Cluster Management
High Availability for PostgreSQL
Cluster management is an essential component of any large-scale PostgreSQL deployment. EDB offers multiple failover and replication solutions to meet your requirements.
Ensure high availability
Detect issues quickly and automatically failover
Preserve cluster
consistency
Avoid split-brain scenarios without risking data loss
Gain critical insights
Be alerted of cluster events and easily access status information
EDB helps ensure high availability of your PostgreSQL clusters
The following tools are all fully supported by EDB in addition to being tested and released as part of the EDB portfolio.
High Availability
Router for Postgres
HARP
- An EDB tool for managing high availability for PostgreSQL, EDB Postgres Extended, and EDB Postgres Advanced clusters using Bi-Directional Replication (BDR) on Linux
- Use in the cloud and on-premises without the need for Virtual IP addresses
- Meets 99.999% availability needs when using a BDR architecture
- Ensures all writes target the active BDR node intrinsically ruling out split-brain situations by design
Failover Manager
EFM
- An EDB tool for managing high availability for PostgreSQL and EDB Postgres Advanced clusters using physical streaming replication on Linux
- Supports asynchronous, synchronous, and cascading replication
- Handles automatic failover based on predefined conditions ensuring protection against split-brain scenarios
- Enables customization for sending notifications, managing load balancers, and fencing
- Manages traffic routing using Virtual IPs, PgPool-II, or load balancers
Replication Manager
repmgr
- An open-source tool for creating and managing PostgreSQL and EDB Postgres Extended clusters using physical streaming replication on Linux
- Supports asynchronous, synchronous, and cascading replication
- Handles automatic failover based on custom conditions defined by the customer
- Easily add standbys to a cluster with a single command
Feature Comparison
| Feature | HARP | Failover Manager | repmgr |
|---|---|---|---|
| Automatic failover | |||
| Manages multi-master BDR clusters | |||
| Manages primary/standby streaming replication clusters | |||
| Protection against common split-brain scenarios | * | ||
| Quorum and consensus resolution | |||
| Set up new standbys with a single command | |||
| Switchover with a single command | * | ||
| Enables 99.999% availability requirements | |||
| Configure prioritization rules for which standby to promote | |||
| Reconfigure standbys to follow newly promoted primary | * | ||
| Reconfigure synchronous replication parameters on cluster changes | * | ||
| Event logging for cluster events | |||
| Virtual IP management | |||
| Built-in integration for traffic routing management | |||
| Event notifications using SNMP and email | |||
| Customizable with scripts | |||
| License | EDB | EDB | GPLv3 |
Get Started with Cluster Management
Need help with designing PostgreSQL cluster management solutions to meet your high availability requirements? Our experts can help you with selecting the right architecture, integrating our solutions in your environment, and deploying at scale. Contact us now to learn more.
Related Products
EDB Postgres Advanced Server
Enterprise-ready, Oracle-compatible PostgreSQL
Postgres Enterprise Manager
PostgreSQL GUI for enterprises
Backup and Recovery Tool
Disaster recovery for PostgreSQL
Related Services
Enterprise Architecture
Modernize and transform with an infrastructure that scales
Solution Design
Customized guidance and solution design from Postgres experts