Skip to content

v2.11.0

Latest
Latest

Choose a tag to compare

View all tags
@pchandra19 pchandra19 released this 04 Jun 14:20
· 21 commits to develop since this release
v2.11.0
9bdc840

Release Date: 5th June, 2026

Summary

OpenEBS Replicated PV Mayastor version 2.11.0 introduces new features and several critical functional fixes.

What’s New

DiskPool Monitoring and Error Visibility

Replicated PV Mayastor now exposes DiskPool errors and alerts, providing greater visibility into pool health and simplifying identification and troubleshooting of storage issues. DiskPool status reporting has also been improved to surface clearer error context when a pool is unavailable due to conditions such as a node being offline or the backing device not being found.

DiskPool Probing and Import Improvements

Replicated PV Mayastor now includes enhanced pool probing capabilities that improve detection and recovery of storage pools across a broader range of devices, helping simplify pool management and recovery workflows.

RDMA QoS and DSCP Marking Support

Replicated PV Mayastor now supports configuring transport-level Quality of Service (QoS) settings for RDMA connections through DSCP marking. This enables integration with network QoS policies and allows administrators to prioritize storage traffic in RDMA-enabled environments.

SPDK Interrupt Mode Support (Experimental)

Replicated PV Mayastor now supports SPDK interrupt mode, providing an alternative I/O processing model that can help reduce CPU utilization in suitable environments.

Offline Node Deletion (Node Purge)

Replicated PV Mayastor now supports offline node deletion (purge), allowing administrators to permanently remove an unreachable and unrecoverable node from the control plane without requiring access to the underlying host. Before performing the operation, you can review the expected impact on volumes and snapshots. This capability helps simplify recovery and cleanup workflows following permanent node or infrastructure failures.

Online and Offline DiskPool Deletion

Replicated PV Mayastor now supports deleting both online and offline DiskPools. Administrators can remove empty online pools through the Mayastor plugin or DiskPool custom resources and safely purge unrecoverable offline pools from the control plane after reviewing the impact on affected volumes and snapshots.

RWX Block Volume Support for KubeVirt Live Migration (Experimental)

Replicated PV Mayastor now provides experimental support for native ReadWriteMany (RWX) block volumes to enable KubeVirt Virtual Machine (VM) live migration without requiring an intermediary NFS layer. This capability allows KubeVirt workloads to migrate between nodes while maintaining access to shared block storage. This feature is intended for evaluation and testing in non-production environments.

Affinity Group Support for Scaling to a Single Replica

Affinity groups now support scaling protected volumes down to a single replica, providing greater flexibility when adjusting replica counts for capacity optimization, maintenance, and test environments.

Enhancements

Faster Snapshot and Replica Operations

Snapshot/replica listing performance has been optimized, reducing the time required to retrieve snapshot/replica information in environments with large numbers of snapshots/replicas.

etcd Metrics Enabled by Default

etcd metrics are now enabled by default in the Helm chart, simplifying monitoring integration and improving visibility into control-plane health and performance.

Configurable RDMA QoS Settings

The Helm chart now exposes RDMA Quality of Service (QoS) configuration options, allowing administrators to configure transport-level traffic prioritization for RDMA-enabled environments.

Global Helm Configuration Values

Support for global Helm values has been added, simplifying configuration management and enabling more consistent deployment settings across OpenEBS components.

Issues Resolved

Disk Hot-Remove Detection with Ongoing I/O

Resolved an issue where hot-removal of a disk during active I/O could go undetected, potentially leaving the system in an inconsistent state. Hot-remove events are now reliably detected even under sustained I/O load.

Disk I/O Stall Detection

Resolved an issue where I/O stalls on underlying storage devices could go undetected for extended periods. The system now identifies stalled I/O paths more promptly, enabling faster fault isolation and recovery. Such Pools are not considered for replica placement until stalled IO completes.

Thread Pool Memory Exhaustion on Large Environments

Resolved an issue where thread pool memory consumption could grow unbounded in large-scale environments, potentially leading to resource exhaustion. Memory allocation within thread pools is now bounded to prevent excessive growth under high concurrency.

Node Unpublish Reliability

Resolved an issue where node unpublish operations could fail when the target path existed as an empty file.

DiskPool Resource Stability

Resolved issues that could cause unnecessary DiskPool custom resource updates during migration-related operations, reducing control-plane churn and improving stability.

Improved Handling of NVMe-oF Connectivity Issues

Resolved multiple issues related to NVMe-oF target connectivity and host management, improving stability during connection failures and recovery scenarios.

Nexus Stability During Device Removal

Resolved issues that could affect nexus stability when underlying storage devices were unexpectedly removed or became unavailable.

Nexus Stability During Device Removal

Resolved issues that could affect nexus stability when underlying storage devices were unexpectedly removed or became unavailable.

Improved Reliability of Logical Volume Operations

Resolved issues affecting logical volume lifecycle operations when underlying storage devices encountered I/O errors.

Storage Scheduling Reliability

Resolved an issue where replicas could be scheduled on pools in a critical state. Scheduling now correctly avoids unhealthy pools.

Cross-Filesystem Restore Validation

Resolved an issue where restore operations could proceed between incompatible filesystem types. Restore requests are now validated to prevent unsupported cross-filesystem restores.

OpenShift Upgrade Reliability

Resolved issues that could prevent successful upgrades on OpenShift environments.

Kubernetes Context Handling During Upgrades

Resolved an issue where upgrade operations could incorrectly use the default Kubernetes context instead of the user-specified context.

Testing

Mayastor is subject to extensive unit, component and system-level testing throughout the development and release cycle. Resources for system-level (E2E) testing are currently provided by DataCore Software.

At this time, personnel and hardware resource limitations constrain testing by the maintainers to linux builds on x86. This reflects the primary use-case which the maintainers are currently targeting with the OpenEBS Mayastor project. Therefore, the use of Mayastor with other operating systems and/or architectures, if even possible, should be considered serendipitous and wholly experimental.

This release has been subject to End-to-End testing under Ubuntu 22.04 LTS (kernel: ubuntu-6.8.0-90-generic)

  • Tested k8s versions
    • 1.23.7
    • 1.24.14
    • 1.25.10
    • 1.29.6-1.1
    • 1.32.10-1.1

Known Behavioural Limitations

  • The IO engine fully utilizes all allocated CPU cores regardless of the actual I/O load, as it runs a poller at full speed.

NOTE: An experimental implementation of interrupt mode support is now available.

  • Each DiskPool is limited to a single block device and cannot span across multiple devices.
  • The data-at-rest encryption feature does not support rotation of Data Encryption Keys (DEKs).
  • Volume rebuilds are only performed on published volumes.

Known Issues

  • if a node hosting a pod reboots and the pod lacks a controller (like a Deployment), the volume unpublish operation may not trigger. This causes the control plane to assume the volume is still in use, which leads to fsfreeze operation failure during snapshots.
    Workaround: Recreate or rebind the pod to ensure proper volume mounting.
  • Large pools (Example: 10–20TiB) may take a while during recovery after a dirty shutdown of the node hosting the io-engine.

New Contributors

Full Changelog:
v2.10.0...v2.11.0
openebs/mayastor-control-plane@v2.10.0...v2.11.0
openebs/mayastor-extensions@v2.10.0...v2.11.0

Contributors

jr42, jleeh, and 3 other contributors
Assets 7
Loading