System and method for providing failover between controllers in a storage array

• US 8,832,489 B2
• Filed: 04/26/2011
• Issued: 09/09/2014
• Est. Priority Date: 04/26/2011
• Status: Active Grant

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Dawkins, William P., Luning, Stephen G.
• Primary Examiner(s): Riad, Amine
• Application Number:
  US13/094,509
• Publication Number:
  US 20120278652A1
• Time in Patent Office: 1,232 Days
• Field of Search: 714/6, 714/4.11
• US Class Current: 714/4.11
• CPC Class Codes: G06F 11/2092 Techniques of failing over between control units
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 2201/815 Virtual middleware or OS functionality using virtual machines to implement generic software techniques for error detection or fault masking G06F11/1484

Abstract

An information handling system includes first and second memories, and first, second, and third servers. The first server includes a first active application for servicing first input/output requests associated with a first disk, and for storing first state information in the first memory. The second server includes a second active application for servicing second input/output requests associated with a second disk, and for storing second state information in the second memory. The third server includes first and second standby applications, and is configured to monitor the first and second active applications, to detect a failure of the first active application, to suspend the second standby application and to activate the first standby application in response to the failure of first active application.

Claims

1. An information handling system comprising:

a first memory for storing first state information;

a second memory for storing second state information;

a first server including a first active application for servicing first input/output requests associated with a first disk, and for storing the first state information in the first memory;

a second server including a second active application for servicing second input/output requests associated with a second disk, and for storing the second state information in the second memory; and

a third server including a first standby application and a second standby application, the third server configured to monitor the first active application and the second active application, to detect a failure of the first active application, to suspend the second standby application running on a first virtual machine in response to the failure of the first active application, and to activate the first standby application running on a second virtual machine in response to the failure of the first active application, to provide memory, processing capabilities, and other resources of the server to the first standby application in response to the failure of the first active application, wherein the first standby application is configured to actively service the first input/output requests directed to the first active application in response to the failure of the first active application and in response to the memory, the processing capabilities, and the other resources of the server being provided to the second standby application.

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

a switch to direct the first input/output request to the first active application and the second input/output request to the second active application.

3. The information handling system of claim 2 further comprising:

a management module in communication with the switch, the management module configured to reconfigure the switch to direct the first input/output requests previously directed to the first active application to the first standby application.

4. The information handling system of claim 3 wherein the management module is further configured to reconfigure the switch to direct the first input/output requests to the first active applications instead of the first standby application in response to the first active application being restarted.

5. The information handling system of claim 1 wherein the first standby application is further configured to retrieve the first state information from the first memory, to determine whether a current input/output request is in progress based on the first state information, and if the current input/output request is in progress, processing and completing the input/output request.

6. The information handling system of claim 1 wherein the third server is further configured to determine that the first active application has been restarted, and to restart the second standby application in response to the one active application being restarted.

7. The information handling system of claim 1 wherein the first memory and the second memory are non-volatile random access memories.

8. The information handling system of claim 1 wherein the first input/output requests are selected from a group consisting of a request to read data from a disk and a request to write data to the disk.

9. A method comprising:

monitoring, by a server including a plurality of standby applications, a plurality of active applications;

detecting a failure of one of the active applications;

suspending a first virtual machine running a first standby application of the standby applications in response to the failure of the one active application;

activating a second standby application of the standby applications running on a second virtual machine in response to the failure of the one active application;

providing memory, processing capabilities, and other resources of the server to the second standby application in response to the failure of the one active application;

reconfiguring a switch to direct input/output requests from a client previously directed to the one active application to the second standby application; and

actively servicing, by the second standby application, the input/output requests from the client in response detecting the failure of the one active application and in response to the memory, the processing capabilities, and the other resources of the server being provided to the second standby application.

10. The method of claim 9 further comprising:

retrieving, by the second standby application, state information from a memory;

determining whether a current input/output request is in progress based on the state information; and

if the current input/output request is in progress, processing and completing, by the second standby application, the input/output request.

11. The method of claim 10 further comprising:

determining that the one active application has been restarted;

reconfiguring the switch to direct the input/output requests to the one active application instead of the second standby application in response to the one active application being restarted; and

restarting the first virtual machine running the first standby application in response to the one active application being restarted.

12. The method of claim 9 wherein the failure of the one active application is caused by a server crashing, the server losing power, or any combination thereof.

13. The method of claim 9 wherein the plurality of standby applications are virtual applications located on the server.

14. The method of claim 9 wherein the memory is a non-volatile random access memory.

15. The method of claim 9 wherein the current input/output request is to read data from a disk, to write data to the disk, or any combination thereof.

16. A method comprising:

monitoring, by a server including a plurality of standby applications, a plurality of active applications;

detecting a failure of one of the active applications;

suspending a first virtual machine running a first standby application of the standby applications in response to the failure of the one active application;

activating a second standby application of the standby applications running on a second virtual machine in response to the failure of the one active application;

providing memory, processing capabilities, and other resources of the server to the second standby application;

reconfiguring a switch to direct input/output requests from a client previously directed to the one active application to the second standby application;

actively servicing, by the second standby application, the input/output requests from the client in response to the memory, the processing capabilities, and the other resources of the server being provided to the second standby application;

retrieving, by the second standby application, state information from a memory;

determining whether a current input/output request is in progress based on the state information; and

if the current input/output request is in progress, processing and completing, by the second standby application, the input/output request.

17. The method of claim 16 further comprising:

determining that the one active application has been restarted;

reconfiguring the switch to direct the input/output requests to the one active application instead of the second standby application in response to the one active application being restarted; and

restarting the first virtual machine running the first standby application in response to the one active application being restarted.

18. The method of claim 16 further comprising:

suspending a third virtual machine running a third standby application of the standby applications in response to the failure of the one active application; and

restarting the third virtual machine running the third standby application in response to the one active application being restarted.

19. The method of claim 16 wherein each of the active applications are located on separate servers.

20. The method of claim 16 wherein the plurality of standby applications are virtual applications located on the server.