Archived from groups: comp.sys.ibm.pc.hardware.storage (
More info?)
"Zvi Netiv" <support@replace_with_domain.com> wrote in message news:5gmil0175f723h2kh76sntqd3uian3ilf9@4ax.com
> "Folkert Rienstra" <see_reply-to@myweb.nl> wrote:
> > "Zvi Netiv" <support@replace_with_domain.com> wrote in message
> > > "ep" <eps57@hotmail.com> wrote:
> > >
> > > > I have a drive that has been diagnosed as having a corrupted/missing MBR
> > > > (master boot record).
> > > >
> > > > This drive was being used for about 3 months (it is a laptop drive, by the
> > > > way) without problems. Then all of a sudden it failed to be recognized.
> > >
> > > "Failed to be recognized" means that the BIOS doesn't recognize the drive
> > > hardware.
> >
> > That is what you make of it. That is the hardware part. There
> > is also the content part of the bios that is looking for executable
> > bootsectors. If it doesn't find one, that can be construed as the
> > drive not being recognized when you know that there should be one.
>
> The BIOS will recognize the presence of the drive (hardware) regardless
> of whether it contains a valid / executable MBR and bootsector(s) or not.
Yes, no one contents that.
He said it failed to be recognized without saying what 'it' was or what it was
that didn't 'recognize' it. 'It' and 'not recognized' therefor can be several things.
> A test for BIOS recognition is FDISK being able to see the drive.
It is stil possible that a drive is (hardware) recognized by POST but not by
FDISK but then that would be of ones own doing (bios setup).
>
> > > The next part of your post suggests that the drive IS recognized, but
> > > its content cannot be accessed normally.
> >
> > > This is quite different from a drive that isn't recognized anymore by the BIOS.
> >
> > So obviously that wasn't the context in wich he
> > (meant to) say(d) that the drive wasn't recognized.
>
> I understood what the OP meant, I was just pointing out the difference to him.
Well, "content cannot be accessed normally" is quite vague in pointing out a 'difference'.
To all intends and purposes it could actually mean the same thing as 'not recognized by bios'.
>
> > > > Using some demo recovery tools to find out what was wrong, it said the data
> > > > was there but the MBR was damaged.
> > >
> > > What recovery tool did you use?
> > >
> > > > Now, can I just use fdisk /mbr to restore it? Or does NTFS and/or XP do
> > > > something different? If not, does it matter what DOS version I use fdisk
> > > > from?
> > >
> > > FDISK is not the tool for restoring the MBR of a drive that has NTFS. If
> > > the drive had just a single partition that occupied the entire drive capacity,
> > > then you can try FIXMBR after having booted from the XP setup CD, into
> > > "restore console".
> >
> > What would be different if it didn't have
> > "a single partition that occupied the entire drive capacity"?
[...]
> >
> > > If only the MBR got damage then it may work.
> >
> > Will that restore the partiton table too or ...?
>
> FIXMBR works differently than 'FDISK with the /MBR argument'. FIXMBR will
> write a default partition table if none exists or if the existing one is corrupted.
> Unlike FDISK /MBR - the latter will write a default DOS partition only if the
> boot signature is 00 00 rather than 55 AA.
Something seems amiss with that.
Presumably that is to read as :
Unlike FIXMBR, FDISK /MBR will write a default DOS par-
tition only if the boot signature is 00 00 rather than 55 AA.
What if the signature bytes are neither?
>
> > > If not,
> >
> > ... guess not then.
>
> FIXMBR doesn't check for actually existing partitions before rewriting the MBR.
> If the partition table is good, then it will use that data, if not, then it will
> write a default partition table.
> Quite a good chance to fix things on first attempt, depending
> on the drive configuration before the MBR was damaged.
MBR's don't get damaged, they get overwritten completely.
There are two possibilities: It gets overwritten with garbage by a programming
error, destructive virus or crash, or it gets manipulated by a virus or partitioning
software (overlays included) or fix that reads it, changes it, and writes it back.
So FIXMBR and MBR/FIX only work successfully on the latter case.
Q:How big a chance that it will overwrite the existing but slightly unconventional
partition table that nonetheless still works? Or in other words, do you know what
it checks to determine whether the table is good? Are there risks of loosing it?
>
> Regards, Zvi