v2.21.11

What's Changed

No changes in this release.

v2.23.0

New Features

• Add new option to restrict project deletion in the admin settings (#5925)
• Introduce Enable Share Cluster settings to toggle the share cluster feature from Admin panel (#5764)
• Add an option in admin settings to enable/enforce auto upgrades for machine deployments (#5893)
• Add support to disable changelog popup (#5905)
• Add support to import digitalocean KubeOne cluster (#5827)
• Add support to import hetzner KubeOne cluster (#5830)
• Add support to import openstack kubeone cluster (#5951)
• Add support to import VSphere kubeone cluster (#5989)
• Configure Ingress Hostname cluster settings of OpenStack provider (#5861)
• Configure report types in schedule configuration (#5894)
• Do not set Assign Public IP by default for AWS and Azure providers (#5938)
• Set Azure data disk size default value to 0 (#5987)
• Support to enable accelerated networking for machines on Azure (#5906)
• The context name for OIDC Kubeconfig has been changed to the cluster ID from default (#5810)
• VMware Cloud Director now supports authentication using API Token (#5885)

Bugfixes

• UI/UX improvements for vSphere credentials in provider settings step (#5959)
  • By default, username/password will be configured and dedicated credentials will be used to configure infra management user for vSphere
• Add cache busting mechanism for theme styles (#5943)
• Allow removing cluster label when PodNodeSelector admission plugin and clusterDefaultNodeSelector namespace are set (#5981)
• Allow updating of the clusterNetwork.proxyMode via the KKP API (PATCH endpoint) (#5803)
• AWS subnets are fetched correctly if credentials are provided directly instead of using a preset (#5883)
• Fix cluster wizard not selecting a default version if custom versions are configured in KubermaticConfiguration (#5879)
• Fix Datacenter MachineFlavorFilter not used (#5787)
• Machine Deployments are initialized without waiting for all cluster details to finish loading (#5922)
• Show correct health information for Machine Deployments with no replicas (#5837)

Design

• Add an option to clear VSphere tags category so it doesn't get stuck when there are no tags (#5940)
• Add color to required indicator of untouched and empty required form fields (#5937)
• Add indicator of what was changed on editing dialogs (#5843)
• Add warning message in the cluster list page in case some seeds are not reachable (#5982)
• Allow selection of items per page under every table along with user settings page (#5954)
• Improve page responsiveness for smaller screen sizes (#5801)
• Update Dialogs to follow latest material design specifications (#5927)
• Update the notification design and improve user experience (#5970)

Updates

• Update to Go 1.20.5 (#6025)
• Use Alpine Linux 3.17 for container images (#5814)

v2.22.4

Bugfixes

• Allow removing cluster label when PodNodeSelector admission plugin and clusterDefaultNodeSelector namespace are set (#5984)

v2.21.10

What's Changed

No changes in this release.

v2.22.3

Bugfixes

• Machine Deployments are initialized without waiting for all cluster details to finish loading (#5922)
• Fix a bug that lead to metering reports overwriting each other when used with multiple seeds. Report names now include the seed name as a prefix (#12222)

Updates

• Patch cilium v1.12 and v1.11 to latest patch releases (v1.12.9) (v1.11.16) (#12264)
• Addons: (#12184)
  • kube-state-metrics updated to v2.5.0
  • node-exporter updated to v1.4.0
• Update Anexia CCM (cloud-controller-manager) to version 1.5.4 (#12212)

New Feature

• New alert for NodeTimeDrift (#12275)

v2.21.9

Bugfixes

• Machine Deployments are initialized without waiting for all cluster details to finish loading (#5922)

Updates

• Patch cilium v1.12 and v1.11 to latest patch releases (v1.12.9) (v1.11.16) (#12272)
• Update Anexia CCM (cloud-controller-manager) to version 1.5.4 (#12214)

New Feature

• New alert for NodeTimeDrift (#12275)

v2.20.16

Updates

• Patch cilium v1.11 to latest patch release (v1.11.16) (#12273)

v2.22.2

Bugfixes

• AWS subnets are fetched correctly if credentials are provided directly instead of using a preset (#5883)
• Fix cluster wizard not selecting a default version if custom versions are configured in KubermaticConfiguration (#5879)
• Show correct health information for Machine Deployments with no replicas (#5837)

Chore

• Update golang version to 1.19.8 (#5887)

New Feature

• Configure Ingress Hostname cluster settings of OpenStack provider (#5861)

v2.21.8

Bugfixes

• Show correct health information for Machine Deployments with no replicas (#5840)

New Feature

• Configure Ingress Hostname cluster settings of OpenStack provider (#5865)

v2.20.15

What's Changed

No changes in this release.