In AWS database clusters, failover is a mechanism that automatically repairs cluster availability when a primary database instance becomes unavailable. This process ensures minimal downtime by promoting a replica instance to become the new primary.
In an Amazon Aurora database cluster, failover works by electing an Aurora Replica to become the new primary DB instance, ensuring the cluster maintains maximum availability for read-write operations.
In Amazon RDS Multi-AZ DB clusters, failover automatically promotes one of the two readable standby instances to become the new primary when the original primary fails.