--- On Tue, 9/2/10, Dawning Sky <the.dawning.sky@xxxxxxxxx> wrote:
> From: Dawning Sky <the.dawning.sky@xxxxxxxxx>
> Subject: Re: Disk I/O error while rebuilding an md raid-5 array
> To: "Greg Freemyer" <greg.freemyer@xxxxxxxxx>
> Cc: linux-raid@xxxxxxxxxxxxxxx
> Date: Tuesday, 9 February, 2010, 0:45
> On Mon, Feb 8, 2010 at 4:04 PM, Greg
> Freemyer <greg.freemyer@xxxxxxxxx>
> wrote:
> >>
> >> PS, if in the end I have to build a new array,
> I'll probably go with a
> >> raid 6 instead.
> >
> > Agreed, someone recently posted that for a raid-5
> composed of 1TB
> > drives the odds of a rebuild failure are 1 in 67 even
> if the remaining
> > drives are within spec. (ie. the unrecoverable bit
> error rate is
> > slowing succumbing to the ever increasing size of
> drives.)
>
> Has anyone done similar calculation for raid-6?
>
> >
> > Greg
I am not sure of the math, but i was under the impression the odds were calculated PER DRIVE, not combining the array size. I HAD a 6 drive array comprised of 4x500GB and 2x750 (with some wasted space) as I wanted to push the new drives into 1~+ TB drives to make use of this lost space I decided to go RAID6 so i now have a 7 drive array.. 4x500GB 2x750 1x1TB. Now i am happy it is ok (see lots of earlier posts about mismatches) I am gearing up to replace 2 of the 500GB's with 1TB drives, and in 6 months I will repeat.
Bottom line, my 'gut' says if your drives are going to stray over 1TB go to RAID6 also, I have staggered DAILY smart reads (i think each drive gets 1 day a week 'off' and the rest are either long or short but never all doing 1 type) i also do a weekly 'scrub' of my array (this is so amazingly important).
My only downside is due to FSB speed issues i dont get more than 30m/s when accessing all 7 drives simultaneously but that is enough for RAID.
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
