Subject: Re: twe status queries?
To: der Mouse <mouse@Rodents.Montreal.QC.CA>
From: Thor Lancelot Simon <tls@rek.tjls.com>
List: tech-kern
Date: 12/02/2005 16:30:32
On Fri, Dec 02, 2005 at 04:16:50PM -0500, der Mouse wrote:
> > Or use Raid 10
> 
> RAID 10 would give only 6 drives' worth of space.  What I have gives
> 10; RAID 5 across all 12 would give 11.
> 
> > I don't know if I would run a Raid5 with 12 drives (Raid5 tolerate
> > only one drive failure).
> 
> That's why we have a cold spare.

That's not a good idea.  A common drive failure mode is to not spin up
when power is cycled -- a cold spare protects against two _successive_
failures, as would likely be the case if the two failures occurred with
the drives spinning, but not against two _simultaneous_ failures, as is
the case if two drives don't come back online after a cold restart.

I've had two clients (who both ignored my advice as well as the advice
of other consultants and their own more clueful employees) experience
major data loss because they put too many disks in a single RAID 5 set.
Some large RAID subsystems will now _refuse_ to put more than 8 or so
disks in a single underlying RAID 5 object precisely so their users
don't lose in this way.

RAIDframe has code for P+Q parity.  I don't know if anyone has ever
tested it.  If it works, you can use it to build large parity RAID
sets safely -- but the performance of two RAID 5 sets, striped, may
be better.

Thor