I've read various articles along those lines. My understanding is that a 500GB odd raid-z / raid-5 array has around a 1 in 10 chance of loosing at least some data during a rebuild.
I'd have raid-5 arrays fail at least 4 times, twice during a rebuild. In most cases I've been able to recover the data (once by re-attaching the original failed drive since it proved more reliable than the 2nd one that failed). However on more than one occasion I've had to revert to backups. Raid-6 was something I was waiting a long time for. Now I use dual parity for everything I buy. At home I've a six drive raid-z2 box, at work the main server is a 16 drive 2 way mirror setup. When using SATA drives capacity is cheap enough (that work server is still 2.5TB for around £2,500) and the peace of mind, particularly on the company servers is worth every penny. If you're stuck with single parity raid-z, my advice would be to simply take a good set of backups and leave it at that until you can upgrade to dual parity. At the end of the day, the risk is relatively slight, and you're data's probably as much risk if you try to pro-actively replace a drive as if you just replace one when it fails. Just scrub every so often, and make sure you've got good backups. I don't expect you'll see too many problems. This message posted from opensolaris.org _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
