Failover database

By default, the database is created automatically when the cluster is initialized and its agents are launched on the master servers. Database agents are automatically configured when the cluster is initialized.

Database agents are a key component of a failover clustered DBMS. Each database record is first received by the master agent and then automatically replicated to one or two standby agents. This ensures high availability and protects data from loss in case of failure of one of the servers. The system supports streaming replication. Data written to the master agent is synchronously replicated to the standby agents. This provides data redundancy at the agent level and allows the database to be restored without data loss in case of failure of the master node.

You can manage database agents in the supervisor interface. To do this:

1. Go to the CONFIGURATION tab.
2. Click the DB AGENTS button.
   
3. The Port field specifies the value of the port that the database agents use. If the value is 0, the database agents use dynamic ports.

4. The icon next to the server indicates the current status of the agent. 

   Icon	Status
   	Launch expected
   	Launched
   	Stopped
   	Unavailable

5. Click to stop or launch the agent and select the required action.