...is that double disk failures do occur in the same array - we had two at the same time and another one a few months later. This was nothing to do with ignoring "replace-me" messages, just really bad coincidence of hardware failures. It's why we take and check regular backups so that we can rebuild the entire array from bare metal if we have to. Things got better when we bought Netapp filers with dual redundancy arrays; we are still waiting for three drives to fail at the same time in the same array.