Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Problem getting drive bootable

1 view
Skip to first unread message

DManzaluni

unread,
Oct 22, 2009, 2:08:16 PM10/22/09
to
I have a dying drive on an IBM Transnote which has no CD drive (and
needless to say I have no restore discs). It still works and Maxtor's
quick check tells me that it is OK but it is on the way out
DEFINITELY. Hitachi DFT reports excessive disk shock and doesnt seem
to care that the drive still boots into WIndows 2000 and works.

So I am trying a trick I haveused numerous times before, Copy
Commander to copy the whole drive to a new one, including MBRs
Bootsectors etc etc. This always works

This time it isnt workinng. Even though the drive works and the
computer boots, when I take the drive out and put it in another
computer and boot off the Partition Commander boot disc, it sees the
drive as empty space with a tiny red partition at the very top and a
report of damaged partition. Needless to say nothing i can do can
revive it or repair it. There is a CC command TOGGLE ACTIVE/BOOTABLE
and if you leave a drive not active or bootable, it warns you but when
I exit, i dont get any message telling me that no drive is active or
bootable so Iassume the new drive is seen as being bootable

Does anyone have any ideas as to what I can do please?

I have tried copying all the files on it to a new drive but it,
obviously, simply isnt bootable. I have also tried copying the whole
of a Windows 2000 CD to a reformatted flash mem drive and making that
one bootable (so that i can run fixmbr or fixboot) but still the
computer (which CAN boot from a flash drive) reports no OS, presumably
on the flash drive.

I dont think there is any utlity on UBCD4WIN which can make a drive
bootable, nor can I imagine there are any Linux distros which would
let me copy the whole drive like Copy Commander does? I am running out
of ideas here!

Rod Speed

unread,
Oct 22, 2009, 3:13:29 PM10/22/09
to
DManzaluni wrote:

> I have a dying drive on an IBM Transnote which has no CD drive
> (and needless to say I have no restore discs). It still works and
> Maxtor's quick check tells me that it is OK but it is on the way out
> DEFINITELY. Hitachi DFT reports excessive disk shock and doesnt
> seem to care that the drive still boots into WIndows 2000 and works.

> So I am trying a trick I haveused numerous times before,
> Copy Commander to copy the whole drive to a new one,
> including MBRs Bootsectors etc etc. This always works

> This time it isnt workinng. Even though the drive works and the
> computer boots, when I take the drive out and put it in another
> computer and boot off the Partition Commander boot disc, it
> sees the drive as empty space with a tiny red partition at the
> very top and a report of damaged partition. Needless to say
> nothing i can do can revive it or repair it.

Presumably the damage didnt affect what was used to boot
Win 2K, but does affect what you are trying to do now.

> There is a CC command TOGGLE ACTIVE/BOOTABLE and
> if you leave a drive not active or bootable, it warns you but when
> I exit, i dont get any message telling me that no drive is active or
> bootable so Iassume the new drive is seen as being bootable

> Does anyone have any ideas as to what I can do please?

Replace the dying drive with a good one.

> I have tried copying all the files on it to a new
> drive but it, obviously, simply isnt bootable.

Presumably because what you are copying from is corrupted.

> I have also tried copying the whole of a Windows 2000 CD
> to a reformatted flash mem drive and making that one bootable
> (so that i can run fixmbr or fixboot) but still the computer (which CAN
> boot from a flash drive) reports no OS, presumably on the flash drive.

Nope, its talking about the hard drive.

> I dont think there is any utlity on UBCD4WIN which can make a
> drive bootable, nor can I imagine there are any Linux distros which
> would let me copy the whole drive like Copy Commander does?

There are. But that wont necessarily help if the source is corrupted.

> I am running out of ideas here!

One obvious approach is to get someone to make a copy from an uncorrupted Transnote etc.


Arno

unread,
Oct 22, 2009, 4:04:30 PM10/22/09
to
DManzaluni <dmanz...@googlemail.com> wrote:
> I have a dying drive on an IBM Transnote which has no CD drive (and
> needless to say I have no restore discs). It still works and Maxtor's
> quick check tells me that it is OK but it is on the way out
> DEFINITELY.

The quick check does not do a surface scan and can miss things.

> Hitachi DFT reports excessive disk shock and doesnt seem
> to care that the drive still boots into WIndows 2000 and works.

DFT is a risk management tool. It does not care that the drive is
not yet completely dead. It tells you that it will likely
be dead soon. It also does not care about what data is on
that disk and it is not its job to care.

> So I am trying a trick I have used numerous times before, Copy


> Commander to copy the whole drive to a new one, including MBRs
> Bootsectors etc etc. This always works

> This time it isnt workinng. Even though the drive works and the
> computer boots, when I take the drive out and put it in another
> computer and boot off the Partition Commander boot disc, it sees the
> drive as empty space with a tiny red partition at the very top and a
> report of damaged partition. Needless to say nothing i can do can
> revive it or repair it. There is a CC command TOGGLE ACTIVE/BOOTABLE
> and if you leave a drive not active or bootable, it warns you but when
> I exit, i dont get any message telling me that no drive is active or
> bootable so Iassume the new drive is seen as being bootable

> Does anyone have any ideas as to what I can do please?

Try a real disk imager and try to make a sector image of the
drive. This will show unreadable areas as well. It is just
possible that too mauch is damaged. Tools than can copy
disks with read errors are rare, but for example Linux
dd_rescue can.

> I have tried copying all the files on it to a new drive but it,
> obviously, simply isnt bootable. I have also tried copying the whole
> of a Windows 2000 CD to a reformatted flash mem drive and making that
> one bootable (so that i can run fixmbr or fixboot)

Sorry, you seem to be missing fundamental understanding here.
That is not how a boot process works. It needs an assembler program
in the first sector or the boot sector of the drive. The BIOS
then loads this assembler program and it takes control. Simply
copying files to a target filesystem is entirely unsuitable for
making something bootable.

> but still the
> computer (which CAN boot from a flash drive) reports no OS, presumably
> on the flash drive.

You should find out.

> I dont think there is any utlity on UBCD4WIN which can make a drive
> bootable, nor can I imagine there are any Linux distros which would
> let me copy the whole drive like Copy Commander does? I am running out
> of ideas here!

Linux full disk 1:1 copy:

dd <source> <target>

or

cat <source> > <destination>

or numerous other ways.

With progress indicator and read errors on the source:

dd_rescue <source> <target>

On partition level (likely not what you need, target has to be
made bootable manually) with graphical user interface:
Use gparted, which also comes in a mini distro as bootable CD or
memory-stick.

Raw parted (commandline) boot floppies/CD images/USB stick
images are also available on the web.

Also, I hope you have backup of any files on the damaged drive
that you want to keep. A drive damaged by mechanical shock
can die any minute without warning. I also would advise you
to make that sector image now, before the drive is completely
dead.

Arno

holarchy

unread,
Oct 22, 2009, 4:41:38 PM10/22/09
to

Nope, just some code.

> The BIOS then loads this assembler program

Nope, just some code.

Yousuf Khan

unread,
Oct 23, 2009, 12:05:02 PM10/23/09
to
DManzaluni wrote:
> I have a dying drive on an IBM Transnote which has no CD drive (and
> needless to say I have no restore discs). It still works and Maxtor's
> quick check tells me that it is OK but it is on the way out
> DEFINITELY. Hitachi DFT reports excessive disk shock and doesnt seem
> to care that the drive still boots into WIndows 2000 and works.

Does the Transnote have a USB 2.0 connector, at least? You can get put a
second drive into a USB enclosure and copy between the internal and
external.

You mentioned that when you take the Transnote's drive out and put it
into another computer, that it can't see it, yet it works in the
Transnote. Depending on how old the BIOS is between the Transnote and
the other computer, they may not be compatible.

> So I am trying a trick I haveused numerous times before, Copy
> Commander to copy the whole drive to a new one, including MBRs
> Bootsectors etc etc. This always works

Don't know Copy Commander, there are alternatives. In the free and
shareware category there is BootIt NextGen (BING), and XXClone.

Of course, you might try installing your own Copy Commander onto the
Transnote and see if it can work directly from there.

> I have tried copying all the files on it to a new drive but it,
> obviously, simply isnt bootable. I have also tried copying the whole
> of a Windows 2000 CD to a reformatted flash mem drive and making that
> one bootable (so that i can run fixmbr or fixboot) but still the
> computer (which CAN boot from a flash drive) reports no OS, presumably
> on the flash drive.

XXClone fits into this category. It works through the operating system
rather than outside it to copy the partition. It does normal
filesystem-based copies of the data between the source and destination
drives, and then in the end it makes the destination drive bootable by
setting all partition flags properly and especially by fixing up the
registry. One advantage of working through the OS is that you can mix
and match the source and destination filesystem types, such as FAT and
NTFS. Secondly, sizes don't have to match, the destination can even be
smaller than the source, as long as the destination is large enough to
fit all of the files. You can even continue to work on the machine while
it's copying. The disadvantage is that it can be a bit slower than
non-OS tools, but it can also be faster sometimes.

> I dont think there is any utlity on UBCD4WIN which can make a drive
> bootable, nor can I imagine there are any Linux distros which would
> let me copy the whole drive like Copy Commander does? I am running out
> of ideas here!

Linux can do raw device copies of drives, using the "dd" utility. But
I'm not sure Linux can fix up the Windows registry as required to make
it bootable into Windows. Unlike in the old days of DOS, Windows is not
directly copyable because its registry is involved in the boot process
and it is very hardware-dependent.

Yousuf Khan

DManzaluni

unread,
Oct 23, 2009, 2:05:55 PM10/23/09
to
On Oct 22, 4:41 pm, "holarchy" <holar...@nospam.com> wrote:
> Arno wrote:
> > Arno- Hide quoted text -
>
> - Show quoted text -- Hide quoted text -
>
> - Show quoted text -

Thanks for your help guys and a lot of what you say is very helpful,
especially about the linux utility. I didnt actually understand the
bit where you say that when I got to BOOT FROM and select USB MEDIA,
the report that I have no OS is actually coming from the hard drive:
Are you assuming this from a failure to boot from the designated
media, - that it goes on to try to boot from the next available
media, the HDD?

But I think a lot of it misses the point completely: Look at the
subject matter. Obviously I know that you cant make a disk image by
just copying files to a new drive and hoping against all logic that
you have copied the boot sector / MBR! Similarly obviously I have
copied everything on the drive to a new drive, - I said so. And as
the old drive does still boot into Windoews, I dont think that what I
am copying from is corrupted

What I am trying to do is to make that drive or the USB flas drive
containing Windows 2000 bootable. I thought that by going through the
Windows 2000 install process and getting into the Restore Console and
running fixboot or fixMBR, I could do that. I said so in the OP.
Certain people her do not seem to have noticed this?

And obviously I know that this drive which I describe carefully as
"DYING" could die at any moment????? Personally i am amazed that it
has lasted this long but I am now trying to concentrate on getting the
new drive bootable where I dont have a CD to install from.

Does anyone have any suggestions on this please or is there no way of
doing this?

Rod Speed

unread,
Oct 23, 2009, 3:03:44 PM10/23/09
to
DManzaluni wrote

>> Nope, just some code.

>> Nope, just some code.

>>> You should find out.

>>> dd <source> <target>

>>> or

>>> cat <source> > <destination>

>>> or numerous other ways.

>>> dd_rescue <source> <target>

> Thanks for your help guys and a lot of what you say is very helpful,
> especially about the linux utility. I didnt actually understand the
> bit where you say that when I got to BOOT FROM and select
> USB MEDIA, the report that I have no OS is actually coming
> from the hard drive: Are you assuming this from a failure to
> boot from the designated media, - that it goes on to try to
> boot from the next available media, the HDD?

Most likely what you put on the USB media is code that
boots from the hard drive, so its complaining about what
it finds isnt on the hard drive, not the USB media.

In other words you cant just copy stuff from a hard
drive to USB media and boot from that USB media.

> But I think a lot of it misses the point completely: Look at the
> subject matter. Obviously I know that you cant make a disk
> image by just copying files to a new drive and hoping against
> all logic that you have copied the boot sector / MBR!

That will work of you copy all the sectors from the old to the
new drive and the new drive is physically identical to the old drive.

> Similarly obviously I have copied everything on the drive to a new drive, - I said so.

Yes, but you dont know that the drive isnt corrupted before the copy.

> And as the old drive does still boot into Windoews,
> I dont think that what I am copying from is corrupted

> What I am trying to do is to make that drive or the
> USB flas drive containing Windows 2000 bootable.
> I thought that by going through the Windows 2000
> install process and getting into the Restore Console
> and running fixboot or fixMBR, I could do that.

Its more complicated than that.

> I said so in the OP. Certain people her do not seem to have noticed this?

Your english leaves a bit to be desired.

Its not always completely clear exactly what you are saying.

> And obviously I know that this drive which I describe carefully as
> "DYING" could die at any moment????? Personally i am amazed
> that it has lasted this long but I am now trying to concentrate on
> getting the new drive bootable where I dont have a CD to install from.

> Does anyone have any suggestions on this please or is there no way of doing this?

I'd try something different to copy the drive to a new drive and see if that will boot.

I dont know enough about Copy Commander to know if its a got a problem
or why the sort of copy you have done in the past isnt working for you now.

Its very likely that it isnt an exact copy of the drive, due to the bad sectors,
and its that difference that is stopping the new drive from being bootable.

Try doing the copy using say xxclone or linux dd.


Arno

unread,
Oct 24, 2009, 12:33:51 AM10/24/09
to
DManzaluni <dmanz...@googlemail.com> wrote:

> On Oct 22, 4:41?pm, "holarchy" <holar...@nospam.com> wrote:
>> Arno wrote:
>> > DManzaluni <dmanzal...@googlemail.com> wrote:
>> >> I have a dying drive on an IBM Transnote which has no CD drive (and
>> >> needless to say I have no restore discs). It still works and Maxtor's
>> >> quick check tells me that it is OK but it is on the way out
>> >> DEFINITELY.
>>
>> > The quick check does not do a surface scan and can miss things.
>>
>> >> Hitachi DFT reports excessive disk shock and doesnt seem
>> >> to care that the drive still boots into WIndows 2000 and works.
>>
>> > DFT is a risk management tool. It does not care that the drive is
>> > not yet completely dead. It tells you that it will likely
>> > be dead soon. It also does not care about what data is on
>> > that disk and it is not its job to care.
>>
>> >> So I am trying a trick I have used numerous times before, Copy
>> >> Commander to copy the whole drive to a new one, including MBRs
>> >> Bootsectors etc etc. ?This always works

>>
>> >> This time it isnt workinng. Even though the drive works and the
>> >> computer boots, when I take the drive out and put it in another
>> >> computer and boot off the Partition Commander boot disc, it sees the
>> >> drive as empty space with a tiny red partition at the very top and a
>> >> report of damaged partition. ?Needless to say nothing i can do can
>> >> revive it or repair it. ? There is a CC command TOGGLE

>> >> ACTIVE/BOOTABLE and if you leave a drive not active or bootable, it
>> >> warns you but when I exit, i dont get any message telling me that no
>> >> drive is active or bootable so Iassume the new drive is seen as
>> >> being bootable
>>
>> >> Does anyone have any ideas as to what I can do please?
>>
>> > Try a real disk imager and try to make a sector image of the
>> > drive. This will show unreadable areas as well. It is just
>> > possible that too mauch is damaged. Tools than can copy
>> > disks with read errors are rare, but for example Linux
>> > dd_rescue can.
>>
>> >> I have tried copying all the files on it to a new drive but it,
>> >> obviously, simply isnt bootable. ?I have also tried copying the whole

>> >> of a Windows 2000 CD to a reformatted flash mem drive and making that
>> >> one bootable (so that i can run fixmbr or fixboot)
>>
>> > Sorry, you seem to be missing fundamental understanding here.
>> > That is not how a boot process works. It needs an assembler
>> > program in the first sector or the boot sector of the drive.
>>
>> Nope, just some code.
>>
>> > The BIOS then loads this assembler program
>>
>> Nope, just some code.

Original meaning of "code": Short form of "assembler code". You fail.

>>
>>
>> > and it takes control. Simply
>> > copying files to a target filesystem is entirely unsuitable for
>> > making something bootable.
>> >> but still the
>> >> computer (which CAN boot from a flash drive) reports no OS,
>> >> presumably on the flash drive.
>>
>> > You should find out.
>>
>> >> I dont think there is any utlity on UBCD4WIN which can make a drive
>> >> bootable, nor can I imagine there are any Linux distros which would
>> >> let me copy the whole drive like Copy Commander does? I am running
>> >> out of ideas here!
>>
>> > Linux full disk 1:1 copy:
>>

>> > ?dd <source> <target>
>>
>> > or
>>
>> > ?cat <source> > <destination>


>>
>> > or numerous other ways.
>>
>> > With progress indicator and read errors on the source:
>>

>> > ?dd_rescue <source> <target>


>>
>> > On partition level (likely not what you need, target has to be

>> > ?made bootable manually) with graphical user interface:
>> > ?Use gparted, which also comes in a mini distro as bootable CD or
>> > ?memory-stick.
>>
>> > ?Raw parted (commandline) boot floppies/CD images/USB stick
>> > ?images are also available on the web.


>>
>> > Also, I hope you have backup of any files on the damaged drive
>> > that you want to keep. A drive damaged by mechanical shock
>> > can die any minute without warning. I also would advise you
>> > to make that sector image now, before the drive is completely
>> > dead.
>>
>> > Arno- Hide quoted text -
>>
>> - Show quoted text -- Hide quoted text -
>>
>> - Show quoted text -

> Thanks for your help guys and a lot of what you say is very helpful,
> especially about the linux utility. I didnt actually understand the
> bit where you say that when I got to BOOT FROM and select USB MEDIA,
> the report that I have no OS is actually coming from the hard drive:
> Are you assuming this from a failure to boot from the designated
> media, - that it goes on to try to boot from the next available
> media, the HDD?

That is the usual process. It tries to boot from a sequence
of sources. If you have disabled booting from HDD or have not
connected the HDD then it really is the USB stick.

> But I think a lot of it misses the point completely: Look at the
> subject matter. Obviously I know that you cant make a disk image by
> just copying files to a new drive and hoping against all logic that
> you have copied the boot sector / MBR!

Have a look at the group history. The assumption that you did
just copy the files is not so far fetched. No insult to you
was intended by me.

> Similarly obviously I have
> copied everything on the drive to a new drive, - I said so. And as
> the old drive does still boot into Windoews, I dont think that what I
> am copying from is corrupted

I don't think that is enough of a test.

> What I am trying to do is to make that drive or the USB flas drive
> containing Windows 2000 bootable. I thought that by going through the
> Windows 2000 install process and getting into the Restore Console and
> running fixboot or fixMBR, I could do that. I said so in the OP.
> Certain people her do not seem to have noticed this?

Noticed, but doubtful that this is the issue.

> And obviously I know that this drive which I describe carefully as
> "DYING" could die at any moment?????

Again, have a look at the group history. Do you _have_ a sector
level backup of the dying drive? If not, that should be your
first priority.

> Personally i am amazed that it
> has lasted this long but I am now trying to concentrate on getting the
> new drive bootable where I dont have a CD to install from.

> Does anyone have any suggestions on this please or is there no way of
> doing this?

There likely is a way, but at this time there is not enough
knowledge about the actual problem to identify it.

Arno

DManzaluni

unread,
Oct 24, 2009, 12:40:22 AM10/24/09
to

What I did was to use the HP flash drive utility to format the USB
drive and then use the utility which makes the drive bootable to make
that drive bootable, Then copy the Windows 2000 CD onto it

I now note that xxclone can indeed make this drive bootable and wonder
if this might do the trick? otherwise I will just try to use xxclone
to copy the whole drive and then make that version bootable

holarchy

unread,
Oct 24, 2009, 4:51:27 AM10/24/09
to

> Original meaning of "code":

Wrong, as always.

> Short form of "assembler code".

Wrong, as always. Before that there was no assembler at all.

> You fail.

Nope, you do, as always.

Rod Speed

unread,
Oct 24, 2009, 4:54:32 AM10/24/09
to
DManzaluni wrote

That isnt going to make a USB drive bootable.

> I now note that xxclone can indeed make this drive
> bootable and wonder if this might do the trick?

Not with a USB drive.

> otherwise I will just try to use xxclone to copy the
> whole drive and then make that version bootable

And this is where your english falls down again. The last bit doesnt make any sense.


DManzaluni

unread,
Oct 24, 2009, 10:36:34 AM10/24/09
to

OK So maybe I should have clarified the use of the words USB DRIVE
because obviously what meant in the circumstances of trying to clone
a drive was 'the new drive presently connected to the USB port' as
opposed to the one in the hard drive bay. But I thought that
obvious. This is a dying drive. It could die at any moment. Why
would I want to bother copying its whole contents f to a flash memory
USB drive, thereby reading from every segment and try to make that
bootable? The contents need putting on a new drive to reinsert in the
computer. Why the intermediate step? Are you trying to make some
point here which I cant see? I have a cable which connects a 2.5 inch
drive to a USB port on the computer with the dying drive.

In any event what is so difficult to understand about making a copy of
a drive which obviously isnt bootable (because I said that all I had
done was to copy the files) and then making that version of the drive
bootable?

By the way, your suggestion as to xxclone was a good one, I am trying
it. And again, obviously I know that there is corruptoin in the
source drive which I said is failing, I just hope that as the system
boots properly into Windows and all programs seem to execute, the
corruption should be in files which arent critical to the OS. In any
event I will run SFC /scannow so where am I going wrong assuming
xxclone can sector copy the drive and can make it bootable when the
clone function has concluded?

Rod Speed

unread,
Oct 24, 2009, 1:59:37 PM10/24/09
to

Yes, but doesnt change anything. Booting from a USB connnected
drive is more complicated than just copying what is on the internal
drive to the USB connected drive. Its a non trivial exercise to boot
XP from a USB connected drive on most systems and presumably
that applys to Win 2K as well, for the same reasons.

Essentially the problem is that the early boot phase doesnt understand
what USB drives are about driver wise. When you boot from an internal
drive, it loads USB drivers later in the boot and so you can use USB
drives fine once its booted, but thats an entirely separate issue to
actually booting from an USB connected drive.

> This is a dying drive. It could die at any moment.

Yes, that was always clear.

> Why would I want to bother copying its whole contents f to a flash memory
> USB drive, thereby reading from every segment and try to make that bootable?

Because you dont have the contents of the drive before it started to die and as
you said in your original, you dont have a CD drive to install Win 2K from again
and no restore disks either. So basically what is on the dying drive is all you have.

> The contents need putting on a new drive to reinsert in the computer.

Yes. But it isnt clear that the contents havent got corrupted by the dying.

Just because it does boot Win 2K, even tho its dying, doesnt prove that
Copy Commander can copy Win 2K from it to another drive. It looks very
like the fact that its dying is confusing Copy Commander during the copy.

> Why the intermediate step?

Because it might let you do what you are trying to do.

> Are you trying to make some point here which I cant see?

Yes, that the problem may well be Copy Commander and a dying drive.

> I have a cable which connects a 2.5 inch drive to
> a USB port on the computer with the dying drive.

Thats fine for copying, but not necessarily for booting from that config.

> In any event what is so difficult to understand about making a copy of
> a drive which obviously isnt bootable (because I said that all I had done
> was to copy the files) and then making that version of the drive bootable?

What is so difficult to understand about the fact that while the Copy Commander
copy should be bootable, and used to be bootable before the drive died, that it isnt
anymore and that the only thing that has changed is that the drive is dying now ?

There's only two logical possibilitys. Copy Commander is having a problem
copying the contents of that drive now that its dying, or you are remembering
it wrong when you said that you used to be able to make a bootable copy
using Copy Commander before the drive was dying.

> By the way, your suggestion as to xxclone was a good one, I am trying
> it. And again, obviously I know that there is corruptoin in the source
> drive which I said is failing, I just hope that as the system boots
> properly into Windows and all programs seem to execute, the
> corruption should be in files which arent critical to the OS.

And that is quite likely given that it does still boot the dying drive.

> In any event I will run SFC /scannow so where am I going


> wrong assuming xxclone can sector copy the drive

xxclone isnt a sector copier, it works at the file level, not the sector level.

It also does a bit more than just file copying, to make the drive bootable.

> and can make it bootable when the clone function has concluded?

Yes, thats what it adds to file copying.


DManzaluni

unread,
Nov 1, 2009, 9:34:42 PM11/1/09
to
> ...
>
> read more »
ro
This is becoming really annoying: I have now used xxclone to clone the
drive. I watched after the copying of files was concluded and saw it
copying the boot.ini and the MBR and making it bootable. At the end
of the process xxclone recommends you to go through the process again
in case what you have cloned isnt bootable. So I did this again.

But still the drive wont boot. I went through the process again and
still it wont boot.

So I know that the old drive may be corrupted adn I suspect where the
corruption may be (in folders like downloaded program files etc).where
corruption crops up in chkdsk but in essence the old drive DOES boot
so I suspect the problem isn't in the MBR or i nboot.ini.

The problem is slightly better than it was after I had copies all
files in linux and made it bootable in copy commander when I got the
error message "cannot find operating system'. Now it bets out of the
BIOS and looks on the drive and still doesnt see much, leaving me with
a simple cursor at the top of the screen.

Does this 'message' (actually clearly a lack of message) mean anything
to anyone? Is it seeing the boot,ini and not getting the right
information out of it? Or is it not seeing the master boot record? Or
is the file allocation table wrong in some way? I cant find anywhere
in the BIOS where the drive is actually identified but have tried
loading defaults with the new drive in place: That should suffice
shouldnt it?

Rod Speed

unread,
Nov 2, 2009, 12:04:18 AM11/2/09
to
>> read more �

> This is becoming really annoying:

Yeah, it can get frustrating.

> I have now used xxclone to clone the drive. I watched after the copying
> of files was concluded and saw it copying the boot.ini and the MBR and
> making it bootable. At the end of the process xxclone recommends you
> to go through the process again in case what you have cloned isnt
> bootable. So I did this again.

> But still the drive wont boot. I went through
> the process again and still it wont boot.

Presumably the drive has failed in a way that affects
the files involved in the early part of the boot phase.

I'd normally do a repair install on the new drive at this stage,
but it isnt that easy to do without a cd drive in the laptop.

> So I know that the old drive may be corrupted

Yes. In fact you know its got problems.

> adn I suspect where the corruption may be (in folders like downloaded program files etc).

Corruption there wouldnt stop it booting.

> where corruption crops up in chkdsk but in essence the old drive


> DOES boot so I suspect the problem isn't in the MBR or in boot.ini.

Or something else used in the early boot phase which works well enough
when booting the old drive, but which stops it copying to the new drive.

Thats why I would normally do a repair install, so those files are replaced,
but it isnt easy to do a repair install with your hardware config.

> The problem is slightly better than it was after I had copies
> all files in linux and made it bootable in copy commander
> when I got the error message "cannot find operating system'.
> Now it bets out of the BIOS and looks on the drive and still doesnt
> see much, leaving me with a simple cursor at the top of the screen.

Which is more evidence of a problem with one
of the files involved in the early boot phase.

> Does this 'message' (actually clearly a lack of message) mean anything to anyone?

Yes, that there is a problem with the early boot phase.

> Is it seeing the boot,ini and not getting the right information out of it?

Unlikely that is the problem. You should be able to check the boot.ini
and see whats in it, its just a normal text file and the format of the
entrys is pretty obvious and well documented on the net anyway.

> Or is it not seeing the master boot record?

Thats unlikely and you can check that with a hex editor
or dumper, its the first physical sector on the hard drive.

> Or is the file allocation table wrong in some way?

Wont be that.

> I cant find anywhere in the BIOS where the drive is actually identified but have
> tried loading defaults with the new drive in place: That should suffice shouldnt it?

Yes. And it does boot from the old drive as you say, so that should be setup fine.


Rod Speed

unread,
Nov 2, 2009, 12:55:22 AM11/2/09
to

And Win2K doesnt have a repair install.

DManzaluni

unread,
Nov 2, 2009, 5:29:45 PM11/2/09
to
"something else used in the early boot phase which works well enough
when booting the old drive, but which stops it copying to the new
drive" Copying OR re-creating!!

This is exceptionally troubling, as you seem to imply there is no
cure: This problem with early boot phase was why I was trying to get
the windows 2000 install files onto a flash memory drive and make THAT
bootable and try to correct things with that but I couldnt get that
solution to m did treat my drive as an HP one but still it wouldnt
boot off it. I hope this isnt a hardware problem!

Infuriating as I suspect that all problems with the drive can be
overcome except for these mysterious early boot phase files.

Curiously enough I have often had the opposite problem with dying
drives: They all usually go through that early boot phase and THEN
wont let the system boot into the OS

Is there really nothing in UBCD4WIN which might fix those early boot
phase files?

andy

unread,
Nov 2, 2009, 5:39:05 PM11/2/09
to

If the new drive is formatted FAT32, you can't overwrite the boot
sector, since the geometry of the partition is contained in the boot
sector.

The following steps should result in a working clone:
1. Use MBRWizard 2.0b to read the disk signature of the parent disk;
write it down. <http://mbrwizard.com/download.php>
2. Perform a new Windows 2000 install (just the text phase of setup)
on the new disk, using Windows 2000 setup to partition and format the
drive. This will create a properly functioning MBR and boot sector on
the new drive. Abort setup when it wants to boot from the hard drive.
3. Use xxclone to copy the OS from the parent to new drive. Do not
copy the MBR or boot sector.
4. Use MBRWizard to write the disk signature recorded in step 1 to the
new disk.
5. The new drive should now boot.

Rod Speed

unread,
Nov 2, 2009, 7:36:37 PM11/2/09
to
DManzaluni wrote:

> "something else used in the early boot phase which works
> well enough when booting the old drive, but which stops it
> copying to the new drive" Copying OR re-creating!!

Stops it being copied properly.

> This is exceptionally troubling, as you seem to imply there is no cure:

Thats overstating it. You should be able to do a clean Win2K
install to the new drive, then use xxcopy, not xxclone, to copy
the files from the old drive and that total should boot fine.

> This problem with early boot phase was why I was trying to get the windows
> 2000 install files onto a flash memory drive and make THAT bootable

Yeah but that adds another complication, booting a flash drive with 2K.

And there is no reason to believe that a flash drive would be any better than a new hard drive.

> and try to correct things with that but I couldnt get that
> solution to m did treat my drive as an HP one but still it
> wouldnt boot off it. I hope this isnt a hardware problem!

Thats unlikely given that it does still boot the old drive.

Assuming it still will boot the old drive, thats worth confirming.

> Infuriating as I suspect that all problems with the drive can be
> overcome except for these mysterious early boot phase files.

Yes, but they should be fixable with a clean install of Win 2K on the new drive.

That wouldnt even need to be done on the laptop, the system you are
using to copy from the old drive to the new drive should be fine for that.

Then use xxcopy to copy the non early boot files from
the old drive to the new drive and that should boot.

> Curiously enough I have often had the opposite problem
> with dying drives: They all usually go through that early
> boot phase and THEN wont let the system boot into the OS

Yeah, because its the other stuff thats got corrupted by the drive dying.

> Is there really nothing in UBCD4WIN which might fix those early boot phase files?

If they cant be copied by the two things you have tried copying them
with already, its unlikely that anything else off that will succeed.

Guess you could try either a forensic cloner or Spinrite. They try
a lot harder to copy files that cant be read error free the first time.
Not sure that Spinrite can be used for copying tho, so try a forensic cloner.

Try clonedisk from
http://invircible.com/resq.php
Its not free tho, not that expensive either.


Rod Speed

unread,
Nov 2, 2009, 7:37:48 PM11/2/09
to

xxclone allows for that.

DManzaluni

unread,
Nov 4, 2009, 7:41:32 AM11/4/09
to
> Try clonedisk fromhttp://invircible.com/resq.php

> Its not free tho, not that expensive either.

OK Things are becoming clearer, though I now have a new problem (to
which I have already referred) with this solution: No CD drive to
install from, which is what all that playing around with the USB drive
was all about. My conclusion that you cant install through the USB
port may be not incorrect.

Playing around in the BIOS I did discover an area letting a user do a
network install of an OS. This is probably how the OS was installed
originally. I am wondering how I can do this or if there is some way
of rebuilding those files over a network? I have to say that most of
the rest of my network is XP. Am I kidding myself by wondering
whether I can rebuild those early boot sector files with XP ones over
the network somehow or are they so different that I will have all the
correct files and a correct mbr but the wrong early boot phase files?

Does this add 371 additional layers of complication to the problem of
getting those early boot phase files installed properly? (Personally
I cant see how you could install an OS without being able to see a
network on the target drive).

i wonder why xxclose takes such trouble to let you know that it is
making the cloned drive bootable when it is doing no such thiing?

Rod Speed

unread,
Nov 4, 2009, 12:51:46 PM11/4/09
to

I did cover that, albeit a bit cryptically.

You should be able to do the clean install of Win2K
on the new drive in the non laptop you have been
doing the copy from the old drive to the new drive
on, with a CD drive in THAT system.

Then after thats been done and you have checked
that that new drive is bootable in the laptop, you
then xxcopy, not xxclone, the contents of the old
drive to the new drive in that non laptop again,
and that should give you a bootable new drive
in the laptop with what you want on it.

> which is what all that playing around with the USB
> drive was all about. My conclusion that you cant
> install through the USB port may be not incorrect.

That last is a double negative, not clear what you mean.

You should be able to do a clean install of Win2K onto
the new drive in the laptop, via the USB port, tho I
would personally do it the other way I spelt out above.

Note that that isnt installing Win2K on the USB stick,
but using it to get the Win2K files from the CD onto
the new drive and then installing Win2K on the new
drive with the Win2K install files on the new drive.

I'll spell this out in more detail if you want to go that route.

> Playing around in the BIOS I did discover an area
> letting a user do a network install of an OS.

You sure thats a network INSTALL, or just a network BOOT
of an OS ? The second one is much more common in BIOs.

> This is probably how the OS was installed originally.

Its much more likely to have just had the contents
of the hard drive copied onto it, with the drive out
of the laptop, before it was installed in the factory.

> I am wondering how I can do this

Yes, it is possible to do a network install of Win2K

> or if there is some way of rebuilding those files over a network?

Not rebuilding so much as restoring an image of the old drive over the network.

But you dont have the image to restore.

And now that the drive is dying, you probably cant create an image now.

> I have to say that most of the rest of my network is XP. Am I
> kidding myself by wondering whether I can rebuild those early
> boot sector files with XP ones over the network somehow

Yes, that is not feasible.

> or are they so different that I will have all the correct files
> and a correct mbr but the wrong early boot phase files?

Yes. Thats the reason for the clean install of Win2K to get the
early phase Win2K boot files on the new drive, since you dont
appear to be able to copy them from the old drive successfully.

> Does this add 371 additional layers of complication to the
> problem of getting those early boot phase files installed properly?

Yes.

> (Personally I cant see how you could install an OS
> without being able to see a network on the target drive).

Correct. And thats why its much more likely that the bios
entry is actually talking about BOOTING an OS from the
network, not INSTALLING an OS from the network.

However, it would certainly be possible to put something
on the new drive that would allow the drive to be booted
in the laptop and be able to see the network and THEN
install Win2K over the network. No real point tho over
doing the clean install of Win2K on the new drive with the
new drive in the non laptop you were doing the copys in.

> i wonder why xxclose takes such trouble to let you know that it is
> making the cloned drive bootable when it is doing no such thiing?

Most likely it doesnt realise that whatever early boot file
is preventing a successful boot is corrupt, it thinks its a
good file when its not been copied successfully.

Could also be a bug in xxclone that he hasnt noticed yet too.


DManzaluni

unread,
Nov 4, 2009, 2:00:58 PM11/4/09
to
Thanks again for that very complete response but I am back at square
one again!

At least I now know what to do: Get all the OS install files onto the
USB drive and make it bootable (You are right, the BIOS entry only
says start from USB, not install from USB) Try to start the install
process as you mention, either let it conclude or stop it at some
stage and then wipe the files on the drive and copy over the working
Transnote files one more time

There may be something wrong with the utility I am using to make the
flash memory drive bootable which prevents it from trying to install
the OS from the files copied from the CD?

Rod Speed

unread,
Nov 4, 2009, 4:05:50 PM11/4/09
to
DManzaluni wrote:

> Thanks again for that very complete response
> but I am back at square one again!

Why cant you do a clean install of Win2K on the non laptop you used
to copy the old drive to the new drive in, with a CD drive in that system ?

Or did I get that story wrong, and you only ever copied in the laptop ?

> At least I now know what to do: Get all the OS
> install files onto the USB drive and make it bootable

Thats much more difficult to do.

> (You are right, the BIOS entry only says start from USB, not install from USB)

> Try to start the install process as you mention, either let it conclude

Thats best.

> or stop it at some stage and then wipe the files on the drive
> and copy over the working Transnote files one more time

Yes, and use xxcopy to do that, not xxclone, because xxclone will fiddle with the boot files.

> There may be something wrong with the utility I am using to
> make the flash memory drive bootable which prevents it from
> trying to install the OS from the files copied from the CD?

The problem is that that its a non trivial exercise to boot 2K and XP from a flash memory drive.

Thats why you should concentrate on making the new drive
bootable, by doing a clean install of Win2K on that, and then
copying the non boot files on the old drive over that clean install.


DManzaluni

unread,
Nov 7, 2009, 10:55:04 AM11/7/09
to
On Nov 4, 4:05 pm, "Rod Speed" <rod.speed....@gmail.com> wrote:
> DManzaluni wrote:
> > Thanks again for that very complete response
> > but I am back at square one again!
>
> Why cant you do a clean install of Win2K on the non laptop you used
> to copy the old drive to the new drive in, with a CD drive in that system ?
>
> Or did I get that story wrong, and you only ever copied in the laptop ?
>
> > At least I now know what to do: Get all the OS
> > install files onto the USB drive and make it bootable
>
> Thats much more difficult to do.
>
> > (You are right, the BIOS entry only says start from USB, not install from USB)
> > Try to start the install process as you mention, either let it conclude
>
> Thats best.
>
> > or stop it at some stage and then wipe the files on the drive
> > and copy over the working Transnote files one more time
>
> Yes, and use xxcopy to do that, not xxclone, because xxclone will fiddle with the boot files.
>
> > There may be something wrong with the utility I am using to
> > make the flash memory drive bootable which prevents it from
> > trying to install the OS from the files copied from the CD?
> o

> The problem is that that its a non trivial exercise to boot 2K and XP from a flash memory drive.
>
> Thats why you should concentrate on making the new drive
> bootable, by doing a clean install of Win2K on that, and then
> copying the non boot files on the old drive over that clean install.

I will clearly have to try this but am not sure that I have a Windows
2000 install CD which isnt tied to some other manufacturer and wont
install unless it sees some wrong name in the BIOS computer
identification: I certainly dont have the Transnote install CD any
more or I wouldnt be asking these questions. I wonder if I can by-
pass all these questions by making a new linux flash install USB drive
from some new Linux download and then installing some Smart Boot
Manager on the drive. Does this avoid the question of the wrong early
boot files and substitute it's own early install files?

Rod Speed

unread,
Nov 7, 2009, 12:35:22 PM11/7/09
to
DManzaluni wrote

>> Thats best.

There arent very many of those around at all.

> I certainly dont have the Transnote install CD
> any more or I wouldnt be asking these questions.

I wouldnt hesitate to use a pirate copy if I had legally owned a legitimate copy and had lost it.

Another legal copy wouldnt cost much.

> I wonder if I can by-pass all these questions by making a new


> linux flash install USB drive from some new Linux download and
> then installing some Smart Boot Manager on the drive.

Nope, those just hand the boot to Win2K when they have
done what they need to do, so that still wont boot Win2K

> Does this avoid the question of the wrong early
> boot files and substitute it's own early install files?

Nope.


0 new messages