MBR trashed

[philo]

I was working on a machine last week that I thought had a HD problem.

The bios "hung" right at the point of loading the OS.
With the drive removed...the bios did not hang, so I replaced the HD
and reloaded the OS. Then cloned the installation to another drive and set
the cloned drive aside.

To see if I could possibly retrieve any data, I hooked the original drive up
as slave...
and was happy that the bios did not "hang"...and that all the data were
there . ( I copied it over the the new drive)

I then ran the machine on the bench for a few days and booted up and
rebooted it several times
and was satisfied that all was OK.

However, just as I was about to put the case back together...
the same problem happened. The bios "hung" just at the point of loading the
OS
and did the same thing with the cloned drive that I had previously confirmed
as good.

At that point I decided the the mobo must have a bad IDE controller...

so I replaced the mobo. (BTW: both mobo's were P-IV's with Intel chipsets
but not identical)

I figured I could perform a repair install and get the machine going
*however*

neither of the drives would boot...ever after attempting a repair install.

The mbr's were trashed to the point where even the "fixmbr" command from
the recovery console did not get things
going.

I had to completely delete the drives and start over to get them working
again.


Interestingly enough, the original drive (which I had thought was bad)

works just fine with the new mobo. All I had to do was a repair install.

Anyway, even though I have things under control now...I never saw a mobo
trash an mbr before...
just wondering if anyone else has ever seen that?

 

[Flasherly]

I was working on a machine last week that I thought had a HD problem.

The bios "hung" right at the point of loading the OS.
With the drive removed...the bios did not hang, so I replaced the HD
and reloaded the OS. Then cloned the installation to another drive and set
the cloned drive aside.

To see if I could possibly retrieve any data, I hooked the original drive up
as slave...
and was happy that the bios did not "hang"...and that all the data were
there . ( I copied it over the the new drive)

I then ran the machine on the bench for a few days and booted up and
rebooted it several times
and was satisfied that all was OK.

However, just as I was about to put the case back together...
the same problem happened. The bios "hung" just at the point of loading the
OS
and did the same thing with the cloned drive that I had previously confirmed
as good.

At that point I decided the the mobo must have a bad IDE controller...

so I replaced the mobo. (BTW: both mobo's were P-IV's with Intel chipsets
but not identical)

I figured I could perform a repair install and get the machine going
*however*

neither of the drives would boot...ever after attempting a repair install.

The mbr's were trashed to the point where even the "fixmbr" command from
the recovery console did not get things
going.

I had to completely delete the drives and start over to get them working
again.

Interestingly enough, the original drive (which I had thought was bad)

works just fine with the new mobo. All I had to do was a repair install.

Anyway, even though I have things under control now...I never saw a mobo
trash an mbr before...
just wondering if anyone else has ever seen that?

yeppers - i used to plug in memory, each chip individually, rows of
2/4/8meg to populate a mb, 9 chips to a row, top being parity. EEMS
taskswapping through 640-1024. $600US to populate memory fully

run a strange program and bingo, bang, blow off a (single) memory
chip. kept a baggie of spares

blown off more mbr-s than i want to think of -- Ghost = best buddy

hardware didn't start getting *real* stable until after/during Win95
revisions, imo -- used to be something of a deal, getting known good
mb in order to run microsoft. by 98se, wow, couldn't believe all the
hardware problems i used to fight with were largely gone

enter the era of "make or break 'em" games

 

[philo]

yeppers - i used to plug in memory, each chip individually, rows of
2/4/8meg to populate a mb, 9 chips to a row, top being parity. EEMS
taskswapping through 640-1024. $600US to populate memory fully

run a strange program and bingo, bang, blow off a (single) memory
chip. kept a baggie of spares

blown off more mbr-s than i want to think of -- Ghost = best buddy

hardware didn't start getting *real* stable until after/during Win95
revisions, imo -- used to be something of a deal, getting known good
mb in order to run microsoft. by 98se, wow, couldn't believe all the
hardware problems i used to fight with were largely gone

enter the era of "make or break 'em" games

Well the good news behind my story is that the mobo had 1 gig of RDRAM on
it.

Four, 256 meg sticks. I can actually sell if for a few bucks on ebay.

Since the machine I was working on is for a friend of mine who does not have
any money...
I'll at least be able to recoup a little bit of payment for all the time I
spent on that thing !

Anyway...It's been years since I trashed an MBR

 

[Franc Zabkar]

I was working on a machine last week that I thought had a HD problem.

The bios "hung" right at the point of loading the OS.
With the drive removed...the bios did not hang, so I replaced the HD
and reloaded the OS. Then cloned the installation to another drive and set
the cloned drive aside.

To see if I could possibly retrieve any data, I hooked the original drive up
as slave...
and was happy that the bios did not "hang"...and that all the data were
there . ( I copied it over the the new drive)

I then ran the machine on the bench for a few days and booted up and
rebooted it several times
and was satisfied that all was OK.

So both the MBR code and the partition table were intact.

However, just as I was about to put the case back together...
the same problem happened. The bios "hung" just at the point of loading the
OS
and did the same thing with the cloned drive that I had previously confirmed
as good.

At that point I decided the the mobo must have a bad IDE controller...

so I replaced the mobo. (BTW: both mobo's were P-IV's with Intel chipsets
but not identical)

I figured I could perform a repair install and get the machine going
*however*

neither of the drives would boot...ever after attempting a repair install.

The mbr's were trashed to the point where even the "fixmbr" command from
the recovery console did not get things
going.

How did you determine that the MBR was trashed? Did you attempt to
read the MBR in a known good machine? Why would the boot process
*write* to the MBR?

I had to completely delete the drives and start over to get them working
again.


Interestingly enough, the original drive (which I had thought was bad)

works just fine with the new mobo. All I had to do was a repair install.

Anyway, even though I have things under control now...I never saw a mobo
trash an mbr before...
just wondering if anyone else has ever seen that?

Did you check the IDE cable?

Not related, but here's a weird problem where sound card software made
my IDE subsystem look bad:
http://tinyurl.com/ywopv6
http://groups.google.com/group/comp...2937e44?hl=en&lnk=st&q=&#doc_a484246f62937e44

- Franc Zabkar

 

[Flasherly]

Well the good news behind my story is that the mobo had 1 gig of RDRAM on
it.

Your friend's lucky. I tend to go hard on people who can't spring a
few bucks to update for a lower end system. Rubbing up on me for Mr.
Computer Genius just won't get them anywhere no more. Days of pulling
hair are gone. And if I wanted it, I know just where to go to in the
stock market.

Intel's MB Memory Architecture - purposely would be the reason I ran
Cyrix and then AMD until buying my first Intel implementation since a
386/SX in the form of a Socket 478 Duron -- with industry standard
acceptance for about any stick of RAM thrown at it. Intel, although a
good development board, I stayed away from while priced too high and
while running RAMBUS. Damn, must've been a decade.

Anyway...It's been years since I trashed an MBR

Me, too -- since last I ran into controller configurations, an
especially wicked early IBM Thinkpad, where remapping a memory manager
through the upper memory registers conflicted with IBM's deisgn for
Cyrix or AMD 486 50Mhz Thinkpad. Only way back was IBM Web HotLine
Solutions -- and only time I've seen a factory LLF routine take eight
hours to format a drive. Also, only time someone has effectively
screamed at me, come to think of it. Bought it refurbished with a
year contract for IBM support, which is what the IBM technician
proceeded to do when I called. Left to myself to figure out why
QuarterDeck's QEMM memory manager would blow through linked IBM IDE
ROM proceedures in upper memory blocks - which I did. The screaming
came, first thing he picked up the phone, "YOU DID WHAT!!?", after
first thing out of my mouth - "Hey, I cleaned Windows 95 off the
HD." (. . .which was shipped factory installed with that particular
model).

 

[SteveH]

Intel's MB Memory Architecture - purposely would be the reason I ran
Cyrix and then AMD until buying my first Intel implementation since a
386/SX in the form of a Socket 478 Duron

Socket 478 Duron eh?

Anyway...It's been years since I trashed an MBR

Me, too -- since last I ran into controller configurations, an
especially wicked early IBM Thinkpad, where remapping a memory manager
through the upper memory registers conflicted with IBM's deisgn for
Cyrix or AMD 486 50Mhz Thinkpad. Only way back was IBM Web HotLine
Solutions -- and only time I've seen a factory LLF routine take eight
hours to format a drive. Also, only time someone has effectively
screamed at me,

I can see why people would scream at you.

 

[Flasherly]

Socket 478 Duron eh?

Really, really impressed at the time, coming back an actual Intel
after so long -- had them up and overclocked to 3Ghz - built a couple
and sold one. One left hooked to a NEC 37" commercial environment TTL/
HDTV flatpannel for my MM center. Middle of the road, dollar cost
averager, so I'll angle in for the biggest bang on the buck. See me
on the road, I'll be the guy in a slow, 25mpg 5-spd Ford Ranger
playing with a Garmin c340.

 

[philo]

So both the MBR code and the partition table were intact.


How did you determine that the MBR was trashed? Did you attempt to
read the MBR in a known good machine? Why would the boot process
*write* to the MBR?


Did you check the IDE cable?

I think the problems with the board was worse than just the IDE channel.

I tried three different IDE cables, four different HD's
and even tried a PCI, IDE controller.

The board is going into the recycle bin!

 

[philo]

Your friend's lucky. I tend to go hard on people who can't spring a
few bucks to update for a lower end system. Rubbing up on me for Mr.
Computer Genius just won't get them anywhere no more. Days of pulling
hair are gone. And if I wanted it, I know just where to go to in the
stock market.

Well, I no longer worry about pulling my hair out.
there is not much left.
My friend is utterly penniless...
he has been working off his debts by doing work for me.
Since my house is 110 years old it always needs a lot of work...
and he is not foolish enough the think he can get something for nothing.

Intel's MB Memory Architecture - purposely would be the reason I ran
Cyrix and then AMD until buying my first Intel implementation since a
386/SX in the form of a Socket 478 Duron -- with industry standard
acceptance for about any stick of RAM thrown at it. Intel, although a
good development board, I stayed away from while priced too high and
while running RAMBUS. Damn, must've been a decade.

I assure you I never would have purchased a machine with RAMBUS...
but all the machines I use and work on were someone else's discards...
so I make use of whatever I can get my hands on.

As to Intel...
though I've never had any trouble with it...
my first choice has always been AMD.

I have a nice 386 up in my attic that has an AMD 386-40
I think Intel's fastest one was a 33mhz.
The 386 runs win95 very well.
I even did some serious hacking and got win98 running on it
http://www.plazaearth.com/philo/win98_386.jpg

Had to first load win98_lite on a p-1,
then move the drive to the 386 and begin a few more tweaks/hacks
to actually get it to run.

Me, too -- since last I ran into controller configurations, an
especially wicked early IBM Thinkpad, where remapping a memory manager
through the upper memory registers conflicted with IBM's deisgn for
Cyrix or AMD 486 50Mhz Thinkpad. Only way back was IBM Web HotLine
Solutions -- and only time I've seen a factory LLF routine take eight
hours to format a drive. Also, only time someone has effectively
screamed at me, come to think of it. Bought it refurbished with a
year contract for IBM support, which is what the IBM technician
proceeded to do when I called. Left to myself to figure out why
QuarterDeck's QEMM memory manager would blow through linked IBM IDE
ROM proceedures in upper memory blocks - which I did. The screaming
came, first thing he picked up the phone, "YOU DID WHAT!!?", after
first thing out of my mouth - "Hey, I cleaned Windows 95 off the
HD." (. . .which was shipped factory installed with that particular
model).

Aah, sounds like it was a lot of fun <G>

 

[Franc Zabkar]

I was working on a machine last week that I thought had a HD problem.

The mbr's were trashed to the point where even the "fixmbr" command from
the recovery console did not get things going.

The following MSKB article describes a bug in FIXMBR. This bug causes
FIXMBR to report good MBR code as bad. The OS is Win2K, but I recently
experienced exactly the same problem with XP Home.

===================================================================
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 ..."

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 ...

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