Auto Piloting

Self-Healing HA Architecture, Cold backups & WAL Archive, Steady as she goes!

Taking PostgreSQL as an example, Pigsty creates a database cluster that is distributed and highly available database cluster. As long as any instance of the cluster survives, the cluster can provide complete read-write service and read-only service to the outside world.

Pigsty’s high availability architecture has been tested in production environments. Pigsty uses Patroni + Consul for fault detection, Fencing, automatic failover, and HAProxy, VIP, or DNS for automatic traffic switching, achieving a complete high availability solution at a very low complexity cost, allowing the master-slave architecture of the database to be used with a cloth-like experience. Database-like experience.

The database cluster can automatically perform fault detection and master-slave switching, and common faults can be self-healing within seconds to tens of seconds: RTO < 1min for master failure, read-only traffic is almost unaffected, sync standby cluster RPO = 0 without data loss.

Each database instance in the database cluster is idempotent in use, and any instance can provide full read and write services through the built-in load balancing component HAProxy. Anyone or more Haproxy instances can act as a load balancer for the cluster and distribute traffic through health checks, shielding the cluster members from the outside world. Users can flexibly define services through config and access through various optional methods.

Last modified 2022-06-04: fill en docs (5a858d3)