StackGres integrates the most renowned and production-tested high availability software for Postgres: Patroni.
StackGres fully integrates Patroni, and for you there’s nothing else to do. If any pod, any node, anything fails, the cluster will perform a failover and re-heal automatically in a matter of seconds, without human intervention.
StackGres exposes one read-write (master) and one read-only (replicas) connection for the applications via Kubernetes services, that will automatically be updated after any disruptive event happens.
Have a look at the High Availability Guide to learn more about Patroni is integrated into StackGres, and how to perform manual switchovers, or how test failovers.