Fixing RAID redundancy failures

If the status of any volume reports Failed Redundancy, Failed Redundancy (At Risk), or just Failed, use one of the following procedures to fully recover it:

Procedure 1 β€”To reactivate a volume in Failed Redundancy state, try the following:

1. In the Disk Management snap-in, switch to Graphical View by selecting View O Bottom O Graphical View.

2. If a disk hosting the volume is listed in Missing, Offline, or Online (Errors) state, right-click it and select Reactivate Disk. If the reactivation succeeds, the disk returns to Online and the volume to Healthy status.

Procedure 2 β€” When a Failed Redundancy volume does not recover using the previous approach, try the following:

1. If the disk returned to the Online status, attempt to reactivate the volume manually. This is done using the Reactivate Volume option (although typically this should take place automatically after performing the steps in Procedure 1).

2. If the disk is still listed as Offline or Missing, then the problem is most likely related to a nonfunctioning disk or loose or failed connectors. To replace a failed mirror, right-click any of disk areas participating in the mirror, select the Remove Mirror option, and follow the wizard. Replace the bad disk and follow the instructions for creating mirrors using existing dynamic volumes provided in the section "Creating and managing RAID-1 volumes," earlier in this chapter. To repair a RAID-5 volume, first eliminate the cause of the problem. If the disk needs to be replaced, do so, initialize it, and convert it to dynamic. Then, use the Repair Volume command from the context menu of the RAID-5 volume.

Procedure 3 β€” To attempt the reactivation of a volume in a Failed Redundancy (At Risk) state, try the following:

1. Try to reactivate the disk hosting the volume (typically listed with Online [Errors] status) in the same fashion as in Procedure 1. If necessary, attempt to reactivate the volume manually using Procedure 2.

2. A change of status to Failed Redundancy (without At Risk) after the first step is usually a good sign, and repeating the steps in Procedures 1 and 2 (if needed) should return the disk to Healthy status. The likely problem is that the volume data is out of sync (for a mirror) or parity information needs to be regenerated (for RAID 5). Use the Reactivate or Resynchronize commands and then run chkdsk against the volume. If this fails, you may need to replace the hardware. You will need to have a valid backup.

Procedure 4 β€” To replace disks, do the following:

1. If you have spare dynamic disks, import them using the Import Foreign Disk option. Otherwise, install the disks as basic and convert them to dynamic.

2. Using the methods described earlier in this chapter, return the volume to its original state.

Was this article helpful?

+5 -2

Post a comment