Need urgent help: can't log onto system partition

I

Industrial One

It hangs on the log-on screen and I am provided no module to log in
with like normal. http://i49.tinypic.com/i739ug.jpg

This all happened after a series of failed attempts to clone current
drive to a bigger 2 TB drive. Someone instructed me to make sure the
first 512 bytes of the cloned drive is the same as the source drive.
So when I modified the DIFFERENT drive and restarted computer to see
if I could boot to it now (which I couldn't) I couldn't log on to THIS
drive.

How in the sam hell does modifying the OTHER drive screw up THIS
drive?

Note there is no recovery console or repair option on my Windows XP CD
for some reason, so I really have no idea what to do. I tried
replacing the registry hives with 4-month-old backups and that failed.
I can't boot into safe mode because I can't log in.

I've no idea what's going on.

Thanks in advance for any help.
 
C

Char Jackson

It hangs on the log-on screen and I am provided no module to log in
with like normal. http://i49.tinypic.com/i739ug.jpg

This all happened after a series of failed attempts to clone current
drive to a bigger 2 TB drive. Someone instructed me to make sure the
first 512 bytes of the cloned drive is the same as the source drive.

Very bad advice, I'd say.
So when I modified the DIFFERENT drive and restarted computer to see
if I could boot to it now (which I couldn't) I couldn't log on to THIS
drive.

Disconnect the 2TB drive for now. Ensure that your system once again
works normally. Then do a proper clone operation, with none of the
manual editing that you apparently did before. It's not necessary.

What are you using to make the clone?
 
P

Paul

Industrial said:
It hangs on the log-on screen and I am provided no module to log in
with like normal. http://i49.tinypic.com/i739ug.jpg

This all happened after a series of failed attempts to clone current
drive to a bigger 2 TB drive. Someone instructed me to make sure the
first 512 bytes of the cloned drive is the same as the source drive.
So when I modified the DIFFERENT drive and restarted computer to see
if I could boot to it now (which I couldn't) I couldn't log on to THIS
drive.

How in the sam hell does modifying the OTHER drive screw up THIS
drive?

Note there is no recovery console or repair option on my Windows XP CD
for some reason, so I really have no idea what to do. I tried
replacing the registry hives with 4-month-old backups and that failed.
I can't boot into safe mode because I can't log in.

I've no idea what's going on.

Thanks in advance for any help.

You can make exact copies of drives.

The trick is, to boot the clone *by itself*, immediately after
the exact copy is made.

*******

Not all copy methods are exact copies.

Some copying or cloning methods, will resolve the identity of
the partition and drive, such that there is no upset.

But then, if that happens, the second drive may need to be
"activated".

The first time a cloned drive is booted, if it's an exact copy,
it should be booted by itself. When I've done that here, no
activation was needed.

If I boot the clone, while the original is present, that doesn't
work. My repair procedure at that point (easiest for me), is to
clone the original drive again. And then, remember to boot it
by itself the first time it is booted.

Once a cloned drive is booted by itself, if you wish to connect
the original again, you can. But then, Windows will resolve any
identical things it sees, to its own satisfaction. If it can stand
to see an exact copy, well, it'll leave it alone. If an identifier
is obnoxious enough to bother it, then it gets resolved.

My guess is, if you boot the clone drive, while the original is present,
the clone will start using the pagefile on the original, or do something
else along those lines. Something gets confused, but I don't know
exactly what. Perhaps that's described somewhere.

The purpose of taking care, when cloning, is if you want to avoid
activation. For example, if the reason for copying a disk, is so
you can do a dangerous experiment, then you don't want Microsoft
seeing an activation attempt once a day, for your once-a-day
experiments. The idea behind working out a good method for yourself,
is so the activation file on the disk, remains valid.

One parameter which will be changed, is the physical disk serial number.
So in terms of the activation voting process, that number will be different,
and counts as "one against you". It takes more than that, to tip over
the activation.

http://aumha.org/win5/a/wpa.htm

What's happened to you now, is not activation. Whatever booting a clone,
with original disk present does, screws up something else.

Once a clone has been booted by itself *once*, it is then safe to connect
the other disk. As near as I can tell.

Paul
 
I

Industrial One

Very bad advice, I'd say.


Disconnect the 2TB drive for now. Ensure that your system once again
works normally. Then do a proper clone operation, with none of the
manual editing that you apparently did before. It's not necessary.

What are you using to make the clone?

It already is disconnected. And my system is NOT once again working normally, that's exactly why I made the post. I cannot log into my system.

Once again, my Windows XP installation CD is being worthless and not recognizing the already-installed Windows system on it, thus I have no repair or recovery console. Why the hell is this?

I have a project to do and need my system working again ASAP.
 
I

Industrial One

You can make exact copies of drives.

The trick is, to boot the clone *by itself*, immediately after
the exact copy is made.

Did that, it failed. Now the source drive is ****ed as well for reasons that defy all logic.
*******

Not all copy methods are exact copies.

Some copying or cloning methods, will resolve the identity of
the partition and drive, such that there is no upset.

But then, if that happens, the second drive may need to be
"activated".

The first time a cloned drive is booted, if it's an exact copy,
it should be booted by itself. When I've done that here, no
activation was needed.

If I boot the clone, while the original is present, that doesn't
work. My repair procedure at that point (easiest for me), is to
clone the original drive again. And then, remember to boot it
by itself the first time it is booted.

I can't clone a drive that I don't have access to now.
Once a cloned drive is booted by itself, if you wish to connect
the original again, you can. But then, Windows will resolve any
identical things it sees, to its own satisfaction. If it can stand
to see an exact copy, well, it'll leave it alone. If an identifier
is obnoxious enough to bother it, then it gets resolved.

My guess is, if you boot the clone drive, while the original is present,
the clone will start using the pagefile on the original, or do something
else along those lines. Something gets confused, but I don't know
exactly what. Perhaps that's described somewhere.

The purpose of taking care, when cloning, is if you want to avoid
activation. For example, if the reason for copying a disk, is so
you can do a dangerous experiment, then you don't want Microsoft
seeing an activation attempt once a day, for your once-a-day
experiments. The idea behind working out a good method for yourself,
is so the activation file on the disk, remains valid.

One parameter which will be changed, is the physical disk serial number.
So in terms of the activation voting process, that number will be different,
and counts as "one against you". It takes more than that, to tip over
the activation.

http://aumha.org/win5/a/wpa.htm

What's happened to you now, is not activation. Whatever booting a clone,
with original disk present does, screws up something else.

I never did that.
Once a clone has been booted by itself *once*, it is then safe to connect
the other disk. As near as I can tell.

Paul

It never was bootable to begin with, that's why as a last resort I took that idiot's advice and copied THIS now-inaccessible hard drive's MBR on the clone, and somehow that screwed THIS drive up, not made it unbootable but made it unlogin-able. I'll never be able to figure out the logic of what the hell went wrong there.
 
I

Industrial One

Very bad advice, I'd say.


Disconnect the 2TB drive for now. Ensure that your system once again
works normally. Then do a proper clone operation, with none of the
manual editing that you apparently did before. It's not necessary.

It already is disconnected, and my system is NOT working normally, that's why I made the post asking for help since my piece of shit Windows XP CD is refusing to recognize the windows installation that exists on the drive andthus I cant repair or use a recovery console, they dont show up.
 
C

Char Jackson

It never was bootable to begin with, that's why as a last resort I took that idiot's advice and copied THIS now-inaccessible hard drive's MBR on the clone, and somehow that screwed THIS drive up, not made it unbootable but made it unlogin-able. I'll never be able to figure out the logic of what the hell went wrong there.

I wonder if you copied the MBR in the wrong direction. It's an easy
mistake to make.
 
I

Industrial One

I wonder if you copied the MBR in the wrong direction. It's an easy
mistake to make.

I didn't, I'm positive of that. Even if I did, MBR doesnt seem to be the problem considering I am able to boot just not log in.
 
P

Paul

Industrial said:
I didn't, I'm positive of that. Even if I did, MBR doesnt seem to be the problem considering I am able to boot just not log in.

Well, you have nothing to lose now.

Welcome to Testdisk.

Testdisk is a tool that "recomputes" the MBR. All it does, is work out
the primary partition table entries. It scans the disk, on track boundaries,
looking for partition headers. A 2TB disk, is going to take a long time...
Of course, everything on a 2TB disk takes a long time.

http://www.cgsecurity.org/wiki/TestDisk_Step_By_Step

You, the user, have to judge the answer it comes up with.

First, write down the "current" partition table info.

Connect the borked drive, to a working PC, and use PTEDIT32 to read out
the MBR. On Windows 7, you must Run As Administrator or otherwise PTEDIT32
gets an "error 5". If you don't edit anything in here, it's not
going to change anything.

ftp://ftp.symantec.com/public/english_us_canada/tools/pq/utilities/PTEDIT32.zip (free download)

Your disk, before using TestDisk, will look like this in PTEDIT32.
This is a three partition disk (three primary partitions). The
display shouldn't be red, as the red color is if you attempt to
edit the partition table (don't do that just yet). In this example,
the boot flag is set on the second partition. Only the Windows boot
loader cares about the boot flag - things like Linux GRUB don't
refer to it.

http://www.goodells.net/dellrestore/files/dell-tbl.gif

When you run TestDisk, it's going to attempt to identify the
partitions on the disk. Compare the "new" computed value,
against the PTEDIT32 information.

Perhaps that will tell you, whether the "structure" is damaged.
If Testdisk shows the same info as PTEDIT32, then there's something
else wrong with the content of C:.

If you erase a partition, then run Testdisk, it'll find the erased
partition. This is why you have to be *real* careful with it,
and not accept verbatim, what it finds. If you know the disk has
three partitions, don't accept a "four partition answer". Successful
usage of TestDisk, requires user prior knowledge. TestDisk isn't
very good at guessing.

For WinXP to boot, you'd need the boot flag set on the appropriate
partition. The boot.ini has an ARC (disk path specification), and
that specification has to be consistent with the partition number
that the partition is currently on. If you use a "partition editor"
for cloning, sometimes it can switch entries (move partition 2 to
partition 3 table entry), and then you'll get an appropriate error
at boot time.

Yours isn't doing that, so I doubt that's the problem.

But if you want to check for "MBR damage", you can slave
the borked drive to a working PC, run testdisk and check the
2TB drive, compare against what PTEDIT32 says.

And if it looks like that will take too long, you can just
admire the info PTEDIT32 is showing you. That will only
take you a couple minutes to verify.

If PTEDIT32 is upset with what it finds, I suppose even that
is an answer. It would imply the MBR got "blasted". Maybe there
isn't even an MBR there any more. Testdisk can work out a new
one, but worst case, you'd need to do a FixMBR from recovery
console if the 440 byte MBR boot code was also ruined.
(On a data only disk, the 440 byte area is a don't care.)

MBR = 440 byte boot code (different for each bootable OS)
4*16 bytes primary partition table (includes boot flags)
disk signature byte etc.
Total = 512 bytes

I'm not aware of any tool, that checks or verifies the 440 byte
boot code, or can tell you what's currently loaded. These OSes
have weaknesses in certain areas, when it comes to forensics.
Some things are harder to work on, than others.

*******

Nothing in your symptoms so far, points in the direction of
the MBR. The above is presented to satisfy your curiosity.

HTH,
Paul
 
I

Industrial One

Well, you have nothing to lose now.

Welcome to Testdisk.

Testdisk is a tool that "recomputes" the MBR. All it does, is work out
the primary partition table entries. It scans the disk, on track boundaries,
looking for partition headers. A 2TB disk, is going to take a long time....
Of course, everything on a 2TB disk takes a long time.

http://www.cgsecurity.org/wiki/TestDisk_Step_By_Step

You, the user, have to judge the answer it comes up with.

First, write down the "current" partition table info.

Connect the borked drive, to a working PC, and use PTEDIT32 to read out
the MBR. On Windows 7, you must Run As Administrator or otherwise PTEDIT32
gets an "error 5". If you don't edit anything in here, it's not
going to change anything.

ftp://ftp.symantec.com/public/english_us_canada/tools/pq/utilities/PTEDIT32.zip (free download)

Your disk, before using TestDisk, will look like this in PTEDIT32.
This is a three partition disk (three primary partitions). The
display shouldn't be red, as the red color is if you attempt to
edit the partition table (don't do that just yet). In this example,
the boot flag is set on the second partition. Only the Windows boot
loader cares about the boot flag - things like Linux GRUB don't
refer to it.

http://www.goodells.net/dellrestore/files/dell-tbl.gif

When you run TestDisk, it's going to attempt to identify the
partitions on the disk. Compare the "new" computed value,
against the PTEDIT32 information.

Perhaps that will tell you, whether the "structure" is damaged.
If Testdisk shows the same info as PTEDIT32, then there's something
else wrong with the content of C:.

If you erase a partition, then run Testdisk, it'll find the erased
partition. This is why you have to be *real* careful with it,
and not accept verbatim, what it finds. If you know the disk has
three partitions, don't accept a "four partition answer". Successful
usage of TestDisk, requires user prior knowledge. TestDisk isn't
very good at guessing.

For WinXP to boot, you'd need the boot flag set on the appropriate
partition. The boot.ini has an ARC (disk path specification), and
that specification has to be consistent with the partition number
that the partition is currently on. If you use a "partition editor"
for cloning, sometimes it can switch entries (move partition 2 to
partition 3 table entry), and then you'll get an appropriate error
at boot time.

Yours isn't doing that, so I doubt that's the problem.

But if you want to check for "MBR damage", you can slave
the borked drive to a working PC, run testdisk and check the
2TB drive, compare against what PTEDIT32 says.

And if it looks like that will take too long, you can just
admire the info PTEDIT32 is showing you. That will only
take you a couple minutes to verify.

If PTEDIT32 is upset with what it finds, I suppose even that
is an answer. It would imply the MBR got "blasted". Maybe there
isn't even an MBR there any more. Testdisk can work out a new
one, but worst case, you'd need to do a FixMBR from recovery
console if the 440 byte MBR boot code was also ruined.
(On a data only disk, the 440 byte area is a don't care.)

MBR = 440 byte boot code (different for each bootable OS)
4*16 bytes primary partition table (includes boot flags)
disk signature byte etc.
Total = 512 bytes

I'm not aware of any tool, that checks or verifies the 440 byte
boot code, or can tell you what's currently loaded. These OSes
have weaknesses in certain areas, when it comes to forensics.
Some things are harder to work on, than others.

*******

Nothing in your symptoms so far, points in the direction of
the MBR. The above is presented to satisfy your curiosity.

HTH,
Paul

I don't care for the clone right now. My SOURCE disk is ****ed. The fact that a healthy drive cloned to an empty new drive ended up borked tells me the cloning software I was using was garbage, and as soon as my source drive that's now borked is fixed I can try cloning again and doing it right next time.

Right now I'm interested in my real drive that I've lost access to. Why does it not give me a log-in module?
 
I

Industrial One

Sigh... everything seems to be malfunctioning this week including Google Groups, I have no idea if my last message went thru so:

Paul, I don't care about the 2TB clone, I care about my current, real 250GBdrive that is now borked. If cloning a healthy drive resulted in a borked drive, I dont wanna guess what other problems it has. I can try cloning again later when my real drive that I need NOW is fixed.

Why does my Windows XP CD display no repair option?
 
R

RJK

Sigh... everything seems to be malfunctioning this week including Google
Groups, I have no idea if my last message went thru so:

Paul, I don't care about the 2TB clone, I care about my current, real 250GB
drive that is now borked. If cloning a healthy drive resulted in a borked
drive, I dont wanna guess what other problems it has. I can try cloning
again later when my real drive that I need NOW is fixed.

Why does my Windows XP CD display no repair option?

FWIW ...several times across the past few years, I've recovered boot up
partitions using steve Gibsons Spinrite 6. ...which seems to repair a
multitude of sins all on its' own !

e.g. out of the blue, Dad's PC a couple of years ago, (on power up / after
bios), presented empty black screen with blinking underscore, screen top
left, (corrupt boot sector / start up files etc.),...
....booted from Spinrite 6 cd | waited for it to do its' thing, ...fault
fixed ! ....immediatley followed by a hasty Ghost image out to external
hard disk !

regards, Richard
 
R

RJK

....mmm this nntp bridge is a bit glitchy, ...sorry for dulplicate psot in
this thread.

regards, Richard

....a bad workman always blames his tools !
 
C

Char Jackson

Sigh... everything seems to be malfunctioning this week including Google
Groups, I have no idea if my last message went thru so:

You might want to consider posting through an actual Usenet News
Provider rather than Google Groups. There are several free ones, if
you're worried about the cost. Some people filter posts from Google
Groups because of the spam (and other reasons) coming from there.
 
R

RJK

Char Jackson said:
You might want to consider posting through an actual Usenet News
Provider rather than Google Groups. There are several free ones, if
you're worried about the cost. Some people filter posts from Google
Groups because of the spam (and other reasons) coming from there.

?
I use usenet.news + MS nntp bridge and OE6 i.e. I do not like the web
based NG interfaces.

....recently I did a couple of "Reply to Group" psots without nntp bridge
running, (hence not signed in via my MS live.* account,
and possibly wrongly assumed, that because those 2 posts vanished from my OE
outbox, they had not been posted,
....quite how they got posted without nntp bridge running, eludes me :)
In fact, I've never quite entirely worked out what goes where via the few
accounts I do have in OE !

regards, Richard
 
C

Char Jackson

?
I use usenet.news + MS nntp bridge and OE6 i.e. I do not like the web
based NG interfaces.

Oh, sorry, I typed that without checking your headers, going only on
your mention of Google Groups (which should obviously be avoided like
the plague). Usenet-news.net is a regular provider, so you're good to
go.
...recently I did a couple of "Reply to Group" psots without nntp bridge
running, (hence not signed in via my MS live.* account,

Since Usenet-news.net is a regular NNTP provider, you can dump your
NNTP bridge software. It's not needed. Outlook Express knows how to
talk directly to your news provider without any bridge in the middle.
You also don't need to be signed in to any 'live' account, since that
has nothing to do with Usenet.
and possibly wrongly assumed, that because those 2 posts vanished from my OE
outbox, they had not been posted,
...quite how they got posted without nntp bridge running, eludes me :)
In fact, I've never quite entirely worked out what goes where via the few
accounts I do have in OE !

You do seem quite confused, but we were probably all in your shoes, at
some point. Not at the same time, of course. Shoes ain't that big. :)
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top