Skip to content

1.0.0

Latest

Choose a tag to compare

@guanzhousongmicrosoft guanzhousongmicrosoft released this 19 Oct 11:39
3928a33

[1.0.0] - 2026-10-19

First generally available release of the DocumentDB Kubernetes Operator, for DocumentDB 1.0. The chart deploys DocumentDB engine 1.0.0 (extension schema 1.0-0) by default. The operator and engine keep separate version numbers; see the version compatibility table.

How to install

helm install documentdb-operator oci://ghcr.io/documentdb/documentdb-operator \
  --version 1.0.0 \
  --namespace documentdb-operator \
  --create-namespace \
  --wait

Requires Kubernetes 1.33+ with ImageVolume (on by default from 1.35, beta in 1.35 and GA in 1.36; on 1.33/1.34 enable the ImageVolume feature gate on the apiserver and every kubelet), a containerd or CRI-O runtime, and cert-manager 1.19+.

Upgrading

Follow Upgrading to DocumentDB 1.0 before you run helm upgrade.

  • From 0.3.0: pin spec.documentDBVersion on each cluster and remove spec.schemaVersion: "auto" first. Otherwise, engine pods roll from 0.110.0 to 1.0.0 as soon as the chart upgrades, and auto runs the schema migration immediately. Then move each cluster to 1.0.0 with the two-phase upgrade. The schema step (ALTER EXTENSION documentdb UPDATE) cannot be undone.
  • From 1.0.0-rc1 (DocumentDB 1.0-RC2): in-place upgrade to GA is not supported (#471). Pin RC clusters to 1.0-RC2 before upgrading the chart, then migrate the data to a new 1.0.0 cluster, for example with mongodump/mongorestore.

Major Features

  • DocumentDB 1.0 by default: the chart, operator, and sidecar injector default to the documentdb:1.0.0 and gateway:1.0.0 images.
  • Configurable image registries: an image.registry prefix and per-component repository values re-home every operator and data-plane image to a private or mirrored registry, with validated repository and tag overrides (#464).
  • ImageVolume capability check: the webhook rejects DocumentDB creation with an actionable error when ImageVolume is unavailable, instead of leaving pods stuck (#431, #437). The chart's kubeVersion floor is >= 1.33.0-0.
  • Dedicated monitoring identity: the OpenTelemetry Collector sidecar connects to PostgreSQL as its own login-only role (#420).
  • Restore compatibility validation: the webhook rejects restoring a backup or retained PersistentVolume into a cluster whose engine binary is older than the source schema (#440).
  • PostgreSQL and replication TLS: spec.tls.postgres accepts your own CA and certificate Secrets for PostgreSQL server and replication connections (#416).
  • Database image builds: extension packages come from PGDG, and RC or not-yet-packaged stable releases build from a pinned upstream tag (#456, #470, #478).

Bug Fixes

  • Address dependency vulnerabilities reported by govulncheck, including gRPC (#429, #436, #468, #473).
  • Track the floating PostgreSQL 18 image again after temporarily pinning 18.4 to avoid a crash in the DocumentDB extension (#445, #459).

Breaking Changes

  • The top-level gatewayImagePullPolicy and documentDbImagePullPolicy Helm values moved to image.gateway.pullPolicy and image.documentdb.pullPolicy. The old keys no longer take effect.

Testing and Documentation

  • Expand E2E schema-upgrade coverage (automatic migration, rollback, HA, unpullable version) and add expired-backup cleanup tests (#426, #443, #393).
  • Add long-haul data-protection verification, retention pruning, and operator/primary pod kill operations (#419, #421, #422).
  • Introduce a product-neutral ClusterIntent abstraction with no functional change (#446).
  • Add multi-cloud deployment docs and an Arc + Fleet hybrid setup (#399, #267).

See CHANGELOG.md for details.