Path switching method

• US 20070234113A1
• Filed: 05/26/2006
• Published: 10/04/2007
• Est. Priority Date: 03/29/2006
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Aoki, Makoto, Komatsu, Yuki
• Granted Patent: US 7,634,691 B2
• Time in Patent Office: Days
• US Class Current: 714/8
• CPC Class Codes: G06F 11/201 between storage system components
• CPC Class Codes: G06F 3/0611 in relation to response time
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

Provided is a computer system including at least one host computer; and at least one storage system, characterized in that: the storage system has a disk drive and a disk controller, and provides a storage area of the disk drive as at least one logical unit; upon detecting a failure in a logical path serving as an access route from the host computer to the logical unit, the host computer specifies logical paths for accessing the same logical unit that is connected to the logical path where the failure is detected; the host computer executes failure detecting processing for the specified logical paths to judge whether the specified logical paths are normal or not; the host computer selects normal logical paths out of the specified logical paths; and the host computer accesses the logical unit via the normal logical paths selected.

Claims

1. A computer system, comprising:

at least one host computer having a processor, a memory, and an interface; and

at least one storage system connected to the host computer,

wherein the storage system has a disk drive for storing data that is requested to be written by the host computer and a disk controller for controlling the disk drive, and provides a storage area of the disk drive as at least one logical unit to the host computer, and

wherein upon detecting a failure in a logical path serving as an access route from the host computer to the logical unit, the host computer specifies logical paths for accessing the logical unit shared by the logical path where the failure is detected,

wherein the host computer executes failure detecting processing for the specified logical paths to judge whether the specified logical paths are normal or not,

wherein the host computer selects normal logical paths out of the specified logical paths, and

wherein the host computer accesses the logical unit via the selected normal logical paths.

2. The computer system according to claim 1,

wherein the host computer chooses one of the selected normal logical paths by round robin, and

wherein the host computer accesses the logical unit via the chosen logical path.

3. The computer system according to claim 1,

wherein the host computer stores logical path connection information, which indicates association between the logical path and components passed by the logical path,

wherein the host computer blocks the logical path where a failure is detected and the logical path that is judged as abnormal through the failure detecting processing, and

wherein the host computer estimates, as a cause of the failure, the component passed by only the blocked logical paths by referring to the logical path connection information.

4. The computer system according to claim 3,

wherein the disk controller has at least one channel adapter connected to the host computer,

wherein the host computer has at least one host bus adapter connected to the storage system, and

wherein the host computer estimates the host bus adapter, the channel adapter, or the logical unit as a cause of a failure.

5. The computer system according to claim 3, further comprising a management server connected to the host computer,

wherein the host computer notifies the management server of the estimated cause of the failure.

6. The computer system according to claim 5,

wherein the management server notifies the cause of the failure, which is notified by a first host computer, to a second host computer, which is not the first host computer,

wherein the second host computer specifies the logical paths that pass the notified cause of the failure, and

wherein the second host computer executes failure detecting processing for the specified logical paths to judge whether the specified logical paths are normal or not.

7. The computer system according to claim 6,

wherein the second host computer stores logical path connection information indicating association between the logical path and components passed by the logical path,

wherein the second host computer blocks the logical path that is judged as abnormal through the failure detecting processing,

wherein the second host computer estimates, as a cause of the failure, the component passed by only the blocked logical path by referring to the logical path connection information, and

wherein the second host computer notifies the management server of the estimated cause of the failure.

8. The computer system according to claim 7, wherein the management server judges a validity of the cause of the failure that is notified from the first host computer through comparison with the cause of the failure that is notified from the second host computer.

9. The computer system according to claim 1, wherein the host computer executes failure detecting processing also for the logical path for accessing the logical unit that is different from the logical unit accessed by the logical path where the failure is detected.

10. A logical path switching method for a computer system that includes at least one host computer having a processor, a memory, and an interface, and at least one storage system connected to the host computer,

wherein the storage system has a disk drive for storing data that is requested to be written by the host computer and a disk controller for controlling the disk drive, the path switching method comprising the steps of:

providing, by the storage system, a storage area of the disk drive to the host computer;

upon detecting a failure in a logical path serving as an access route from the host computer to the logical unit, specifying, by the host computer, logical paths for accessing the logical unit shared by the logical path where the failure is detected;

executing, by the host computer, failure detecting processing for the specified logical paths to judge whether the specified logical paths are normal or not;

selecting, by the host computer, normal logical paths out of the specified logical paths; and

accessing, by the host computer, the logical unit via the selected normal logical paths.

11. The logical path switching method according to claim 10, further comprising the steps of:

choosing, by the host computer, one of the selected normal logical paths by round robin, and

accessing, by the host computer, the logical unit via the chosen logical path.

12. The logical path switching method according to claim 10, further comprising the steps of:

storing, by the host computer, logical path connection information, which indicates association between the logical path and components passed by the logical path,

blocking, by the host computer, the logical path where a failure is detected and the logical path that is judged as abnormal through the failure detecting processing, and

estimating, the host computer, as a cause of the failure, the component passed by only the blocked logical paths by referring to the logical path connection information.

13. The logical path switching method according to claim 12,

wherein the disk controller has at least one channel adapter connected to the host computer,

wherein the host computer has at least one host bus adapter connected to the storage system,

The logical path switching method further comprising the step of estimating host bus adapter, the channel adapter, or the logical unit as a cause of a failure.

14. The logical path switching method according to claim 12,

wherein the computer system further includes a management server connected to the host computer, and

The logical path switching method further comprising the step of notifying, by the host computer, the management server of the estimated cause of the failure.

15. The logical path switching method according to claim 14, further comprising the steps of:

notifying, by management server, the cause of the failure, which is notified by a first host computer, to a second host computer, which is not the first host computer,

specifying, by the second host computer, the logical paths that pass the notified cause of the failure, and

executing, by the second host computer, failure detecting processing for the specified logical paths to judge whether the specified logical paths are normal or not.

16. The logical path switching method according to claim 15, further comprising the steps of:

storing, by the second host computer, logical path connection information indicating association between the logical path and components passed by the logical path,

blocking, by the second host computer, the logical path that is judged as abnormal through the failure detecting processing,

estimating, by the second host computer, as a cause of the failure, the component passed by only the blocked logical path by referring to the logical path connection information; and

notifying, by the second host computer, the management server of the estimated cause of the failure.

17. The logical path switching method according to claim 16, further comprising the step of judging, by the management server, a validity of the cause of the failure that is notified from the first host computer through comparison with the cause of the failure that is notified from the second host computer.

18. The logical path switching method according to claim 10, further comprising the step of executing, by the host computer, failure detecting processing also for the logical path for accessing the logical unit that is different from the logical unit accessed by the logical path where the failure is detected.

19. A logical path switching method for a computer system that includes a plurality of host computers, each containing a processor, a memory, and an interface, at least one storage system connected to the host computers, and a management server containing a processor, a memory, and an interface and connected to the host computers,

the storage system having a disk drive for storing data that is requested to be written by the host computers and a disk controller for controlling the disk drive,

the host computers having a first host computer and a second host computer,

the logical path switching method comprising:

providing, by the storage system, a storage area of the disk drive as one or more logical units to the first host computer and the second host computer;

storing, by the first host computer, logical path connection information indicating association between a logical path serving as an access route from the first host computer to the logical unit and components passed by the logical path;

identifying, by the first host computer, which ones of all logical paths are currently blocked;

executing, by the first host computer, failure detecting processing for the logical paths identified as blocked paths to judge whether the logical paths identified as blocked paths are normal or not;

selecting, by the first host computer, normal logical paths out of the logical paths identified as blocked paths;

referring, by the first host computer, to the logical path connection information to estimate, as a component that has recovered from a failure, one of failure causing components passed by the selected normal logical paths;

notifying, by the first host computer, the management server of the estimated failure recovery component; and

notifying, by the management server, the notified failure recovery component to the second host computer.