Method and system for establishing a quorum for a geographically distributed cluster of computers

• US 7,016,946 B2
• Filed: 07/05/2001
• Issued: 03/21/2006
• Est. Priority Date: 07/05/2001
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Inventors: Shirriff, Kenneth W.
• Primary Examiner(s): WINDER, PATRICE L
• Application Number:
  US09/900,298
• Publication Number:
  US 20030023680A1
• Time in Patent Office: 1,720 Days
• Field of Search: 709220-222, 709/224, 714/4, 714/6
• US Class Current: 709/221
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/1425 by reconfiguration of node membership
• CPC Class Codes: G06F 11/1658 Data re-synchronization of a redundant component, or initial sync of replacement, additional or spare unit
• CPC Class Codes: G06F 11/2035 without idle spare hardware
• CPC Class Codes: G06F 11/2048 where the redundant components share neither address space nor persistent storage
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/125 involving control of end-device applications over a network
• CPC Class Codes: H04L 67/14 Session management for real-time applications in data packet communications networks H04L65/1066
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

One embodiment of the present invention provides a system that facilitates establishing a quorum for a cluster of computers that are geographically distributed. The system operates by detecting a change in membership of the cluster. Upon detecting the change, the system forms a potential new cluster by attempting to communicate with all other computers within the cluster. The system accumulates votes for each computer successfully contacted. The system also attempts to gain control of a quorum server located at a site separate from all computers within the cluster. If successful at gaining control, the system accumulates the quorum server'"'"'s votes as well. If the total of accumulated votes is a majority of the available votes, the system forms a new cluster from the potential new cluster.

Claims

1. A method that facilitates establishing a quorum for a cluster within a plurality of computers that are geographically distributed, the method comprising:

detecting a change in membership of the cluster at a computer within the plurality of computers; and

upon detecting the change in membership,

forming a potential new cluster by attempting to communicate with all other computers within the plurality of computers,

accumulating votes for each computer successfully contacted,

attempting to gain control of a quorum server located at a site that is geographically separate from all computers within the plurality of computers,

if successful, accumulating the quorum server'"'"'s votes, and

if a total of accumulated votes includes a majority of available votes, forming a new cluster from the potential new cluster.

2. The method of claim 1, wherein detecting the change in membership involves:

exchanging heartbeat messages with all computers that are part of the cluster; and

upon discovering an absence of heartbeat messages from any computer in the cluster, initiating a cluster membership protocol.

3. The method of claim 1, wherein detecting the change in cluster membership includes detecting that the cluster has not been formed.

4. The method of claim 1, wherein attempting to gain control of the quorum server involves communicating with the quorum server using cryptographic techniques.

5. The method of claim 1, further comprising:

exchanging a status message with each member of the new cluster; and

updating a local status at the computer to a most recent status available within the status message.

6. A method that facilitates establishing a quorum for a cluster within a plurality of computers that are geographically distributed, the method comprising:

providing a quorum server at a site that is geographically separate from a location of a computer within the plurality of computers;

assigning at least one vote to each computer within the plurality of computers;

assigning at least one vote to the quorum server;

attempting to establish communications between each pair of computers within the plurality of computers;

accumulating a count of votes for each computer communicated with at each computer;

attempting to establish control over the quorum server from each computer within the plurality of computers;

if control is established over the quorum server, accumulating the quorum server'"'"'s votes in the count of votes; and

establishing the quorum when a majority of available votes has been accumulated in the count of votes.

7. The method of claim 6, wherein the quorum server grants control to only a first computer attempting to establish control.

8. The method of claim 6, wherein the quorum server grants control to only one computer of all computers attempting to establish control based on a pre-established priority list.

9. The method of claim 6, wherein votes are assigned so that the quorum includes at least one computer that was in an immediately previous cluster, to ensure that a cluster formed from the quorum has current data.

10. The method of claim 6, wherein attempting to establish control over the quorum server involves establishing communications with the quorum server.

11. The method of claim 10, wherein establishing communications with the quorum server involves using cryptographic techniques.

12. A computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method that facilitates establishing a quorum for a cluster within a plurality of computers that are geographically distributed, the method comprising:

detecting a change in membership of the cluster at a computer within the plurality of computers; and

upon detecting the change in membership,

forming a potential new cluster by attempting to communicate with all other computers within the plurality of computers,

accumulating votes for each computer successfully contacted,

attempting to gain control of a quorum server located at a site that is geographically separate from all computers within the plurality of computers,

if successful, accumulating the quorum server'"'"'s votes, and

if a total of accumulated votes includes a majority of available votes, forming a new cluster from the potential new cluster.

13. The computer-readable storage medium of claim 12, wherein detecting the change in membership involves:

exchanging heartbeat messages with all computers that are part of the cluster; and

upon discovering an absence of heartbeat messages from any computer in the cluster, initiating a cluster membership protocol.

14. The computer-readable storage medium of claim 12, wherein detecting the change in cluster membership includes detecting that the cluster has not been formed.

15. The computer-readable storage medium of claim 12, wherein attempting to gain control of the quorum server involves communicating with the quorum server using cryptographic techniques.

16. The computer-readable storage medium of claim 12, the method further comprising:

exchanging a status message with each member of the new cluster; and

updating a local status at the computer to a most recent status available within the status message.

17. A computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method that facilitates establishing a quorum for a cluster within a plurality of computers that are geographically distributed, the method comprising:

providing a quorum server at a site that is geographically separate from a location of a computer within the plurality of computers;

assigning at least one vote to each computer within the plurality of computers;

assigning at least one vote to the quorum server;

attempting to establish communications between each pair of computers within the plurality of computers;

accumulating a count of votes for each computer communicated with at each computer;

attempting to establish control over the quorum server from each computer within the plurality of computers;

if control is established over the quorum server, accumulating the quorum server'"'"'s votes in the count of votes; and

establishing the quorum when a majority of available votes has been accumulated in the count of votes.

18. The computer-readable storage medium of claim 17, wherein the quorum server grants control to only a first computer attempting to establish control.

19. The computer-readable storage medium of claim 17, wherein the quorum server grants control to only one computer of all computers attempting to establish control based on a pre-established priority list.

20. The computer-readable storage medium of claim 17, wherein votes are assigned so that the quorum includes at least one computer that was in an immediately previous cluster, to ensure that a cluster formed from the quorum has current data.

21. The computer-readable storage medium of claim 17, wherein attempting to establish control over the quorum server involves establishing communications with the quorum server.

22. The computer-readable storage medium of claim 21, wherein establishing communications with the quorum server involves using cryptographic techniques.

23. A system that facilitates establishing a quorum for a cluster within a plurality of computers that are geographically distributed, comprising:

the plurality of computers;

a network coupling the plurality of computers;

a quorum server located at a site that is geographically separate from any one computer of the plurality of computers, wherein the quorum server provides one or more votes to a computer within the plurality of computers that is attempting to form a quorum; and

an independent communications link coupling each computer of the plurality of computers and the quorum server.

24. The system of claim 23, wherein the quorum server includes a mechanism for granting control to only one computer of the plurality of computers requesting control.

25. The system of claim 23, wherein the quorum server includes a mechanism for maintaining a list of computers accepted into the cluster.

26. The system of claim 23, wherein the quorum server includes a mechanism for cryptographically ensuring an identity of a computer attempting to establish control.

27. The system of claim 23, wherein the quorum server includes monitoring means to monitor the status of each computer within the plurality of computers.