Corrupted MBR -Regular occurrence

[Bernard]

Hi group,

I've had to format my hard drive around 6 times in the past 6 months
because of a corrupted MBR.

Initially, a few runs of CHKDSK in the recovery console and a running
FIXMBR has fixed the problem for a couple of months only to have the
same problem happen again.

Thinking it was the hard drive, I took it back to the manufacturer for
a warranty replacement. Now the problem seems to be happening again on
the new drive and only 1 month of having it.

Just wondering what could be the problem source of the problem.

There are a few things that I think could be the problem, you might be
able to shed some light on this.

1. Windows XP x64 -I might have a bodgy copy of the O/S, although I
highly doubt it as it's a volume license version.
2. The SATA/IDE bus is somehow not working and therefore corrupting
the hard drive.

The problem seems to pop it's head up straight after a reboot. Either
because the machine has frozen or I get "BSoD".

Is there anyway of testing the SATA/IDE bus or are there other areas
that could be the problem?

Thanks in advanced,
Bernard

 

[Rod Speed]

Hi group,

Lo groupy,

I've had to format my hard drive around 6 times
in the past 6 months because of a corrupted MBR.

Initially, a few runs of CHKDSK in the recovery console
and a running FIXMBR has fixed the problem for a couple
of months only to have the same problem happen again.

Thinking it was the hard drive,

Thats unlikely if only the MBR was affected.

I took it back to the manufacturer for a warranty replacement.
Now the problem seems to be happening again on the new
drive and only 1 month of having it.

Just wondering what could be the problem source of the problem.

Something else in the hard drive subsystem, or its a software problem.

Most likely a software problem.

There are a few things that I think could be the problem,
you might be able to shed some light on this.

1. Windows XP x64 -I might have a bodgy copy of the O/S,
although I highly doubt it as it's a volume license version.

It wont be that.

2. The SATA/IDE bus is somehow not working
and therefore corrupting the hard drive.

Very unlikely to JUST corrupt the MBR.

The problem seems to pop it's head up straight after a reboot.

Thats important.

Either because the machine has frozen or I get "BSoD".

That doesnt normally corrupt the MBR.

Is there anyway of testing the SATA/IDE bus

It wont be that, if that was the problem, it wouldnt JUST corrupt the MBR.

or are there other areas that could be the problem?

Yep, it could be in a wide variety of areas.

Most likely something is causing the freeze and BSODs and when that has happened,
the MBR is corrupted when the system attempts to boot after that problem has shown up.
That might be due to bad caps. These are the usually blue or black plastic covered post
like things that stick up vertically from the motherboard surface. The tops should be flat.
If any have bulged or have leaked, thats likely the problem.

Is there any commonality in what produces the freeze or BSOD ?

 

[Bernard]

Thanks for the quick reply Rod.

Regarding the BSoD, that only happened once. The common denominator
here seems to be a reboot while the hard drive is "doing something".
For example, the last time the hard drive "screwed up", the BSoD was
dumping the physicals and this took around 1hr. So I got impatient and
restarted the machine. While another time, the system just froze up
and the hard drive I/O light was on, so I restarted and BANG. Same
problem.

As for the only the MBR being affected, I forgot to mention on the
original post that the while doing CHKDSK, there were a lot of index,
USN journal fixes, orphan files and also it did a step I've never seen
before "Correcting errors in the Volume Bitmap".
The CHKDSK errors happen every time this problem pops up.

Does this sound more like an onboard controller fault?

Are there any utilities I can use to test the motherboard/chipset and
the controllers?

I thought something that would send large amounts of data to the hard
drive (therefore via the SATA/IDE controller) and checks whether the
data was received as expected.

Thanks again,
Bernard

 

[Bernard]

Thanks Franc,

The reason I run FIXMBR is because recovery console doesn't recognise
the drive. Once I run FIXMBR and restart, the drive becomes available
for me to run CHKDSK.

Cheers,
Bernard

 

[Rod Speed]

Thanks for the quick reply Rod.

Unfortunately this one took longer |-(

Regarding the BSoD, that only happened once.

Thats even worse, freezes can be due to a hell of a lot of things.

The common denominator here seems to be a reboot while the hard drive is "doing something".

OK, then it may not actually be a freeze at all, just the drive retrying on a bad.

Post the Everest SMART report on the drive.
http://www.majorgeeks.com/download.php?det=4181

For example, the last time the hard drive "screwed up", the BSoD was
dumping the physicals and this took around 1hr. So I got impatient and
restarted the machine. While another time, the system just froze up and
the hard drive I/O light was on, so I restarted and BANG. Same problem.

As for the only the MBR being affected, I forgot to mention on the
original post that the while doing CHKDSK, there were a lot of index,
USN journal fixes, orphan files and also it did a step I've never seen
before "Correcting errors in the Volume Bitmap".
The CHKDSK errors happen every time this problem pops up.

OK, thats an important difference.

Does this sound more like an onboard controller fault?

No, just that you do have a problem with the hard drive subsystem.

Are there any utilities I can use to test the motherboard/chipset and the controllers?

No, not in the sense that you can readily check what ends up on the platters.

I thought something that would send large amounts of data
to the hard drive (therefore via the SATA/IDE controller)
and checks whether the data was received as expected.

There isnt anything much like that thats free.

You can obviously do a copy and compare at the file level using a cmd file.

 

[Franc Zabkar]

I've had to format my hard drive around 6 times in the past 6 months
because of a corrupted MBR.

Initially, a few runs of CHKDSK in the recovery console and a running
FIXMBR has fixed the problem for a couple of months only to have the
same problem happen again.

I can't understand why your OS or your software would want to write to
the MBR, unless you were repartitioning the disc or intentionally
refreshing the MBR code. Instead I suspect that your drive is
developing bad sectors.

If you believe that the MBR code is corrupt because FIXMBR is telling
you so, then be aware of the following bug (it is also present in Win
XP Home's recovery console).

====================================================================
Error Message When You Run fixmbr Command:
http://support.microsoft.com/kb/266745/

SYMPTOMS

When you attempt to run the fixmbr command in the Microsoft Windows
2000 recovery console, your computer system may display the following
error message:

This computer appears to have a non-standard or invalid master boot
record. FIXMBR may damage your partition tables if you proceed. This
could cause all the partitions on the current hard disk to become
inaccessible. If you are not having problems accessing your drive, do
not continue. Are you sure you want to write a new MBR?

RESOLUTION

Ignore the error message described in the "Symptoms" section of this
article. The fixmbr command can safely rewrite the MBR.

STATUS

Microsoft has confirmed that this is a problem in the Microsoft
products that are listed at the beginning of this article.

MORE INFORMATION

The fixmbr command causes this error message to be displayed on your
computer system whenever you run the command, regardless of the state
of the Master Boot Record (MBR).
====================================================================

- Franc Zabkar

 

[Franc Zabkar]

Are there any utilities I can use to test the motherboard/chipset and
the controllers?

I thought something that would send large amounts of data to the hard
drive (therefore via the SATA/IDE controller) and checks whether the
data was received as expected.

Check your drive for bad sectors with your HD manufacturer's
diagnostic software. For example, Seagate supplies SeaTools.

If you suspect that something is wrong at the motherboard end, then
stick your drive in an external USB enclosure and test it from a USB
port.

Otherwise you could copy 1GB .VOB files from one location on your hard
drive to another and then compare the source and destination files.

The following Windows commands copy 10GB of data and compare each of
the copies with the original.

for %i in (0 1 2 3 4 5 6 7 8 9) do copy test.vob test%i.vob
for %i in (0 1 2 3 4 5 6 7 8 9) do fc /b test.vob test%i.vob

- Franc Zabkar

 

[Bernard]

The hard drive is stuffed, I've already had 3 different 1's with the
same problems. So I'd like to know what the source of the problem is.

I might just buy a SATA PCI controller and see how I go.

I just don't want to invest more time in rebuilding the system,
installing all the programs and updates for it to then go down again.

I'll let you know how I go with the Everest test. I'm also going to
use Hot CPU Tester to test the chipset and motherboard. I'll let you
know soon.

Cheers,
Bernard

 

[Stretch]

Franc Zabkar wrote in news:[email protected]

Check your drive for bad sectors

Bad sectors don't 'corrupt' the MBR.

 

[Stretch]

Franc Zabkar wrote in news:[email protected]

I can't understand why your OS or your software would want to write
to the MBR, unless you were repartitioning the disc or intentionally
refreshing the MBR code.

If I'm not mistaken NT keeps a signature there.

Instead I suspect that your drive is developing bad sectors.

A corrupt MBR is not a bad sector.
It's a readable one that lost it's expected contents to a certain extend.

 

[Stretch]

Bernard wrote in

Hi group,

I've had to format my hard drive around 6 times in the past 6 months
because of a corrupted MBR.

Initially, a few runs of CHKDSK in the recovery console and a running
FIXMBR has fixed the problem for a couple of months only to have the
same problem happen again.

Thinking it was the hard drive, I took it back to the manufacturer for
a warranty replacement. Now the problem seems to be happening again on
the new drive and only 1 month of having it.

Just wondering what could be the problem source of the problem.

There are a few things that I think could be the problem, you might be
able to shed some light on this.

1. Windows XP x64 -I might have a bodgy copy of the O/S, although I
highly doubt it as it's a volume license version.
2. The SATA/IDE bus is somehow not working and therefore corrupting
the hard drive.

The problem seems to pop it's head up straight

*after a reboot*. Either because the machine has frozen or I get "BSoD".

That is likely the source of the problem.
One likely source could be memory corruption.

 

[Stretch]

Bob Willard wrote in news:[email protected]

Bernard wrote:

So it's not the harddrive(s).

So I'd like to know what the source of the problem is.

Given that history, I'd take a serious look at:
1. Cooling -- open the case, and clean the filters and heatsinks

2. Cooling -- make sure your case fans are cooling the HDs
3. Cooling -- get an app that reports HD temps in realtime
4. Cooling -- make sure that cabling/etc. does not block air flow
5. Cabling -- check signal & power cables to HDs for nicks and seating
6. Power -- make sure your PS has enough juice and low ripple

It's not the harddrives if the MBR is corrupted and FixMBR can still
repair it. That means it's readable but the code has been corrupted.

 

[Mike Tomlinson]

As for the only the MBR being affected, I forgot to mention on the
original post that the while doing CHKDSK, there were a lot of index,
USN journal fixes, orphan files and also it did a step I've never seen
before "Correcting errors in the Volume Bitmap".
The CHKDSK errors happen every time this problem pops up.

Does this sound more like an onboard controller fault?

It could be faulty memory. Download Memtest86+ from www.memtest.org and
run it overnight for a minimum of two full passes.

 

[Rod Speed]

The hard drive is stuffed, I've already had 3 different 1's with the
same problems. So I'd like to know what the source of the problem is.

Most likely the power supply needs replacing.

I might just buy a SATA PCI controller and see how I go.

I'd try the power supply first myself.

 

[Rod Speed]

Franc Zabkar wrote

If I'm not mistaken NT keeps a signature there.

But doesnt keep changing that.

A corrupt MBR is not a bad sector.

Yes, but he does get more than just a corrupt MBR.

It's a readable one that lost it's expected contents to a certain extend.

It isnt clear that that is whats happening to it.

 

[Bernard]

WOW! Thanks guys for the all replies.

At the moment I'm am running Hot CPU Tester which will run for the
next 7hrs testing the chipset, memory and hard drive. I'll see if the
logs bring back anything.

I was going to get a SATA PCI-E controller to give it a go, but now
that we are bringing the power supply into question I'll have to see
which 1 might really be the problem.

Is there any way of testing the amount of wattage all my components
are taking up? And maybe a ripple test too?

I'll also run memtester tonight, see how I go. There probably wont be
any updates until tomorrow.

Thanks again for all your help!
Bernard

 

[Bernard]

Thanks Bob, I've attached my comments below. Can you elaborate on some
points?

Given that history, I'd take a serious look at:
1. Cooling -- open the case, and clean the filters and heatsinks

Because of all the problems I've had in the past the case hasn't been
closed in ages.

2. Cooling -- make sure your case fans are cooling the HDs
3. Cooling -- get an app that reports HD temps in realtime

The HDD temperatures seem good around 36°C

4. Cooling -- make sure that cabling/etc. does not block air flow Nope, very tidy in there
5. Cabling -- check signal & power cables to HDs for nicks and seating

Can you please elaborate on this point?

 

[Rod Speed]

WOW! Thanks guys for the all replies.

At the moment I'm am running Hot CPU Tester which will run for the
next 7hrs testing the chipset, memory and hard drive. I'll see if the
logs bring back anything.

I was going to get a SATA PCI-E controller to give it a go, but now
that we are bringing the power supply into question I'll have to see
which 1 might really be the problem.

Is there any way of testing the amount of wattage all my
components are taking up? And maybe a ripple test too?

Yes, but what you need to do that isnt cheap, particularly if the power
supply is periodically sagging out of spec and its that thats producing
the not very common freezes and resulting hard drive data corruption.

It would be cheaper to get a new power supply instead.

I'll also run memtester tonight, see how I go.
There probably wont be any updates until tomorrow.

Thanks again for all your help!

No problem, thats what these technical newsgroups are for.

 

[Lars]

Is there any way of testing the amount of wattage all my components
are taking up? And maybe a ripple test too?

How about telling us the Wattage of your current PSU, and a list of
your main hardware.

CPU
Video card
Hard drives
Optical drive

etc

A few weeks ago I had a bunch of strange problems after having added a
new HD to my system. All problems went away when I upgraded to a more
powerful PSU.


Lars
Stockholm

 

[Bernard]

How about telling us the Wattage of your current PSU, and a list of
your main hardware. Ok, 450W power supply

Motherboard
ASUS M2NPV-VM

CPU AMD Athlon 64 X2 dual 3600+
Video card Onboard
Hard drives

1x SATA 250GB Seagate
1x SATA 120GB WD

Optical drive

2x DVD-RW
PCI
2x TV Capture Card (1x Analogue, 1x Digital)
USB
Internal multi-card/usb reader (plugs directly into oboard usb pins)
USB Keyboard, Mouse, Printer.

Does it sound like it's overloaded?