April 30th, 2026
vCluster

We're excited to announce vCluster Platform v4.9 and vCluster v0.34, a release focused on resilience and operational simplicity for teams running vCluster as critical production infrastructure.
From active/active Multi-Region Platform deployments and standalone snapshotting to an overhauled template authoring experience and simpler vNode licensing, these updates make Platform easier to run, recover, and scale wherever you deploy it.
vCluster Platform now runs active/active across regions. Both regions serve traffic concurrently behind Route 53 latency-based routing, with leader election coordinating writes to a shared database and automatic DNS failover on regional outage. Launching with AWS support, clusters on this release can use either Postgres or MySQL as backing stores on RDS.
Configuration layers a multiRegion stanza on top of a standard Platform install. Each region gets its own values file specifying its own multiRegion.region and multiRegion.host. All regions share the same config.loftHost and config.database.dataSource:
config: loftHost: multi-region.example.com database: enabled: true dataSource: ${RDS_CONNECTION_STRING} identityProvider: aws extraArgs: - --datastore-max-open-connections=20 - --datastore-max-idle-connections=0 costControl: enabled: false # not compatible with multi-region mode multiRegion: enabled: true region: us-east-1 host: us.multi-region.example.com replicaCount: 3 See the full AWS walkthrough (VPC peering, Route 53 health checks, IAM database auth with Pod Identity, cross-region relay) in the docs.
Templates allow administrators to enforce governance while providing users flexibility through configurable parameters.
Historically, the UI has offered limited support for managing those parameters. Parameters are now first-class in both the built-in YAML editor and the configuration UI, making them easy to create, reference, and edit.
vcluster snapshot create now supports capturing the full control-plane state, including etcd and configuration, of a vCluster deployed directly to baremetal or VMs with use of the --standalone flag. Write the snapshot to S3, OCI, or a local container path, using the same snapshot semantics used for a vCluster deployed on a Kubernetes cluster.
export ACCESS_KEY_ID=... export SECRET_ACCESS_KEY=... export SESSION_TOKEN=... vcluster snapshot create --standalone \\ "s3://my-bucket/my-snapshot.tar.gz?access-key-id=$ACCESS_KEY_ID&secret-access-key=$SECRET_ACCESS_KEY&session-token=$SESSION_TOKEN" Pass credentials inline in the snapshot URL (base64-encoded, as shown) or sourced from local AWS config and environment variables.
The bootstrap pattern. Deploy vCluster to baremetal or VMs, install vCluster Platform on top, then snapshot. You end up with a portable, restorable image of a fully-configured Platform control plane that is usable for disaster recovery, promoting from staging to prod, or cloning Platform installs across environments.
For more information please take a look at the docs.
With this release we are significantly improving how vNode handles licensing. Today, you must reference a platform host and platform access key when deploying vNode by its Helm chart. This creates a small, additional management surface area of key issuance / management.
Starting with this release any worker node or tenant control plane cluster attached to the vCluster Platform can deploy vNode with no access key / platform host. You can retrieve its license directly from the connected platform. Learn more in the vNode docs.
Last year we announced our new Rancher OSS integration. With this release we have removed the legacy Rancher integration, including the authentication mechanism that allowed users to log in to our Platform with Rancher. Users must migrate to our new integration before upgrading.
With the introduction of Multi-Region Platform, the Multi-Region-Mode feature has been renamed to Regional Cluster Endpoints.
Our Backup & Restore feature now supports Azure Blob storage as a backend using both the CLI and our auto-snapshots feature. The feature also supports S3-compatible paths.
Virtual clusters created from the platform now default to embedded etcd as the backing store, replacing sqlite.
For a list of additional fixes and smaller changes, please refer to the vCluster release notes and the vCluster Platform release notes. For detailed documentation and migration guides, visit vcluster.com/docs.