Two-node active-passive configuration is described in Figure 10-1.
In Sun Cluster, Adaptive Server runs as a data service and is managed by the Sun Cluster Resource Group Manager (RGM). Adaptive Server is associated with a resource group that contains the Adaptive Server resource and all other resources it requires, such as the SUNW.HAStorage, SUNW.HAStoragePlus, and SUNW.LogicalHostname resources.
SY.ase is the Adaptive Sever resource and it defines various extension properties. See “Adaptive Server resource extension properties” for more information. See the Sun Cluster documentation for more information on standard resource properties.
Figure 10-1: Active-passive setup in Sun Cluster before failover
In Figure 10-1, the Adaptive Server MONEY is associated with the resource group rg_MONEY, which consists of three resources: the Adaptive Server resource, ase_MONEY, of resource type SY.ase, the storage resource, has_MONEY, of resource type SUNW.HAStorage, and the logical host resource, lh_MONEY, of resource type SUNW.LogicalHostname. has_MONEY manages the global file system /global/node1_share on the shared disk. The logical host resource is associated with the logical host name or floating IP address loghost. ase_MONEY uses has_MONEY and lh_MONEY.
Initially, the Adaptive Server resource group, rg_MONEY, is hosted by the primary node, node1, and Adaptive Server MONEY serves its clients through the logical host name loghost associated with lh_MONEY.
When node1 crashes, the resource group rg_MONEY and all of its resources is relocated and restarted on the secondary node as shown in Figure 10-2.
After failover, the Adaptive Server runs on node2 and continues to serve its clients using the same loghost.
Figure 10-2: Active-passive setup on Sun Cluster after failover
The resource group properties Pingpong_interval and Global_resources_used may affect fail over. For example, in update 1 of Sun Cluster version 3.0 documentation, if the Adaptive Server resource group, rg_MONEY, is moving between the primary and secondary nodes too frequently (within about 300 seconds), the RGM may stop fail over of the Adaptive Server resource group with the following error:
608202 :scha_control: resource group ase_MONEY was frozen on Global_resources_used within the past 300 seconds; exiting
