Troubleshoot RAID Failure

---

Redundant Array of Independent Disks (RAID) is usually configured as a means of protecting data against the risk of a single fixed disk failing.

• data is either copied to a second drive (mirroring) or additional information is recorded on multiple drives to enable them to recover from a device failure (parity)
• RAID can be implemented using hardware controllers or features of the operating system
• made available as a volume, which can be partitioned and formatted in the OS as one or more drives

Causes of RAID Failure

There are two main scenarios for RAID failure:

1. failure of a device within the array
2. failure of the whole array or volume

• If one of the underlying devices fails, the volume will be listed as degraded
  • the data on the volume will still be accessible and it should continue to function as a boot device if so configured
• Most desktop-level RAID solutions can tolerate the loss of only one disk
  • should be replaced as soon as possible
  • If the array supports hot swapping, then the new disk can simply be inserted into the chassis of the computer or into a disk chassis
  • Once this is done, the array can be rebuilt using the RAID configuration utility (if a hardware RAID controller is used) or an OS utility (if you are using software RAID)
  • the rebuilding process is likely to severely affect performance as the controller is probably writing multiple gigabytes of data to the new disk
• If a volume is not available, either more than the tolerated number of disks has failed, or the controller has failed
  • If the boot volume is affected, then the operating system will not start
  • If too many disks have failed, you will have to turn to the latest backup or try to use file recovery solutions
  • If the issue is controller failure, then data on the volume should be recoverable, though there may be file corruption if a write operation was interrupted by the failure
  • Either install a new controller or import the disks into another system