DISK ARRAY CONTROLLER CAPABLE OF DETECTING AND CORRECTING FOR UNEXPECTED DISK DRIVE POWER-ON-RESET EVENTS

• US 20090292875A1
• Filed: 08/05/2009
• Published: 11/26/2009
• Est. Priority Date: 12/05/2003
• Status: Active Grant

• Current Assignee: Summit Data Systems LLC (Acacia Research Corporation)
• Original Assignee: Summit Data Systems LLC (Acacia Research Corporation)
• Inventors: Therene, Christophe, Stonelake, Paul R., Harris, Richard L., Tang, Alex Ga Hing
• Granted Patent: US 8,458,495 B2
• Time in Patent Office: Days
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1441 Resetting or repowering
• CPC Class Codes: G06F 12/0804 with main memory updating G06F12/0806 takes precedence
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache

Abstract

A disk array controller detects disk drive power-on-reset events that may cause a disk drive to lose uncommitted write data stored in its cache. When an unexpected disk drive power-on-reset event is detected, the disk array controller may initiate an appropriate corrective action. For example, the disk array controller may initiate a disk drive rebuild operation, or may re-send a set of write commands to the disk drive.

Claims

1. A disk array controller that is operative to connect to and control a plurality of disk drives, the disk array controller comprising:

a memory that stores program code; and

a processor that is operable to execute said program code, said processor programmed via said program code to perform at least the following functions;

assess whether any of the disk drives have experienced unexpected power-on-reset events; and

when a disk drive is determined to have experienced an unexpected power-on-reset event, initiate a corrective action to correct for a potential data loss caused by the unexpected power-on-reset event.

2. The disk array controller of claim 1, wherein the corrective action is a reconstruction operation in which data written to said disk drive is reconstructed based on data stored on one or more other disk drives of said plurality of disk drives.

3. The disk array controller of claim 1, wherein the corrective action comprises resending a previously-sent set of write commands to the disk drive.

4. The disk array controller of claim 1, wherein the corrective action comprises resending write data to the disk drive to correct for a potential loss of write data that was stored in a cache of the disk drive but not yet committed to disk media at the time of the power-on-reset event.

5. The disk array controller of claim 1, wherein the processor is programmed to assess whether any of the disk drives have experienced unexpected power-on-reset events based at least partly on power cycle count attributes read from the disk drives.

6. The disk array controller of claim 1, wherein the processor is programmed to perform a power-on-reset test of a disk drive in response to detecting disk drive behavior reflective of a possible power-on-reset event.

7. The disk array controller of claim 1, wherein the processor is programmed to detect the unexpected power-on-reset event at least partly by comparing a current power cycle count attribute value of the disk drive to a past power cycle count attribute value of the disk drive in response to detecting an event which indicates that the disk drive may have undergone a power-on-reset event.

8. The disk array controller of claim 1, wherein the processor is programmed to test for said unexpected power-on-reset event in response to detecting that the disk drive has aborted a command after being placed into an unlocked state.

9. A method performed by a disk array controller that controls a plurality of disk drives, the method comprising:

determining, based on communications with the disk drives, whether any of the disk drives have experienced unexpected power-on-reset events; and

when a disk drive is determined to have experienced an unexpected power-on-reset event, initiating a corrective action to correct for a potential data loss caused by the unexpected power-on-reset event.

10. The method of claim 9, wherein the corrective action is a reconstruction operation in which data written to said disk drive is reconstructed based on data stored on one or more other disk drives of said plurality of disk drives.

11. The method of claim 9, wherein the corrective action comprises resending a previously-sent set of write commands to the disk drive.

12. The method of claim 9, wherein the corrective action comprises resending write data to the disk drive to correct for a potential loss of write data cached by the disk drive but not yet committed to by the disk drive to disk media.

13. The method of claim 9, wherein determining whether any of the disk drives have experienced unexpected power-on-reset events comprises comparing current power cycle count attributes of the disk drives to previously read power cycle count attributes of the disk drives.

14. The method of claim 9, wherein determining whether any of the disk drives have experienced unexpected power-on-reset events comprises performing a power-on-reset test of a disk drive in response to detecting disk drive behavior reflective of a possible power-on-reset event.