[Discuss] New SMART disk errors reported

[cra at WPI.EDU (Chuck Anderson)]

On Mon, Jan 23, 2012 at 11:52:17PM -0500, Doug wrote:
> <email>
> Reallocated sector count has increased in the last day.
> 
> Disk  2:
>  Previous count: 54
>  Current count: 58
> 
> Growing SMART errors indicate a disk that may fail soon.  If the
> errors continue to increase, you should be prepared to replace the
> disk.
> </email>
> 
> The pattern has been like so:
> 
> Jan 16: 17
> Jan 17: 34
> Jan 18: 54
> Jan 21: 58
> 
> I have now copied all the videos to another drive. The NAS has been
> powered off, and instructed to look for errors on the next restart.
> 
> I feel like I can run this disk until its death. Should I expect its
> demise soon?

Yes, it is basically dead already and on life support.  Those
reallocated sectors may or may not have the original data intact on
them, so you may have already had data corruption/loss.  When it runs
out of places to move bad sectors to, it will just start telling the
OS about the bad sector read/writes, instead of hiding that
information from the OS block layer and silently reallocating to spare
sectors as it is doing now.

In my experience, most drives have fewer than 100 spare sectors to
reallocate to, and I always replace the drive after even a single
reallocation event.  The only time I would consider continuing to use
a drive with reallocated sectors on it is if the data is non-critial
and the storage is non-primary storage (i.e. I have a copy elsewhere,
or the data is readily available for re-download from the Internet),
and only if the number of reallocated sectors is not increasing
(i.e. if there was a one-time event with a few sectors, and the drive
isn't getting worse over time).