Hard Drive Acting Up
 

So I was going through one of my externals, opening videos at random when I find a significant portion of these videos don't work anymore.

At this point, I cracked open my external (It's a Maxtor OneTouch 4 500GB) and stuck the actual drive itself (a Seagate Barracuda 7200rpm 500GB, go figure) in my computer (with an internal SATA cable) and booted it up. No change at all, files that didn't work still don't work.

Fearing for the worst, I immediately copied everything that still worked from the hard drive and onto another hard drive. I ran chkdsk on the drive and it doesn't seem to have any problems (though it didn't take very long to run, oddly enough, on a 80% full 500GB drive).

Any ideas on what would cause this? Is the drive failing? Or is it some kind of fluke?

Download and install smartmontools. Run the short test and then check your error log. If you see any errors, it time to err on the side of caution and replace the drive.

Seagate owns Maxtor now. They're the same company.

Personally, I will thank my lucky stars that the files were still retrievable, and discard the old drive.

If you wish to keep it, pirate a copy of SpinRite 6 and run it on your drive, while it's hooked up internally.

Smartmontools reports that it has overheated once in the past.

So I suppose that was what happened. The drive still passed inspection, apparently, but it has overheated once.

Debating on whether or not to continue using this drive or not....

And yeah, I guess I should be lucky I can still grab files off of it. I might take a shot at SpinRite 6 actually, since redownloading what I lost might take a while, or be hard to impossible.

Overheated? I'm guessing your external doesn't have a fan?

That is surprising. How would it be able to tell if the drive overheated in the past (do drive keep that type of info somewhere?)?

S.M.A.R.T. is part of the ATA spec that allows the drive to store information about itself. A log is created on the drive that can monitor certain values. Each attribute is assigned an ID. The hard drive maker sets a threshold and if a condition is met, the bit switches from off to on. Once this happens, an error is placed in the log file using the ID number to state what's gone wrong.

There isn't a standard, as certain drives will contain higher number of ID attributes. For example, some drives may no contain a temperature sensor so they won't be able to say if they've overheated or not. Nonetheless, most of the major ones are consistent.

Stuff like SeaTools and something Western Digital puts out uses the same principles as S.M.A.R.T.

That's what it looks like to me:

http://img180.imageshack.us/img180/8420/39877991ma8.jpg

And I didn't see any kind of fan when I popped it open.

EDIT: Now that I look it over again, that Seek_Error_Rate doesn't seem friendly....going to run it a few more times to see if it goes up.

EDIT 2: So I ran it again: New value: 12584722.

Damn, I guess I'm trashing this drive. Still going to try to retrieve as much as I can from the corrupted files. Going to try SpinRite 6.

Sweet, I didn't know smartmontools also works on windows (I have it running on my RAID1 system, contacting me if drives begin to fail)

The Seek Error Rate is problematic. It means that the head of the magnetic reader and the hard drive surface are gradually separating. Not good.

Actually you can ignore the Airflow_Temperature_Cel. For one excessive heat in a hard drive well cause the drive to shut down well before any actual damage can be done. Second looking at the results it just barely went over the drives minimum temp threshold and a lot of drives will actually do this during a SMART test as that is physically straining on a hard drive. Max recorded temp was 43c. That's warm but that hard drive could run a lifetime at that temperature with no problem.

It does look like the drive is failing, but not because of a heat issue.