Hard drive issue

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View

Hey friends, been a while since i had to ask for help but here i am

Ok so here is the problem : i have 2 Seagate hard drives, both at
160gb each, these used to be in a raid 5 with 2 more of the same drive
(4 total drives in the raid) i recently broke the raid up because i went
to SSD's for OS and important applications and am now using 2 drives
with no problem for storage, the other 2 that i am having the problem
with is this : windows reports a cyclic redundancy check error and is
reporting each drive as having 298.10gb of storage when in fact they are
not that size. i have no data of importance that needs to be recovered
from these drives so any kind of data destructive repair is ok. i can
not run a chkdsk on these drives as windows will not assign a drive
letter, disk management is unable to format the drives. these drives are
just shy of a year old. i am needing these drives to do a new OS install
to replace a completely failed drive in a laptop.

So there it is, my problem. if there is a solution i am all for it
if not i will have to RMA these things i guess.

Thanks in advanced for any insight any of you may have.


Re: Hard drive issue

Eidelmaim wrote:
Quoted text here. Click to load it

I could understand the OS seeing 298.10gb if the drives were
in RAID 0, and you were using Intel Matrix RAID.

But a single drive should never be able to report a capacity
that goes past the physical end of the disk.

The ATA/ATAPI command set supports SET_MAX_ADDRESS for
setting a Host Protected Area. That would allow trimming
a 160GB drive so it was 120GB for example. But SET_MAX_ADDRESS
should not support setting the declaration of drive size, past
the physical limits. As indeed, you'd get a CRC or other kind
of error, as soon as the operating system tries to read that
area of the disk. It makes no sense, for a drive to support
such a thing.


What you could try, is to take the "defective" 298.10GB drive
and actually make an HPA on it, use SET_MAX_ADDRESS and define
a smaller drive. Then, the CRC errors will stop.


You boot a Linux LiveCD, and use hdparm. If you attempt
to set an HPA, you can only do this once per boot session.
Additional attempts to define the capacity, will require a
reboot of Linux. I feel this is actually a feature of the
hard drive - the hard drive refuses to accept more than
one of these commands per session.

    hdparm --yes-i_know_what_i_am_doing   -N p# /dev/sdX

To determine what number of sectors to use, examine the
capacity information of the existing disks. You can try
"dd --list" on the working hard drives. This is a Windows
program, and you can run this on the Windows machine having
the remaining two "good and working" disks.



Not every disk interface on a motherboard supports HPA commands.
On my current P5E Deluxe motherboard, only the Jmicron IDE chip
allows it. The BIOS module for the Southbridge SATA ports, appear
to be locking out the command. When I wanted to set an HPA
on a SATA drive here, I used an IDE to SATA adapter dongle
and connected the SATA drive to the IDE controller. And then
I could issue the "hdparm" command from Linux.

The capacity change from SET_MAX_ADDRESS is persistent across
power cycles, so when I trimmed my drives down to 6GB and 4GB
as an experiment, they would have stayed that way until using
the "hdparm" command again later to correct the size.


The ATA/ATAPI spec supports "Secure Erase" option. Under your
current circumstances, I would not try this on the drive unless
you were absolutely sure it wasn't in RAID mode. And all other
attempts to make the drive work, had failed. There is no magic
about this command. If the control information on the drive
is corrupted (unlikely), then a Secure Erase would get just as
many CRC errors as your OS is seeing, and then the Secure Erase
could stay "stuck" in that mode... forever.


Before using that program, *read* all the documentation!
A password is set on the drive when you use Secure Erase,
and using a wax pencil, write the password on the body of
the hard drive for future reference (for the next Secure
Erase attempt, using some other software).


Re: Hard drive issue

Quoted text here. Click to load it


   A couple of possibilites come to mind to get your Seagates to be seen by  
Windows again for drive letter placement. I'd download and run Seagate's  
diagnostic tool 'Seatools for Windows' and check that the drives are OK or  
allow a repair if they aren't. Running a low level format if possible may  
get them back into running order and allow Windows to see each one properly.  
If Sea Tools doesn't pass the drives then I would be inclined to look for a  
RMA for replacement.

The other thought is to get a partitioning program such as Easeus Partition  
Master Free  
and see if it will delete the current partitions to allow creating new ones.

Jan Alter

Quoted text here. Click to load it

Re: Hard drive issue

On 9/30/2013 11:34 AM, Eidelmaim wrote:
Quoted text here. Click to load it
You write that disk management won't format the drives but what will it do?  
For example, does disk management 'see' the drives at all? If it does does  
it show any partitioning? If it shows partitioning of any sort then it  
should be able to kill the partitions and that may give you a starting point.

A quick destructive fallback would be to run Linux from CD and use it to  
zap the drives to a null state. Or use the Seagate tools as someone else  
suggested. Or use DBAN to lobotomize them.

Re: Hard drive issue

On 09/30/2013 10:34 AM, Eidelmaim wrote:
Quoted text here. Click to load it

Can you just plain delete them by using Disk Management?
I'd do that and recreate.

Otherwise I'd disconnect all drives but those two problematic ones and  
boot from a utility cd that contains dban or some other disk erasing  

  these drives are
Quoted text here. Click to load it

Site Timeline