zfs i/o error, no driver error

[Bob Friesenhahn]

On Mon, 7 Jun 2010, Martin Simmons wrote:
>
> It doesn't conflict if you agree that freshly written data is more likely to
> be readable that data written long ago (with some curve in between).

Depending on the actual failure mechanism, the inverse may actually be 
true.  Freshly written data may be trash while old data still reads 
fine.

> I don't know if there is any science behind that theory...

The science is continually changing.  A study done even 5 or 7 years 
ago may no longer be relevant.  Regardless, actual results seen in the 
field count more than any theory.

Bob
-- 
Bob Friesenhahn
bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/
GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/