Handling Failed Disks

A member’s disk space is protected with RAID and one or two spare disks, except in the case of no-spare RAID policies, as described in Introduction to Member RAID Policies.

If a member has experienced a disk failure, the Member Disks window (Figure 9: Member Disks - PS5000 Array ) shows the failure, and the Alarms panel shows an alarm.

Note: Be sure to replace failed disks as soon as possible. Avoid running with a degraded RAID set. Disk failure in a RAID set that is already degraded will result in loss of data.

When a disk in a RAID set fails, a member behaves as follows:

• Spare disk is available. Data from the failed disk is reconstructed on the spare. During the reconstruction, the RAID set that contains the failed disk will be temporarily degraded. After reconstruction, there is no impact on performance.
• Spare disk is not available, but the RAID set is intact. The RAID set that contains the failed disk is degraded. For RAID 5, performance is diminished.
• Spare disk is not available, and the RAID set is degraded. The member is set offline, and any volumes and snapshots that have data located on the member are set offline. Data may be lost and must be recovered from a backup or replica.

For disk installation information, see Installing and Removing Disks.

When you replace a failed disk, a member behaves as follows:

• Spare disk was used. The new disk automatically becomes a spare. There is no impact on performance.
• RAID set was degraded. Data is automatically reconstructed on the new disk and performance returns to normal after reconstruction.
• Member was offline because of multiple RAID set disk failures. Any volumes and snapshots with data on the member will be offline and the data may be lost.

If you have backups, you may be able to recover volume data, as follows:

1. Remove the offline member from the group, as described in Removing a Member from a Group. Once removed, the array will be automatically reset to its original condition. This will remove all volume data and group configuration information from the array.
2. Delete the affected volumes. The volume snapshots will also be deleted.
3. Recreate the affected volumes.
4. Recover the volume data from a backup or replica.

After installing a disk, check the status to ensure that the disk is either online or a spare, as shown in the Member Disks window (Figure 9: Member Disks - PS5000 Array ).