[arch-general] [OT]Disk showing too many bad sectors - is it going to fail ?

[Mauro Santos]

On 10/16/2010 08:46 AM, Partha Chowdhury wrote:
> I have a Seagate Barracuda 7200.9 family harddisk of 160gb capacity
> which is more than 3 years old. recently I installed gsmartcontrol and
> after running it, it showed number of current pending sector count and
> offline uncorrectable sectors as -- 4294967295.Fearing the disk going
> bad, i backed up data and started monitoring it. This is the latest
> gsmartcontrol shows - 
> http://img820.imageshack.us/img820/5707/201010161300051280x1024.png
> 
> After two weeks, the no. of bad sectors is the same and the disk shows
> no sign of failure (at the least there is no problem to read or write
> files from or to the disk).hence I am confused .is the disk going bad ?
> if it is fine why can't it remap all those bad sectors ?
> 

That looks a little odd, however the drive does not seem to report that
those attributes are failing (value is above threshold). If I'm not
mistaken, some drives report strange information in some of the smart
fields, it might be your case.

I have an HD that once reported non zero pending sector count and
offline uncorrectable sectors. In my case I believe the sectors were not
damaged, just contained erroneous data for some reason, because writing
over them solved the problem, the reallocated sector count didn't
increase while doing so and the pending sector count and
offline uncorrectable sectors went back to zero after that.

If you can, try overwriting the whole disk with dd, or test it with
badblocks using the write and readback test and see if anything changes.
The not so worst case is that you caught an impending disk faillure
before it caused trouble and you already have a backup, the best case is
that you find out those values are bogus and should not be taken into
account.

-- 
Mauro Santos