Testing a dodgy hard drive
Righto.
I'm going to abuse the fact I'm a mod (and megla did it"!£$) and post a hardware/tech support thingy here!
Right.
I've got a computer here that is behaving nortily. More specifically, random crashes. Now, I've tracked the problem to being HD corruption - this explains the crashes (page file gets corrupt, swaps back in, random application crashes).
I've brought the 'beast' back here for some heavy testing. I want to establish which component is failing. As far as I can figure out, there are four potential areas in which it may have failed.
a.) The hard drive is knackered.
b.) The IDE controller on the mobo is knackered
c.) The CPU is knackered.
d.) The IDE cableing is damaged
Now, of these, I see (a) as being the most likely, and (b) and (c) will mean scrapping the entire system. (d) should be obvious as soon as I examine the system closely.
I intend taking the hd out of the computer and setting it up in my own computer to test (so that I am only testing the hd, not any of the other components).
What I'm after is how YOU would test the HD. I was thinking writing a large amount of binary data to the HD, all 0, and then reading it back and seeing if anything changes. Critiques?
__________________
#linux : Home of Genius
<idimmu> ok i was chained to a desk with this oriental dude
|