Failover

The automatic process of switching incoming traffic from a failed primary server, data centre, or service instance to a healthy standby or backup. Failover is triggered by health check failures — when a backend stops responding to health probes within a configured timeout and retry count, the load balancer removes it from the active pool and redistributes its connections. Active-active configurations handle failure by rebalancing across remaining healthy nodes; active-passive configurations promote a standby node. DNS-based failover at the GSLB level redirects entire regions. Recovery and fail-back policies determine when a recovered server re-enters the active pool.