blog dds

2005.10.04

Hard Disk Failure

I tell everybody that the question is not whether your hard drive will fail, but when it will fail. My laptop's drive started emmitting a loud grinding sound last Saturday.

Interestingly, the SMART details for the drive, as presented by SpeefFan continued to look healthy. Nevertheless I started backing up my files: my last full backup was two weeks old, and the last incremental one was two days old.. A few hours later I saw the first hard error in the Windows event log:

The device, \Device\Harddisk0\D, has a bad block.
Windows did not provide any other data, such as the block's address, and SMART continued to claim that my drive was fine.

The next morning I had on my hands a new hard disk drive, and a USB disk enclosure. Because the original hard disk was still working, I decided to create an image of it, rather than chance a backup/restore with tools I did not completely trust. If you feel confident about your backup procedures, read Zwicky's LISA 2003 paper, and reconsider. Unfortunatelly, both a version of Knoppix, and FreeBSD 4.11 bootable CDs I had at hand failed to recognise the USB hard disk enclosure, so I had to work with Windows-based tools. I first used John Newbigin's dd for Windows version 0.3. This could image complete partitions: I could then mount them under Linux with the loopback interface and use them without a problem. However, it failed imaging the complete drive (all the partitions), which made me nervous, because my drive contained a small unknown partition identified with the laptop manufactorer's name; I wanted to ensure I would get a correct copy of this partition. I then tried the dd version that comes with the Forensic Acquisition Utilities written by George M. Garner Jr., which worked like a charm.

The final stumbling block I encountered was using the old hard disk through the USB enclosure. I wanted to compare the two disk contents, to ensure that the transfer of my data was complete and correct. Windows (XP, service pack 2) would refuse to recognise it when I connected it, and would crash when if it was connected during boot time. After a couple of hours of debugging, I solved this mystery. I had installed on the hard drive an ATAPI password, as specified in the ATAPI security mode feature set. Although my laptop's BIOS supports this feature, and asks for the hard disk password before booting up, it appears that Windows doesn't.

Read and post comments, or share through   


Creative Commons License Last modified: Tuesday, October 4, 2005 11:14 am
Unless otherwise expressly stated, all original material on this page created by Diomidis Spinellis is licensed under a Creative Commons Attribution-Share Alike 3.0 Greece License.