Method and apparatus for building and managing multi-clustered computer systems

A method of managing a clustered computer system having at least one node, said method comprising the steps of:

• (a) establishing a multi-cluster comprising said at least one node and at least one shared resource;
• (b) managing said at least one node with a cluster services program;
• (c) returning said system to a consistent state after a failover event;
• and
• (d) adding a hidden resource into a resource group on each node.

A data processing system for executing a method having a plurality of paths, wherein the data processing system executes native machine code, the data processing system comprising:

• (a) means for establishing a multi-cluster computer system comprising at least two subclusters wherein each of the subclusters comprises at least one node;
• (b) means for managing said at least one node with a cluster services program;
• (c) means for failing over between a first node in a first one of the at least two subclusters and any other node within the multi-cluster computer system in response to a failover event;
• and
• (d) means for returning said system to a consistent state after the failover event.

A data processing system for executing a method having a plurality of paths, wherein the data processing system executes native machine code, the data processing system comprising:

• (a) means for establishing a multi-cluster comprising said at least one node and at least one shared resource;
• (b) means for managing said at least one node with a cluster services program;
• (c) means for returning said system to a consistent state after a failover event;
• and
• (d) means for adding a hidden resource into a resource group on each node.

A computer program product in a computer readable medium for executing a method in a data processing system,

• (a) first instructions for establishing a multi-cluster computer system comprising at least two subclusters wherein each of the subclusters comprises at least one node;
• (b) second instructions for managing said at least one node with a cluster services program;
• (c) third instructions for failing over between a first node in a first one of the at least two subclusters and any other node within the multi-cluster computer system in response to a failover event;
• and
• (d) fourth instructions for returning said system to a consistent state after the failover event.

A method of managing a clustered computer system having at least one node, said method comprising the steps of:

• (a) establishing a multi-cluster comprising said at least one node and at least one shared resource;
• (b) managing said at least one node with a cluster services program;
• wherein said cluster services program manages using a resource API within the at least one node;
• including managing a heartbeat signal sent between said at least one node and any other node within the multi-cluster;
• (c) failing over between a first node and any other node within the multi-cluster;
• (d) comprises updating a cluster wide data file;
• (e) returning said system to an initial state after a failover event.