"NTLDR IS MISSING" (Win XP-3)

B

Barry Bruyea

Failing with what you have checked here below is the 10 reasons for the NTLDR
issue.I would check each one against your error. I have seen this error many times
posted and usually it is number 8 which caused it for most.

I can't dispute what you are saying about No. 8, but given the
Computer stops dead even before the initial boot screen has completed,
I find that hard to understand.

Previously I had asked if the SATA socket can be removed and didn't
get an answer. I'm not clear if it is soldered to the main board, or
removable, as I keep coming back to the cable problem.
 
P

Paul

Failing with what you have checked here below is the 10 reasons for the
NTLDR issue.I would check each one against your error. I have seen this
error many times posted and usually it is number 8 which caused it for
most.

1.. Computer is booting from a non-bootable source.
2.. Computer hard disk drive is not properly setup in BIOS.
3.. Corrupt NTLDR and NTDETECT.COM file.
4.. Misconfiguration with the boot.ini file.
5.. Attempting to upgrade from a Windows 95, 98, or ME computer that is
using FAT32.
6.. New hard disk drive being added.
7.. Corrupt boot sector or master boot record.
8.. Seriously corrupted version of Windows 2000 or Windows XP.
9.. Loose or Faulty IDE/EIDE hard disk drive cable.
10.. Failing to enable USB keyboard support in the BIOS.

http://www.computerhope.com/issues/ch000465.htm

1. Implies the BIOS is selecting the wrong disk. But the MBR on that
disk, points to a partition which has the boot flag set, and that
partition also happens to have a partition boot sector (but none of
the files needed for booting). Then later, the BIOS selects the
correct disk ?

3. Corrupt, but then later uncorrupt. Hard to explain, as a file system issue.

4. Like an ARC path problem. But then, how does it correct itself ?

I'm having trouble making a credible theory from those. Malware
can certainly do anything it wants, but that's a pretty lazy answer :)

Paul
 
T

Tim Meddick

Surely, if the problem is [still] that the OP receives the errormessage:
NTLDR IS MISSING... then, is it going to be the result of faulty cabling,
internal connections (SATA / IDE / or otherwise) or anything else which
results in any disk [storage volume] failing to be enumerated properly
(except for the disk containing the system or boot volume)...

My thinking is this; if it were any of those things, how are you left with
the "NTLDR" error message? The system looks for the boot-loader file:
NTLDR, because the code contained within the bootsector of the active
boot-volume ( normally; physical disk0 partition0 )! Ergo, if the
boot-drive was failing / failed, then how would it know what boot-files to
load? If you replaced the operating system on the same computer, to Win98,
and if the same problem occurred, it would tell you ;

Invalid system disk
Disk I/O error
Replace the disk, and then press any key

....instead of ;

NTLDR is missing or corrupted...(etc.,)

==

Cheers, Tim Meddick, Peckham, London. :)
 
G

glee

Barry Bruyea said:
On Wed, 02 May 2012 19:25:29 -0400, Barry Bruyea

I downloaded the Disk Diagnostics program from Seagate and all of my
disks passed.

You did the Long Test, not just the Quick Test, yes?
 
T

Tim Meddick

All I'm saying was that the boot-disk must be accessible for the system to
return the error message : "NTLDR is missing or corrupted..." as this is a
result of executing the code contained in the disk's first sector [0]
instructing the system as to what boot-files should be loaded...

==

Cheers, Tim Meddick, Peckham, London. :)
 
G

glee

Tim Meddick said:
All I'm saying was that the boot-disk must be accessible for the
system to return the error message : "NTLDR is missing or
corrupted..." as this is a result of executing the code contained in
the disk's first sector [0] instructing the system as to what
boot-files should be loaded...

You can have a damaged or failing disk that still boots, but gives an
"file missing" error. A failing disk can still be accessible. A failed
disk is what would not be. Just saying...
 
B

Barry Bruyea

For all those who have tried to help, I thank you. My weekend will be
spent formatting Disk C: and reloading Windows.

Barry said:
On Wed, 02 May 2012 19:25:29 -0400, Barry Bruyea

I downloaded the Disk Diagnostics program from Seagate and all of my
disks passed.

So what can we observe then ?

http://technet.microsoft.com/en-us/library/cc977213.aspx

On my machine, if I copy the MBR (sector 0) of the disk, it has
messages like this. The 440 bytes of code in the MBR, can be
repaired with FIXMBR, which should not change the primary partition
table entries coming just after it. The 440 bytes of code, just
before these messages, looks for the boot flag on one
of the partitions, to determine which is the boot partition.

Invalid partition table.
Error loading operating system.
Missing operating system.

In the partition corresponding to C:, I'm supposed to be
able to find these messages in the Partition Boot Sector. This
is a sector (or sectors) at the front of the partition, before
the file system proper begins. The Partition Boot Sector is placed
there by commands such as FIXBOOT (available from recovery console).

A disk read error occurred.
NTLDR is missing.
NTLDR is compressed.

I can find "NTLDR is missing", at byte address 0x7E00 area of the disk.
Converting that to sectors, that's sector 63. Allowing sectors 0..62 to
hold MBR and 62 sectors of slack, that implies my partition boot sector
is very near the beginning of my FAT32 WinXP partition. The actual
"NTLDR is missing", is at 0x7FAE. That might be slightly different than
how and where, an NTFS partition stores it. But I could find it relatively
easily, by copying a portion of the raw sectors on the disk, with "dd".
For security reasons, "dd" was not allowed to copy that area, relative
to the beginning of C:, and instead, I had to tell it to copy
Partition0, which is relative to the whole disk.

http://www.chrysocome.net/downloads/dd-0.5.zip

Another way to examine that, is with TestDisk. Using the
option in the interface called [Boot], you can eventually
get to examine what the tool thinks is the partition boot
sector. For some reason, the hex editor in here, traverses
about three 512 byte sectors, when looking in that area.

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

The only problem with that tool, is the number of steps to get down
there for a look. And in this case, I don't think it really helps
anyway, except to satisfy curiosity.

This is my partition boot sector and backup boot sector. Notice my
backup boot sector is not valid, and I can't remember the last
time I actually found the backup boot sector to be valid. I don't
know what step was supposed to put valid info into it, but it's never
seemed to work.

http://img528.imageshack.us/img528/4760/cfirstsector.gif

Since my partition is FAT32, I'm guessing that's why
I don't have the same error messages as the article I used above
to get the three lines that are supposed to be in a PBS.
It's possible NTFS and FAT32 don't have exactly the same
partition boot sector (because the file systems are different,
and finding files would require different code).

*******

So that's where the "NTLDR is missing" could come from. And
then we need a mechanism to make NTLDR actually go missing...
And then, come back again.

Paul
 
P

Paul

Barry said:
I did the long test..twice.

Tell me, on the second attempt to boot, is CHKDSK running ?

Can the dirty bit get set, if the first boot fails, such that
the second boot repairs the partition with CHKDSK, and then
it boots ?

Maybe the partition is broken in such a way, that CHKDSK tries
to repair it, but doesn't actually finish the job properly. Leading
to the same apparent fault happening over and over again.

*******

I still don't know if your partition is NTFS or FAT32, but one
thing I do here about every three months, is copy the files off
C: onto another partition, reformat the partition, copy the files
back, then do a FIXBOOT to put back the partition boot sector. This
is faster than running the terrible built-in defragmenter, which
on occasion has taken longer than eight hours to run to completion.
If I copy files instead, it might take me 40 minutes to have a minty fresh C:.

To do that, I need a second bootable OS. One disk has WinXP. The
other disk has Win2K. I use a copy of Robocopy from Microsoft,
to copy the partition over, from WinXP to an empty partition.
If I'm booted from the Win2K partition, the WinXP partition
is not "busy", and that's why all the files copy correctly.

To do the FIXBOOT, I boot my WinXP SP3 installer CD, and run
the recovery console from there, after the C: partition has
been put back. If you have multiple Windows partitions, you
have to be a little careful with FIXBOOT, because of the difficulty
of determining which partition is which. (You're supposed to "log in"
to the partition, and I have different administrator passwords on each
one, so I can tell when I've selected the wrong one.)

It's possible to "install" the recovery console, onto C:, and make it
a boot time option, but I'm too lazy to do that.

My purpose in mentioning this silly idea, is you could try
copying the files off, in the hopes that a "fresh new file system"
would repair the problem. (I.e. CHKDSK isn't fixing it, so the
theory goes that a fresh file system might help.) If the copy stage
fails, in copying from the source partition, to the temporary partition,
then you know the thing is too badly busted anyway, and the experiment
stops there (time to "level and reload").

If all the files come off successfully, then you can format C: (since
you're booted from another OS), and then copy the files back. When you
format a partition, it erases the partition boot sector (the one that
says "NTLDR is missing"). If they didn't overwrite the damn thing, the
ugly part of this recipe would not be needed.

For a FAT32 formatter (to wipe C:, before copying the files back,
and having a "minty fresh" file system), I use this. I notice they
have a GUI version, and that's overkill :) The nice thing about
this utility, is it does FAT32 on larger than 32GB partitions.
My WinXP partition is around 72GB or so. And this thing is fast,
because it doesn't erase every data sector, just writes a new FAT.
In roughly one second, you have a new, empty C:.

http://www.ridgecrop.demon.co.uk/index.htm?fat32format.htm
http://www.ridgecrop.demon.co.uk/download/fat32format.zip

If your WinXP is NTFS, that wouldn't change things too much. You
could use Disk Management to reformat C: in that case.
NTFS is supposed to be a bit more robust (harder to damage) than FAT32.
But any file system can break, given multiple faults and enough time.

If you don't have a second OS to use, that procedure would be a
bit more difficult. I suppose you could build a BartPE disc and
use that (put Robocopy and other tools on it), but that is
"cruel and unusual punishment". That would take you all next
week to get running.

This is the Robocopy I use:

http://technet.microsoft.com/en-us/magazine/2006.11.utilityspotlight.aspx

Wikipedia lists the various releases of it - it copies files, but
does a better job than your average utility. The one I use is XP026
version.

http://en.wikipedia.org/wiki/Robocopy

This is the format of the command I use, to copy the entire partition.
(This is actually running in "mirror" mode, but since the destination
is always completely empty, I like to think of it as a "copy" operation.)
I have the tool create a log for each run, in case I suspect something
has gone wrong. I don't think this has a "verify" option, and if I
really needed to verify everything copied correctly, that would be
a lot more work. Since I'm booted from my second OS, the partition
letters are all screwed up.

robocopy J:\ E:\ /mir /copy:datso /dcopy:t /r:3 /w:2 /zb /np /tee /v /log:robocopy_J_to_E.log

Don't feel obliged to do that. Reinstalling will undoubtedly fix it,
or will identify that the disk is at fault. And if you're more
comfortable doing it that way (reinstall), then that's the way to go.

HTH,
Paul
 
P

Paul

rb said:
Not a common occurrence, but I have encountered two 'power' related
situations that resulted in NTLDR error message:

1- one or more of the cylindrical sockets in the molex connector were
not making good contact with the corresponding pins in the
harddrive...drive would 'spin-up' then momentarily lose power during
boot process (Note...jiggled wires at molex connector and would hear
drive start to 'spin-up' and then 'spin-down'. Connected drive to a
different molex connector...drive booted up normal...no further ntldr
issues).

2- faulty power supply caused similar situation to above...drive would
start to spin-up then spin-down during boot process (Note...replaced
power supply...drive booted up normal...no further ntldr issues).

Drives are voltage sensitive. I've experienced that fault here
as well. In my case, it happened because a video card was powered
from the same harness as the SATA drive. The SATA drive stayed
in a "spin-up,spin-down" loop. The controller seems to have
a voltage sensor, if the 12V rail drops to say 11V, the drive will
spin down. So you can't have the voltage too far out of spec,
or the controller board on the hard drive will react.

The other extreme, is what happens on overvoltage. A poster here,
reported some components burning on the controller. It seems
the controllers come with surge arrestors on +5V and +12V. If
you were to take the +12V up to +15V and hold it for half a
minute, the surge arrestor burns up. It was only intended to
stop a "transient" if you hot-plug or hot-unplug a drive (a
short pulse of excess voltage, caused by inductance in the power
path). When faced with a long term voltage surge, the power
dissipation in the arresting device, burns it.

So that means there are two extremes. Too little voltage (sometimes
caused by another, significant load on the same cable), and the drive
goes into one of those spin-down and spin-up loops. Too much voltage
(not a common occurrence) and two components on the controller
can burn. Which, I suppose, is better than the whole thing being
fried.

The spec on power, is 12V +/- 5%, which would range from 11.4 to 12.6.
You don't have to drop too far below 11.4, before the spin-down thing
can happen.

One reason they may have set it up that way, is to ensure
the heads end up on the landing ramp, if the power is going
off. By having "advanced power fail" detection, it improves
the odds of a proper head retraction and landing. If it wasn't
for that, they could probably afford to allow the 12V rail
to sag lower than that, without reacting to it.

Paul
 
D

David H. Lipman

From: "Paul said:
Drives are voltage sensitive. I've experienced that fault here
as well. In my case, it happened because a video card was powered
from the same harness as the SATA drive. The SATA drive stayed
in a "spin-up,spin-down" loop. The controller seems to have
a voltage sensor, if the 12V rail drops to say 11V, the drive will
spin down. So you can't have the voltage too far out of spec,
or the controller board on the hard drive will react.

The other extreme, is what happens on overvoltage. A poster here,
reported some components burning on the controller. It seems
the controllers come with surge arrestors on +5V and +12V. If
you were to take the +12V up to +15V and hold it for half a
minute, the surge arrestor burns up. It was only intended to
stop a "transient" if you hot-plug or hot-unplug a drive (a
short pulse of excess voltage, caused by inductance in the power
path). When faced with a long term voltage surge, the power
dissipation in the arresting device, burns it.

So that means there are two extremes. Too little voltage (sometimes
caused by another, significant load on the same cable), and the drive
goes into one of those spin-down and spin-up loops. Too much voltage
(not a common occurrence) and two components on the controller
can burn. Which, I suppose, is better than the whole thing being
fried.

The spec on power, is 12V +/- 5%, which would range from 11.4 to 12.6.
You don't have to drop too far below 11.4, before the spin-down thing
can happen.

One reason they may have set it up that way, is to ensure
the heads end up on the landing ramp, if the power is going
off. By having "advanced power fail" detection, it improves
the odds of a proper head retraction and landing. If it wasn't
for that, they could probably afford to allow the 12V rail
to sag lower than that, without reacting to it.

Paul

If it was a power issue then it would have caused SeaTools to fail with an error.
 
U

Unk

On Wed, 02 May 2012 19:25:29 -0400, Barry Bruyea

I downloaded the Disk Diagnostics program from Seagate and all of my
disks passed.


A quick, TEMPORY, fix is to format a floppy on a good XP machine, and
copy "Boot.ini", "Ntdetect.com", "Ntldr" to it. Boot your system with the
floppy in the drive.. For a permanent fix........

Repair the Boot Sector using the Recovery console on the CD.
<http://www.webtree.ca/windowsxp/repair_xp.htm#How to Repair the Boot Sector:>
(Copy & paste the ENTIRE above line into the url box)

To repair a damaged Boot Sector, run the Recovery console from the CD. At
the command prompt type FIXBOOT and press Enter. Then answer "Y"
 

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

Similar Threads

NTLDR is missing 6
NTLDR is Missing 7
"NTLDR is missing" 6
NTLDR is Missing 8
Dual boot Vista and XP (NTLDR is missing) 4
NTLDR is missing 1
Ntldr is missing 3
NTLDR - MISSING ERROR 33

Top