High availability for intelligent applications in storage networks
Methods and apparatus provide formation and management of intelligent application clusters in a storage area network. Disk arbitration mechanisms ensure that a cluster is owned by a single member. In the event of a network partition, each cluster member involved arbitrates to gain ownership of a cluster. High availability mechanisms allow monitoring of system resources and effective failover capabilities.
1. A method for providing high availability at a virtual logical unit number server (VS), the method comprising:
detecting a heartbeat timeout between a first VS and a second VS in a service-group cluster, the first VS having a first VS write region, the second VS having a second VS write region, and the first VS and the second VS sharing an ownership region;
writing an identifier associated with the first VS into a first VS write region when it is determined that the second VS write region is clear;
wherein the writing the first VS write region is triggered upon detecting a the heartbeat timeout at a the first VS;
writing the ownership region when it is determined that the ownership region is clear;
operating the first VS as the standalone master VS upon writing the ownership region; and
wherein the second VS operates in the isolated state upon operating the first VS as the standalone master VS.
2. The method of claim 1 , wherein the first VS is a storage area network switch.
3. The method of claim 1 , wherein the first VS and the second VS are associated with a single storage area network switch.
4. The method of claim 1 , wherein a mechanism associated with the heartbeat timeout is Virtual Router Redundancy Protocol (VRRP).
5. The method of claim 1 , wherein the service group includes a plurality of disk groups.
6. The method of claim 1 , wherein the first VS is included in multiple service groups.
7. The method of claim 1 , wherein the first VS operates in a network having multiple overlapping clusters.
8. The method of claim 1 , wherein the first VS operates in a network having a service group.
9. The method of claim 8 , wherein the service group is a group of resource made highly available by the first VS.
10. A system for providing high availability, the system comprising:
means for detecting a heartbeat timeout between a first virtual logical unit number server (VS) and a second VS in a service-group cluster, the first VS having a first VS write region, the second VS having a second VS write region, and the first VS and the second VS sharing an ownership region;
means for writing an identifier associated with the first VS into a first VS write region when it is determined that the second VS write region is clear;
wherein the writing the first VS write region is triggered upon detecting a the heartbeat timeout at a the first VS;
means for writing the ownership region when it is determined that the ownership region is clear;
means for operating the first VS as the standalone master VS upon writing the ownership region; and
wherein the second VS operates in the isolated state upon operating the first VS as the standalone master VS.
11. The system of claim 10 , wherein the first VS is a storage area network switch.
12. The system of claim 10 , wherein the first VS and the second VS are associated with a single storage area network switch.
13. The system of claim 10 , wherein a mechanism associated with the heartbeat timeout is Virtual Router Redundancy Protocol (VRRP).
14. The system of claim 10 , wherein the service group includes a plurality of disk groups.