Failover
Automatically sending new traffic to a healthy backup when the primary server or region stops working.
See it
What it is
Failover is the automatic switch from a failed primary origin, region, or service to a healthy backup. A health check detects the failure, and a load balancer or DNS service stops directing new traffic to the primary.
Reach for it when downtime costs more than keeping spare capacity and a second copy of the data. Load-balancer failover can react quickly. DNS failover is useful across regions, but resolvers may keep the old answer until its TTL expires.
A backup that has never carried production traffic is a hope, not a plan. It may be cold, undersized, missing recent data, or dependent on the same failed database. This is the whole reason Netflix built Chaos Monkey and pointed it at production: the only failover you can trust is one that has already happened by surprise. Exercise the switch on a schedule, and verify both replication lag and the route back to the recovered primary.
Ask AI for it
Configure Amazon Route 53 failover routing with one PRIMARY record and one SECONDARY record, unique SetIdentifier values, and a Route 53 HTTPS health check for the primary's /healthz path. Set a 60-second TTL on non-alias records, keep the secondary independent of the primary region, and include a scripted failover test that proves DNS returns the secondary after the primary health check fails.