Release 2.13.0 (2026 Q1)

What’s new #

Infrastructure #

• shtil management cluster utility improved:
  • YAML manifest console installation available;
  • command line flag installation without GUI available.
• Nvidia GPU integration with automatic node configuration.
• Migrated to Containerd 2.x as container runtime.
• Kubernetes version updated to 1.34.3.
• Shturval v2 provider host minimum CPU and Memory system requirements check.
• Client cluster configuration profile management.

GUI #

• Cluster logs GUI optimized:
  • logs grouped by stream;
  • log parameter search with available field list.
• Cluster configuration GUI improved:
  • Ingress DNS record configuration when using internal Ingress load balancer for clusters with providers without Cloud Controller Manager;
  • default resource configuration in client clusters aligned with minimum cluster installation requirements.
• Shturval v2 provider hosts with CPU and Memory below minimum requirements blocked.
• Node configuration (NCI) GUI improved:
  • Kube API Server parameter configuration in NCI GUI;
  • all configurable NCI sections available via modal windows;
  • Pause image address specification excluded in “Container Runtime” section.
• Three-position switch for cluster service operation mode management optimized.
• Cluster node configuration (NodeConfigs) state display added.
• Platform-wide summary reports added:
  • Kyverno security policies;
  • Trivy container image vulnerability scan results.
• User filter, pagination, and search settings preserved in GUI when navigating between pages.

Bugs fixed #

• Access error when adding storage in namespace by namespace administrator.
• Node changes application after NCI data update in cluster.
• Incorrect server addition to inactive service configuration for time synchronization when both chronyd and ntpd present on host.
• Cluster administrator access error when modifying recipient data in “Alerts” GUI section.
• Incorrect values from another service in service configuration available parameters GUI.
• Page load hang after editing alert rule in GUI.
• Cluster administrator cluster deletion access error.
• Missing InfraMachineTemplate data in Worker node group configuration for vSphere provider cluster in GUI.
• Shell field requirement when adding user via NCI in GUI.
• Incorrect NodeConfig status display on node GUI when cluster node configuration access unavailable.
• Automatic saved account data fill in GUI sections with similar fields.

Enterprise ML edition #

• Kubeflow platform integrated for ML workflow organization in Kubernetes.
• MLflow added for ML model lifecycle management.