Locality based quorum eligibility

Patent metadata

Field	Value
Publication number	US-11442824-B2
Application number	US-201715650054-A
Country	US
Kind code	B2
Filing date	Jul 14, 2017
Priority date	Dec 13, 2010
Publication date	Sep 13, 2022
Grant date	Sep 13, 2022

Abstract

Disclosed are various embodiments for distributing data items. A plurality of nodes forms a distributed data store. A new master candidate is determined through an election among the plurality of nodes. Before performing a failover from a failed master to the new master candidate, a consensus is reached among a locality-based failover quorum of the nodes. The quorum excludes any of the nodes that are in a failover quorum ineligibility mode.

First claim

Therefore, the following is claimed:

1. A method comprising:
receiving, by a computing device, a temporary maintenance event indicating that the computing device is to enter a failover-quorum-non-eligibility mode;
transitioning, by the computing device, to the failover-quorum-non-eligibility mode from a failover-quorum-eligibility mode in response to receiving the temporary maintenance event;
ceasing to acknowledge, by the computing device, a replication request received from a master computing device, wherein the master computing device and a plurality of computing devices form a distributed data store, wherein the plurality of computing devices reside in a plurality of data centers and includes the computing device;
notifying, by the computing device, the master computing device and a remainder of the plurality of computing devices that the failover-quorum-non-eligibility mode has been entered by the computing device, wherein the computing device is temporarily excluded during the failover-quorum-non-eligibility mode from participating in an election for a new master computing device, the election being based at least in part on a consensus among a quorum of the remainder of the plurality of computing devices;
receiving, by the computing device, a transition event indicating that the computing device is to re-enter the failover-quorum-eligibility mode;
transitioning, by the computing device, to the failover-quorum-eligibility mode responsive to receiving the transition event;
notifying, by the computing device, the master computing device and the remainder of the plurality of computing devices that the failover-quorum-eligibility mode has been re-entered by the computing device; and
receiving, by the computing device, an acknowledgement from at least one of the remainder of the plurality of computing devices, wherein the computing device is included as part of the quorum of the plurality of computing devices that are eligible to participate in a subsequent election for a subsequent master computing device after receiving the acknowledgement.

2. The method of claim 1 , wherein the temporary maintenance event is a message received as part of a graceful shutdown of the computing device.

3. The method of claim 1 , wherein the temporary maintenance event is a result of system operator intervention.

4. The method of claim 1 , further comprising during the failover-quorum-non-eligibility mode, ceasing to acknowledge a data item replication request received from the master computing device.

5. The method of claim 1 , wherein the transition event is a result of a graceful shutdown of the computing device.

6. The method of claim 1 , wherein the transition event is a result of system operator intervention.

7. The method of claim 1 , further comprising after receiving the transition event and before sending a notification of entering the failover-quorum-eligibility mode, processing at least one data item replication request received from the master computing device during the failover-quorum-non-eligibility mode.

8. The method of claim 7 , further comprising determining that the computing device is up to date in processing data item replication requests as compared to other computing devices in the distributed data store.

9. The method of claim 1 , wherein each of the remainder of the plurality of computing devices in the distributed data store is notified that the computing device has entered the failover-quorum-non-eligibility mode.

10. A system comprising:
a computing device comprising a processor and memory, the computing device configured to at least:
receive a temporary maintenance event indicating that the computing device is to enter a failover-quorum-non-eligibility mode;
transition to the failover-quorum-non-eligibility mode from a failover-quorum-eligibility mode in response to receiving the temporary maintenance event;
cease to acknowledge a replication request received from a master computing device, wherein the master computing device and a plurality of computing devices form a distributed data store, wherein the plurality of computing devices reside in a plurality of data centers and includes the computing device;
notify the master computing device and remaining ones of the plurality of computing devices that the failover-quorum-non-eligibility mode has been entered by the computing device, wherein the computing device is temporarily excluded during the failover-quorum-non-eligibility mode from participating in an election for a master node among a quorum of the remaining ones of the plurality of computing devices, the election being based at least in part on a consensus among the quorum of the remaining ones of the plurality of computing devices;
receive a transition event indicating that the computing device is to re-enter the failover-quorum-eligibility mode;
transition to the failover-quorum-eligibility mode responsive to receiving the transition event;
notify the master computing device and at least one of a remainder of the plurality of computing devices that the failover-quorum-eligibility mode has been re-entered by the computing device; and
receive an acknowledgement from at least one of the remaining ones of the plurality of computing devices, wherein the computing device is included as part of the quorum of the plurality of computing devices that are eligible to participate in a subsequent election for a subsequent master computing device after receiving the acknowledgement.

11. The system of claim 10 , wherein the temporary maintenance event is a result of a graceful shutdown of the computing device.

12. The system of claim 10 , wherein the temporary maintenance event is a result of system operator intervention.

13. The system of claim 10 , wherein the computing device is further configured to cease to acknowledge a data item replication request received from the master computing device during the failover-quorum-non-eligibility mode.

14. The system of claim 10 , wherein the transition event is a result of a graceful shutdown of the computing device.

15. The system of claim 10 , wherein the transition event is a result of system operator intervention.

16. The system of claim 10 , wherein the computing device is further configured to after receiving the transition event and before sending a notification of entering the failover-quorum-eligibility mode, process at least one data item replication request received from the master computing device during the failover-quorum-non-eligibility mode.

17. The system of claim 16 , wherein the computing device is further configured to determine that the computing device is up to date in processing data item replication requests as compared to other computing devices in the distributed data store.

18. The system of claim 10 , wherein each of the remaining ones of the plurality of computing devices in the distributed data store is notified that the computing device has re-entered the failover-quorum-non-eligibility mode.

19. A method comprising:
receiving a first notification from a computing device of a plurality of computing devices in a quorum that the computing device is temporarily entering a failover-quorum-non-eligibility mode from a failover-quorum-eligibility mode;
excluding the computing device from being eligible to be part of the quorum during an election of a new master computing device among the quorum of the plurality of computing devices, wherein the computing device, while excluded from being eligible to be part of the quorum, is configured to receive and pr

Assignees

• Amazon Tech Inc

Inventors

• Helmick Michael T
• Kulesza Jakub
• Stefani Stefano
• Lutz David A

Classifications

• G06F11/1425 Primary

  by reconfiguration of node membership · CPC title

• G06F11/2028 Primary

  eliminating a faulty processor or activating a spare · CPC title

• G06F11/2097

  maintaining the standby controller/processing unit updated (initialisation or re-synchronisation thereof G06F11/1658 and subgroups) · CPC title

• G06F16/1844

  Management specifically adapted to replicated file systems · CPC title

• G06F16/273

  Asynchronous replication or reconciliation · CPC title