On 10/18/21 9:25 AM, Thomas Anderson wrote:
Hello Gurus,
I have been having problems with a drive (non-SSD) for a while now, but
I would like to "identify" the problem specifically, so that I may
perhaps be able to get the drive replaced.
Essentially, I have been experienced data loss, where nodes become
unreadable, when I try to "fix it" with fdisk, it says it moves
unreadable to the trash, basically deleting data.
I have tried reformatting the drive, writing zeros to it, and then
temporarily the drive looks fine; but it's not! it's as if the drive is
some how decaying: where it looks good, but then when you go to use it,
the data later becomes unreadable.
Is there any kind of test I can execute on this drive to show it needs
to be replaced??
Thanks!
The canonical testing tool is the one made by the drive manufacturer.
This is the definitive way to obtain a warranty RMA. For example, Seagate:
https://www.seagate.com/support/downloads/seatools/
As suggested by other readers, smartctl(8) can be used to obtain
information from the drive controller, direct the controller to perform
short and/or long self-tests, etc.. If your drive is out of warranty,
you want a second opinion, you want to do periodic monitoring/ logging,
etc., this tool works well.
But before you blame the drive, you also have to eliminate the SATA
interface (typically on the motherboard), the SATA cable, the power
supply, and all connections. This means you need at least two
interfaces, two cables, and two drives; it is best to have a second
computer with two interfaces to confirm. Test all of the permutations,
work up a truth table of results, and find the bad component(s) by a
process of elimination. Beware that you may have more than one bad
component, and that the failures may be intermittent.
David
