Listener failover scenario 1: Non-master listener node fails

This topic pertains to a clustered listener configuration. In this scenario, a non-master listener node is unresponsive. Either the node is down or is not reachable due to network issues.

A node is determined to be unresponsive based on a limited number of retries within a certain time period.

In this case, the master listener concludes that the node is down. During the node down-time, the master listener stops routing requests to that node. Instead, requests are routed to any remaining listeners in the cluster, based on the assigned masterListenerPriority and loadBalanceWeight. If there are no other viable listeners, the single remaining listener processes all requests itself.

When the unresponsive node comes back up, requests are again routed to it. In this scenario, the interruption and subsequent recovery are recorded in the masterlistener.log. If the listener node comes back up before user takes any action, the user will not notice any disruption, because the connection has been restored. If the user takes action while the listener node is down, failover will occur and flowchart will be moved to another listener. In this case, the user is alerted with a message.