System and method for storage area network switch port failure recovery

• US 7,398,421 B1
• Filed: 12/03/2004
• Issued: 07/08/2008
• Est. Priority Date: 12/03/2004
• Status: Active Grant

• Current Assignee: Veritas Technologies, LLC (Whitehouse Group Ltd.)
• Original Assignee: Symantec Operating Corporation (Gen Digital Inc.)
• Inventors: Sharma, Gopal, Limaye, Prasad, Guha, Amitava
• Primary Examiner(s): Chu; Gabriel L
• Application Number:
  US11/004,525
• Time in Patent Office: 1,313 Days
• Field of Search: 714/16, 714/18
• US Class Current: 714/18
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

Systems and methods for storage area network switch port failure recovery. A system may include several hosts, storage devices and a storage area network (SAN) switch. The SAN switch may include several ports, where each host is coupled to a respective host port and each storage device is coupled to a respective storage port. The SAN switch may also include virtualization logic configured to map several of the storage devices as a volume presented to the hosts via the respective host ports, and an input/output (I/O) tracking mechanism configured to store information about write operations in progress at each of the ports. In response to a failure of a given port occurring while write operation(s) to the volume received via one or more of the respective host ports are in progress, the SAN switch may be configured to initiate consistency recovery of the volume using the stored information.

Claims

1. A system comprising:

a plurality of hosts;

a plurality of storage devices; and

a storage area network switch comprising;

a plurality of ports, wherein each of said plurality of hosts is coupled to a respective host one of said plurality of ports, and wherein each of said plurality of storage devices is coupled to a respective storage one of said plurality of ports;

virtualization logic configured to map two or more of said storage devices as a volume presented to one or more of said hosts via said respective host ports, wherein said volume comprises a plurality of data blocks;

an input/output (I/O) tracking mechanism configured to store information about write operations in progress at each of said plurality of ports;

wherein in response to a failure of a given one of said plurality of ports occurring while one or more write operations to said volume received via one or more of said respective host ports are in progress, said storage area network switch is configured to initiate consistency recovery of said volume using said stored information about write operations in progress at said given port.

2. The system as recited in claim 1, wherein for a given write operation, said stored information about write operations in progress at each of said plurality of ports comprises a volume identifier corresponding to said volume and an indication of one or more data blocks being written to said volume.

3. The system as recited in claim 2, wherein recovering consistency of said volume comprises performing a recovery operation only on said one or more data blocks indicated as being written to said volume.

4. The system as recited in claim 2, wherein said volume is configured as a mirrored volume comprising two or more replicas of each of said plurality of data blocks.

5. The system as recited in claim 4, wherein recovering consistency of said volume comprises, for each of said one or more data blocks indicated as being written to said volume, selecting one of said replicas of said indicated block and copying said selected replica to each other replica of said indicated block.

6. The system as recited in claim 1, wherein said storage area network switch is further configured to recover consistency of said volume transparently to said plurality of hosts.

7. The system as recited in claim 1, wherein said I/O tracking mechanism is included in a master one of said plurality of ports.

8. A method comprising:

a storage area network switch mapping two or more of a plurality of storage devices as a volume presented to one or more of a plurality of hosts via respective host ones of a plurality of ports, wherein said volume comprises a plurality of data blocks;

said storage area network switch storing information about write operations in progress at each of said plurality of ports;

in response to a failure of a given one of said plurality of ports occurring while one or more write operations to said volume received via one or more of said respective host ports are in progress, said storage area network switch initiating consistency recovery of said volume using said stored information about write operations in progress at said given port.

9. The method as recited in claim 8, wherein for a given write operation, said stored information about write operations in progress at each of said plurality of ports comprises a volume identifier corresponding to said volume and an indication of one or more data blocks being written to said volume.

10. The method as recited in claim 9, wherein recovering consistency of said volume comprises performing a recovery operation only on said one or more data blocks indicated as being written to said volume.

11. The method as recited in claim 9, wherein said volume is configured as a mirrored volume comprising two or more replicas of each of said plurality of data blocks.

12. The method as recited in claim 11, wherein recovering consistency of said volume comprises, for each of said one or more data blocks indicated as being written to said volume, selecting one of said replicas of said indicated block and copying said selected replica to each other replica of said indicated block.

13. The method as recited in claim 8, wherein said storage area network switch is further configured to recover consistency of said volume transparently to said plurality of hosts.

14. The method as recited in claim 8, wherein said stored information about write operations in progress at each of said plurality of ports is stored in a master one of said plurality of ports.

15. A computer-accessible storage medium, storing program instructions, wherein the program instructions are computer-executable to:

map two or more of a plurality of storage devices as a volume presented to one or more of a plurality of hosts via respective host ones of a plurality of ports of a storage area network switch, wherein said volume comprises a plurality of data blocks;

in response to a failure of a given one of said plurality of ports, access information about write operations in progress at said given port, wherein said information is stored by said storage area network switch; and

in response to determining that said failure occurred while one or more write operations to said volume are in progress, as indicated by said write operation information, initiate consistency recovery of said volume using said write operation information.

16. The computer-accessible storage medium as recited in claim 15, wherein for a given write operation, said write operation information comprises a volume identifier corresponding to said volume and an indication of one or more data blocks being written to said volume.

17. The computer-accessible storage medium as recited in claim 16, wherein recovering consistency of said volume comprises performing a recovery operation only on said one or more data blocks indicated as being written to said volume.

18. The computer-accessible storage medium as recited in claim 16, wherein said volume is configured as a mirrored volume comprising two or more replicas of each of said plurality of data blocks.

19. The computer-accessible storage medium as recited in claim 18, wherein recovering consistency of said volume comprises, for each of said one or more data blocks indicated as being written to said volume, selecting one of said replicas of said indicated block and copying said selected replica to each other replica of said indicated block.

20. The computer-accessible storage medium as recited in claim 15, wherein said program instructions are further executable to recover consistency of said volume transparently to said plurality of hosts.

21. The computer-accessible storage medium as recited in claim 15, wherein said write operation information is stored in a master one of said plurality of ports.