System and method for RAID recovery arbitration in shared disk applications

• US 7,363,426 B2
• Filed: 04/26/2005
• Issued: 04/22/2008
• Est. Priority Date: 04/26/2005
• Status: Active Grant

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Fernandez, Anthony, Ahmadian, Mahmoud
• Primary Examiner(s): Elmore; Stephen C.
• Application Number:
  US11/114,675
• Publication Number:
  US 20060242360A1
• Time in Patent Office: 1,092 Days
• Field of Search: 711/114, 711/158, 711/170, 714/3, 714/6, 713/2, 719/325
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1092 Rebuilding, e.g. when physically replacing a failing disk
• CPC Class Codes: G06F 11/1662 the resynchronized component or unit being a persistent storage device re-synchronization of failed mirror storage G06F11/2082; rebuild or reconstruction of parity RAID storage G06F11/1008
• CPC Class Codes: G06F 11/2089 Redundant storage control functionality

Abstract

A RAID controller is provided for each host sharing a RAID. Each RAID controller can determine whether another host is sharing the RAID and assume a master or slave status with respect to rebuild operations for the shared disk. The master controller may then manage any rebuild operations on rebuild disks within the RAID.

Claims

1. An information handling system comprising:

a redundant array of independent disks (RAID) having a plurality of storage disks;

a first host in operative communication with the RAID and having a first RAID controller; and

a second host having a second RAID controller ready for connection to the RAID;

wherein during connection of the second host to the RAID, and before the second RAID controller has assumed a master controller status or a slave controller status, the second RAID controller is operable to;

determine whether another host is sharing the RAID;

assume a master controller status in response to determining that no other host is sharing the RAID;

assume a slave controller status in response to determining that another host is sharing the RAID; and

the master controller operable to manage a rebuild operation on a rebuild disk within the RAID.

2. The information handling system of claim 1 further comprising the determining step conducted during the initialization of firmware associated with the second RAID controller.

3. The information handling system of claim 1 further comprising:

the first RAID controller comprising firmware for facilitating the determination and rebuild management steps; and

the second RAID controller comprising firmware for facilitating the determination and rebuild management steps.

4. The information handling system of claim 1 further comprising the determining step utilizing a SCSI read buffer command and a SCSI write buffer command.

5. The information handling system of claim 1 further comprising the master controller operable to:

determine a transaction level for a selected period for the first RAID controller and the second RAID controller; and

instruct the RAID controller having a lower transaction level to manage the rebuild operation.

6. The information handling system of claim 5 further comprising the master RAID controller monitoring a transaction count for the first RAID controller and a transaction count for the second RAID controller.

7. The information handling system of claim 1 further comprising:

the first host assigned a first SCSI ID;

the second host assigned a second SCSI ID;

the master controller further operable to;

determine a transaction level for a selected period for the first RAID controller and a transaction level for the second RAID controller;

instruct the RAID controller having a substantially lower transaction level to manage a rebuild operation; and

if the first RAID controller and the second RAID controller have a substantially equal transaction level, utilize the first SCSI ID and the second SCSI ID to determine which RAID controller will manage a rebuild operation.

8. The information handling system of claim 1 wherein the RAID comprises a plurality of virtual disks.

9. The information handling system of claim 8 wherein each virtual disk comprises two or more storage disks.

10. A redundant array of independent disks (RAID) controller operable to communicate with a RAID in a shared disk environment, the RAID controller comprising:

firmware operable to, during connection of the RAID controller to the RAID, and before the RAID controller has assumed a master controller status or a slave controller status;

determine whether another host is sharing the RAID;

assume a master controller status in response to determining that no other host is sharing the RAID;

assume a slave controller status in response to determining that another host is sharing the RAID; and

if the firmware assumes the master controller status, manage a rebuild operation on a rebuild disk within the RAID.

11. The RAID controller of claim 10 further comprising the determining step conducted during the initialization of firmware.

12. The RAID controller of claim 10 further comprising the determining step utilizing SCSI read buffer command and a SCSI write buffer command.

13. The RAID controller of claim 10 further comprising the master controller operable to:

determine a transaction level for a selected period for the first RAID controller and a transaction level for a second RAID controller associated with the RAID; and

instruct the second RAID controller to manage a rebuild operation in response to determining that the second RAID controller has a substantially lower transaction level.

14. The RAID controller of claim 13 further comprising the master RAID controller monitoring a transaction count for the first RAID controller and the second RAID controller.

15. The RAID controller of claim 10 further comprising the master controller operable to:

determine a transaction level of the RAID controller;

compare the transaction level of the RAID Controller to the transaction level of a second RAID controller associated with the RAID;

instruct the second RAID controller to manage a rebuild operation in response to determining that the second RAID controller has a substantially lower transaction level; and

utilize a first SCSI ID assigned to the RAID controller and a second SCSI ID assigned to the second RAID controller to determine to select a rebuild controller in response to determining that the second RAID controller and the RAID controller have substantially equal transaction levels.

16. A method comprising:

in a network including a redundant array of independent disks (RAID) having a plurality of storage disks, providing a first host to be connected to the RAID and having a first RAID controller;

during connection of the first host to the RAID, and before the first RAID controller has assumed a master controller status or a slave controller status;

the first RAID controller determining whether another host is sharing the RAID;

assuming a master controller status in response to determining that no other host is sharing the RAID;

assuming a slave controller status in response to determining that another host is sharing the RAID; and

if the first RAID controller assumes the master controller status, the first RAID controller managing a rebuild operation on a rebuild disk within the RAID.

17. The method of claim 16 further comprising the determining step conducted during the initialization of firmware associated with the first RAID controller.

18. The method of claim 16 further comprising the determining step utilizing SCSI read buffer and SCSI write buffer commands.

19. The method of claim 16 further comprising:

determining a transaction level for a selected period for the first RAID controller and a second RAID controller associated with the RAID; and

instructing the RAID controller having a lower transaction level to manage a rebuild operation.

20. The method of claim 19 further comprising:

monitoring a transaction count for the first RAID controller; and

monitoring a transaction count for the second RAID controller.