Just what it says on the tin; the failure of more than one hard drive in any storage system that consists of more than one hard disk.
It would be extremely rare for two or more hard drives to fail in the same system simply because the shelf life of their MTBF (mean time before failure) had expired. Usually, it takes something serious such as a power spike or physical knock (such as dropping a storage device).
RAID 0, JBOD and Spanned arrays require that for operation, all drives must be working. RAID 1 through 5 will usually carry on working with a single drive failure, whereas RAID 6 (and many hybrid RAIDs) will support the simultaneous failure of two drives.
We have had numerous RAID 5 arrays in with multiple drive failure; when one drive fails, the array runs in degraded mode until the failed drive is replaced with a good drive, at which point the (intelligent) RAID controller will start rebuilding the array, using parity from the remaining good disks, and incorporate the new disk into the rebuilt array. However, RAID rebuilds are intensive and long-winded, and place far more stress on hard drives than normal operation, and it is during this rebuild that another hard drive can fail – and this is not an uncommon scenario.
