EDB Postgres® AI for CloudNativePG™ Global Cluster 2.0.0 release notes v2.0.1

Released: 26 March 2026

This is the first major stable release of EDB Postgres® AI for CloudNativePG™ Global Cluster, a Kubernetes operator to deploy and manage EDB Postgres Distributed versions 6.x clusters.

The operator continues to implement the PGDGroup custom resource in the API group pgd.k8s.enterprisedb.io. This resource can be used to create and manage EDB Postgres Distributed versions 6.x clusters inside Kubernetes. The additional features and improvements included in this release are listed below.

Supported versions

  • Kubernetes: 1.33, 1.34, and 1.35
  • PostgreSQL: 18, 17, 16, 15, and 14

This release of EDB Postgres® AI for CloudNativePG™ Global Cluster includes the following:

Features

DescriptionAddresses
Deployment of EDB Postgres Distributed clusters with versions 6.x and later
Connection Manager support for PGD versions 6.x#1187, #1163, #1930
Support for upgrading from PGD 5 to PGD 6#1538
Default monitoring support with predefined set of metrics#1923, #1934
Support for showing parent group UUID in status and labels for better PGD cluster management and monitoring#1894
Global routing support for PGD versions 6.x#1886, #1803
Managed role and managed application user support for PGD versions 6.x#1248, #1432, #1435
Replaced `nodeMaintenanceWindow` with `enablePGD` in PG4K cluster#1910
Enhanced the `pgdgroupcleanup` phase#1617
Support for declarative way to depart a PGDGroup node#1323
Support for configuring `initDB` options#1302
Support for barman cloud plugin for backup and restore#1068
Support for custom `postgresUID` and `postgresGID` for PGD nodes#1974

Changes

DescriptionAddresses
Update the default PGD image to PostgreSQL 18.3 and PGD 6.2.0 expanded (image `postgresql-pgd:18.3-pgd620-expanded-ubi9`)
The tested PG4K version is v1.28.1

Bug Fixes

DescriptionAddresses
Start physical join if CNP cluster is hibernated but failed to set phase#1906
Use remote database connection string to do the physical join#1659
Cleanup the stale recovery settings in `postgresql.auto.conf` after physical join#2115
Wait for local partition to be created before calling `bdr.create_node_group`
Increase max_active_replication_origins to 30 for PostgreSQL 18 to support more replication slots for PGD v6