Promoting device level error to RAIDset level error to restore redundancy in a RAID array data storage system

摘要

Consistency in data in multiple data blocks in a RAID array is restored under the control of metadata recorded in the RAID array. The RAID array has a plurality of members, each member being a storage device. The metadata includes device metadata for data blocks recorded on each member and RAIDset metadata for RAID protected data blocks recorded across the members of the RAID array. The RAID protected data blocks include user data blocks, metadata data blocks and parity data blocks. Consistency is restored in the multiple data blocks by marking each inconsistent block in a RAID protected sliver of blocks. An inconsistent block contains data inconsistent with RAID protection of other blocks in the RAID protected sliver. When a plurality of blocks marked inconsistent in the RAID protected sliver of blocks are detected, each of the inconsistent blocks is written with predetermined data to provide predetermined-data blocks in place the inconsistent blocks. The parity data block in the RAID protected sliver is reconstructed based on the predetermined-data blocks and consistent data in other blocks of the RAID protected sliver. The device FE bits corresponding to said predetermined-data blocks and said parity data block are cleared to indicate all blocks in the RAID protected sliver are now consistent. The RAIDset metadata for the predetermined-data blocks is set to a state to indicate the data in the pre-determined data block is not valid.