Goldkin <goldkin@[EMAIL PROTECTED]
'tspam.goldkin.net> writes:
> Something sounds odd, though. If a 12-drive array did indeed fail in
> that manner, it sounds to me that Aeto was using something like RAID0
> or a six-drive RAID1, both of which provide fairly minimal redundancy
> (but quite a bit of storage capacity). I'd be curious to know the
> state of his backups, and how far from "the end of the MUCK" they
> fall.
You mean a 6-drive RAID5, right? 6-drive RAID1 would have 5-disk
redundancy, while 6-drive RAID5 would have single disk redundancy.
My guess is, it was 2 6-disk RAID5 arrays combined in a RAID1. That
gives 6-disk redundancy in the best case, and 2-disk redundancy in the
worst, while still giving you 5d storage (where "d" is "the size of any
one disk in the system). It's probably what I would do with 12 hard
drives... my co-workers 3.6 TB fileserver is actually set up almost
exactly like that (only 5 disks in each RAID5).
--
darkside
DC2.Da G"two-spirit" L W-- T- Phvlwlt Sks,wl Cwl,pu B- A(rv+++)
Fr^++"Linux nut" Nw M+++! O H-- $ Fv R++ Ac~ J++ S- U I--# V+
Q+ Tc+++![sw] E++
|
