High Availability overview

High Availability for Stateful services overview

PostgreSQL High Availability

Postgres in high-availability mode uses three Postgres instances and one PgPool instance. For HA, Postgres uses RepMgr, part of the bitnami/postgresql-repmgr image.

RepMgr ensures HA using automatic membership control. If the primary is down, any of the replica nodes will be promoted as primary to avoid data loss.

PgPool handles the connection to the Postgres instances and is responsible for spreading the queries among these instances.

MinIO High Availability

MinIO HA mode uses two standalone instances with Site replication and a single load balancer (Kubernetes service) in front of them.

MinIO peer sites can proxy GET/HEAD requests for an object to other peers to check if it exists. This allows a site that is healing or lagging behind other peers to return an object that persisted in other sites.