Chapter 10: Solving Problems ● 85Recovering from a Disk Drive FailureThis section explains how to recover when a disk drive fails:● If the array was protected by a hot spare (see page 85).● If the array was not protected by a hot spare (see page 85).● If there is a disk drive failure in more than one array simultaneously (see page 85).● If it is a RAID 0 array (see page 86).● If multiple disk drives fail within the same array (see page 86).● If the drive is part of the maxCache pool (see page 86).Note: Adaptec Storage Manager uses the term logical drives or logical devices when referringto arrays (see page 13).Failed Disk Drive Protected by a Hot SpareWhen an array is protected by a hot spare, if a disk drive in that array fails the hot spare isautomatically incorporated into the array and takes over for the failed drive.To recover from the failure:1 Remove and replace the failed disk drive.2 If copyback is not enabled—In Adaptec Storage Manager, remove the ‘hot spare’designation from the original hot spare (the disk drive that was built into the array). Then,designate a new hot spare to protect the arrays on that controller.If copyback is enabled—Data is automatically moved back to its original location once thecontroller detects that the failed drive has been replaced. No action is required.Failed Disk Drive Not Protected by a Hot SpareWhen a array is not protected by a hot spare, if a disk drive in that array fails, remove andreplace the failed disk drive. The controller detects the new disk drive and begins to rebuild thearray.If the controller fails to rebuild the array, check that the cables, disk drives, and controllers areproperly installed and connected. Make sure that the new disk drive is equal or greater in sizethan the failed disk drive. Then, if necessary, use Adaptec Storage Manager to rebuild the array.For instructions, refer to the Adaptec Storage Manager User’s Guide or online Help.Failure in Multiple Arrays SimultaneouslyIf there’s a disk drive failure in more than one array at the same time (one failure per array),and the arrays have hot spares protecting them, the controller rebuilds the arrays with theselimitations:● A hot spare must be of equal or greater size than the failed disk drive it’s replacing.● Failed disk drives are replaced with hot spares in the order in which they failed. (The arraythat includes the disk drive that failed first is rebuilt first, assuming an appropriate hotspare is available—see bullet above.)
