Reconstructing data blocks in a raid array data storage system having storage device metadata and raid set metadata

• US 5,826,001 A
• Filed: 10/13/1995
• Issued: 10/20/1998
• Est. Priority Date: 10/13/1995
• Status: Expired due to Term

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Digital Equipment Corporation (HP Inc.)
• Inventors: Sicola, Stephen J., Ellis, Robert A., Lubbers, Clark E., Jackson, James Perry, McLean, Ronald H.
• Primary Examiner(s): Beausoliel, Jr., Robert W.
• Assistant Examiner(s): Do, Luan C.
• Application Number:
  US08/542,670
• Time in Patent Office: 1,103 Days
• Field of Search: 395/182.03, 395/182.04, 395/182.18, 395/404, 395/441, 395/438, 395/439, 371/37.7, 371/40.4, 371/51.1
• US Class Current: 714/6.21
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G11B 20/1833 by adding special lists or symbols to the coded information G11B20/1806, G11B20/1866 take precedence

Abstract

A data block in a RAID array is reconstructed under the control of metadata recorded on the RAID array. The RAID array has a plurality of members, each member being a data 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, RAIDset metadata blocks and parity data blocks. The data blocks are reconstructed by detecting from a device FE bit in the device metadata that a bad data block corresponding to or associated with the device FE bit needs to be reconstructed. The data is read from each data block, other than the bad data block, in the same RAID sliver with bad data block. A RAID sliver of data blocks includes all the data blocks in a RAID protected sliver of data blocks. From the data read from the other data blocks in the RAID sliver with the bad data block, the bad data block is regenerated to produce a new data block. The new data block is written in the RAID array as the reconstructed block in place of the bad data block. The device FE bit for the reconstructed data block is cleared to indicate the reconstructed data block is consistent with the other data blocks in the RAID protected sliver containing the reconstructed data block.

Claims

1. A method for reconstructing a data block in a RAID array, the RAID array having a plurality of data storage devices for storing user data blocks and metadata blocks, the metadata blocks including device metadata blocks associated with data blocks recorded on each storage device and RAIDset metadata blocks associated with RAID protected data blocks recorded across the storage devices of the RAID array, the RAID protected data blocks including the user data blocks and the RAIDset metadata blocks, said method comprising the steps of:

scanning forced error (FE) bits that each correspond to one of the RAIDset metadata blocks;

after scanning the FE bits that each correspond to one of the RAIDset metadata blocks, scanning other FE bits that each correspond to one of the user data blocks;

determining from each scanned FE bit whether a block corresponding to each scanned FE bit needs to be reconstructed; and

reconstructing all RAIDset metadata blocks determined to need reconstructing before reconstructing any user data block that is determined to need reconstructing.

2. The method of claim 1, further comprising the steps of:

reading data from each other block in a RAID protected strip that has a particular block determined to need reconstructing; and

regenerating a new data block from the data read from each other block.

3. The method of claim 2 further comprising the steps of:

writing the new data block in the RAID array as a reconstructed block in place of the particular block determined to need reconstructing; and

clearing the FE bit corresponding to the particular block to indicate that the reconstructed block is consistent with the other blocks in the RAID protected strip that has the reconstructed block.

4. The method of claim 3 further comprising the steps of:

storing another FE bit that corresponds to the particular block in one of the RAIDset metadata blocks; and

clearing the FE bit stored in one of the RAIDset metadata blocks to indicate that the reconstructed block contains valid data.

5. The method of claim 2 wherein said regenerating step comprises the step of:

bitwise exclusive ORing the bits of each other block in the RAID protected strip to produce the new data block.

6. The method of claim 2, further comprising the step of:

storing the FE bits that each correspond to one of the RAIDset metadata blocks in the device metadata blocks.

7. The method of claim 1 wherein the step of scanning FE bits that correspond to the RAIDset metadata blocks scans the FE bits corresponding to the RAIDset metadata blocks in sequence from a first RAIDset metadata block to a last RAIDset metadata block.

8. The method of claim 1 wherein the step of scanning FE bits that correspond to the user data blocks scans the FE bits corresponding to the user data blocks in sequence from a first user data block to a last user data block.

9. The method of claim 1 where each of the scanning steps comprises the step of:

scanning FE Dir bits corresponding to a block of FE bits to determine if any FE Bit in the block of FE bits indicates a block needs to be reconstructed.

10. A method for reconstructing a data block in a RAID protected strip in a RAID array comprising the steps of:

scanning forced error (FE) bits that each correspond to a block of RAIDset metadata;

determining from each scanned FE bit whether the corresponding block of RAIDset metadata needs to be repaired;

repairing each RAIDset metadata block determined to need repair until all RAIDset metadata blocks needing repair are repaired;

after repairing all RAIDset metadata blocks that need repair, scanning other FE bits that each correspond to a block of user data;

determining from each scanned other FE bit whether the corresponding block of user data needs to be repaired; and

repairing each user data block determined to need repair.