MANAGEMENT SYSTEM OF SERVER SYSTEM INCLUDING A PLURALITY OF SERVERS

• US 20160314050A1
• Filed: 01/16/2014
• Published: 10/27/2016
• Est. Priority Date: 01/16/2014
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: CHO, RAI, NAKAYAMA, Jun
• Granted Patent: US 9,921,926 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/16 Error detection or correction of the data by redundancy in hardware
• CPC Class Codes: G06F 11/20 using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
• CPC Class Codes: G06F 11/2023 Failover techniques
• CPC Class Codes: G06F 11/2041 with more than one idle spare processing component
• CPC Class Codes: G06F 11/2046 where the redundant components share persistent storage G06F11/2043 takes precedence
• CPC Class Codes: G06F 2201/845 Systems in which the redundancy can be transformed in increased performance

Abstract

For each of N active servers (N being an integer equal to or larger than 2), a management system performs, on at least one of M standby servers (M being an integer equal to or larger than 2), a full test for determining whether a failover is executable by performing a failover from the active server to the standby server, and performs, on at least one of the standby servers that is different from the standby server on which the full test is performed, a simplified test for determining whether the failover is executable without performing the failover from the active server to the standby server, the number of standby servers on which the simplified test is performed being larger than the number of standby servers on which the full test is performed.

Claims

1. A management system comprising:

an interface device coupled to a server system including N active servers (N being an integer equal to or larger than

     2) and M standby servers (M being an integer equal to or larger than

     2), and a storage system that provides a plurality of LUs (Logical Units) to the N active servers; and

a processor coupled to the interface device and a storage device, the storage device being configured to store server management information including information indicating for each server whether the server is an active server or a standby server,

the processor being configured to determine a plurality of server combinations based on the server management information,

each of the server combinations being a combination between any one of the N active servers and any one of the M standby servers,

the processor being configured to execute diagnosis processing for performing a test to determine whether a failover is executable for each of the plurality of server combinations,

the test being a full test for determining whether the failover is executable by performing a failover from the active server to the standby server or a simplified test for determining whether the failover is executable without performing the failover from the active server to the standby server, and

in the diagnosis processing, the processor being configured to perform, for each of the N active servers, the full test on at least one of the standby servers, and the simplified test on at least one of the standby servers that is different from the standby server on which the full test is performed, the number of standby servers on which the simplified test is performed being larger than the number of standby servers on which the full test is performed.

2. The management system according to claim 1, wherein the processor is configured to perform tests of the same type in parallel on M of the N active servers, with respect to the M standby servers.

3. The management system according to claim 1, wherein when any one of the N active servers fails, the processor is configured to preferentially select, as a failover destination of a failed server that is an active server that has failed, a standby server on which the simplified test has succeeded among a plurality of standby servers associated with the failed server.

4. The management system according to claim 1, wherein each time the test is performed on the M standby servers, the processor is configured to determine whether the test is executable on the standby servers, and perform the test on the standby server for which a result of the determination is true.

5. The management system according to claim 1, wherein

at least one simplified test is an LU reference test that is a test including determining whether the standby server is able to read data from an LU recognized by the active server, and

in the LU reference test, for the storage system, the processor is configured to permit a standby server, in a server combination on which the LU reference test is executed, to only read an LU recognized by the active server in the server combination on which the LU reference test is performed.

6. The management system according to claim 1, wherein

the storage device is configured to store diagnosis management information,

the diagnosis management information includes a test result for each of the plurality of server combinations, and

the processor is configured to

display first information indicating an active server for which, among the standby servers associated therewith, the number of standby servers with a failed test result is equal to or larger than a predetermined number, based on the diagnosis management information,

receive selection of the active server in the first information, and

further display second information indicating, for the selected active server, a test result for each standby server on which the test is performed, based on the diagnosis management information.

7. The management system according to claim 1, wherein

the storage device is configured to store diagnosis management information,

the diagnosis management information includes a test result for each of the plurality of server combinations, and

the processor is configured to display information indicating an active server associated with standby servers that have all failed in the full test, based on the diagnosis management information.

8. The management system according to claim 1, wherein

the storage device is configured to store diagnosis management information,

the diagnosis management information includes a test result for each of the plurality of server combinations, and

the processor is configured to display summary information that is information including the number of active servers for which the number of standby servers with a failed test result is equal to or larger than a predetermined number, based on the diagnosis management information.

9. The management system according to claim 4, wherein

when the full test is performed on at least one of the standby servers for each of the N active servers, the processor is configured to start the server system even when the diagnosis processing is in process, and preform rest of the diagnosis processing after the server system is started, and

the processor is configured to perform the full test before the simplified test for each of the N active servers.

10. The management system according to claim 1, wherein in the full test, the processor is configured to perform the failover from the active server to the standby server, and then perform a failback from the standby server to the active server.

11. The management system according to claim 1, wherein there are, as the simplified tests, a first simplified test that has a lower reliability than the full test but is completed in a shorter time than the full test and a second simplified test that has a lower reliability than the first simplified test but is completed in a shorter time than the first test.

12. The management system according to claim 11, wherein, for each of the N active servers, the number of standby servers on which the first simplified test is performed is larger than the number of standby servers on which the full test is performed, and the number of standby servers on which the second simplified test is performed is larger than the number of standby servers on which the first simplified test is performed.

13. The management system according to claim 11, wherein when any one of the N active servers fails, the processor is configured to preferentially select, as a failover destination of a failed server that is an active server that has failed, a standby server with a lowest reliability for a test that has succeeded among a plurality of standby servers associated with the failed server.

14. The management system according to claim 1, wherein

the first simplified test is an LU reference test that is a test including determining whether the standby server is able to read data from an LU recognized by the active server, and

the second simplified test is an LU recognition test that is a test including determining whether the standby server is able to recognize the LU recognized by the active server, and not including determining whether the standby server is able to read data from the LU recognized by the active server.

15. A computer program executed on a computer coupled to a server system including N active servers (N being an integer equal to or larger than 2) and M standby servers (M being an integer equal to or larger than 2), and a storage system that provides a plurality of LUs (Logical Units) to the N active servers, the computer program comprising:

performing, on at least one of the standby servers for each of the N active servers, a full test for determining whether a failover is executable by performing a failover from the active server to the standby server; and

performing, on at least one of the standby servers for each of the N active servers, the standby servers being different from the standby server on which the full test is performed, a simplified test for determining whether the failover is executable without performing the failover from the active server to the standby server, the number of standby servers on which the simplified test is performed being larger than the number of standby servers on which the full test is performed.