Replication Agent provides support for Oracle 10g and 11g Real Application Cluster (RAC) environments.
When you initialize a Replication Agent for Oracle instance, the Oracle database is queried to determine how many nodes are supported by the cluster. Based on this information, Replication Agent automatically configures itself to process the redo log information from all nodes.
To process the redo log data from all nodes in an Oracle RAC cluster, the Replication Agent must execute from a location that has access to the same shared storage used by the Oracle nodes to store their redo data. The Replication Agent must have read access to the shared storage where both the online and archived redo logs exist.
You can configure Replication Agent to connect to a single Oracle instance by supplying the required host, port, and Oracle SID values to the pds_host_name, pds_port_number and pds_database_name configuration parameters. In an Oracle RAC environment, Replication Agent must be able to connect to any node in the cluster in the event that a node fails or becomes unavailable.
To support the configuration of multiple node locations, Replication Agent supports connectivity to all possible RAC nodes by obtaining needed information from an Oracle tnsnames.ora file for one specified entry. As a result, instead of configuring individual host, port, and instance names for all nodes, Replication Agent requires only the location of a tnsnames.ora file and the name of the TNS connection to use.
For more information about Replication Agent support for Oracle RAC, see the Replication Agent Primary Database Guide.