the LVM operation tries to add in a USB floppy drive, when you already
have a floppy drive in the Intel 915 motherboard unit. That is how the
TRAP "D" happens I was seeing here.
Notice the /REMOVABLES:1 above? That is make sure there is only ONE
removable device. It will be the USB 2.0 hard disk when needed.
With these changes, even with a brand new not-used hard disk in the
Intel 915 case, the floppy diskettes will boot to the A: prompt. Do not
plug in the USB 2.0 drive unit during this floppy diskette boot run if
you are doing this.
After the A: prompt is there, use the Utility Disk #3 to use the LVM
program. Plug it in the A: drive on the Intel 915 case. Then plug in
the USB 2.0 Hard Disk unit into with the complete DFSEE image of the
working OS/2 system into a USB port in the Intel 915 case. I use one of
the front ports.
Then run the LVM program like the below from the OS/2 command prompt:
A:LVM /rediscoverPRM
You should see the USB device disk light flash as LVM finds it and
attaches it to your OS/2 command prompt operation.
Then plug in your floppy diskette with DFSOS2 and the few needed other
OS/2 DLL's in drive A: Run DFSOS2 there.
You will see both the SATA hard drive in the Intel 915 case as well as
the USB 2.0 hard drive in the DFSEE work list. You may the copy the
entire USB 2.0 disk to the new SATA hard disk in the Intel case. Shut
down DFSEE. Turn off the Intel computer. Remove the USB hard drive.
If you have the Intel 915 case set to boot from the SATA hard drive, it
will boot right up when you power up the next time.
If you wish, you can plug in the USB 2.0 drive before power up on the
Intel case. You can power it on and with the <F2> key change the Intel
BIOS to boot from a USB device and the USB 2.0 plugged in hard drive.
The Intel motherboard can see it if it is plugged in at power up. In
that way, you'll get to see the Intel case actually able to boot from
the external USB 2.0 unit!
I hope this is clear enough to cross the Atlantic, grin!
Oliver Rick
> Now if someone will help me unravel how to keep the USB enabled floppy
> diskette set from TRAP "D" with a completely barren virgin hard disk in
> that box so we don't have to use a double dip boot, I can finish this
> deal.
The first thing which comes to my mind: in the older code parts OS/2 relies
on a drive C: being present. Maybe the drive count for removable drives
relies on that or the like ...
Mike Luther
Oliver Rick wrote:
> On Thu, 24 Aug 2006 Mike Luther wrote:
>
>
>> Now if someone will help me unravel how to keep the USB enabled floppy
>> diskette set from TRAP "D" with a completely barren virgin hard disk in
>> that box so we don't have to use a double dip boot, I can finish this
>> deal.
>>
>
> The first thing which comes to my mind: in the older code parts OS/2 relies
> on a drive C: being present. Maybe the drive count for removable drives
> relies on that or the like ...
>
> /Olli/
>
No, I've found that in the case of the ThinkPad, because you are looking
at using a plug-in USB floppy, plus whatever may be a USB other device,
you have to use the /FLOPPIES:1 and /REMOVABLES:2 calling convention.
But when we get to the Intel 915 boxes, with their built in floppy
drive, if you do this then the only way to get past the Trap "D" is
simply to back out the floppy allowable and reduce the removables
allowable to only one unit! If I do that the box works fine and even
with a virgin non-used SATA drive in the box with which to later work
with LVM and so on.
Thanks to your earlier post I found the(my) error!
Rainer Stroebel
>
>
> If you wish, you can plug in the USB 2.0 drive before power up on the
> Intel case. You can power it on and with the <F2> key change the
> Intel BIOS to boot from a USB device and the USB 2.0 plugged in hard
> drive. The Intel motherboard can see it if it is plugged in at power
> up. In that way, you'll get to see the Intel case actually able to
> boot from the external USB 2.0 unit!
>
>
> I hope this is clear enough to cross the Atlantic, grin!
>
thanks for the crossing :-)
Back to the test lab to recreate your success with the booting from USD HDD
I just come from an TeamOS/2 meeting.
Here are the newest test results from the hardware test lab.
1. Change the config.sys - all USB statements without the /V parameter
Test with the T23 and the 16 MB CF Card mini System / USB 1.1 reader
no changes --> "OS/2 is unable...."
2. My mini OS/2 System on the 16 MB CF Card boots without any problem
in an ThinClients Modell "IBM NetworkStation", Typ "8364-S20" in
the CF slot ( no other boot device / IDE device connected).
For Infos about the ThinClient see
http://p2066.typo3server.info/85.html
- sorry German language only - some nice pictures!
3. ThinkPad T40 (with USB 2.0 ports ) and USB 1.1 Card Reader with
the 16 CF Card
The USB HDD is the first device in the boot startup sequence and
the USB BIOS support is active
The T40 does not start booting from the CF Card - BIOS hang up
- no OS/2 blob
Now I am waiting for the parcel with the USB 2.0 Card Reader and
an CF to IDE 44 (2.5" HDD connector type ) .
kind regards
Rainer
Rainer Stroebel
4. ThinkPad R51 (with USB 2.0 ports )
4.1 with USB 1.1 Card Reader and the 16 CF Card
4.2 with the Kingston FCR-HS2/ATA USB 2.0 ATA Card Reader
and the (slow) 16 MB CF Card
The USB HDD is the first device in the boot startup sequence and
the USB BIOS support is active
The R51 does not start booting from the CF Card - BIOS hang up -
no OS/2 blob
Kind Regards
Rainer
Rainer Stroebel
The CF to IDE 44 adapter arrived!
Here are the first test results:
5. ThinkPad T23 and 2nd HDD UltraBay Adapter
with the CF to IDE 44 Adapter and the 16MB CF Card
( OS/2 mini system with booting to the command prompt ) in the
Ultra Bay 2000
BIOS Setup
Now there is a second HDD in the Startup to select
5.1 BIOS Setup Boot sequence
first the HDD in the UltraBay
second the primary HDD of the T23
Test Result:
The T23 does not boot directly form the CF HHD,
which is the first HDD in the boot sequence!!
The boot manager from the primary HDD is booted!!
The C: drive of the CF Card System is one of boot options
in the BMR.
( I have not checked - do I added the C: drive to the Boot Menu
by DFSEE in the past or is it automatically added to the menu by
the BMR Code? )
Select the Boot Menu with the C: drive ( System on CF-Card )
an the system boots to the command prompt!
5.2 Disable the primary HDD on the BIOS,
the CF HDD in the UltraBay is now the only boot device in
the boot sequence.
Now the system boots from the CF CARD correctly.
===> Obviously case 5.1 is an BIOS BUG of the ThinkPad T23 BIOS
The bios startup sequence is not correctly working!!!
If the CF Card HDD is before the Primary HDD in the boot sequence,
the system should boot from the CF Card and not from the
Primary HDD !!!
Waiting for the USB 2.0 CF Card Reader
kind regards
Rainer
William L. Hartzell
Rainer Stroebel wrote:
> Here are the additional test results from the hardware test lab.
<snip>
> 5.2 Disable the primary HDD on the BIOS,
> the CF HDD in the UltraBay is now the only boot device in the
> boot sequence.
>
> Now the system boots from the CF CARD correctly.
>
>
> ===> Obviously case 5.1 is an BIOS BUG of the ThinkPad T23 BIOS
>
> The bios startup sequence is not correctly working!!!
>
> If the CF Card HDD is before the Primary HDD in the boot sequence,
> the system should boot from the CF Card and not from the
> Primary HDD !!!
>
The Boot Manager will always be booted first, even if it is on the
second in order boot device. You would have to change the MBR on both
drives to remove the Boot Manager to get the behavior that you expect.
--
Bill
Thanks a Million!
Rainer Stroebel
Hello William,
thanks for looking on the problem.
There is an CF Card in the ThinkPad T23 Ultrabay and a primary HDD.
The CF-Card has on physical partition with drive letter C: and
the partition is flagged active in the MBR (master boot record ) tables.
The HDD in the primary HDD slot has a
1. phyiscal partiton for W2K
2. physical partition with the BMR ( bootmanager ) - this
patition is flagged active
3. extended partition with several logical partitions - with
OS/2 systems drives and data drives
The boot startup sequence in the BIOS is
HDD with the CF Card
HDD primary T23 Slot
So the system BIOS has to look first on the CF HDD for a active
partition flag
if it does not find a active partition flag ,
the BIOS has to look at the next device in the boot sequence.
If the BIOS run out of boot devices in the startup up sequence and does
not found a partition
with an active flag --> bios error message : no system found
The BIOS does found an active flag on the CF Card HDD and should boot
from this partition - independent from the kind of system in the
partition
Do you see an other logic?
kind regards
Rainer
Rainer Stroebel
> Here are the additional test results from the hardware test lab.
>
>
> Waiting for the USB 2.0 CF Card Reader
>
> kind regards
The USB 2.0 CF Card reader arrived and the first test with the ThinkPad
T23 USB 1.1
has passed.
The Card Reader passed the USB 2.0 device downgrade test to an USB 1.1
controller.
Result as expected: "OS/2 is unable ....."
Finally I am read for the test
USB 2.0 reader to USB 2.0 controller with the usbehcd.sys device
driver.
Where is an ThinkPad with an USB 2.0 port located nearest? :-)
kind regards
Rainer
Rainer Stroebel
> Rainer Stroebel schrieb:
>
>> Here are the additional test results from the hardware test lab.
>>
>>
>> Waiting for the USB 2.0 CF Card Reader
>>
>> kind regards
>
>
>
> The USB 2.0 CF Card reader arrived and the first test with the
> ThinkPad T23 USB 1.1
> has passed.
>
> The Card Reader passed the USB 2.0 device downgrade test to an USB 1.1
> controller.
>
> Result as expected: "OS/2 is unable ....."
>
addendum:
USB 2.0 CF Card Reader
Vendor : 0x058F Alcor Mico Corp
Product: 0x6366 ??? Rev. 01.00
Bug: After the eject command the green led does not switch off !!!!!!
( the USB 1.1 CF Card Reader
Vendor: 0x0C45 Micordia
Product: 0x1050
Device Release# : 0110
the LED is switched off at eject command correctly ! )
Rainer