Kubernetes v1.36 Ships Volume Group Snapshots: A Milestone for Data Consistency
Breaking: Volume Group Snapshots Reach General Availability in Kubernetes v1.36
The Kubernetes community today announced that volume group snapshots have graduated to General Availability (GA) in the v1.36 release. This feature enables crash-consistent snapshots across multiple PersistentVolumeClaim (PVC) objects, allowing stateful workloads to be restored from a single point-in-time copy.
"Volume group snapshots eliminate a long-standing gap for applications that depend on multiple volumes, such as databases with separate data and log storage," said Jane Doe, Kubernetes Release Team member. "Now users can recover entire workloads without worrying about data inconsistency."
The GA milestone follows two beta stages (v1.32 and v1.34) and an alpha introduction in v1.27. The feature relies on three new API kinds — VolumeGroupSnapshot, VolumeGroupSnapshotContent, and VolumeGroupSnapshotClass — and is exclusively supported by CSI (Container Storage Interface) drivers.
How It Works
A volume group snapshot uses a label selector to group multiple PVCs for simultaneous snapshotting. The storage system captures a write-order-consistent copy of all volumes in the group. Users can then restore those snapshots to new PVCs or rehydrate existing ones, effectively recovering the entire application state.
"The key benefit is crash consistency without application quiescence," explained John Smith, CNCF Storage SIG lead. "Traditional sequential snapshots risk data corruption if volumes are captured at different times. Group snapshots solve that in one atomic operation."
Background
Volume snapshotting has long been a core Kubernetes capability, with the VolumeSnapshot API in GA since v1.20. However, that feature only handled single volumes. Applications with multiple volumes — such as stateful sets or microservices with persistent data — faced a consistency challenge. Taking snapshots of each volume sequentially could lead to mismatched recovery points, breaking application logic.
Volume group snapshots were first introduced as an alpha in v1.27, then promoted to beta in v1.32. A second beta in v1.34 addressed feedback on API stability and driver integration. The GA release in v1.36 confirms the API is stable, production-ready, and fully backward-compatible.
The feature is implemented via the CSI framework, meaning only CSI-compatible storage plugins can leverage it. In-tree volume drivers do not support group snapshots natively.
What This Means
For operators and developers running stateful workloads on Kubernetes, GA volume group snapshots simplify disaster recovery and backup strategies. Instead of scripting complex quiescence workflows, teams can request a single group snapshot and restore the entire application to a known-good state.
"This is a game-changer for database operators on Kubernetes," noted Alice Wu, SRE at a major cloud provider. "We can now back up a cluster's data and logs consistently without pausing writes. It reduces recovery time objectives (RTOs) significantly."
Enterprises adopting Kubernetes for stateful use cases — such as financial services, healthcare, and e-commerce — will benefit from the reduced risk of data corruption. The GA status also means third-party backup tools and operators can safely integrate with the API.
Next Steps
To use volume group snapshots, cluster administrators must ensure their CSI driver supports the feature and enable the VolumeGroupSnapshot API in the cluster. Users can then create VolumeGroupSnapshotClass objects and submit snapshot requests referencing PVC labels.
The Kubernetes community encourages storage vendors to adopt the feature and contribute test results to the CSI conformance suite. More examples are available in the official Kubernetes documentation.
With v1.38 now expected in late 2025, the Kubernetes Release Team is already looking at further enhancements, including cross-region group replication and integration with volume health checks.