HDD Becoming Unstable ?

[Sandgroper]

Hi ,
I am wondering if my HDD is about to go belly up.

I recently upgraded my dual boot system Win98 SE / Fedora Core 3 with Win
XP Pro to make the dual boot system of Win XP Pro /Fedora Core 3 using GRUB
as the bootloader and everything has been working good.
However , lately on odd occasions on boot up , I get the message :

"One of your disks needs to be checked , you may cancel the disk check , but
it is strongly recommended that you continue."
After the check is done , the report shows no problems with the HDD.

My HDD is only about 2 years old and is a 40 Gb Seagate Barracuda
partitioned as 2 x 20 GB disks and I am running FAT 32 with the Win XP Pro ,
after the second time I got this message , I defragged the HDD and hadn't
had a problem until today.

I am not sure but I could be getting the above message every time or on
occasions after I boot into FC3 and then later on boot into XP , so does
anybody know if my HDD is failing or that it is a problem cause by the GRUB
bootloader after I boot to FC3 ?

Thanks.


--

 

[DL]

Visit Seagate site and download/use their checking utility

 

[Unk]

Hi ,
I am wondering if my HDD is about to go belly up.

I recently upgraded my dual boot system Win98 SE / Fedora Core 3 with Win
XP Pro to make the dual boot system of Win XP Pro /Fedora Core 3 using GRUB
as the bootloader and everything has been working good.
However , lately on odd occasions on boot up , I get the message :

"One of your disks needs to be checked , you may cancel the disk check , but
it is strongly recommended that you continue."
After the check is done , the report shows no problems with the HDD.

My HDD is only about 2 years old and is a 40 Gb Seagate Barracuda
partitioned as 2 x 20 GB disks and I am running FAT 32 with the Win XP Pro ,
after the second time I got this message , I defragged the HDD and hadn't
had a problem until today.

I am not sure but I could be getting the above message every time or on
occasions after I boot into FC3 and then later on boot into XP , so does
anybody know if my HDD is failing or that it is a problem cause by the GRUB
bootloader after I boot to FC3 ?

Thanks.

Seagate Diagnostic Software (SeaTools)
http://www.seagate.com/support/seatools/

 

[Sandgroper]

Visit Seagate site and download/use their checking utility

Thanks ,
I will visit the site and download their checking utility.



--

 

[Sandgroper]

Seagate Diagnostic Software (SeaTools)
http://www.seagate.com/support/seatools/

Thanks for the link , it saves me searching for it.


--

 

[Unk]

You're welcome.

Thanks for the link , it saves me searching for it.


------------------------------------------------------------------------------------------------------------------
True Multitasking is having three computers and a chair with wheels.


Sandgroper

 

[cquirke (MVP Windows shell/user)]

On Fri, 7 Jul 2006 14:54:07 +0800, "Sandgroper"

I am wondering if my HDD is about to go belly up.

If the system is still stable, then I'd download HD Tune from
www.hdtune.com and use that to do the following tests:
- temperature; below 40C is good
- SMART detail; good early warning but hard to interpret
- surface test (the "slow" one); any bad sectors = Bad

If the system is not stable, i.e. is prone to lockups or random
errors, then I would not run any OS that writes to the hard drive
automatically, as Windows does. You can still use HD Tune, but from a
Bart PE boot CDR rather than your own Windows installation.

Also, you want to ensure other hardware is OK, such as RAM. Suspect
this if errors and instabilities pervade both XP and Linux! Bad RAM
can cause data written to HD to be corrupted, or written to the wrong
location on the HD, so that file system errors arise, and if the
system hangs or resets, secondary file system damage will follow as a
result of interrupted disk writes.

The next Windows boot will check the file system because the "dirty"
bit will have been left set within the file system, indicating that
writes to the hard drive were still in progress when Windows "ended".

I recently upgraded my dual boot system Win98 SE / Fedora Core 3 with Win
XP Pro to make the dual boot system of Win XP Pro /Fedora Core 3 using GRUB
as the bootloader and everything has been working good.
However , lately on odd occasions on boot up , I get the message :

"One of your disks needs to be checked , you may cancel the disk check , but
it is strongly recommended that you continue."

As above.

After the check is done , the report shows no problems with the HDD.

My HDD is only about 2 years old and is a 40 Gb Seagate Barracuda
partitioned as 2 x 20 GB disks and I am running FAT 32 with the Win XP Pro ,
after the second time I got this message , I defragged the HDD and hadn't
had a problem until today.

Don't defrag if there are doubts about the health of the system - like
running a marathon, defrag will make a healthy system fitter but may
kill an unhealthy system stone dead.

Think about what defrag does; it reads potentially all and any file
content off the HD and writes it back somewhere else. Anything that
can cause errors in this process - e.g. bad RAM or sick HD - can
inflict damage to material that may have been fine if left alone.

I am not sure but I could be getting the above message every time or on
occasions after I boot into FC3 and then later on boot into XP , so does
anybody know if my HDD is failing or that it is a problem cause by the GRUB
bootloader after I boot to FC3 ?

I wouldn't think Grub was doing this, but if Linux's file system
handling code did not properly maintain the FAT32 free space value,
then you may get this mileage after every time that Linux has written
to the "Microsoft" volumes.

With the small FATs of FAT16 and FAT12, it was comparitively easy to
deduce free space directly, from first principles as it were, by
counting up all clusters set to chain to 0 in the FAT.

But FAT32 has far larger FATs, so this process can have a material
impact on performance. So the value of free space was stored in an
extension of the volume's boot record, and updated "every now and
then" to keep it in sync with what was really going on.

It's up to the code that manages the file system to do this, and it
may be that your particular Linux doesn't do this properly, if at all.
For example, a re-write of FAT16 code to handle FAT32 may simply have
left out this particular aspect of FAT32.

If that's the case, then whenever Linux has written to these FAT32
volumes, it will have left the free space value stored in the boot
record at variance with the real free space as derived from the FAT.
Anything that prompts the OS to check this, will cause it to fix it.

Another possibility is that Linux sets the dirty bit in the file
system, but doesn't clear it when the volume is unmounted or when the
OS shuts down. In that case, the dirty bit will prompt Windows to
check the volume for errors, even though there aren't any.

------------ ----- --- -- - - - -

Drugs are usually safe. Inject? (Y/n)