Feature Gates

This page contains an overview of the various feature gates an administrator can specify on different Kubernetes components.

See feature stages for an explanation of the stages for a feature.

Overview

Feature gates are a set of key=value pairs that describe Kubernetes features. You can turn these features on or off using the --feature-gates command line flag on each Kubernetes component.

Each Kubernetes component lets you enable or disable a set of feature gates that are relevant to that component. Use -h flag to see a full set of feature gates for all components. To set feature gates for a component, such as kubelet, use the --feature-gates flag assigned to a list of feature pairs:

--feature-gates=...,GracefulNodeShutdown=true

The following tables are a summary of the feature gates that you can set on different Kubernetes components.

The "Since" column contains the Kubernetes release when a feature is introduced or its release stage is changed.
The "Until" column, if not empty, contains the last Kubernetes release in which you can still use a feature gate.
If a feature is in the Alpha or Beta state, you can find the feature listed in the Alpha/Beta feature gate table.
If a feature is stable you can find all stages for that feature listed in the Graduated/Deprecated feature gate table.
The Graduated/Deprecated feature gate table also lists deprecated and withdrawn features.

Note:

For a reference to old feature gates that are removed, please refer to feature gates removed.

Feature gates for Alpha or Beta features

Feature gates for features in Alpha or Beta states
Feature	Default	Stage	Since	Until
`AnonymousAuthConfigurableEndpoints`	`false`	Alpha	1.31	–
`AnyVolumeDataSource`	`false`	Alpha	1.18	1.23
`AnyVolumeDataSource`	`true`	Beta	1.24	–
`APIResponseCompression`	`false`	Alpha	1.7	1.15
`APIResponseCompression`	`true`	Beta	1.16	–
`APIServerIdentity`	`false`	Alpha	1.20	1.25
`APIServerIdentity`	`true`	Beta	1.26	–
`APIServerTracing`	`false`	Alpha	1.22	1.26
`APIServerTracing`	`true`	Beta	1.27	–
`AuthorizeNodeWithSelectors`	`false`	Alpha	1.31	–
`AuthorizeWithSelectors`	`false`	Alpha	1.31	–
`CloudControllerManagerWebhook`	`false`	Alpha	1.27	–
`ClusterTrustBundle`	`false`	Alpha	1.27	–
`ClusterTrustBundleProjection`	`false`	Alpha	1.29	–
`ComponentSLIs`	`false`	Alpha	1.26	1.26
`ComponentSLIs`	`true`	Beta	1.27	–
`ConsistentListFromCache`	`false`	Alpha	1.28	1.30
`ConsistentListFromCache`	`true`	Beta	1.31	–
`ContainerCheckpoint`	`false`	Alpha	1.25	1.29
`ContainerCheckpoint`	`true`	Beta	1.30	–
`ContextualLogging`	`false`	Alpha	1.24	–
`ContextualLogging`	`true`	Beta	1.30	–
`CoordinatedLeaderElection`	`false`	Alpha	1.31	–
`CPUManagerPolicyAlphaOptions`	`false`	Alpha	1.23	–
`CPUManagerPolicyBetaOptions`	`true`	Beta	1.23	–
`CPUManagerPolicyOptions`	`false`	Alpha	1.22	1.22
`CPUManagerPolicyOptions`	`true`	Beta	1.23	–
`CRDValidationRatcheting`	`false`	Alpha	1.28	1.29
`CRDValidationRatcheting`	`true`	Beta	1.30	–
`CronJobsScheduledAnnotation`	`true`	Beta	1.28	–
`CrossNamespaceVolumeDataSource`	`false`	Alpha	1.26	–
`CSIMigrationPortworx`	`false`	Alpha	1.23	1.24
`CSIMigrationPortworx`	`false`	Beta	1.25	–
`CSIVolumeHealth`	`false`	Alpha	1.21	–
`CustomCPUCFSQuotaPeriod`	`false`	Alpha	1.12	–
`CustomResourceFieldSelectors`	`false`	Alpha	1.30	1.30
`CustomResourceFieldSelectors`	`true`	Beta	1.31	–
`DisableCloudProviders`	`false`	Alpha	1.22	1.28
`DisableCloudProviders`	`true`	Beta	1.29	–
`DisableKubeletCloudCredentialProviders`	`false`	Alpha	1.23	1.28
`DisableKubeletCloudCredentialProviders`	`true`	Beta	1.29	–
`DisableNodeKubeProxyVersion`	`false`	Alpha	1.29	1.30
`DisableNodeKubeProxyVersion`	`true`	Beta	1.31	–
`DRAControlPlaneController`	`false`	Alpha	1.26	–
`DynamicResourceAllocation`	`false`	Alpha	1.30	–
`EventedPLEG`	`false`	Alpha	1.25	–
`GracefulNodeShutdown`	`false`	Alpha	1.20	1.20
`GracefulNodeShutdown`	`true`	Beta	1.21	–
`GracefulNodeShutdownBasedOnPodPriority`	`false`	Alpha	1.23	1.23
`GracefulNodeShutdownBasedOnPodPriority`	`true`	Beta	1.24	–
`HonorPVReclaimPolicy`	`false`	Alpha	1.23	1.30
`HonorPVReclaimPolicy`	`true`	Beta	1.31	–
`HPAScaleToZero`	`false`	Alpha	1.16	–
`ImageMaximumGCAge`	`false`	Alpha	1.29	1.29
`ImageMaximumGCAge`	`true`	Beta	1.30	–
`ImageVolume`	`false`	Alpha	1.31	–
`InPlacePodVerticalScaling`	`false`	Alpha	1.27	–
`InTreePluginPortworxUnregister`	`false`	Alpha	1.23	–
`JobBackoffLimitPerIndex`	`false`	Alpha	1.28	1.28
`JobBackoffLimitPerIndex`	`true`	Beta	1.29	–
`JobManagedBy`	`false`	Alpha	1.30	–
`JobPodReplacementPolicy`	`false`	Alpha	1.28	1.28
`JobPodReplacementPolicy`	`true`	Beta	1.29	–
`JobSuccessPolicy`	`false`	Alpha	1.30	1.30
`JobSuccessPolicy`	`true`	Beta	1.31	–
`KubeletCgroupDriverFromCRI`	`false`	Alpha	1.28	1.30
`KubeletCgroupDriverFromCRI`	`true`	Beta	1.31	–
`KubeletInUserNamespace`	`false`	Alpha	1.22	–
`KubeletPodResourcesDynamicResources`	`false`	Alpha	1.27	–
`KubeletPodResourcesDynamicResources`	`false`	Alpha	1.27	–
`KubeletPodResourcesGet`	`false`	Alpha	1.27	–
`KubeletSeparateDiskGC`	`false`	Alpha	1.29	1.30
`KubeletSeparateDiskGC`	`true`	Beta	1.31	–
`KubeletTracing`	`false`	Alpha	1.25	1.26
`KubeletTracing`	`true`	Beta	1.27	–
`LoadBalancerIPMode`	`false`	Alpha	1.29	1.30
`LoadBalancerIPMode`	`true`	Beta	1.30	–
`LocalStorageCapacityIsolationFSQuotaMonitoring`	`false`	Alpha	1.15	1.30
`LocalStorageCapacityIsolationFSQuotaMonitoring`	`false`	Beta	1.31	–
`LoggingAlphaOptions`	`false`	Alpha	1.24	–
`LoggingBetaOptions`	`true`	Beta	1.24	–
`MatchLabelKeysInPodAffinity`	`false`	Alpha	1.29	1.30
`MatchLabelKeysInPodAffinity`	`true`	Beta	1.31	–
`MatchLabelKeysInPodTopologySpread`	`false`	Alpha	1.25	1.26
`MatchLabelKeysInPodTopologySpread`	`true`	Beta	1.27	–
`MaxUnavailableStatefulSet`	`false`	Alpha	1.24	–
`MemoryManager`	`false`	Alpha	1.21	1.21
`MemoryManager`	`true`	Beta	1.22	–
`MemoryQoS`	`false`	Alpha	1.22	–
`MultiCIDRServiceAllocator`	`false`	Alpha	1.27	1.30
`MultiCIDRServiceAllocator`	`false`	Beta	1.31	–
`MutatingAdmissionPolicy`	`false`	Alpha	1.30	–
`NFTablesProxyMode`	`false`	Alpha	1.29	1.30
`NFTablesProxyMode`	`true`	Beta	1.31	–
`NodeInclusionPolicyInPodTopologySpread`	`false`	Alpha	1.25	1.25
`NodeInclusionPolicyInPodTopologySpread`	`true`	Beta	1.26	–
`NodeLogQuery`	`false`	Alpha	1.27	1.29
`NodeLogQuery`	`false`	Beta	1.30	–
`NodeSwap`	`false`	Alpha	1.22	1.27
`NodeSwap`	`false`	Beta	1.28	1.29
`NodeSwap`	`true`	Beta	1.30	–
`OpenAPIEnums`	`false`	Alpha	1.23	1.23
`OpenAPIEnums`	`true`	Beta	1.24	–
`PodAndContainerStatsFromCRI`	`false`	Alpha	1.23	–
`PodDeletionCost`	`false`	Alpha	1.21	1.21
`PodDeletionCost`	`true`	Beta	1.22	–
`PodIndexLabel`	`true`	Beta	1.28	–
`PodLifecycleSleepAction`	`false`	Alpha	1.29	1.29
`PodLifecycleSleepAction`	`true`	Beta	1.30	–
`PodReadyToStartContainersCondition`	`false`	Alpha	1.28	1.28
`PodReadyToStartContainersCondition`	`true`	Beta	1.29	–
`PortForwardWebsockets`	`false`	Alpha	1.30	1.30
`PortForwardWebsockets`	`true`	Beta	1.31	–
`ProcMountType`	`false`	Alpha	1.12	–
`QOSReserved`	`false`	Alpha	1.11	–
`RecoverVolumeExpansionFailure`	`false`	Alpha	1.23	–
`RecursiveReadOnlyMounts`	`false`	Alpha	1.30	1.30
`RecursiveReadOnlyMounts`	`true`	Beta	1.31	–
`RelaxedEnvironmentVariableValidation`	`false`	Alpha	1.30	–
`ResilientWatchCacheInitialization`	`true`	Beta	1.31	–
`ResourceHealthStatus`	`false`	Alpha	1.31	–
`RetryGenerateName`	`false`	Alpha	1.30	–
`RetryGenerateName`	`false`	Alpha	1.30	–
`RotateKubeletServerCertificate`	`false`	Alpha	1.7	1.11
`RotateKubeletServerCertificate`	`true`	Beta	1.12	–
`RuntimeClassInImageCriApi`	`false`	Alpha	1.29	–
`SchedulerQueueingHints`	`true`	Beta	1.28	1.28
`SchedulerQueueingHints`	`false`	Beta	1.29	–
`SELinuxMount`	`false`	Alpha	1.30	–
`SELinuxMountReadWriteOncePod`	`false`	Alpha	1.25	1.26
`SELinuxMountReadWriteOncePod`	`false`	Beta	1.27	1.27
`SELinuxMountReadWriteOncePod`	`true`	Beta	1.28	–
`SeparateTaintEvictionController`	`true`	Beta	1.29	–
`ServiceAccountTokenJTI`	`false`	Alpha	1.29	1.29
`ServiceAccountTokenJTI`	`true`	Beta	1.30	–
`ServiceAccountTokenNodeBinding`	`false`	Alpha	1.29	1.30
`ServiceAccountTokenNodeBinding`	`true`	Beta	1.31	–
`ServiceAccountTokenNodeBindingValidation`	`false`	Alpha	1.29	1.29
`ServiceAccountTokenNodeBindingValidation`	`true`	Beta	1.30	–
`ServiceAccountTokenPodNodeInfo`	`false`	Alpha	1.29	1.29
`ServiceAccountTokenPodNodeInfo`	`true`	Beta	1.30	–
`ServiceTrafficDistribution`	`false`	Alpha	1.30	1.30
`ServiceTrafficDistribution`	`true`	Beta	1.31	–
`SidecarContainers`	`false`	Alpha	1.28	1.28
`SidecarContainers`	`true`	Beta	1.29	–
`SizeMemoryBackedVolumes`	`false`	Alpha	1.20	1.21
`SizeMemoryBackedVolumes`	`true`	Beta	1.22	–
`StatefulSetAutoDeletePVC`	`false`	Alpha	1.23	1.26
`StatefulSetAutoDeletePVC`	`true`	Beta	1.27	–
`StorageVersionAPI`	`false`	Alpha	1.20	–
`StorageVersionHash`	`false`	Alpha	1.14	1.14
`StorageVersionHash`	`true`	Beta	1.15	–
`StorageVersionMigrator`	`false`	Alpha	1.30	1.32
`StructuredAuthenticationConfiguration`	`false`	Alpha	1.29	1.29
`StructuredAuthenticationConfiguration`	`true`	Beta	1.30	–
`StructuredAuthorizationConfiguration`	`false`	Alpha	1.29	1.29
`StructuredAuthorizationConfiguration`	`true`	Beta	1.30	–
`SupplementalGroupsPolicy`	`false`	Alpha	1.31	–
`TopologyAwareHints`	`false`	Alpha	1.21	1.22
`TopologyAwareHints`	`false`	Beta	1.23	1.23
`TopologyAwareHints`	`true`	Beta	1.24	–
`TopologyManagerPolicyAlphaOptions`	`false`	Alpha	1.26	–
`TopologyManagerPolicyBetaOptions`	`false`	Beta	1.26	1.27
`TopologyManagerPolicyBetaOptions`	`true`	Beta	1.28	–
`TopologyManagerPolicyOptions`	`false`	Alpha	1.26	1.27
`TopologyManagerPolicyOptions`	`true`	Beta	1.28	–
`TranslateStreamCloseWebsocketRequests`	`true`	Beta	1.30	–
`UnauthenticatedHTTP2DOSMitigation`	`false`	Beta	1.28	1.28
`UnauthenticatedHTTP2DOSMitigation`	`true`	Beta	1.29	–
`UnknownVersionInteroperabilityProxy`	`false`	Alpha	1.28	–
`UserNamespacesPodSecurityStandards`	`false`	Alpha	1.29	–
`UserNamespacesSupport`	`false`	Alpha	1.28	1.29
`UserNamespacesSupport`	`false`	Beta	1.30	–
`VolumeAttributesClass`	`false`	Alpha	1.29	1.30
`VolumeAttributesClass`	`false`	Beta	1.31	–
`VolumeCapacityPriority`	`false`	Alpha	1.21	–
`WatchCacheInitializationPostStartHook`	`false`	Beta	1.31	–
`WatchFromStorageWithoutResourceVersion`	`false`	Beta	1.30	–
`WatchList`	`false`	Alpha	1.27	–
`WindowsHostNetwork`	`true`	Alpha	1.26	–
`WinDSR`	`false`	Alpha	1.14	–
`WinOverlay`	`false`	Alpha	1.14	1.19
`WinOverlay`	`true`	Beta	1.20	–

Feature gates for graduated or deprecated features

Feature Gates for Graduated or Deprecated Features
Feature	Default	Stage	Since	Until
`AdmissionWebhookMatchConditions`	`false`	Alpha	1.27	1.27
`AdmissionWebhookMatchConditions`	`true`	Beta	1.28	1.29
`AdmissionWebhookMatchConditions`	`true`	GA	1.30	–
`AggregatedDiscoveryEndpoint`	`false`	Alpha	1.26	1.26
`AggregatedDiscoveryEndpoint`	`true`	Beta	1.27	1.29
`AggregatedDiscoveryEndpoint`	`true`	GA	1.30	–
`AllowServiceLBStatusOnNonLB`	`false`	Deprecated	1.29	–
`APIListChunking`	`false`	Alpha	1.8	1.8
`APIListChunking`	`true`	Beta	1.9	1.28
`APIListChunking`	`true`	GA	1.29	–
`AppArmor`	`true`	Beta	1.4	1.30
`AppArmor`	`true`	GA	1.31	–
`CloudDualStackNodeIPs`	`false`	Alpha	1.27	1.28
`CloudDualStackNodeIPs`	`true`	Beta	1.29	1.29
`CloudDualStackNodeIPs`	`true`	GA	1.30	–
`CPUManager`	`false`	Alpha	1.8	1.9
`CPUManager`	`true`	Beta	1.10	1.25
`CPUManager`	`true`	GA	1.26	–
`DefaultHostNetworkHostPortsInPodTemplates`	`false`	Deprecated	1.28	–
`DevicePluginCDIDevices`	`false`	Alpha	1.28	1.28
`DevicePluginCDIDevices`	`true`	Beta	1.29	1.30
`DevicePluginCDIDevices`	`true`	GA	1.31	–
`EfficientWatchResumption`	`false`	Alpha	1.20	1.20
`EfficientWatchResumption`	`true`	Beta	1.21	1.23
`EfficientWatchResumption`	`true`	GA	1.24	–
`ElasticIndexedJob`	`true`	Beta	1.27	1.30
`ElasticIndexedJob`	`true`	GA	1.31	–
`ExecProbeTimeout`	`true`	GA	1.20	–
`HPAContainerMetrics`	`false`	Alpha	1.20	1.26
`HPAContainerMetrics`	`true`	Beta	1.27	1.29
`HPAContainerMetrics`	`true`	GA	1.30	–
`JobPodFailurePolicy`	`false`	Alpha	1.25	1.25
`JobPodFailurePolicy`	`true`	Beta	1.26	1.30
`JobPodFailurePolicy`	`true`	GA	1.31	–
`KMSv1`	`true`	Deprecated	1.28	1.28
`KMSv1`	`false`	Deprecated	1.29	–
`KMSv2`	`false`	Alpha	1.25	1.26
`KMSv2`	`true`	Beta	1.27	1.28
`KMSv2`	`true`	GA	1.29	–
`KMSv2KDF`	`false`	Beta	1.28	1.28
`KMSv2KDF`	`true`	GA	1.29	–
`KubeProxyDrainingTerminatingNodes`	`false`	Alpha	1.28	1.30
`KubeProxyDrainingTerminatingNodes`	`true`	Beta	1.30	1.30
`KubeProxyDrainingTerminatingNodes`	`true`	GA	1.31	–
`LegacyServiceAccountTokenCleanUp`	`false`	Alpha	1.28	1.28
`LegacyServiceAccountTokenCleanUp`	`true`	Beta	1.29	1.29
`LegacyServiceAccountTokenCleanUp`	`true`	GA	1.30	–
`LogarithmicScaleDown`	`false`	Alpha	1.21	1.21
`LogarithmicScaleDown`	`true`	Beta	1.22	1.30
`LogarithmicScaleDown`	`true`	GA	1.31	–
`MinDomainsInPodTopologySpread`	`false`	Alpha	1.24	1.24
`MinDomainsInPodTopologySpread`	`false`	Beta	1.25	1.26
`MinDomainsInPodTopologySpread`	`true`	Beta	1.27	1.29
`MinDomainsInPodTopologySpread`	`true`	GA	1.30	–
`NewVolumeManagerReconstruction`	`false`	Beta	1.27	1.27
`NewVolumeManagerReconstruction`	`true`	Beta	1.28	1.29
`NewVolumeManagerReconstruction`	`true`	GA	1.30	–
`NodeOutOfServiceVolumeDetach`	`false`	Alpha	1.24	1.25
`NodeOutOfServiceVolumeDetach`	`true`	Beta	1.26	1.27
`NodeOutOfServiceVolumeDetach`	`true`	GA	1.28	–
`PDBUnhealthyPodEvictionPolicy`	`false`	Alpha	1.26	1.26
`PDBUnhealthyPodEvictionPolicy`	`true`	Beta	1.27	1.30
`PDBUnhealthyPodEvictionPolicy`	`true`	GA	1.31	–
`PersistentVolumeLastPhaseTransitionTime`	`false`	Alpha	1.28	1.28
`PersistentVolumeLastPhaseTransitionTime`	`true`	Beta	1.29	1.30
`PersistentVolumeLastPhaseTransitionTime`	`true`	GA	1.31	–
`PodDisruptionConditions`	`false`	Alpha	1.25	1.25
`PodDisruptionConditions`	`true`	Beta	1.26	1.30
`PodDisruptionConditions`	`true`	GA	1.31	–
`PodHostIPs`	`false`	Alpha	1.28	1.28
`PodHostIPs`	`true`	Beta	1.29	1.30
`PodHostIPs`	`true`	GA	1.30	–
`PodSchedulingReadiness`	`false`	Alpha	1.26	1.26
`PodSchedulingReadiness`	`true`	Beta	1.27	1.29
`PodSchedulingReadiness`	`true`	GA	1.30	–
`RemainingItemCount`	`false`	Alpha	1.15	1.15
`RemainingItemCount`	`true`	Beta	1.16	1.28
`RemainingItemCount`	`true`	GA	1.29	–
`ServerSideApply`	`false`	Alpha	1.14	1.15
`ServerSideApply`	`true`	Beta	1.16	1.21
`ServerSideApply`	`true`	GA	1.22	–
`ServerSideFieldValidation`	`false`	Alpha	1.23	1.24
`ServerSideFieldValidation`	`true`	Beta	1.25	1.26
`ServerSideFieldValidation`	`true`	GA	1.27	–
`StableLoadBalancerNodeSet`	`true`	Beta	1.27	1.29
`StableLoadBalancerNodeSet`	`true`	GA	1.30	–
`StatefulSetStartOrdinal`	`false`	Alpha	1.26	1.26
`StatefulSetStartOrdinal`	`true`	Beta	1.27	1.30
`StatefulSetStartOrdinal`	`true`	GA	1.31	–
`ValidatingAdmissionPolicy`	`false`	Alpha	1.26	1.27
`ValidatingAdmissionPolicy`	`false`	Beta	1.28	1.29
`ValidatingAdmissionPolicy`	`true`	GA	1.30	–
`WatchBookmark`	`false`	Alpha	1.15	1.15
`WatchBookmark`	`true`	Beta	1.16	1.16
`WatchBookmark`	`true`	GA	1.17	–
`ZeroLimitedNominalConcurrencyShares`	`false`	Beta	1.29	1.29
`ZeroLimitedNominalConcurrencyShares`	`true`	GA	1.30	–

Using a feature

Feature stages

A feature can be in Alpha, Beta or GA stage. An Alpha feature means:

Disabled by default.
Might be buggy. Enabling the feature may expose bugs.
Support for feature may be dropped at any time without notice.
The API may change in incompatible ways in a later software release without notice.
Recommended for use only in short-lived testing clusters, due to increased risk of bugs and lack of long-term support.

A Beta feature means:

Usually enabled by default. Beta API groups are disabled by default.
The feature is well tested. Enabling the feature is considered safe.
Support for the overall feature will not be dropped, though details may change.
The schema and/or semantics of objects may change in incompatible ways in a subsequent beta or stable release. When this happens, we will provide instructions for migrating to the next version. This may require deleting, editing, and re-creating API objects. The editing process may require some thought. This may require downtime for applications that rely on the feature.
Recommended for only non-business-critical uses because of potential for incompatible changes in subsequent releases. If you have multiple clusters that can be upgraded independently, you may be able to relax this restriction.

Note:

Please do try Beta features and give feedback on them! After they exit beta, it may not be practical for us to make more changes.

A General Availability (GA) feature is also referred to as a stable feature. It means:

The feature is always enabled; you cannot disable it.
The corresponding feature gate is no longer needed.
Stable versions of features will appear in released software for many subsequent versions.

List of feature gates

Each feature gate is designed for enabling/disabling a specific feature.

AdmissionWebhookMatchConditions: Enable match conditions on mutating & validating admission webhooks.
AggregatedDiscoveryEndpoint: Enable a single HTTP endpoint /discovery/<version> which supports native HTTP caching with ETags containing all APIResources known to the API server.
AllowServiceLBStatusOnNonLB: Enables .status.ingress.loadBalancer to be set on Services of types other than LoadBalancer.
AnonymousAuthConfigurableEndpoints: Enable configurable endpoints for anonymous auth for the API server.
AnyVolumeDataSource: Enable use of any custom resource as the DataSource of a PVC.
APIListChunking: Enable the API clients to retrieve (LIST or GET) resources from API server in chunks.
APIResponseCompression: Compress the API responses for LIST or GET requests.
APIServerIdentity: Assign each API server an ID in a cluster, using a Lease.
APIServerTracing: Add support for distributed tracing in the API server. See Traces for Kubernetes System Components for more details.
AppArmor: Enable use of AppArmor mandatory access control for Pods running on Linux nodes. See AppArmor Tutorial for more details.
AuthorizeNodeWithSelectors: Make the Node authorizer use fine-grained selector authorization. Requires AuthorizeWithSelectors to be enabled.
AuthorizeWithSelectors: Allows authorization to use field and label selectors. Enables fieldSelector and labelSelector fields in the SubjectAccessReview API, passes field and label selector information to authorization webhooks, enables fieldSelector and labelSelector functions in the authorizer CEL library, and enables checking fieldSelector and labelSelector fields in authorization webhook matchConditions.
CloudControllerManagerWebhook: Enable webhooks in cloud controller manager.
CloudDualStackNodeIPs: Enables dual-stack kubelet --node-ip with external cloud providers. See Configure IPv4/IPv6 dual-stack for more details.
ClusterTrustBundle: Enable ClusterTrustBundle objects and kubelet integration.
ClusterTrustBundleProjection: clusterTrustBundle projected volume sources.
ComponentSLIs: Enable the /metrics/slis endpoint on Kubernetes components like kubelet, kube-scheduler, kube-proxy, kube-controller-manager, cloud-controller-manager allowing you to scrape health check metrics.
ConsistentListFromCache:
Enhance Kubernetes API server performance by serving consistent list requests directly from its watch cache, improving scalability and response times. To consistent list from cache Kubernetes requires a newer etcd version (v3.4.31+ or v3.5.13+), that includes fixes to watch progress request feature. If older etcd version is provided Kubernetes will automatically detect it and fallback to serving consistent reads from etcd. Progress notifications ensure watch cache is consistent with etcd while reducing the need for resource-intensive quorum reads from etcd.
See the Kubernetes documentation on Semantics for get and list for more details.
ContainerCheckpoint: Enables the kubelet checkpoint API. See Kubelet Checkpoint API for more details.
ContextualLogging: Enables extra details in log output of Kubernetes components that support contextual logging.
CoordinatedLeaderElection: Enables the behaviors supporting the LeaseCandidate API, and also enables coordinated leader election for the Kubernetes control plane, deterministically.
CPUManager: Enable container level CPU affinity support, see CPU Management Policies.
CPUManagerPolicyAlphaOptions: This allows fine-tuning of CPUManager policies, experimental, Alpha-quality options This feature gate guards a group of CPUManager options whose quality level is alpha. This feature gate will never graduate to beta or stable.
CPUManagerPolicyBetaOptions: This allows fine-tuning of CPUManager policies, experimental, Beta-quality options This feature gate guards a group of CPUManager options whose quality level is beta. This feature gate will never graduate to stable.
CPUManagerPolicyOptions: Allow fine-tuning of CPUManager policies.
CRDValidationRatcheting: Enable updates to custom resources to contain violations of their OpenAPI schema if the offending portions of the resource update did not change. See Validation Ratcheting for more details.
CronJobsScheduledAnnotation: Set the scheduled job time as an annotation on Jobs that were created on behalf of a CronJob.
CrossNamespaceVolumeDataSource: Enable the usage of cross namespace volume data source to allow you to specify a source namespace in the dataSourceRef field of a PersistentVolumeClaim.
CSIMigrationPortworx: Enables shims and translation logic to route volume operations from the Portworx in-tree plugin to Portworx CSI plugin. Requires Portworx CSI driver to be installed and configured in the cluster.
CSIVolumeHealth: Enable support for CSI volume health monitoring on node.
CustomCPUCFSQuotaPeriod: Enable nodes to change cpuCFSQuotaPeriod in kubelet config.
CustomResourceFieldSelectors: Enable selectableFields in the CustomResourceDefinition API to allow filtering of custom resource list, watch and deletecollection requests.
DefaultHostNetworkHostPortsInPodTemplates:
This feature gate controls the point at which a default value for .spec.containers[*].ports[*].hostPort is assigned, for Pods using hostNetwork: true. The default since Kubernetes v1.28 is to only set a default value in Pods.
Enabling this means a default will be assigned even to the .spec of an embedded PodTemplate (for example, in a Deployment), which is the way that older releases of Kubernetes worked. You should migrate your code so that it does not rely on the legacy behavior.
DevicePluginCDIDevices: Enable support to CDI device IDs in the Device Plugin API.
DisableCloudProviders: Disables any functionality in kube-apiserver, kube-controller-manager and kubelet related to the --cloud-provider component flag.
DisableKubeletCloudCredentialProviders: Disable the in-tree functionality in kubelet to authenticate to a cloud provider container registry for image pull credentials.
DisableNodeKubeProxyVersion: Disable setting the kubeProxyVersion field of the Node.
DRAControlPlaneController: Enables support for resources with custom parameters and a lifecycle that is independent of a Pod. Allocation of resources is handled by a resource driver's control plane controller.
DynamicResourceAllocation: Enables support for resources with custom parameters and a lifecycle that is independent of a Pod. Allocation of resources is handled by the Kubernetes scheduler based on "structured parameters".
EfficientWatchResumption: Allows for storage-originated bookmark (progress notify) events to be delivered to the users. This is only applied to watch operations.
ElasticIndexedJob: Enables Indexed Jobs to be scaled up or down by mutating both spec.completions and spec.parallelism together such that spec.completions == spec.parallelism. See docs on elastic Indexed Jobs for more details.
EventedPLEG: Enable support for the kubelet to receive container life cycle events from the container runtime via an extension to CRI. (PLEG is an abbreviation for “Pod lifecycle event generator”). For this feature to be useful, you also need to enable support for container lifecycle events in each container runtime running in your cluster. If the container runtime does not announce support for container lifecycle events then the kubelet automatically switches to the legacy generic PLEG mechanism, even if you have this feature gate enabled.
ExecProbeTimeout: Ensure kubelet respects exec probe timeouts. This feature gate exists in case any of your existing workloads depend on a now-corrected fault where Kubernetes ignored exec probe timeouts. See readiness probes.
GracefulNodeShutdown: Enables support for graceful shutdown in kubelet. During a system shutdown, kubelet will attempt to detect the shutdown event and gracefully terminate pods running on the node. See Graceful Node Shutdown for more details.
GracefulNodeShutdownBasedOnPodPriority: Enables the kubelet to check Pod priorities when shutting down a node gracefully.
HonorPVReclaimPolicy: Honor persistent volume reclaim policy when it is Delete irrespective of PV-PVC deletion ordering. For more details, check the PersistentVolume deletion protection finalizer documentation.
HPAContainerMetrics: Allow HorizontalPodAutoscalers to scale based on metrics from individual containers within target pods.
HPAScaleToZero: Enables setting minReplicas to 0 for HorizontalPodAutoscaler resources when using custom or external metrics.
ImageMaximumGCAge: Enables the kubelet configuration field imageMaximumGCAge, allowing an administrator to specify the age after which an image will be garbage collected.
ImageVolume: Allow using the image volume source in a Pod. This volume source lets you mount a container image as a read-only volume.
InPlacePodVerticalScaling: Enables in-place Pod vertical scaling.
InTreePluginPortworxUnregister: Stops registering the Portworx in-tree plugin in kubelet and volume controllers.
JobBackoffLimitPerIndex: Allows specifying the maximal number of pod retries per index in Indexed jobs.
JobManagedBy: Allows to delegate reconciliation of a Job object to an external controller.
JobPodFailurePolicy: Allow users to specify handling of pod failures based on container exit codes and pod conditions.
JobPodReplacementPolicy: Allows you to specify pod replacement for terminating pods in a Job
JobSuccessPolicy: Allow users to specify when a Job can be declared as succeeded based on the set of succeeded pods.
KMSv1: Enables KMS v1 API for encryption at rest. See Using a KMS Provider for data encryption for more details.
KMSv2: Enables KMS v2 API for encryption at rest. See Using a KMS Provider for data encryption for more details.
KMSv2KDF: Enables KMS v2 to generate single use data encryption keys. See Using a KMS Provider for data encryption for more details. If the KMSv2 feature gate is not enabled in your cluster, the value of the KMSv2KDF feature gate has no effect.
KubeletCgroupDriverFromCRI: Enable detection of the kubelet cgroup driver configuration option from the CRI. You can use this feature gate on nodes with a kubelet that supports the feature gate and where there is a CRI container runtime that supports the RuntimeConfig CRI call. If both CRI and kubelet support this feature, the kubelet ignores the cgroupDriver configuration setting (or deprecated --cgroup-driver command line argument). If you enable this feature gate and the container runtime doesn't support it, the kubelet falls back to using the driver configured using the cgroupDriver configuration setting. See Configuring a cgroup driver for more details.
KubeletInUserNamespace: Enables support for running kubelet in a user namespace. See Running Kubernetes Node Components as a Non-root User.
KubeletPodResourcesDynamicResources: Extend the kubelet's pod resources gRPC endpoint to to include resources allocated in ResourceClaims via DynamicResourceAllocation API. See resource allocation reporting for more details. with information about the allocatable resources, enabling clients to properly track the free compute resources on a node.
KubeletPodResourcesDynamicResources: Extend the kubelet's pod resources gRPC endpoint to to include resources allocated in ResourceClaims via DynamicResourceAllocation API. See resource allocation reporting for more details. with information about the allocatable resources, enabling clients to properly track the free compute resources on a node.
KubeletPodResourcesGet: Enable the Get gRPC endpoint on kubelet's for Pod resources. This API augments the resource allocation reporting.
KubeletSeparateDiskGC: The split image filesystem feature enables kubelet to perform garbage collection of images (read-only layers) and/or containers (writeable layers) deployed on separate filesystems.
KubeletTracing: Add support for distributed tracing in the kubelet. When enabled, kubelet CRI interface and authenticated http servers are instrumented to generate OpenTelemetry trace spans. See Traces for Kubernetes System Components for more details.
KubeProxyDrainingTerminatingNodes: Implement connection draining for terminating nodes for externalTrafficPolicy: Cluster services.
LegacyServiceAccountTokenCleanUp: Enable cleaning up Secret-based service account tokens when they are not used in a specified time (default to be one year).
LoadBalancerIPMode: Allows setting ipMode for Services where type is set to LoadBalancer. See Specifying IPMode of load balancer status for more information.
LocalStorageCapacityIsolationFSQuotaMonitoring: When LocalStorageCapacityIsolation is enabled for local ephemeral storage, the backing filesystem for emptyDir volumes supports project quotas, and UserNamespacesSupport is enabled, project quotas are used to monitor emptyDir volume storage consumption rather than using filesystem walk, ensuring better performance and accuracy.
LogarithmicScaleDown: Enable semi-random selection of pods to evict on controller scaledown based on logarithmic bucketing of pod timestamps.
LoggingAlphaOptions: Allow fine-tuning of experimental, alpha-quality logging options.
LoggingBetaOptions: Allow fine-tuning of experimental, beta-quality logging options.
MatchLabelKeysInPodAffinity: Enable the matchLabelKeys and mismatchLabelKeys fields for pod (anti)affinity.
MatchLabelKeysInPodTopologySpread: Enable the matchLabelKeys field for Pod topology spread constraints.
MaxUnavailableStatefulSet: Enables setting the maxUnavailable field for the rolling update strategy of a StatefulSet. The field specifies the maximum number of Pods that can be unavailable during the update.
MemoryManager: Allows setting memory affinity for a container based on NUMA topology.
MemoryQoS: Enable memory protection and usage throttle on pod / container using cgroup v2 memory controller.
MinDomainsInPodTopologySpread: Enable minDomains in Pod topology spread constraints.
MultiCIDRServiceAllocator: Track IP address allocations for Service cluster IPs using IPAddress objects.
MutatingAdmissionPolicy: In Kubernetes 1.31, this feature gate has no effect. A future release of Kubernetes may use this feature gate to enable the MutatingAdmissionPolicy in admission chain.
NewVolumeManagerReconstruction:
Enables improved discovery of mounted volumes during kubelet startup. Since the associated code had been significantly refactored, Kubernetes versions 1.25 to 1.29 allowed you to opt-out in case the kubelet got stuck at the startup, or did not unmount volumes from terminated Pods.
This refactoring was behind the SELinuxMountReadWriteOncePod feature gate in Kubernetes releases 1.25 and 1.26.
NFTablesProxyMode: Allow running kube-proxy in nftables mode.
NodeInclusionPolicyInPodTopologySpread: Enable using nodeAffinityPolicy and nodeTaintsPolicy in Pod topology spread constraints when calculating pod topology spread skew.
NodeLogQuery: Enables querying logs of node services using the /logs endpoint.
NodeOutOfServiceVolumeDetach: When a Node is marked out-of-service using the node.kubernetes.io/out-of-service taint, Pods on the node will be forcefully deleted if they can not tolerate this taint, and the volume detach operations for Pods terminating on the node will happen immediately. The deleted Pods can recover quickly on different nodes.
NodeSwap: Enable the kubelet to allocate swap memory for Kubernetes workloads on a node. Must be used with KubeletConfiguration.failSwapOn set to false. For more details, please see swap memory
OpenAPIEnums: Enables populating "enum" fields of OpenAPI schemas in the spec returned from the API server.
PDBUnhealthyPodEvictionPolicy: Enables the unhealthyPodEvictionPolicy field of a PodDisruptionBudget. This specifies when unhealthy pods should be considered for eviction. Please see Unhealthy Pod Eviction Policy for more details.
PersistentVolumeLastPhaseTransitionTime: Adds a new field to PersistentVolume which holds a timestamp of when the volume last transitioned its phase.
PodAndContainerStatsFromCRI: Configure the kubelet to gather container and pod stats from the CRI container runtime rather than gathering them from cAdvisor. As of 1.26, this also includes gathering metrics from CRI and emitting them over /metrics/cadvisor (rather than having cAdvisor emit them directly).
PodDeletionCost: Enable the Pod Deletion Cost feature which allows users to influence ReplicaSet downscaling order.
PodDisruptionConditions: Enables support for appending a dedicated pod condition indicating that the pod is being deleted due to a disruption.
PodHostIPs: Enable the status.hostIPs field for pods and the downward API. The field lets you expose host IP addresses to workloads.
PodIndexLabel: Enables the Job controller and StatefulSet controller to add the pod index as a label when creating new pods. See Job completion mode docs and StatefulSet pod index label docs for more details.
PodLifecycleSleepAction: Enables the sleep action in Container lifecycle hooks.
PodReadyToStartContainersCondition:
Enable the kubelet to mark the PodReadyToStartContainers condition on pods.
This feature gate was previously known as PodHasNetworkCondition, and the associated condition was named PodHasNetwork.
PodSchedulingReadiness: Enable setting schedulingGates field to control a Pod's scheduling readiness.
PortForwardWebsockets: Allow WebSocket streaming of the portforward sub-protocol (port-forward) from clients requesting version v2 (v2.portforward.k8s.io) of the sub-protocol.
ProcMountType: Enables control over the type proc mounts for containers by setting the procMount field of a SecurityContext.
QOSReserved: Allows resource reservations at the QoS level preventing pods at lower QoS levels from bursting into resources requested at higher QoS levels (memory only for now).
RecoverVolumeExpansionFailure: Enables users to edit their PVCs to smaller sizes so as they can recover from previously issued volume expansion failures. See Recovering from Failure when Expanding Volumes for more details.
RecursiveReadOnlyMounts: Enables support for recursive read-only mounts. For more details, see read-only mounts.
RelaxedEnvironmentVariableValidation: Allow almost all printable ASCII characters in environment variables.
RemainingItemCount: Allow the API servers to show a count of remaining items in the response to a chunking list request.
ResilientWatchCacheInitialization: Enables resilient watchcache initialization to avoid controlplane overload.
ResourceHealthStatus: Enable the allocatedResourcesStatus field within the .status for a Pod. The field reports additional details for each container in the Pod, with the health information for each device assigned to the Pod. See Device plugin and unhealthy devices for more details.
RetryGenerateName: Enables retrying of object creation when the API server is expected to generate a name. When this feature is enabled, requests using generateName are retried automatically in case the control plane detects a name conflict with an existing object, up to a limit of 8 total attempts.
RetryGenerateName: Enables retrying of object creation when the API server is expected to generate a name. When this feature is enabled, requests using generateName are retried automatically in case the control plane detects a name conflict with an existing object, up to a limit of 8 total attempts.
RotateKubeletServerCertificate: Enable the rotation of the server TLS certificate on the kubelet. See kubelet configuration for more details.
RuntimeClassInImageCriApi: Enables images to be pulled based on the runtime class of the pods that reference them.
SchedulerQueueingHints: Enables the scheduler's queueing hints enhancement, which benefits to reduce the useless requeueing. The scheduler retries scheduling pods if something changes in the cluster that could make the pod scheduled. Queueing hints are internal signals that allow the scheduler to filter the changes in the cluster that are relevant to the unscheduled pod, based on previous scheduling attempts.
SELinuxMount:
Speeds up container startup by allowing kubelet to mount volumes for a Pod directly with the correct SELinux label instead of changing each file on the volumes recursively. It widens the performance improvements behind the SELinuxMountReadWriteOncePod feature gate by extending the implementation to all volumes.
Enabling the SELinuxMount feature gate requires the feature gate SELinuxMountReadWriteOncePod to be enabled.
SELinuxMountReadWriteOncePod: Speeds up container startup by allowing kubelet to mount volumes for a Pod directly with the correct SELinux label instead of changing each file on the volumes recursively. The initial implementation focused on ReadWriteOncePod volumes.
SeparateTaintEvictionController: Enables running TaintEvictionController, that performs Taint-based Evictions, in a controller separated from NodeLifecycleController. When this feature is enabled, users can optionally disable Taint-based Eviction setting the --controllers=-taint-eviction-controller flag on the kube-controller-manager.
ServerSideApply: Enables the Sever Side Apply (SSA) feature on the API Server.
ServerSideFieldValidation: Enables server-side field validation. This means the validation of resource schema is performed at the API server side rather than the client side (for example, the kubectl create or kubectl apply command line).
ServiceAccountTokenJTI: Controls whether JTIs (UUIDs) are embedded into generated service account tokens, and whether these JTIs are recorded into the Kubernetes audit log for future requests made by these tokens.
ServiceAccountTokenNodeBinding: Controls whether the API server allows binding service account tokens to Node objects.
ServiceAccountTokenNodeBindingValidation: Controls whether the apiserver will validate a Node reference in service account tokens.
ServiceAccountTokenPodNodeInfo: Controls whether the apiserver embeds the node name and uid for the associated node when issuing service account tokens bound to Pod objects.
ServiceTrafficDistribution: Allows usage of the optional spec.trafficDistribution field in Services. The field offers a way to express preferences for how traffic is distributed to Service endpoints.
SidecarContainers: Allow setting the restartPolicy of an init container to Always so that the container becomes a sidecar container (restartable init containers). See Sidecar containers and restartPolicy for more details.
SizeMemoryBackedVolumes: Enable kubelets to determine the size limit for memory-backed volumes (mainly emptyDir volumes).
StableLoadBalancerNodeSet: Enables less load balancer re-configurations by the service controller (KCCM) as an effect of changing node state.
StatefulSetAutoDeletePVC: Allows the use of the optional .spec.persistentVolumeClaimRetentionPolicy field, providing control over the deletion of PVCs in a StatefulSet's lifecycle. See PersistentVolumeClaim retention for more details.
StatefulSetStartOrdinal: Allow configuration of the start ordinal in a StatefulSet. See Start ordinal for more details.
StorageVersionAPI: Enable the storage version API.
StorageVersionHash: Allow API servers to expose the storage version hash in the discovery.
StorageVersionMigrator: Enables storage version migration. See Migrate Kubernetes Objects Using Storage Version Migration for more details.
StructuredAuthenticationConfiguration: Enable structured authentication configuration for the API server.
StructuredAuthorizationConfiguration: Enable structured authorization configuration, so that cluster administrators can specify more than one authorization webhook in the API server handler chain.
SupplementalGroupsPolicy: Enables support for fine-grained SupplementalGroups control. For more details, see Configure fine-grained SupplementalGroups control for a Pod.
TopologyAwareHints: Enables topology aware routing based on topology hints in EndpointSlices. See Topology Aware Hints for more details.
TopologyManagerPolicyAlphaOptions: Allow fine-tuning of topology manager policies, experimental, Alpha-quality options. This feature gate guards a group of topology manager options whose quality level is alpha. This feature gate will never graduate to beta or stable.
TopologyManagerPolicyBetaOptions: Allow fine-tuning of topology manager policies, experimental, Beta-quality options. This feature gate guards a group of topology manager options whose quality level is beta. This feature gate will never graduate to stable.
TopologyManagerPolicyOptions: Enable fine-tuning of topology manager policies.
TranslateStreamCloseWebsocketRequests: Allow WebSocket streaming of the remote command sub-protocol (exec, cp, attach) from clients requesting version 5 (v5) of the sub-protocol.
UnauthenticatedHTTP2DOSMitigation: Enables HTTP/2 Denial of Service (DoS) mitigations for unauthenticated clients. Kubernetes v1.28.0 through v1.28.2 do not include this feature gate.
UnknownVersionInteroperabilityProxy: Proxy resource requests to the correct peer kube-apiserver when multiple kube-apiservers exist at varied versions. See Mixed version proxy for more information.
UserNamespacesPodSecurityStandards: Enable Pod Security Standards policies relaxation for pods that run with namespaces. You must set the value of this feature gate consistently across all nodes in your cluster, and you must also enable UserNamespacesSupport to use this feature.
UserNamespacesSupport: Enable user namespace support for Pods.
ValidatingAdmissionPolicy: Enable ValidatingAdmissionPolicy support for CEL validations be used in Admission Control.
VolumeAttributesClass: Enable support for VolumeAttributesClasses. See Volume Attributes Classes for more information.
VolumeCapacityPriority: Enable support for prioritizing nodes in different topologies based on available PV capacity.
WatchBookmark: Enable support for watch bookmark events.
WatchCacheInitializationPostStartHook: Enables post-start-hook for watchcache initialization to be part of readyz (with timeout).
WatchFromStorageWithoutResourceVersion: Enables watches without resourceVersion to be served from storage.
WatchList: Enable support for streaming initial state of objects in watch requests.
WindowsHostNetwork: Enables support for joining Windows containers to a hosts' network namespace.
WinDSR: Allows kube-proxy to create DSR loadbalancers for Windows.
WinOverlay: Allows kube-proxy to run in overlay mode for Windows.
ZeroLimitedNominalConcurrencyShares: Allow priority & fairness in the API server to use a zero value for the nominalConcurrencyShares field of the limited section of a priority level.

What's next

The deprecation policy for Kubernetes explains the project's approach to removing features and components.
Since Kubernetes 1.24, new beta APIs are not enabled by default. When enabling a beta feature, you will also need to enable any associated API resources. For example, to enable a particular resource like storage.k8s.io/v1beta1/csistoragecapacities, set --runtime-config=storage.k8s.io/v1beta1/csistoragecapacities. See API Versioning for more details on the command line flags.

Last modified June 20, 2024 at 12:44 PM PST: Sync changest from andygol/k8s-website (36d05bc8a1)

Feature Gates

Overview

Note:

Feature gates for Alpha or Beta features

Feature gates for graduated or deprecated features

Using a feature

Feature stages

Note:

List of feature gates

What's next

Feedback