Features and capabilities

Percona Operator for PostgreSQL is a Kubernetes-native controller that automatically manages the full lifecycle of Percona Distribution for PostgreSQL clusters. The Operator offloads your team from manual day-to-day database management operations. This enables you to focus on tasks that matter instead. To learn how the Operator fits into Kubernetes, see Kubernetes Operator concepts.

Here’s what the Operator brings to your infrastructure:

High availability and failover

Run PostgreSQL with confidence: Patroni provides automatic leader election, failover, and coordination so your cluster stays available through node and Pod failures. For architecture details, see Cluster architecture.

• Automatic failover — Patroni manages leader election and failover to ensure the cluster always has a healthy primary. See High availability.
• Zero data loss failover — WAL-based replication limits data loss during failover; synchronous replication is available when you need stronger guarantees.
• Health monitoring — Continuous health checks trigger failover when PostgreSQL is not ready to serve traffic.
• Manual switchover — Promote a replica to primary in a controlled way for maintenance.

Automated backup and restore flows

Safeguard your data at any scale: the Operator automates backups and restores using pgBackRest , a robust open source solution trusted for PostgreSQL in production. Read About backups for the full workflow.

Also, leverage Kubernetes PersistentVolumeClaim snapshots for rapid, consistent backup and restore operations. It is especially valuable for large database clusters.

• Full, incremental, and differential backups — Select the backup strategy that matches your recovery objectives and storage requirements.
• Point-in-time recovery (PITR) — Achieve low Recovery Point Objectives (RPO) by restoring to any specific time using WAL archives.
• Scheduled backups — Automate backups on your chosen schedule with cron-like expressions.
• Flexible storage — Store backups in S3-compatible object storage or on local PersistentVolumes for hybrid strategies.
• PVC snapshot support — Boost backup and restore performance for large datasets with a point-in-time snapshot of your data volume.
• Encryption — Secure backups at rest where your storage backends and configuration allow it.
• Retention — Manage backup lifecycle and automate old backup cleanup to prevent storage sprawl.

Connection pooling with pgBouncer

Reduce connection churn and spread read load without extra operational burden.

• Efficient pooling — Lower PostgreSQL connection overhead by pooling client connections
• Transaction-level pooling — Manage connections at the transaction level efficiently
• Read balancing — Distribute read queries across replicas where configured
• High availability — Replica pgBouncer instances provide high availability
• Integrated lifecycle — Automatically configured and managed by the Operator

Automated scaling and resource management

Scale your cluster up or down to match demand while keeping changes declarative.

• Declarative clusters — Describe desired cluster state in YAML; the Operator automatically reconciles Kubernetes resources to match.
• Replica scaling — Adjust replica count in the Custom Resource to scale horizontally.
• Dynamic configuration — Update PostgreSQL parameters without a full cluster restart.
• Self-healing — The Operator automatically detects and recovers from Pod crashes, node issues, and common network problems.
• Rolling updates — Apply configuration and image updates with controlled rollouts.
• Storage expansion — Automatically increase storage size for PostgreSQL instances when supported by your environment and configuration.

PostgreSQL-specific features

Use PostgreSQL capabilities that operators expect in production.

• WAL storage — Optional dedicated volumes for Write-Ahead Logs when you want to separate I/O.
• Tablespaces — Custom tablespaces with dedicated storage.
• Extensions — Built-in support for extensions such as pg_stat_monitor, pgAudit, set_user, wal2json, plus ability to extend PostgreSQL with custom extensions.
• Users and databases — Automatically create users, databases, and manage credentials.
• Init SQL — Execute custom SQL scripts during cluster initialization.

Standby clusters for disaster recovery

Leverage disaster-recovery topologies that fit your RTO and RPO.

• Backups or streaming — Deploy your standby cluster based on backups or streaming replication, depending on your architecture
• Cross-namespace or cross-cluster — Primary and standby clusters can run in different namespaces or Kubernetes clusters
• Promotion — Promote a standby to primary when you need to recover from an outage or drill a failover

Security and compliance

Keep traffic and data protected with encryption and flexible TLS workflows.

• TLS for connections — Encrypt client traffic and traffic between cluster components
• Certificates — Comply with your security policy via custom certificates or automated certificate generation with cert-manager with configurable lifecycle management.

Monitoring and observability

Understand performance and troubleshoot faster with metrics and optional Percona tooling.

• PMM integration — Connect the cluster to Percona Monitoring and Management (PMM) for dashboards and alerting.
• pg_stat_monitor — Get query performance insights with fingerprinting when you enable the extension.
• Broad metrics — Track connection counts, transaction rates, cache hit ratios, replication lag, and more.
• Query analytics — Deeper query analysis in PMM. See Query Analytics in the PMM documentation.

Operator capabilities

Operate at the scale of your platform with flexible reconciliation scope.

• Selective namespaces — Reconcile clusters in a single namespace or multi-namespace mode. See cluster-wide deployment to learn more.
• Concurrent reconciliation — Run concurrent reconciliations to manage many clusters efficiently.

---

Last update: March 26, 2026
Created: March 26, 2026