Advertisements It's my understanding, that the most likely failure cause for another drive, beyond the first to fail, is the extra load on all the other drives, necessary to rebuilt the one that failed, to a replacement, or a hot spare, drive. It's also my understanding that if not for the added workload to rebuild, another drive failure was not necessarily "imminent", under the usual working load, but became so due to the change in workload. I'll go out on a limb, and posit that there's really no way to know that one of the remaining drives would fail under the extra load. This is the reason why many say "RAID is not a backup system/solution". I'll make a WAG, that using drives rated for a much heavier workload, than you normally require, when not factoring rebuild load, could lessen, but not eliminate, the threat. That kind of throws the "inexpensive" drive component out of the "I" in RAID. I would think it would be wise to buy drives based on peak "normal" load, plus peak "rebuilding" load, plus room for more (not just "enough"). I'm only "book smart" on this. So, I don't claim to know everything, or know that what I think to be the best approach, actually is, or will even be enough.