> > "Power off during write operations may make an incomplete sector which
> > will report hard data error when read. The sector can be recovered by a
> > rewrite operation."
>
> So the proper defect management would be to simply initialize the broken
> sector once a fsck hits it (still, I've never seen disks develop so many
> bad blocks so quickly as those failed DTLA-307045 drives I had).
>
> Note, the specifications say that the write cache setting is ignored
> when the drive runs out of spare blocks for reassignment after defects
> (so that the drive can return the error code right away when it cannot
> guarantee the write actually goes to disk).
They should turn off write-back after number-of-spare-block < cache-size,
otherwise they are not safe.
Pavel
-- Philips Velo 1: 1"x4"x8", 300gram, 60, 12MB, 40bogomips, linux, mutt, details at http://atrey.karlin.mff.cuni.cz/~pavel/velo/index.html.- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/