Norton Ghost Image Drive RAID0 weirdness

[David Shorthouse]

Hello folks,

I run a RAID0 for my System partition and periodically image the
partition onto a third IDE partition using Norton Ghost, assuming that in
the event of a stripe failure, I'd be able to boot off the IDE drive and
re-image onto the RAID0. Just for kicks, I disconnected the RAID0 drives and
booted off the IDE imaged partition and get the following error:

Cannot verify the license, blah, blah, blah with an error code 0x80090006.

I guess my fail-safe plan is not so fail-safe afterall. I tried booting the
IDE in Safe Mode and it just won't get past the Windows splash screen.

What the heck is going on here and how do I fix this such that I have a
fail-safe back-up of my RAID0 system partition in the event of failure?

Thanks for any ideas,

Dave

[Ron]

Hmm. I'm sure you're 'way past this, but what the heck;

Did you alter the IF cable routing AND the jumpers on the backup drive AND
the boot sequence in the BIOS?

Back to you.
Ron

[David Shorthouse]

Ron,

Thanks for the post. As I mentioned in an earlier post, I was attempting
a Ghost clone from a RAID0 system partition (2 physical drives in an array)
to an active IDE partition on a third disk. There is no need for me to mess
around with cables or jumpers in this case, and yes I did change the boot
sequence- I wouldn't have got as far as I did!

The problem occurs when I make the clone and then disconnect the RAID0
drives from the motherboard to test the "back-up". When I attempt to boot
off the now cloned partition, I get as far as the Windows splash screen when
I am then presented with an error about Windows being unable to verify the
activation license. I have tried booting into safe mode and this doesn't
help. There's must be something else going on here that I can't put my
finger on. The drives will indeed have different serial numbers, but this
shouldn't cause the problem; people clone their disks all the time without
issue. Maybe you can only clone a disk and not just the system partition
onto another active partition. I dunno.

Dave

> Hmm. I'm sure you're 'way past this, but what the heck;
>
> Did you alter the IF cable routing AND the jumpers on the backup drive AND
> the boot sequence in the BIOS?
>
> Back to you.
> Ron
>
>

[Pete Baker]

David

Are you aware of an example of this type of image (of a RAID0 set) being
able to immediately boot from an IDE partition?

Since the system you are imaging is configured to boot from the RAID set, it
appears to me that when it instead finds itself on an IDE that is when the
boot process stops as it imagines you have transferred it to a new hardware
configuration (which in effect you have).

I realise this does not offer the quick recovery that you require but one
solution may be to perform a repair installation (an in-place upgrade) to
reconfigure the system files to the IDE disk - at this point you should be
able to boot successfully - and then image back to the RAID0 set followed
by another repair installation (not sure if you need to re-introduce RAID
drivers). You'd need to re-install all necessary updates after that as
well. A re-activation may be required at this point, but since you're back
to the original configuration there shouldn't be a problem.

By the way, I'd back up important files before trying this out.

As I said it's not a quick recovery, but it should only be necessary in
extreme conditions.

Hope that helps
Pete


"David Shorthouse" > wrote in message
a...
> Hello folks,
>
> I run a RAID0 for my System partition and periodically image the
> partition onto a third IDE partition using Norton Ghost, assuming that in
> the event of a stripe failure, I'd be able to boot off the IDE drive and
> re-image onto the RAID0. Just for kicks, I disconnected the RAID0 drives
and
> booted off the IDE imaged partition and get the following error:
>
> Cannot verify the license, blah, blah, blah with an error code 0x80090006.
>
> I guess my fail-safe plan is not so fail-safe afterall. I tried booting
the
> IDE in Safe Mode and it just won't get past the Windows splash screen.
>
> What the heck is going on here and how do I fix this such that I have a
> fail-safe back-up of my RAID0 system partition in the event of failure?
>
> Thanks for any ideas,
>
> Dave
>
>

[Ron]

Hmm. This is an excellent question, Dave. I don't recall seeing it
before...although it is VERY reasonable. I regret that I am unable to
assist you further, but I AM keenly interested in the outcome. And so...

Please do post a follow-up when you arrive at a solution!
Ron

[David Shorthouse]

Hmm. Seems to me it would have been cheaper and quicker in the long run to
just make periodic Ghost CD's then to have a back-up on a third drive. Pity.

Here's a wrench, though. If I leave my RAID0 drives attached and the boot
off the IDE cloned partition, I get into Windows no problem. It's only when
I detach the RAID0 drives and then try to boot off the IDE drive that things
get messed up. This to me would mean that there's something improperly
referenced in the registry or something else happening here.

Dave

> David
>
> Are you aware of an example of this type of image (of a RAID0 set) being
> able to immediately boot from an IDE partition?
>
> Since the system you are imaging is configured to boot from the RAID set,
it
> appears to me that when it instead finds itself on an IDE that is when the
> boot process stops as it imagines you have transferred it to a new
hardware
> configuration (which in effect you have).
>
> I realise this does not offer the quick recovery that you require but one
> solution may be to perform a repair installation (an in-place upgrade) to
> reconfigure the system files to the IDE disk - at this point you should be
> able to boot successfully - and then image back to the RAID0 set followed
> by another repair installation (not sure if you need to re-introduce RAID
> drivers). You'd need to re-install all necessary updates after that as
> well. A re-activation may be required at this point, but since you're
back
> to the original configuration there shouldn't be a problem.
>
> By the way, I'd back up important files before trying this out.
>
> As I said it's not a quick recovery, but it should only be necessary in
> extreme conditions.
>
> Hope that helps
> Pete

[Pete Baker]

Dave

Well, which alternative to the original suggested method of recovery you
choose is entirely up to you...

As for being able to boot when the RAID0 set is still active, and I'm sure
that you've taken all necessary precautions to ensure that you are actually
booting from the partition clone (just checking)... given that the error
that appears is the verification of the license it supports the suggestion I
made that the changes to the hardware (removal of the RAID set, physical
removal of drives and the BIOS changes) may be enough to convince the
license verification that it has been cloned onto a new system... Those
changes are not happening if you leave the RAID0 intact...

It's a theory that supports the evidence... it may not be the whole story
here.

And, since you do already have an alternative recovery method that works in
all circumstances, you have the time to do further research.... and I hope
you find a definitive answer, one way or another.

Good luck
Pete




"David Shorthouse" > wrote in message
. ca...
> Hmm. Seems to me it would have been cheaper and quicker in the long run to
> just make periodic Ghost CD's then to have a back-up on a third drive.
Pity.
>
> Here's a wrench, though. If I leave my RAID0 drives attached and the boot
> off the IDE cloned partition, I get into Windows no problem. It's only
when
> I detach the RAID0 drives and then try to boot off the IDE drive that
things
> get messed up. This to me would mean that there's something improperly
> referenced in the registry or something else happening here.
>
> Dave
>
> > David
> >
> > Are you aware of an example of this type of image (of a RAID0 set) being
> > able to immediately boot from an IDE partition?
> >
> > Since the system you are imaging is configured to boot from the RAID
set,
> it
> > appears to me that when it instead finds itself on an IDE that is when
the
> > boot process stops as it imagines you have transferred it to a new
> hardware
> > configuration (which in effect you have).
> >
> > I realise this does not offer the quick recovery that you require but
one
> > solution may be to perform a repair installation (an in-place upgrade)
to
> > reconfigure the system files to the IDE disk - at this point you should
be
> > able to boot successfully - and then image back to the RAID0 set
followed
> > by another repair installation (not sure if you need to re-introduce
RAID
> > drivers). You'd need to re-install all necessary updates after that as
> > well. A re-activation may be required at this point, but since you're
> back
> > to the original configuration there shouldn't be a problem.
> >
> > By the way, I'd back up important files before trying this out.
> >
> > As I said it's not a quick recovery, but it should only be necessary in
> > extreme conditions.
> >
> > Hope that helps
> > Pete
>
>

[David Shorthouse]

Pete,

Thanks for the great dialog run. I find it troubling that removing a
RAID0 array would be enough to flag XP's license-check and yet upgrading to
a new processor and many other devices does nothing.
I still think there is something else going on here. I just did a bit of
reading about the XP boot process and see that when it is installed on a
SCSI disk (i.e. RAID0 partition), a system file called ntbootdd.sys is
created in the root folder. Could be that when I boot off the IDE drive, XP
is loading up this system file, then promptly gets hung-up when the RAID0
isn't found. So, you may have a point afterall. I wonder if deleting that
system file might allow the OS to then boot. Worth a try... Another test
would be to hook up some blank drives to the RAID ports on the motherboard.

Before I get completely absorbed and obsessed with this issue, perhaps I
should "back-up" a little. Is it possible for one to clone a system
partition off an IDE drive onto another IDE drive's active partition and
then boot off the clone with the source detached? I'm not talking about
cloning the whole drive to another, just the system partition.

Will post back with results if anyone else is following this,

Dave


> Well, which alternative to the original suggested method of recovery you
> choose is entirely up to you...
>
> As for being able to boot when the RAID0 set is still active, and I'm sure
> that you've taken all necessary precautions to ensure that you are
actually
> booting from the partition clone (just checking)... given that the error
> that appears is the verification of the license it supports the suggestion
I
> made that the changes to the hardware (removal of the RAID set, physical
> removal of drives and the BIOS changes) may be enough to convince the
> license verification that it has been cloned onto a new system... Those
> changes are not happening if you leave the RAID0 intact...
>
> It's a theory that supports the evidence... it may not be the whole story
> here.
>
> And, since you do already have an alternative recovery method that works
in
> all circumstances, you have the time to do further research.... and I hope
> you find a definitive answer, one way or another.
>
> Good luck
> Pete

[Pete Baker]

Dave

For the other upgrades you mention (processor, PCI cards, devices etc.) it
wouldn't make sense for MS not to allow those within the license - they'd
have to leave headroom for those changes to a system - change the
motherboard however and you're in a different game entirely .... Stepping
down from a RAID set to IDE would be unusual outside of the recovery
scenario you have in mind (keep in mind that such a change has a ripple
effect) and there are alternatives to your original plan as you have
discovered...

I'm not saying definitely that this is the case - only that the
circumstances strongly suggest it.... remember also that (although it's
difficult to be sure of how the points are calculated), according to the
info I'm aware of, the changes allowed from the activation point are
accumulative..

As for the testing you suggest.... it's the best way to learn what works and
what doesn't (as long as you can figure out why it doesn't)... the IDE clone
you suggest doesn't really help very much though... by all means check to be
sure that the cloning process is viable, but for what you intend to be able
to do, it doesn't move things forward...

Just one note of caution though, don't delete that system file, rename it
instead.

In all this it's worth remembering that the situation you're planning for is
a catastrophic failure... if it occurs then the priority should be the
safest recovery.. not the quickest recovery.

cheers
Pete

"David Shorthouse" > wrote in message
...
> Pete,
>
> Thanks for the great dialog run. I find it troubling that removing a
> RAID0 array would be enough to flag XP's license-check and yet upgrading
to
> a new processor and many other devices does nothing.
> I still think there is something else going on here. I just did a bit
of
> reading about the XP boot process and see that when it is installed on a
> SCSI disk (i.e. RAID0 partition), a system file called ntbootdd.sys is
> created in the root folder. Could be that when I boot off the IDE drive,
XP
> is loading up this system file, then promptly gets hung-up when the RAID0
> isn't found. So, you may have a point afterall. I wonder if deleting that
> system file might allow the OS to then boot. Worth a try... Another test
> would be to hook up some blank drives to the RAID ports on the
motherboard.
>
> Before I get completely absorbed and obsessed with this issue, perhaps I
> should "back-up" a little. Is it possible for one to clone a system
> partition off an IDE drive onto another IDE drive's active partition and
> then boot off the clone with the source detached? I'm not talking about
> cloning the whole drive to another, just the system partition.
>
> Will post back with results if anyone else is following this,
>
> Dave
>
>
> > Well, which alternative to the original suggested method of recovery you
> > choose is entirely up to you...
> >
> > As for being able to boot when the RAID0 set is still active, and I'm
sure
> > that you've taken all necessary precautions to ensure that you are
> actually
> > booting from the partition clone (just checking)... given that the error
> > that appears is the verification of the license it supports the
suggestion
> I
> > made that the changes to the hardware (removal of the RAID set,
physical
> > removal of drives and the BIOS changes) may be enough to convince the
> > license verification that it has been cloned onto a new system... Those
> > changes are not happening if you leave the RAID0 intact...
> >
> > It's a theory that supports the evidence... it may not be the whole
story
> > here.
> >
> > And, since you do already have an alternative recovery method that works
> in
> > all circumstances, you have the time to do further research.... and I
hope
> > you find a definitive answer, one way or another.
> >
> > Good luck
> > Pete
>
>

[David Shorthouse]

Well, removing that ntbootdd.sys did nothing to solve the problem. It still
got hung up on something and then promptly complained about not being able
to verify the license. Just for kicks, I booted off the XP CD to see if the
boot partition on the lone IDE drive was being assigned the wrong letter
(i.e. other than 'C'). It was designated 'C' just as it should be.

Now this is really starting to irk me. All I am attempting to do is a
legitimate clone of my 'C' partition, which happens to reside on a RAID0
array, onto a regular IDE partition in the event that a stripe in the RAID0
array should break. I do have image files of the C drive on CD-R, but what's
to guarantee that re-imaging these to the source will allow me to boot-up
and have everything running as before a broken stripe? Perhaps I am a little
paranoid over this, but as it stands, I am not convinced that my back-ups
will ever be functional (unless of course I have to resort to a "repair
install" and then spend 2 days updating everything).

I am also not convinced that having installed XP on a RAID0 partition was
such a good idea. Since most of the system files are small, the RAID0
partition wouldn't really offer any improved boot time or general speed
throughout a session. I should have installed it on a regular IDE drive
(preferably one with a large buffer) and kept the larger, personal files on
a RAID0 array. However, it seems that now I can't just image or clone my
current 'C' partition onto my IDE drive and expect to boot.

Dismayed,

Dave

> Pete,
>
> Thanks for the great dialog run. I find it troubling that removing a
> RAID0 array would be enough to flag XP's license-check and yet upgrading
to
> a new processor and many other devices does nothing.
> I still think there is something else going on here. I just did a bit
of
> reading about the XP boot process and see that when it is installed on a
> SCSI disk (i.e. RAID0 partition), a system file called ntbootdd.sys is
> created in the root folder. Could be that when I boot off the IDE drive,
XP
> is loading up this system file, then promptly gets hung-up when the RAID0
> isn't found. So, you may have a point afterall. I wonder if deleting that
> system file might allow the OS to then boot. Worth a try... Another test
> would be to hook up some blank drives to the RAID ports on the
motherboard.
>
> Before I get completely absorbed and obsessed with this issue, perhaps I
> should "back-up" a little. Is it possible for one to clone a system
> partition off an IDE drive onto another IDE drive's active partition and
> then boot off the clone with the source detached? I'm not talking about
> cloning the whole drive to another, just the system partition.
>
> Will post back with results if anyone else is following this,
>
> Dave