Replication Server and RepAgent thread both support the Adaptive Server shared-disk cluster environment. In a Sybase shared-disk cluster, a database can be either a replication source or a replication destination. You can perform all of the tasks, such as configuring RepAgent or marking tables for replication, from any instance in the cluster. Replication status is coherent across the entire cluster.
When adding new connections from or to an Adaptive Server cluster environment, the servername in the connection syntax must be the clustername and not the instancename. Use select @@servername to retrieve the clustername.
By default, the RepAgent starts on the cluster coordinator; However, you can configure it to start on any instance in the cluster. For example, to configure the RepAgent on the primary database pdb to always start on the “ase2” instance, enter:
sp_config_rep_agent pdb, "cluster instance name" "ase2"
For a new configuration to take effect, restart the RepAgent using sp_start_rep_agent. To return to the default behavior with the RepAgent starting on the cluster coordinator, enter:
sp_config_rep_agent pdb, "cluster instance name", "coordinator"
When an instance starts, it checks if there are RepAgents configured to start on its node. If there are, and if the database is marked to start automatically, the RepAgent starts.
When the cluster coordinator starts, it also starts all RepAgents that are not configured to start on a specific instance. If the coordinator node fails, or is stopped with a graceful shutdown, a RepAgent starts on the new coordinator node.
If the RepAgent is configured to start on an instance other than the coordinator node, and this instance fails, the RepAgent starts on the coordinator.
See the Replication Server Reference Manual for information about the cluster instance name configuration parameter.
