XP new installation - three hard drives show 'Unmountable Boot Volume' message

On Wed, 11 Jan 2017 14:10:17 +0000, Ian Jackson
wrote:

In message , Shadow
writes



If it's reading the HD, it's not a hardware problem, it's a
driver. Might even be a conflict.

There's no problem seeing the HD - either when you're installing from
the CD or for the first 15 seconds when XP is booting up. I think that
after that when other parameters start to get loaded - including the
what the screen resolution has been set to (which, until you have set
it, will be the initial default settings). I presume that if a different
(and incompatible) HD driver is one of the things that gets loaded, the
HD could be cast adrift - and hence the catastrophic failure to
continue.

See if you can get some diagnostic
running long enough to write a log you can read from a live-cd.
Hint - you can boot from the Kaspersky Rescue Disk and use the
Registry Editor to disable services/drivers from loading at startup. I
once had to do that to remove a particularly obnoxious registry item
that Acronis put there and would not allow me to boot.
[]'s

As this is more a 'labour of love' than something which is actually
causing me a problem, I think that it's maybe time to put things on the
back burner for now. [I really ought to attend to some more-pressing
matter!] However, I'll do one more simple test, ie as I suggested, clone
the trouble-free virgin installation on the WD400 disk to one of the
no-go Seagates - and then to use the Seagate as the C-drive.

Actually I didn't say that. I said a "driver" conflict, not a
"drive" conflict. If you could get a diagnostic to run at startup you
could probably spot the conflicting driver and rename it on the HD, so
it won't load next boot-up.
[]'s
--
Don't be evil - Google 2004
We have a new policy - Google 2012

In message , Shadow
writes
On Wed, 11 Jan 2017 14:10:17 +0000, Ian Jackson
wrote:

In message , Shadow
writes



If it's reading the HD, it's not a hardware problem, it's a
driver. Might even be a conflict.

There's no problem seeing the HD - either when you're installing from
the CD or for the first 15 seconds when XP is booting up. I think that
after that when other parameters start to get loaded - including the
what the screen resolution has been set to (which, until you have set
it, will be the initial default settings). I presume that if a different
(and incompatible) HD driver is one of the things that gets loaded, the
HD could be cast adrift - and hence the catastrophic failure to
continue.

See if you can get some diagnostic
running long enough to write a log you can read from a live-cd.
Hint - you can boot from the Kaspersky Rescue Disk and use the
Registry Editor to disable services/drivers from loading at startup. I
once had to do that to remove a particularly obnoxious registry item
that Acronis put there and would not allow me to boot.
[]'s

As this is more a 'labour of love' than something which is actually
causing me a problem, I think that it's maybe time to put things on the
back burner for now. [I really ought to attend to some more-pressing
matter!] However, I'll do one more simple test, ie as I suggested, clone
the trouble-free virgin installation on the WD400 disk to one of the
no-go Seagates - and then to use the Seagate as the C-drive.

Actually I didn't say that. I said a "driver" conflict, not a
"drive" conflict. If you could get a diagnostic to run at startup you
could probably spot the conflicting driver and rename it on the HD, so
it won't load next boot-up.
[]'s

I'm not sure what you think you said - but didn't.

However, this might be a better clue.

I cloned the working WD400 disk to the Seagate (using AOMEI's 'Copy'
function). When the operation ended, I got the message "Sorry. The
current operation has been cancelled. Information Code 17. Description:
Failed to update boot.ini"

For luck, I did a Repair MBR (executed OK) - so I connected the Seagate
disk to the ASUS PC, and booted up.

The boot started OK - then crashed (as before), but immediately started
to reboot. This time, I got the option of various boot options, so I
chose Safe Mode - and it did boot OK.

Using Start, Run, msconfig, I found my way to the System Cinfiguration
Utility screen, and selected the BOOT.INI tab. When compared with this
machine, the boot.ini file looks normal - apart from this addition to
the last line (after fastdetect):
/NoExecute=OptIn
Does this look normal (and what does it do)?

Finally, just for the record, a "Check all Boot Paths" gives an 'all
OK'.

So..... does any of this make sense - and any suggestions?
--
Ian

Ian Jackson wrote:
In message , Shadow
writes
On Wed, 11 Jan 2017 14:10:17 +0000, Ian Jackson
wrote:

In message , Shadow
writes



If it's reading the HD, it's not a hardware problem, it's a
driver. Might even be a conflict.

There's no problem seeing the HD - either when you're installing from
the CD or for the first 15 seconds when XP is booting up. I think that
after that when other parameters start to get loaded - including the
what the screen resolution has been set to (which, until you have set
it, will be the initial default settings). I presume that if a different
(and incompatible) HD driver is one of the things that gets loaded, the
HD could be cast adrift - and hence the catastrophic failure to
continue.

See if you can get some diagnostic
running long enough to write a log you can read from a live-cd.
Hint - you can boot from the Kaspersky Rescue Disk and use the
Registry Editor to disable services/drivers from loading at startup. I
once had to do that to remove a particularly obnoxious registry item
that Acronis put there and would not allow me to boot.
[]'s

As this is more a 'labour of love' than something which is actually
causing me a problem, I think that it's maybe time to put things on the
back burner for now. [I really ought to attend to some more-pressing
matter!] However, I'll do one more simple test, ie as I suggested, clone
the trouble-free virgin installation on the WD400 disk to one of the
no-go Seagates - and then to use the Seagate as the C-drive.

Actually I didn't say that. I said a "driver" conflict, not a
"drive" conflict. If you could get a diagnostic to run at startup you
could probably spot the conflicting driver and rename it on the HD, so
it won't load next boot-up.
[]'s

I'm not sure what you think you said - but didn't.

However, this might be a better clue.

I cloned the working WD400 disk to the Seagate (using AOMEI's 'Copy'
function). When the operation ended, I got the message "Sorry. The
current operation has been cancelled. Information Code 17. Description:
Failed to update boot.ini"

For luck, I did a Repair MBR (executed OK) - so I connected the Seagate
disk to the ASUS PC, and booted up.

The boot started OK - then crashed (as before), but immediately started
to reboot. This time, I got the option of various boot options, so I
chose Safe Mode - and it did boot OK.

Using Start, Run, msconfig, I found my way to the System Cinfiguration
Utility screen, and selected the BOOT.INI tab. When compared with this
machine, the boot.ini file looks normal - apart from this addition to
the last line (after fastdetect):
/NoExecute=OptIn
Does this look normal (and what does it do)?

Finally, just for the record, a "Check all Boot Paths" gives an 'all OK'.

So..... does any of this make sense - and any suggestions?

/NoExecute=OptIn uses the NX bit.

https://en.wikipedia.org/wiki/Nx_bit

It uses PAE to provide a place in the TLB entries to store
the state of the NX bit in mappings. PAE is likely to be a
default option in SP3, not sure about the others.

The description here is a bit confused as to whether
all the versions of WinXP have PAE enabled by default. You
couldn't Opt In to No Execute, unless the page table operating
format was PAE.

https://en.wikipedia.org/wiki/Physic...rosoft_Windows

I think this is a red herring.

*******

Your system comes up in Safe Mode.

Your system does not work, in regular boot.

One difference is drivers.

What are you using for a video card ?

What are the AGP settings in the BIOS ?

Are you operating the Northbridge AGP rate
faster than that chipset properly supports ?

I generally recollect that SIS wasn't that bad.
Not as bad as the board I bought with the ALI chipset,
which never had a stable AGP interface.

When the system comes up in Safe Mode, the video
card runs in VESA mode, displaying a frame buffer of content
at a known address. There is no acceleration to
speak of, and if the video card is in the AGP slot,
the AGP slot supports two protocols at the same time.
PCI protocol is used for register access. Whereas
DMA/DIME transfers operate the AGP slot in AGP burst
mode. With 8X giving 2100MB/sec or so. There is still
plenty of PCI accesses in normal usage and when the
system is booted in normal mode and playing a game,
there would be some PCI cycles (running at 1X) and
AGP cycles (running at a higher speed).

I seem to remember SIS had a little utility for
adjusting something AGP related. Perhaps in Safe
Mode, you could fire up that utility and see
whether the utility will even work without
the driver running. It's possible that utility
leaves settings in the Registry, so that they
could be applied to your next Normal Mode boot.

https://forum-en.msi.com/index.php?topic=17012.0

While you can enable boot logging so that routine
names are printed on the screen as it boots,
the last name you see is not the offender. The
one that crashed, the name of that is not on the
screen. And for me at least, I cannot predict what
will show in that boot process, to say "oh, this
one is the next one in your list, and that's
the driver that is crashing". So for me personally,
I would not get any diagnostic info from watching
boot log on the screen.

Paul

On Wed, 11 Jan 2017 21:03:21 +0000, Ian Jackson
wrote:

Actually I didn't say that. I said a "driver" conflict, not a
"drive" conflict. If you could get a diagnostic to run at startup you
could probably spot the conflicting driver and rename it on the HD, so
it won't load next boot-up.
[]'s

I'm not sure what you think you said - but didn't.

However, this might be a better clue.

I cloned the working WD400 disk to the Seagate (using AOMEI's 'Copy'
function). When the operation ended, I got the message "Sorry. The
current operation has been cancelled. Information Code 17. Description:
Failed to update boot.ini"

For luck, I did a Repair MBR (executed OK) - so I connected the Seagate
disk to the ASUS PC, and booted up.

The boot started OK - then crashed (as before), but immediately started
to reboot. This time, I got the option of various boot options, so I
chose Safe Mode - and it did boot OK.

Using Start, Run, msconfig, I found my way to the System Cinfiguration
Utility screen, and selected the BOOT.INI tab. When compared with this
machine, the boot.ini file looks normal - apart from this addition to
the last line (after fastdetect):
/NoExecute=OptIn
Does this look normal (and what does it do)?

Normal, you can change it to
/noexecute=AlwaysOff

But I don't think it'll make a difference.

Still think XP is recognizing the Seagate as something else
and loading the wrong DRIVER (or service)
You being able to boot into safemode practically confirms it.
[]'s


Finally, just for the record, a "Check all Boot Paths" gives an 'all
OK'.
--
Don't be evil - Google 2004
We have a new policy - Google 2012

In message , Paul
writes




/NoExecute=OptIn uses the NX bit.

https://en.wikipedia.org/wiki/Nx_bit

It uses PAE to provide a place in the TLB entries to store
the state of the NX bit in mappings. PAE is likely to be a
default option in SP3, not sure about the others.

The description here is a bit confused as to whether
all the versions of WinXP have PAE enabled by default. You
couldn't Opt In to No Execute, unless the page table operating
format was PAE.

https://en.wikipedia.org/wiki/Physic...rosoft_Windows

I think this is a red herring.

*******

Your system comes up in Safe Mode.

Your system does not work, in regular boot.

One difference is drivers.

What are you using for a video card ?

What are the AGP settings in the BIOS ?

Are you operating the Northbridge AGP rate
faster than that chipset properly supports ?

I generally recollect that SIS wasn't that bad.
Not as bad as the board I bought with the ALI chipset,
which never had a stable AGP interface.

When the system comes up in Safe Mode, the video
card runs in VESA mode, displaying a frame buffer of content
at a known address. There is no acceleration to
speak of, and if the video card is in the AGP slot,
the AGP slot supports two protocols at the same time.
PCI protocol is used for register access. Whereas
DMA/DIME transfers operate the AGP slot in AGP burst
mode. With 8X giving 2100MB/sec or so. There is still
plenty of PCI accesses in normal usage and when the
system is booted in normal mode and playing a game,
there would be some PCI cycles (running at 1X) and
AGP cycles (running at a higher speed).

I seem to remember SIS had a little utility for
adjusting something AGP related. Perhaps in Safe
Mode, you could fire up that utility and see
whether the utility will even work without
the driver running. It's possible that utility
leaves settings in the Registry, so that they
could be applied to your next Normal Mode boot.

https://forum-en.msi.com/index.php?topic=17012.0

While you can enable boot logging so that routine
names are printed on the screen as it boots,
the last name you see is not the offender. The
one that crashed, the name of that is not on the
screen. And for me at least, I cannot predict what
will show in that boot process, to say "oh, this
one is the next one in your list, and that's
the driver that is crashing". So for me personally,
I would not get any diagnostic info from watching
boot log on the screen.

I'm digesting your questions and comments. I'll try and get back
tomorrow with some sensible answers!
--
Ian

In message , Shadow
writes
On Wed, 11 Jan 2017 21:03:21 +0000, Ian Jackson
wrote:


When compared with this
machine, the boot.ini file looks normal - apart from this addition to
the last line (after fastdetect):
/NoExecute=OptIn
Does this look normal (and what does it do)?

Normal, you can change it to
/noexecute=AlwaysOff

But I don't think it'll make a difference.

Yes - I'll try that.

Still think XP is recognizing the Seagate as something else
and loading the wrong DRIVER (or service)
You being able to boot into safemode practically confirms it.
[]'s


Finally, just for the record, a "Check all Boot Paths" gives an 'all
OK'.

The question is then, "Which driver does it really need"? In the safe
mode, I can't access the internet to try an online driver update. Even
with a monitor running, it might not be obvious which drivers were
destined for the HD. [I'll need to try and do this.]
--
Ian

On Thu, 12 Jan 2017 19:42:36 +0000, Ian Jackson
wrote:

In message , Shadow
writes
On Wed, 11 Jan 2017 21:03:21 +0000, Ian Jackson
wrote:


When compared with this
machine, the boot.ini file looks normal - apart from this addition to
the last line (after fastdetect):
/NoExecute=OptIn
Does this look normal (and what does it do)?

Normal, you can change it to
/noexecute=AlwaysOff

But I don't think it'll make a difference.

Yes - I'll try that.

Still think XP is recognizing the Seagate as something else
and loading the wrong DRIVER (or service)
You being able to boot into safemode practically confirms it.
[]'s


Finally, just for the record, a "Check all Boot Paths" gives an 'all
OK'.

The question is then, "Which driver does it really need"? In the safe
mode, I can't access the internet to try an online driver update. Even
with a monitor running, it might not be obvious which drivers were
destined for the HD. [I'll need to try and do this.]

It probably doesn't (other than the default), and is
recognizing your Seagate as something that does ..... which is why it
crashes. It's loading an un-necessary driver/service, which it does
not do in safe mode.
[]'s
--
Don't be evil - Google 2004
We have a new policy - Google 2012

In message , Shadow
writes
On Thu, 12 Jan 2017 19:42:36 +0000, Ian Jackson
wrote:

In message , Shadow
writes
On Wed, 11 Jan 2017 21:03:21 +0000, Ian Jackson
wrote:


When compared with this
machine, the boot.ini file looks normal - apart from this addition to
the last line (after fastdetect):
/NoExecute=OptIn
Does this look normal (and what does it do)?

Normal, you can change it to
/noexecute=AlwaysOff

But I don't think it'll make a difference.

Yes - I'll try that.

Still think XP is recognizing the Seagate as something else
and loading the wrong DRIVER (or service)
You being able to boot into safemode practically confirms it.
[]'s


Finally, just for the record, a "Check all Boot Paths" gives an 'all
OK'.

The question is then, "Which driver does it really need"? In the safe
mode, I can't access the internet to try an online driver update. Even
with a monitor running, it might not be obvious which drivers were
destined for the HD. [I'll need to try and do this.]

It probably doesn't (other than the default), and is
recognizing your Seagate as something that does ..... which is why it
crashes. It's loading an un-necessary driver/service, which it does
not do in safe mode.
[]'s

I got what I think is the correct Seagate IDE driver (exe) file off the
internet, and ran it (while in the safe mode). It made no difference.

However, things are going from bad to worse. I'm rather busy, but I
quickly tried an XP Home plus SP2a installation on a 160GB Seagate
ST3160215ACE (DB35.3). This failed as before, with the 'Unmountable Boot
Volume' message. I then tried a second disk (same type). It too failed.
So that's two more Seagates it doesn't like.

I was beginning to wonder if the PC had become faulty, so I plugged in
the original old (2001) 120GB Seagate disk (an ST3120026A Barracuda
7200.7) which has accumulated 16 years-worth of junk. It booted OK like
it normally does. [I suppose can now check which drivers are being
used.]

After I've done some essential chores, my next test will be to clone the
120GB to one of the 160GB disks.
--
Ian

In message , Ian Jackson
writes

After I've done some essential chores, my next test will be to clone
the 120GB to one of the 160GB disks.

As it takes a while, overnight I cloned the old original 120GB Seagate
ST3120026A (very slow, but working) from my ancient ASUS PC to the 160GB
Seagate ST3160215ACE (one that, with a new XP installation, gave the
'Unmountable Boot Volume' message when rebooting).

[To do the cloning, I used a different PC and Easeus, with the source
and target hard drives connected via USB adapters]

After putting the target the PC and booting up, I got (as before) the
usual message 'Unmountable Boot Volume' (but, again, as before, I could
boot up in the Safe mode).

I powered down the PC, and booted up again from Minitool Partition
Wizard (bootable, on a CD). I the did a Rebuild MBR. [This is something
I've occasionally found necessary after doing a clone.] The PC then
booted correctly.

Finally, I ran CHKDSK /F, and it showed no errors. Mind you, I didn't
get any errors when I did this previously during the past few days
(after my several failed installations) - and, of course GSmartControl
wasn't showing any errors either.

So it seems to me that the problem I am having does seem to associated
with the hard drives (rather than being a driver problem).

Also, it has occurred to me that a common factor is that a did a
low-level format on all of the problem HDs (using HDDGuru's HDD LLF Low
Level Format Tool). After some Googling, I see that there is sometimes a
mention that a low-level format can (and, indeed, will) 'destroy' a hard
drive - but it isn't clear whether this means that it will no longer be
usable, or only means that it will wipe all data. There is also mention
that a low-level format can make blocks which have been marked as 'bad'
usable again (at least temporarily) - which is not a good thing.

Any further advice on the saga would be appreciated!
--
Ian

On Sat, 14 Jan 2017 11:54:25 +0000, Ian Jackson
wrote:

So it seems to me that the problem I am having does seem to associated
with the hard drives (rather than being a driver problem).

No, you booted perfectly in safe mode, no disk errors, so the
drives are working fine.
[]'s
--
Don't be evil - Google 2004
We have a new policy - Google 2012

In message , Shadow
writes
On Sat, 14 Jan 2017 11:54:25 +0000, Ian Jackson
wrote:

So it seems to me that the problem I am having does seem to associated
with the hard drives (rather than being a driver problem).

No, you booted perfectly in safe mode, no disk errors, so the
drives are working fine.
[]'s

Noted.

Even though an MBR rebuild did fix one of my 160GB Seagates, what about
the suggestions that a low-level format could possibly permanently wreck
some hard drives?

Incidentally, I've just tried yet again to do virgin installation to my
second 160GB Seagate - and unlike the other (cloned-to) 160GB Seagate,
an MBR doesn't fix the 'Unmountable Boot Volume' message. As I said, my
AOMEI and GSmartControl (and CHKDSK) tests indicate that it has no
flaws. Can you suggest anything better?


--
Ian

On Sat, 14 Jan 2017 20:44:31 +0000, Ian Jackson
wrote:

In message , Shadow
writes
On Sat, 14 Jan 2017 11:54:25 +0000, Ian Jackson
wrote:

So it seems to me that the problem I am having does seem to associated
with the hard drives (rather than being a driver problem).

No, you booted perfectly in safe mode, no disk errors, so the
drives are working fine.
[]'s

Noted.

Even though an MBR rebuild did fix one of my 160GB Seagates, what about
the suggestions that a low-level format could possibly permanently wreck
some hard drives?

Incidentally, I've just tried yet again to do virgin installation to my
second 160GB Seagate - and unlike the other (cloned-to) 160GB Seagate,
an MBR doesn't fix the 'Unmountable Boot Volume' message. As I said, my
AOMEI and GSmartControl (and CHKDSK) tests indicate that it has no
flaws. Can you suggest anything better?

You did try to install using a slipstreamed XP SP3 disk,as we
suggested ? It's very easy to make with nlite (last version to support
XP was 1.4.9.3)

Win XP (first release) only supported disks up to 137GB.
Support for 137GB was only introduced with SP1.

https://support.microsoft.com/en-us/kb/303013

I was beginning to wonder if the PC had become faulty, so I plugged in
the original old (2001) 120GB Seagate disk (an ST3120026A Barracuda
7200.7) which has accumulated 16 years-worth of junk. It booted OK like
it normally does. [I suppose can now check which drivers are being
used.]

16 years old ? Yes, that would NOT support drives 137GB
A DRIVER problem.
[]'s
--
Don't be evil - Google 2004
We have a new policy - Google 2012

In message , Shadow
writes
On Sat, 14 Jan 2017 20:44:31 +0000, Ian Jackson
wrote:

In message , Shadow
writes
On Sat, 14 Jan 2017 11:54:25 +0000, Ian Jackson
wrote:

So it seems to me that the problem I am having does seem to associated
with the hard drives (rather than being a driver problem).

No, you booted perfectly in safe mode, no disk errors, so the
drives are working fine.
[]'s

Noted.

Even though an MBR rebuild did fix one of my 160GB Seagates, what about
the suggestions that a low-level format could possibly permanently wreck
some hard drives?

Incidentally, I've just tried yet again to do virgin installation to my
second 160GB Seagate - and unlike the other (cloned-to) 160GB Seagate,
an MBR doesn't fix the 'Unmountable Boot Volume' message. As I said, my
AOMEI and GSmartControl (and CHKDSK) tests indicate that it has no
flaws. Can you suggest anything better?

You did try to install using a slipstreamed XP SP3 disk,as we
suggested ? It's very easy to make with nlite (last version to support
XP was 1.4.9.3)

No (see below). But I'll probably make one later, as it could be useful.

Win XP (first release) only supported disks up to 137GB.
Support for 137GB was only introduced with SP1.

The original CD (which would have been used when the PC was born) is XP
Home plus SP1a. However, the CD I've been using for these tests is
usually an XP Home plus SP2.

I get the same problem with either CD when installing to three 40GB
Seagates, one 120GB Seagate, and now three (or is it four?) 160GB
Seagates.

The PC's original HD is a 120GB Seagate - so presumably XP installed OK
in 2001 - and that would be using the XP plus SP1a.

https://support.microsoft.com/en-us/kb/303013

I was beginning to wonder if the PC had become faulty, so I plugged in
the original old (2001) 120GB Seagate disk (an ST3120026A Barracuda
7200.7) which has accumulated 16 years-worth of junk. It booted OK like
it normally does. [I suppose can now check which drivers are being
used.]

16 years old ? Yes, that would NOT support drives 137GB
A DRIVER problem.
[]'s

Just to recap.....
Yesterday, when I cloned the original (and working) 120GB Seagate to one
of my 160GB Seagates, the PC did eventually boot up correctly (it did
need a 'rebuild MBR' to get it to go). But note that this is the same
160GB that, when I had done a normal installation, I got the usual boot
problem. However, when I repeated the exercise with another 160GB drive,
it didn't work.

I also realised that it's only a few weeks ago that I did a normal (and
trouble-free) installation to (yet another) 160GB Seagate (probably
never low-level formatted), and then installed the unofficial SP4. So it
would seem that some of my Seagates ARE OK - despite all of them
apparently having no flaws.

Finally, this morning I did yet another XP Home plus SP2 installation -
this time to a Hitachi 320GB disk (which has not been subjected to a
low-level format). This time, there were no problems.

Anyway, I think I had better put these investigations aside for the
moment. While it is certainly instructive, I've already taken up too
much of my time - and also that of those offering advice (to whom I
express my thanks).


--
Ian