Cluster membership monitor

• US 20020042693A1
• Filed: 05/02/2001
• Published: 04/11/2002
• Est. Priority Date: 05/02/2000
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Inventors: Kampe, Mark A., Penkler, David, Ramer, Rebecca A.
• Granted Patent: US 7,039,694 B2
• Time in Patent Office: Days
• US Class Current: 702/186
• CPC Class Codes: H04L 41/0618 based on the physical or logical position
• CPC Class Codes: H04L 41/085 Retrieval of network configuration; Tracking network configuration history
• CPC Class Codes: H04L 41/0856 by backing up or archiving configuration information
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 43/12 Network monitoring probes
• CPC Class Codes: H04L 67/104 Peer-to-peer [P2P] networks
• CPC Class Codes: H04L 67/1044 Group management mechanisms management of multicast group membership H04L12/185; reconfiguring of node membership in a computing system to eliminate errors G06F11/1425
• CPC Class Codes: H04L 67/1046 Joining mechanisms
• CPC Class Codes: H04L 67/1048 Departure or maintenance mechanisms
• CPC Class Codes: H04L 67/1051 Group master selection mechanisms
• CPC Class Codes: H04L 67/1093 Some peer nodes performing special functions

Abstract

The present invention provides a system and method within a high availability network for monitoring and managing cluster membership. The cluster membership monitor provides the ability to maintain a list of current cluster members, monitor status of each peer node on the cluster, stay apprised of each nodes viability, elect a master node for the cluster when necessary, and coordinate reformation as members join and leave the cluster.

Claims

1. A system for monitoring cluster memberships within a clustered computer network, comprising:

at least one node;

a membership monitor entity located on the at least one of each node.

2. The system of claim 1, further comprising at least one node data entry on each node.

3. The system of claim 2, wherein the node data entry further comprises a node id entry, a node type entry and a master priority value entry.

4. The system of claim 1, further comprising a high resolution timer located on each node.

5. The system of claim 1, further comprising a log for logging.

6. The system of claim 1, further comprising a repository for storing cluster configuration data.

7. The system of claim 1, further comprising a hold off timer.

8. A method for joining a cluster with a cluster membership monitor within a clustered computer network, comprising the steps of:

starting a local cluster membership monitor entity on a node when it boots;

establishing contact with peer cluster membership monitor entities; and

wherein if peers are found a consensus protocol is performed.

9. The method of claim 8, further comprising the step of determining if a master node exists, wherein if a master exists the node joins the cluster.

10. The method of claim 9, further comprising the step of obtaining configuration data from a master cluster membership monitor entity located on the master node.

11. The method of claim 10, further comprising the step of logging in a local system log of the master node a cluster joint event.

13. The method of claim 12, further comprising the steps of

verifying that each peer node has responded to the heartbeat;

and wherein if all nodes listed in a master list of cluster members have responded, another heartbeat can be provided, and wherein otherwise a node not responding is removed from the list of cluster members and a cluster change request is made.

14. The method of claim 13, wherein the cluster change request further comprises notifying a cluster coordination function.

15. The method of claim 12, wherein the cluster reformation further comprises the steps of:

verifying if cluster reformation has already been initiated; and

wherein if cluster reformation has not been initiated, cluster reformation is initiated;

the highest priority node becomes the master node; and

the master heartbeat can be provided.

12. A method for monitoring the viability of cluster members comprising the steps of

providing a heartbeat by a master node;

and wherein if the heartbeat is within a peer node time-out period, the peer node responds, otherwise a cluster reformation occurs.