Methods for transitioning control between two controllers of a storage system

Patent metadata

Field	Value
Publication number	US-9348682-B2
Application number	US-201414470885-A
Country	US
Kind code	B2
Filing date	Aug 27, 2014
Priority date	Aug 30, 2013
Publication date	May 24, 2016
Grant date	May 24, 2016

Abstract

Described herein are techniques for transitioning control between a first and second controller of a storage system. In such transition, the first controller transmits a message to a memory element shared by the first and second controllers, the message capable of notifying the second controller of an imminent failure of the first controller. The second controller receives the message from the shared memory element, the message notifying the second controller of an imminent failure of the first controller. Subsequent to transmitting the message to the shared memory element, the first controller becomes unavailable to facilitate access to the storage devices of the storage system. Subsequent to receiving the message from the shared memory element, the second controller becomes available to facilitate access to the storage devices of the storage system.

First claim

What is claimed is:

1. A method for a storage system, the storage system including a first controller, a second controller and a plurality of storage devices, each of the first and second controllers communicatively coupled to each one of the storage devices, the method comprising:
by the first controller,
(i) starting a first timer that expires after a first time period;
(ii) subsequent to starting the first timer, transmitting a first message from the first controller to a memory element shared by the first and second controllers; and
(iii) subsequent to transmitting the first message to the shared memory element and before the first timer expires, entering a failure state in which the first controller has failed and is unavailable to facilitate access to the storage devices,
wherein the transmission of the first message is in anticipation of the imminent failure of the first controller and is not reactive to a detection of the failure of the first controller; and

by the second controller,
(i) receiving by the second controller the first message from the shared memory element, the first message notifying the second controller of the imminent failure of the first controller;
(ii) subsequent to receiving the first message, starting a second timer that expires after a second time period; and
(iii) upon the second timer expiring, entering an active state in which the second controller is active and is available to facilitate access to the storage devices.

2. The method of claim 1 , wherein a duration of the first time period is adjustable by the first controller.

3. The method of claim 1 , wherein the first message further notifies the second controller of a duration of the first time period.

4. The method of claim 1 , wherein a duration of the second time period is equal to or greater than a duration of the first time period.

5. The method of claim 1 , wherein after the second time period has elapsed following the reception of the first message, further transmitting, by the second controller, a second message to the shared memory element, the second message indicating that the second controller will start facilitating access to the storage devices.

6. The method of claim 1 , wherein the second controller enters the active state without confirming whether the first controller has failed.

7. The method of claim 6 , wherein the second controller does not confirm whether the first controller has failed based on a heartbeat signal of the first controller.

8. A storage system including a first controller, a second controller and a plurality of storage devices, each of the first and second controllers communicatively coupled to each one of the storage devices,
the first controller configured to:
(i) start a first timer that expires after a first time period;
(ii) subsequent to starting the first timer, transmit a first message from the first controller to a memory element shared by the first and second controllers; and
(iii) subsequent to transmitting the first message to the shared memory element and before the first timer expires, enter a failure state in which the first controller has failed and is unavailable to facilitate access to the storage devices,

wherein the transmission of the first message is in anticipation of the imminent failure of the first controller and is not reactive to a detection of the failure of the first controller; and
the second controller configured to:
(i) receive the first message from the shared memory element, the first message notifying the second controller of the imminent failure of the first controller;
(ii) subsequent to receiving the first message, start a second timer that expires after a second time period; and
(iii) upon the second timer expiring, enter an active state in which the second controller is active and is available to facilitate access to the storage devices.

9. The storage system of claim 8 , wherein a duration of the second time period is equal to or greater than a duration of the first time period.

10. The storage system of claim 8 , wherein the second controller enters the active state without confirming whether the first controller has failed.

11. The storage system of claim 10 , wherein the second controller does not confirm whether the first controller has failed based on a heartbeat signal of the first controller.

12. The storage system of claim 8 , wherein a duration of the first time period is adjustable by the first controller.

13. The storage system of claim 8 , wherein the first message further notifies the second controller of a duration of the first time period.

14. The storage system of claim 8 , wherein the second controller is further configured to, after the second time period has elapsed following the reception of the first message, transmit a second message to the shared memory element, the second message indicating that the second controller will start facilitating access to the storage devices.

15. A non-transitory machine-readable storage medium for a storage system having a first controller, a second controller and a plurality of storage devices, each of the first and second controllers communicatively coupled to each one of the storage devices, the non-transitory machine-readable storage medium comprising:
software instructions that, when executed by the first controller, cause the first controller to:
start a first timer that expires after a first time period;
subsequent to starting the first timer, transmit a first message from the first controller to a memory element shared by the first and second controllers; and
subsequent to transmitting the first message to the shared memory element and before the first timer expires, enter a failure state in which the first controller has failed and is unavailable to facilitate access to the storage devices,

wherein the transmission of the first message is in anticipation of the imminent failure of the first controller and is not reactive to a detection of the failure of the first controller; and
software instructions that, when executed by the second controller, cause the second controller to:
(i) receive the first message from the shared memory element, the first message notifying the second controller of the imminent failure of the first controller;
(ii) subsequent to receiving the first message, start a second timer that expires after a second time period; and
(iii) upon the second timer expiring, enter an active state in which the second controller is active and is available to facilitate access to the storage devices.

16. The non-transitory machine-readable storage medium of claim 15 , wherein a duration of the first time period is adjustable by the first controller.

17. The non-transitory machine-readable storage medium of claim 15 , wherein the second controller enters the active state without confirming whether the first controller has failed.

18. The non-transitory machine-readable storage medium of claim 15 , wherein a duration of the second time period is equal to or greater than a duration of the first time period.

19. The non-transitory machine-readable storage medium of claim 15 , wherein the first message further notifies the second controller of a duration of the first time period.

20. The non-transitory machine-readable storage medium of claim 15 , further comprising software instructions that, when executed by the second controller, cause the second controller to, after the second time period has elapsed following the reception of the first message, tran

Assignees

• Nimble Storage Inc

Inventors

• Gupta Vikas
• Tripathi Ashutosh
• Fozard Bob
• Barszczak Tomasz

Classifications

• G06F11/2092 Primary

  Techniques of failing over between control units · CPC title

• G06F11/0757 Primary

  by exceeding a time limit, i.e. time-out, e.g. watchdogs · CPC title

• G06F2201/81

  Threshold · CPC title

• G06F11/0727

  in a storage system, e.g. in a DASD or network based storage system (drivers for digital recording or reproducing units G06F3/06; circuits for error detection or correction within digital recording or reproducing units G11B20/18; 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], H04L67/1097) · CPC title