Corrupt NTFS filesystem

Citizen Bob wrote
Rod Speed wrote
Citizen Bob wrote
Rod Speed wrote

You still do not appreciate the situation I am in.

Wrong.

Do you depend on your computer every day to
make money? Or do you just use it for recreation?

Irrelevant to what is possible for YOU to do for a TEST.

It is not irrelevant to me. I use my computer in financial transactions
every day, including weekends. I simply cannot stop using it to run tests.

Irrelevant to whether there are plenty of tests you can do in that situation.

In order to test the new install, I would
have to run it full time for several days.

Wrong.

The problem does not manifest itself for days at a time.

Doesnt mean that you have to run the test drive continuously for that time.

Even you should be able to boot it occasionally over that time.

I would have to install all the apps I normally use to perform a valid test

Wrong when it turns out that you can produce the
corrupted MFT when JUST running ImPerfect Disk.

and I would have to do it for at least 1 week.

Wrong again.

I cannot afford to do that.

You dont need to do that.

Although I do use my computer for recreation
like you, I do use it for finanacial transactions.

So do I thanks.

What am I supposed to do about all the apps I normally
run in the course of a day? I can't just abandon my routine
for a test - I need to run the apps every weekday.

I doubt you actually run all that many of them every weekday

You do not know what you are talking about.

We'll see...

How could you possibly know what I run or do not run every weekday?

I do know that you are very unlikely to actually run the
100s of apps you claim to have installed every weekday.

I do financial transactions every day of the week.
The most intense activity is during the week.

Irrelevant to how many apps that actually involves.

and if you do,

There is no "if I do". I do run financial
transactions every day of the week.

Irrelevant to how many apps that actually involves.

you can certainly do the other test, try with the drive
directly connected instead of in a removable drive bay.

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

I have never experienced a corrupt NTFS partition while running.

You dont know that because you have never actually tested that.

ALL you know is that no entrys have showed up in the EV reporting a problem.

That's also why I do not believe that RAM memory is involved.

See above.

Anyway I have run extensive diagnostics on all hard
drives and RAM and nothing shows any signs of failure.

Irrelevant to what is clearly corrupting the DATA in the MFT.

If I do not install enough apps then I can't run the things I need to run.

I doubt that involves all that many apps,

You do not know what you are talking about.

We'll see...

How could you possibly know what I run or do not run every weekday?

I do know that you are very unlikely to actually run the
100s of apps you claim to have installed every weekday.

I do financial transactions every day of the week.
The most intense activity is during the week.

Irrelevant to how many apps that actually involves.

They take quite a few apps to run in entirety.

Irrelevant to how many apps that actually involves.

and if it does,

There is no "if I do". I do need a lot of apps.

Irrelevant to how many apps that actually involves.

you can certainly do the other test, try with the drive
directly connected instead of in a removable drive bay.

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

I have never experienced a corrupt NTFS partition while running.

You dont know that because you have never actually tested that.

ALL you know is that no entrys have showed up in the EV reporting a problem.

But I plan on doing this test anyway, just to
eliminate the possibility however remote it may be.

Yep, its the only sensible approach because its so easy to do.

How am I going to run two versions of Win2K
on two separate partitions at the same time?

You dont have to run them at the same time.

In order to reproduce the conditions that the corruption
occurs I need to run the test 24x7 for at least a week.

Wrong. You can just run ImPerfect disk on the clean 2K
install since that does corrupt the MFT on that current install.

I cannot reboot anytime or the test will not be valid.

Wrong.

Now tell me, genius, how am I going to run my apps
on the other partition if I am running the test 24x7?

You dont need to do that, stupid.

Then you can obviously install what you do need to run,

If I do that then I just as well do a clean
reinstall and not chase this problem down.

Wrong again.

I may have a cabling problem because of
the location of the internal vs removable bay.

Unlikely given that its easier to cable an internal than a removable bay.

Again you do not know what you are talking about.

We'll see...

I have two bays connected to one cable. The second one is near
the top of the computer, whereas the drive I mount permanently
is closer to the bottom. I may not have enough cable to reach both.

You dont need to have both connected to the cable to do the test.

However as I said, I have some hardware for mounting 3.5"
drives in 5.25" bays, so I can mount the drive next to the
removable bay and circumvent any possible cable problems.

And even if you did need to get another cable for the test, that is
well worth doing because its very likely what is corrupting the MFT.

It is not very likely.

Corse it is.

There is no evidence to support that claim.

Wrong again. When running ImPerfect Disk ALONE corrupts the MFT,
there are only two possibilitys now, either its a ****ed install of 2K that
is the problem, or its a hardware problem, the removable drive bay,
the cable currently being used, or the drive or the controller.

You have an intense bigotry against Centronic-based
removable bays that is obsessing you.

I have seen a number of instances where those have caused
problems, QUITE A FEW OF THEM AT BOOT TIME.

Think about it. If the Kingwin KPF style bays I am using are such
crap as you make them out to be, why are they still on the market?

They arent all used with the same motherboard yours is.

If they work fine with some motherboard and not others,
you'd get that effect, they arent a problem in some configs.

Kingwin is still in business, they are still offering that style bay and
I never hear any complaints about them on this forum or any other.

Even you should be able to find plenty using groups.google.

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

I have never experienced a corrupt NTFS partition while running.

You dont know that because you have never actually tested that.

ALL you know is that no entrys have showed up in the EV reporting a problem.

But I plan on doing this test anyway, just to
eliminate the possibility however remote it may be.

Yep, its the only sensible approach because its so easy to do.

I'd make sure it was a proper legal ATA cable too,
a proper 80 wire flat ribbon cable of legal length. No
point in doing the test with a known non standard cable.

I have stated several times that I have
an official ATA133 80-wire ribbon cable.

Not recently you havent. Even you should have noticed that
I do comment on quite a few system configs over that time.

It has the blue connector on one end to ensure proper orientation for CS.

U check EV all the time and have never seen it.

All that means is that the OS hasnt noticed it until boot time.

However it may be that it is not being detected except at boot.

Precisely.

However if the filesystem is corrupt at run time, how could it even function?

There's plenty of blemishes that still allow it to function.

It was designed to be that robust.

I can't run chkdsk whenever I want - it must be run at boot.

No it doesnt need to to just CHECK for corruption,
only for FIXING any corruption seen.

That only happens at boot time.

Wrong.

Is there some other diagnostic that can detect a corrupt NTFS volume
that I could schedule to run periodically while Win2K is running?

You dont need one, chkdsk can do that fine.

But I have to reboot to run chkdsk.

No you dont.

If I don't want to do something it is not because I am lazy or obstinate

That remains to be seen. You've got one hell
of a capacity for refusing to do the obvious tests.

Please stop with the finger wagging. You are not my wife.

Go and **** yourself.

- it's because I believe I have a good reason not to.

Thats just the excuse for the bone headedness.

That's an ad hominem.

reams of your puerile **** flushed where it belongs

Grow up.

Citizen Bob wrote
Rod Speed wrote
Citizen Bob wrote

Something I just thought of. When I first started using PerfectDisk,
every once in a while it would cause the same corruption problem.
I knew because before I ran PD, I would reboot to make sure the
volume was not corrupt and then I would create a clone backup.
Then I would reboot and run PD and then reboot to see if it
corrupted the disk, Sure enough, a couple times it did and I
had to use either the clone I just made to recover or run chkdsk.

That would seem to indicate that its actually
disk activity that produces the corruption,

I use three identical WD 80 GB drives which I have tested in
so many ways that they are known to be good. I ran a full
SpinRight on each of them overnight. They check out perfectly.

Irrelevant to that point.

If it were the disks

I didnt say it was.

why can I go as long as a week without any problems?

You'll find out when you identify what is producing the corruption.

supporting the possibility that is just something as
basic as the removable drive tray thats the problem.

Because it clearly cant be one of the other apps.

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

I have never experienced a corrupt NTFS partition while running.

You dont know that because you have never actually tested that.

ALL you know is that no entrys have showed up in the EV reporting a problem.

There clearly isnt any other app involved
when an ImPerfect Disk run corrupts the drive.

Bet its the removable drive bay or the cable.

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

I have never experienced a corrupt NTFS partition while running.

You dont know that because you have never actually tested that.

ALL you know is that no entrys have showed up in the EV reporting a problem.

But I plan on doing this test anyway, just to
eliminate the possibility however remote it may be.

Yep, its the only sensible approach because its so easy to do.

Speaking of clones I think I mentioned this but sometimes
you may not have picked up on it. If I put the clone in the
D: without changing the signature with Win98SE fdisk /mbr,
it will always BSOD. That's because Win2K tried to mount
the same device to two disks with identical signatures.

I dont believe that claim about 2K, essentially because clones work
fine for others without that abortion involving Win98SE fdisk /mbr

You are wrong again.

We'll see...

Disk Signature Conflict On Identical Clone Drives

http://www.goodells.net/multiboot/partsigs.htm

Doesnt say anything like your claim above.

What are you doing the cloning with again ?

Acronis True Image. I have a boot CD and it runs the clone offline.

If I use the trick of Win98SE fdisk /mbr on the D: disk,
then I do not get the BSOD. Of course Win2K prompts
me to reboot because it has found a "new device".

So let's imaging the scenario where I have a clone in archive

Not clear what you mean by that last.

I always have two clone disks in archive since I have three
identical disks. One is very recent and the other is less recent.

OK, I realised that you had that many physical drives,
just wouldnt have called that an archive myself.

which I use as the boot disk when the original
disk gets corrupted. I mount this archived clone

Or that either.

I just meant the use of the word 'archived' there.

If I get a corrupt disk that cannot be repaired with the automatic
chkdsk that runs at boot time, I have to mount it as D:. So I use the
most recent clone as the boot disk in C:. But they have the same
signature, so get a BSOD. That's why I have to use Win98SE to replace
the first 4 bytes of the signature with zeros, which forces Win2K to
remount it internally.

as the boot disk and mount the bad disk as D: so I can run
chkdsk d: /f on it. Since I changed the signature on the bad
disk to prevent the device conflict and BSOD, and if I don't
reboot to satisfy Win2K's request for a new device, then
chkdsk will screw up the bad disk and it is not recoverable.
IOW, it can't even be mounted any more.

Cant really understand what you mean
config wise with those 'archive' comments.

Archive means the disk sits on a shelf away from the
computer. That's why I use removable drive bays.

On Sun, 29 Oct 2006 00:31:30 -0400, kony wrote:

This idea you have about how valuable your system use is, is
exactly WHY you should never be relying on a system in this
state.

I have two clones, and I know how to recover. That gets me by.

I make sure that I have at least one archive disk (disk on the shelf)
that works. The way I know it works is I reboot it and if it works I
immediately clone it, then I put it on the shelf. Then I boot the
clone I just made and if it works too I know the original I just put
on the shelf works.

If something in that sequence goes wrong, I immediately fix it. For
example if the clone is broken, I fix it and use it to fix the source
from which it is made, then I start over with the source disk and redo
the above procedure.

Yeah, I know - it's a pain in the ass. That's why I am willing to try
your "clean install" procedure after I make damn sure it is going to
work.



--

"Nothing in the world can take the place of perseverence. Talent
will not; nothing is more common than unsuccessful men with talent.
Genius will not; unrewarded genius is almost a proverb. Education
will not; the world is full of educated derelicts. Persistence and
determination alone are omnipotent."
--Calvin Coolidge

On Sun, 29 Oct 2006 16:22:30 +1100, "Rod Speed"
wrote:

The problem does not manifest itself for days at a time.

Doesnt mean that you have to run the test drive continuously for that time.
Even you should be able to boot it occasionally over that time.

I have done that as a test - reboot an uncorrupt disk several times
during the day. That's exactly what I did with both the new FAT32 and
the new NTFS I made the other day. Both passed the test every time.

I did not run the FAT32 for a week so I will not know if it would have
worked. I may go back to FAT32 when I figure out what would happen if
one of my DVD applications built a temp file that is larger than 4 GB.
For all I know, that never happens.

The main reason I went back to NTFS was a comment you made (actually
it was one of your famous pontifications) that if I converted the
FAT32 to a new NTFS filesystem it would not get corrupted.
Unfortunately you were wrong, because when I let the system go about 3
days between reboots, it got corrupted.

I can convert over to FAT32 again (I can do it in my sleep now).

I would have to install all the apps I normally use to perform a valid test

Wrong when it turns out that you can produce the
corrupted MFT when JUST running ImPerfect Disk.

and I would have to do it for at least 1 week.

Wrong again.

You are assuming that the corruption occurs while running instead of
when I shut down for a reboot. There is no evidence to support that.

How could you possibly know what I run or do not run every weekday?

I do know that you are very unlikely to actually run the
100s of apps you claim to have installed every weekday.

That does not mean I do not run many of them.

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

Actually I suspect the corruption occurs at shutdown. I even tried
disabling the write thru cache but it did not help. Then I tried
setting the Registry key that tells Win2K to do a cleanup of the
system files at shutdown, but that did not work either.

I have never experienced a corrupt NTFS partition while running.

You dont know that because you have never actually tested that.
ALL you know is that no entrys have showed up in the EV reporting a problem.

I agree and that's why I will run ImPerfect Disk when you tell me
where to get it.

Irrelevant to what is clearly corrupting the DATA in the MFT.

Is that where Win2K keeps the so-called "security descripters"?

BTW, when a corrupt disk occurs and Win2K runs chkdsk at boot time
automatically, the repair is almost always the same, which seems to
indicate that whatever is causing the corruption is almost always the
same thing.

When chkdsk at reboot is unable to clean up the mess (and I get a
BSOD), or I get a BSOD before it even gets to run chkdsk - and I have
to mount the disk as D: to run chkdsk from inside Win2K, the repair is
considerably more extensive. Chkdsk fills several screens with repair
comments, most of which is fixing security descripters.

I have run System File Checker (SFC) a couple of times in the outside
chance that I had a virus. I have swept the disk several times with
Ad-Aware and Avast. No viruses detected.

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

Yes, but there is even more disk activity when I am running yet there
is no indication of disc corruption.

Actually that is not completely true. There was a brief period when
the system would reboot itself while running. I thought it could be
from a corrupt disk during run time. But it did not run chkdsk when it
rebooted and there was no EV record about a corrupt NTFS volume. I
even forced chkdsk on reboot and nothing was wrong.

But that random rebooting went away months ago and I have never
experienced it since.

But I plan on doing this test anyway, just to
eliminate the possibility however remote it may be.

Yep, its the only sensible approach because its so easy to do.

But first I want to run ImPerfect Disk.

Wrong again. When running ImPerfect Disk ALONE corrupts the MFT,
there are only two possibilitys now, either its a ****ed install of 2K that
is the problem, or its a hardware problem, the removable drive bay,
the cable currently being used, or the drive or the controller.

You want me to run ImPerfect Disk by itself? I can do that overnight
for 12 hours.

I have seen a number of instances where those have caused
problems, QUITE A FEW OF THEM AT BOOT TIME.

Then I need to connect the boot disk directly. I would be very happy
of that turns out to be the problem because I can work around it with
my cloning system. I would have to tell the BIOS to boot off of a disk
in the removable bay so the direct-connected one is D: and therefore I
can run chkdsk on it from inside Win2K.

However if the filesystem is corrupt at run time, how could it even function?

There's plenty of blemishes that still allow it to function.

It was designed to be that robust.

Especially since I am running a 2 GB pagefile in memory. I still think
it's during shutdown that the problem occurs. If I had the time and
patience, I would take every disk I shutdown and before I rebooted it,
I would mount it as D: so I could run chkdks on it. That way I would
find out if the problem occurs at reboot. But that is a lot of work
and I would rather dedicate my limited resources to things that are
more direct.

But I have to reboot to run chkdsk.

No you dont.

How do you propose to run chkdsk without rebooting or without
remounting the disk as D:?

Please stop with the finger wagging. You are not my wife.

Go and **** yourself.

Jeez, you can sure disk it out, but you can't take it. That's a sure
sign of a brittle overblown ego.

Thats just the excuse for the bone headedness.

That's an ad hominem.

reams of your puerile **** flushed where it belongs

Jeez, you can sure disk it out, but you can't take it. That's a sure
sign of a brittle overblown ego.

Grow up.

You grow up - you need to a lot more than me.


--

"Nothing in the world can take the place of perseverence. Talent
will not; nothing is more common than unsuccessful men with talent.
Genius will not; unrewarded genius is almost a proverb. Education
will not; the world is full of educated derelicts. Persistence and
determination alone are omnipotent."
--Calvin Coolidge

On Sun, 29 Oct 2006 16:34:13 +1100, "Rod Speed"
wrote:

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

My vote is that the corruption occurs during shutdown, when Win2K
writes the memory-resident part to the system files, the pagefile and
the MFT.

Disk Signature Conflict On Identical Clone Drives

http://www.goodells.net/multiboot/partsigs.htm

Doesnt say anything like your claim above.

It explains why I must use Win98SE fdisk to clear the signature.

OK, I realised that you had that many physical drives,
just wouldnt have called that an archive myself.

It's shorter than "removable disk I put on the shelf".

I just meant the use of the word 'archived' there.

You speak Oz English, which is like Pom English. I speak Real English,
the same as most of the world's computers. The Real English meaning of
"archive" is found in an Real English American dictionary like
Websters Online:

archive: the material preserved


--

"Nothing in the world can take the place of perseverence. Talent
will not; nothing is more common than unsuccessful men with talent.
Genius will not; unrewarded genius is almost a proverb. Education
will not; the world is full of educated derelicts. Persistence and
determination alone are omnipotent."
--Calvin Coolidge

On Sun, 29 Oct 2006 00:58:37 -0400, kony wrote:

A few of them would tend to be
HKLM-Software, HKCU-Software, and you mentioned classes so
HKCR.

"A few of them" tells me there are more.

Yes that's quite possible.
As I already wrote, the key to doing it is NOT what you are
doing, not trying to think on things.
The key is to actually do it.
Actually.
Not think, do.

I do not have time to run around on merry chases. Either I know what I
am going to do has a good chance of success or I pass it by.

You are spending hours "thinking" on things then telling us
you haven't the spare time to do what wouldn't have taken
this long.

I am doing that in hope of running this problem down. I am confident
that you experts will come across something along the way that points
us to the problem. We already have a lot of data, but there are a
couple things still missing.

As already written, this is a bulk process to get most
things working. Some may not work. Maybe you install 5
things again, or maybe you fire up sysinternal's regmon and
see that when you launch the app, it's trying to read a
particular registry entry, so you merely export that parent
key and merge it.

I just as well reinstall Windows and all my apps as do that.

These are basic concepts, which take more time to think on,
than to do. You may not realize this, and that is why I
continually stress not doing what you are doing, which is
anything except the productive path to get it done. I have
been down this road and have advised what addresses your
expressed need, to have minimal time spent, while you
continue to do the opposite, making it the most drawn out
process possible.

You must think I sit around looking for things to do. That is not the
case. I am always busy with something.

If I had more confidence that this Registry stunt of yours would
really work, then I would give it a try. But it sounds like you are
just throwing **** on the wall in the outside chance it will work,
maybe. I need more confidence before I embark on a long project such
as we discussed.

My problem with what you propose - exporting three Registry keys - is
that the Registry has a lot more configuration information that is
specific to applications than just 3 keys. If I don't export that,
then I am not going to get a "clean install", as you call it. The
search for more keys could take days. Then I could leave behind some
keys that I do not discover are missing until months later, in which
it is too late to go back and export them because the apps have
changed their configuration and the exported keys do not have that
information.

Whoever came up with that Registry crap should be executed so his
screwed up genes do not contaminate the human race. People do not have
this kind of nightmare to deal with on UNIX, because configurations
are file-based. It is much easier to deal with a flat file than a data
base.

Those phrases told me you were not sure, so I didn't take them as a
final statement.

Sure of what? I'm sure you need those keys and I'm sure
it's not guaranteed to make 100% of your apps work. This
was already written, that it is a bulk transfer to get the
majority working, then anything remaining will indicate what
to do next, whether it be more registry entries or files,
but each thing done in turn, NOT trying to do everything at
once is the key.

It is important NOT to do everything at once, because we are
trying to isolate the problem, not duplicate the old
installation perfectly which would naturally reproduce the
problem. Thus, the prudent approach is going to be a
conservative transferral of each type of setting, file, etc.

That makes more sense than your earlier terse comments. But it still
involves a lot of work chasing after things that I know nothing about.

HKCU-software, HKLM- software, HK-Classes-Root.
Is that correct?

Yes, export each of these, but not merging them. Get new
installation working 100% first.

Of course.

Did you do a clean installation yet?

I have not done anything yet because I want to be certain what I am
going to do.

What you need to do is to NOT try to think ahead. It is a
fluid process and you may need adapt to what happens. For
instance, after merging registry keys you might launch an
app and get a message like vbrun*.dll not found (or
similar), meaning you need to install MS's visual basic
package.

So in this example, you might google search;

http://www.google.com/search?q=Windo...basic+download

... and the first hit is the page to download it, then
install.

I kept every app and its support files in a ZIP directory. However if
I did have to reinstall, I would consider getting the latest version
so I can at least be updated.

you only have 3 options left:

You left out going back to a FAT32 system and hoping none of my apps
ever builds a 4 GB file.

Which is what I may do because I never really tested it.

In fact that is exactly what I am going to do before I do anything
else. I can do a FAT32 conversion in my sleep, so it's no big deal. I
actually have the last one but it is dated by now so it would be
better just to make one with this current version of the operating
system and applications. I have installed a lot of new stuff for my
JP1 project I am working on now, including an update of Java.

Let's run the FAT32 for a full week or two and see what happens. I
realize you don't care for FAT32 because it can lead to lost files.
But that's with Win9x or WinME, not Win2K.


--

"Nothing in the world can take the place of perseverence. Talent
will not; nothing is more common than unsuccessful men with talent.
Genius will not; unrewarded genius is almost a proverb. Education
will not; the world is full of educated derelicts. Persistence and
determination alone are omnipotent."
--Calvin Coolidge

On Sun, 29 Oct 2006 01:11:04 -0400, kony wrote:

I will then export the above mentioned keys and save the exports for
later use. I will then copy the new Registry in F: in entirety to the
current partition D:, thereby replacing the entire Registry on D:. I
can copy anything else from the new install F: you think is important.

That is not what I'd suggested, and not what I'd do, but if
it's what you want to do, go ahead- it's a clone so if one
way doesn't work you can always try another.

Then I can import the exported keys into the new Registry on D:.

That way I will have a new install of Win2K without having to copy all
the apps and other stuff.

No, you will have the old install of Win2k, and years worth
of clutter, then merely a slimmer registry. This is
exactly what we wanted to avoid, but maybe you get lucky and
find it (remove the problem) this way regardless.

IMO, the way you're doing it is worse,
I agree. Forget I mentioned it. If I do a clean install I will have to
do it your way to avoid carrying over any contamination.

However, I would rather just do a new install including all my apps
than to go thru all those contortions. That would guarantee no
contamination.

Of course if I did it that way I would first copy over the Profiles
and install a minimum number of apps to keep productive for my most
important projects. The other stuff I would have to put on hold.

But them if I do that, why not install XP Pro and use its wizard to
move everything from Win2K. I suppose I can learn to live with XP if I
use the compatibility mode. I was wanting to wait for Vista but I
never install a new version of Windows without the first SP out and
tested, and that could be a couple years with Vista. Therefore I
should consider XP for the interim. My son has been running XP for
years and he likes it, so it can't be all that bad.

Then I can slowly install my other apps, taking the opportunity to
upgrade them as I install each one.

It sounds like a plan, but first I want to squeeze the last bit of
life out of Win2K. That's why I am going to convert back to FAT32 and
give it a chance to hang itself by running it a couple weeks.


--

"Nothing in the world can take the place of perseverence. Talent
will not; nothing is more common than unsuccessful men with talent.
Genius will not; unrewarded genius is almost a proverb. Education
will not; the world is full of educated derelicts. Persistence and
determination alone are omnipotent."
--Calvin Coolidge

Citizen Bob wrote
kony wrote

This idea you have about how valuable your system use is, is
exactly WHY you should never be relying on a system in this state.

I have two clones, and I know how to recover. That gets me by.

For now, until the MFT gets so corrupted that even chkdsk cant
save you and you lose whats happened since the last clone.

I make sure that I have at least one archive disk (disk on the shelf)
that works. The way I know it works is I reboot it and if it works I
immediately clone it, then I put it on the shelf. Then I boot the
clone I just made and if it works too I know the original I just put
on the shelf works.

Pity about what you do on that system since you do that.

If something in that sequence goes wrong, I immediately fix it.

Remains to be seen if that is always possible.

For example if the clone is broken, I fix it and use it to
fix the source from which it is made, then I start over
with the source disk and redo the above procedure.

Yeah, I know - it's a pain in the ass. That's why I am willing to try your
"clean install" procedure after I make damn sure it is going to work.

We'll see...

Citizen Bob wrote
Rod Speed wrote

The problem does not manifest itself for days at a time.

Doesnt mean that you have to run the test drive continuously for that
time. Even you should be able to boot it occasionally over that time.

I have done that as a test - reboot an uncorrupt disk several times
during the day. That's exactly what I did with both the new FAT32 and
the new NTFS I made the other day. Both passed the test every time.

Doesnt prove much about a problem that you claim can take a week to manifest.

I did not run the FAT32 for a week so I will not know if it would
have worked. I may go back to FAT32 when I figure out what
would happen if one of my DVD applications built a temp file
that is larger than 4 GB. For all I know, that never happens.

Digital TV capture cards will do that, guaranteed.

The main reason I went back to NTFS was a comment you made
(actually it was one of your famous pontifications) that if I converted
the FAT32 to a new NTFS filesystem it would not get corrupted.

I never ever said anything even remotely resembling anything like that.

Unfortunately you were wrong,

Nope, because I never ever said anything
even remotely resembling anything like that.

because when I let the system go about
3 days between reboots, it got corrupted.

All that proves is that the fault has nothing
to do with the file system being used.

I can convert over to FAT32 again (I can do it in my sleep now).

Must be one of those rocket scientist boneheads.

I would have to install all the apps
I normally use to perform a valid test

Wrong when it turns out that you can produce the
corrupted MFT when JUST running ImPerfect Disk.

and I would have to do it for at least 1 week.

Wrong again.

You are assuming that the corruption occurs while
running instead of when I shut down for a reboot.

Nope, that claim you made shows that it cant be
a specific app thats corrupting the MFT since you
claimed that you saw that corruption before running PD.

That is one possibility completely eliminated.

There is no evidence to support that.

Having fun thrashing that straw man ?

How could you possibly know what I run or do not run every weekday?

I do know that you are very unlikely to actually run the
100s of apps you claim to have installed every weekday.

That does not mean I do not run many of them.

Never ever said it did.

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

Actually I suspect the corruption occurs at shutdown. I even
tried disabling the write thru cache but it did not help. Then I
tried setting the Registry key that tells Win2K to do a cleanup
of the system files at shutdown, but that did not work either.

And I have told you how to prove whether
its just bootup or shutdown thats the problem.

I have never experienced a corrupt NTFS partition while running.

You dont know that because you have never actually tested that. ALL
you know is that no entrys have showed up in the EV reporting a problem.

I agree and that's why I will run ImPerfect Disk when you tell me where to get it.

Pathetic, really.

Irrelevant to what is clearly corrupting the DATA in the MFT.

Is that where Win2K keeps the so-called "security descripters"?

BTW, when a corrupt disk occurs and Win2K runs chkdsk
at boot time automatically, the repair is almost always the
same, which seems to indicate that whatever is causing
the corruption is almost always the same thing.

It would be a hell of a lot more surprising if it wasnt.

When chkdsk at reboot is unable to clean up the mess (and I get a
BSOD), or I get a BSOD before it even gets to run chkdsk - and I have
to mount the disk as D: to run chkdsk from inside Win2K, the repair is
considerably more extensive. Chkdsk fills several screens with repair
comments, most of which is fixing security descripters.

I have run System File Checker (SFC) a couple of times in
the outside chance that I had a virus. I have swept the disk
several times with Ad-Aware and Avast. No viruses detected.

It wont be a virus.

Why would the removable bay corrupt an NTFS partition only at boot time?

Because even you should have noticed considerable drive activity at boot time.

Yes, but there is even more disk activity when I am running yet

Wrong.

there is no indication of disc corruption.

You dont know that because you have never actually tested that.

ALL you know is that no entrys have showed up in the EV reporting a problem.

Actually that is not completely true. There was a brief period when
the system would reboot itself while running. I thought it could be
from a corrupt disk during run time. But it did not run chkdsk when it
rebooted and there was no EV record about a corrupt NTFS volume.
I even forced chkdsk on reboot and nothing was wrong.

But that random rebooting went away months
ago and I have never experienced it since.

Unlikely to be relevant.

But I plan on doing this test anyway, just to
eliminate the possibility however remote it may be.

Yep, its the only sensible approach because its so easy to do.

But first I want to run ImPerfect Disk.

Pathetic, really.

Wrong again. When running ImPerfect Disk ALONE corrupts the
MFT, there are only two possibilitys now, either its a ****ed install
of 2K that is the problem, or its a hardware problem, the removable
drive bay, the cable currently being used, or the drive or the controller.

You want me to run ImPerfect Disk by itself?

Nope, I want you to connect the boot drive directly, and run PD
repeatedly and see if you get any corruption of the MFT when you do that.

I can do that overnight for 12 hours.

It would be a better test to check for corruption of the MFT after each PD run.

I have seen a number of instances where those have caused
problems, QUITE A FEW OF THEM AT BOOT TIME.

Then I need to connect the boot disk directly.

What I said months ago.

I would be very happy of that turns out to be the problem because
I can work around it with my cloning system. I would have to tell the
BIOS to boot off of a disk in the removable bay so the direct-connected
one is D: and therefore I can run chkdsk on it from inside Win2K.

Pointless worrying about what to do until you
work out what is producing the corruption.

However if the filesystem is corrupt at run time, how could it even function?

There's plenty of blemishes that still allow it to function.

It was designed to be that robust.

Especially since I am running a 2 GB pagefile in memory.

Fark. What else are you doing like that that you havent even mentioned ?

I still think it's during shutdown that the problem occurs. If I had
the time and patience, I would take every disk I shutdown and
before I rebooted it, I would mount it as D: so I could run chkdks
on it. That way I would find out if the problem occurs at reboot.
But that is a lot of work and I would rather dedicate my limited
resources to things that are more direct.

Doesnt matter when it happens, what matters is what is causing it.

But I have to reboot to run chkdsk.

No you dont.

How do you propose to run chkdsk without
rebooting or without remounting the disk as D:?

You cant actually be THAT stupid.

Please stop with the finger wagging. You are not my wife.

Go and **** yourself.

Jeez, you can sure disk it out, but you can't take it.

I can take it fine, I choose to tell you go and ****
yourself when you try puerile stuff like that.

reams of your puerile **** flushed where it belongs

Citizen Bob wrote
Rod Speed wrote

Why would the removable bay corrupt
an NTFS partition only at boot time?

Because even you should have noticed
considerable drive activity at boot time.

My vote is that the corruption occurs during shutdown,

Irrelevant whether its shutdown or bootup, what matters is why it happens.

when Win2K writes the memory-resident part
to the system files, the pagefile and the MFT.

It doesnt do that either.

Disk Signature Conflict On Identical Clone Drives

http://www.goodells.net/multiboot/partsigs.htm

Doesnt say anything like your claim above.

It explains why I must use Win98SE fdisk to clear the signature.

No it doesnt.

OK, I realised that you had that many physical drives,
just wouldnt have called that an archive myself.

It's shorter than "removable disk I put on the shelf".

But less obvious what you meant.

I just meant the use of the word 'archived' there.

You speak Oz English, which is like Pom English.

Wrong, as always.

I speak Real English,

Wrong, as always.

the same as most of the world's computers. The Real
English meaning of "archive" is found in an Real English
American dictionary like Websters Online:

archive: the material preserved

Pathetic, really.