SageMaker

40 resource kinds. Generated from the operator's CRD schemas.

API Group: aws.konfig.io/v1alpha1

SageMakerAction # CLOUD CONTROL · AWS::SageMaker::Action

SageMakerAction manages AWS::SageMaker::Action through the AWS Cloud Control API.

kubectl get sagemakeractions · example on GitHub

Example

# SageMakerAction — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakeraction
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerAction
metadata:
  name: example-sagemakeraction
  namespace: default
spec:
  actionName: example-sagemakeraction
  actionType: example
  description: Managed by konfig-konector
  metadataProperties:
    generatedBy: example
    repository: example
  properties:
    key: value
  providerRef:
    name: example-sagemakeraction
    region: us-east-1
  source:
    sourceType: example
    sourceUri: example
  status: Unknown
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`actionName`	string	yes	The name of the action.
`actionType`	string	yes	The action type.
`description`	string	—	The description of the action.
`metadataProperties`	object	—	Metadata properties of the tracking entity, trial, or trial component.
`metadataProperties.commitId`	string	—	The commit ID.
`metadataProperties.generatedBy`	string	—	The entity this entity was generated by.
`metadataProperties.projectId`	string	—	The project ID.
`metadataProperties.repository`	string	—	The repository.
`properties`	map<string>	—	A list of properties to add to the action.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`source`	object	yes	The source type, ID, and URI.
`source.sourceId`	string	—	The ID of the source.
`source.sourceType`	string	—	The type of the source.
`source.sourceUri`	string	yes	The URI of the source.
`status`	enum	—	The status of the action. Allowed: `Unknown, InProgress, Completed, Failed, Stopping, Stopped`
`tags`	array<object>	—	A list of tags to apply to the action.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`arn`	string	—	The Amazon Resource Name (ARN) of the action.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	When the action was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	When the action was last modified.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerAlgorithm # CLOUD CONTROL · AWS::SageMaker::Algorithm

SageMakerAlgorithm manages AWS::SageMaker::Algorithm through the AWS Cloud Control API.

kubectl get sagemakeralgorithms · example on GitHub

Example

# SageMakerAlgorithm — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakeralgorithm
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerAlgorithm
metadata:
  name: example-sagemakeralgorithm
  namespace: default
spec:
  algorithmDescription: Managed by konfig-konector
  algorithmName: example-sagemakeralgorithm
  certifyForMarketplace: false
  inferenceSpecification:
    containers:
    - containerHostname: example-sagemakeralgorithm
      framework: example
      frameworkVersion: '1.0'
      image: example
      imageDigest: example
      isCheckpoint: false
      nearestModelName: example-sagemakeralgorithm
    supportedContentTypes:
    - example
    supportedRealtimeInferenceInstanceTypes:
    - '2026-01-01T00:00:00Z'
    supportedResponseMIMETypes:
    - example
    supportedTransformInstanceTypes:
    - example
  providerRef:
    name: example-sagemakeralgorithm
    region: us-east-1
  tags:
  - key: example
    value: example
  trainingSpecification:
    metricDefinitions:
    - name: example-sagemakeralgorithm
      regex: example
    supportedHyperParameters:
    - defaultValue: example
      description: Managed by konfig-konector
      isRequired: false
      isTunable: false
      name: example-sagemakeralgorithm
      type: Integer
    supportedTrainingInstanceTypes:
    - example
    supportedTuningJobObjectiveMetrics:
    - metricName: example-sagemakeralgorithm
      type: Maximize
    supportsDistributedTraining: false
    trainingChannels:
    - description: Managed by konfig-konector
      isRequired: false
      name: example-sagemakeralgorithm
      supportedCompressionTypes:
      - example
      supportedContentTypes:
      - example
      supportedInputModes:
      - example
    trainingImage: example
    trainingImageDigest: example

Spec

Field	Type	Required	Description
`algorithmDescription`	string	—	A description of the algorithm.
`algorithmName`	string	yes	The name of the algorithm.
`certifyForMarketplace`	boolean	—	Whether to certify the algorithm so that it can be listed in AWS Marketplace.
`inferenceSpecification`	object	—	Create-only in AWS: changing it requires replacing the resource.
`inferenceSpecification.containers`	array<object>	yes
`inferenceSpecification.containers[].containerHostname`	string	—
`inferenceSpecification.containers[].environment`	map<string>	—
`inferenceSpecification.containers[].framework`	string	—
`inferenceSpecification.containers[].frameworkVersion`	string	—
`inferenceSpecification.containers[].image`	string	yes
`inferenceSpecification.containers[].imageDigest`	string	—
`inferenceSpecification.containers[].isCheckpoint`	boolean	—
`inferenceSpecification.containers[].modelInput`	object	—	SageMakerAlgorithmModelInput is a nested property type of AWS::SageMaker::Algorithm.
`inferenceSpecification.containers[].modelInput.dataInputConfig`	string	yes
`inferenceSpecification.containers[].nearestModelName`	string	—
`inferenceSpecification.supportedContentTypes`	array<string>	—
`inferenceSpecification.supportedRealtimeInferenceInstanceTypes`	array<string>	—
`inferenceSpecification.supportedResponseMIMETypes`	array<string>	—
`inferenceSpecification.supportedTransformInstanceTypes`	array<string>	—
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes
`trainingSpecification`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`trainingSpecification.metricDefinitions`	array<object>	—
`trainingSpecification.metricDefinitions[].name`	string	yes
`trainingSpecification.metricDefinitions[].regex`	string	yes
`trainingSpecification.supportedHyperParameters`	array<object>	—
`trainingSpecification.supportedHyperParameters[].defaultValue`	string	—
`trainingSpecification.supportedHyperParameters[].description`	string	—
`trainingSpecification.supportedHyperParameters[].isRequired`	boolean	—
`trainingSpecification.supportedHyperParameters[].isTunable`	boolean	—
`trainingSpecification.supportedHyperParameters[].name`	string	yes
`trainingSpecification.supportedHyperParameters[].range`	object	—	SageMakerAlgorithmParameterRange is a nested property type of AWS::SageMaker::Algorithm.
`trainingSpecification.supportedHyperParameters[].range.categoricalParameterRangeSpecification`	object	—	SageMakerAlgorithmCategoricalParameterRangeSpecification is a nested property type of AWS::SageMaker::Algorithm.
`trainingSpecification.supportedHyperParameters[].range.categoricalParameterRangeSpecification.values`	array<string>	yes
`trainingSpecification.supportedHyperParameters[].range.continuousParameterRangeSpecification`	object	—	SageMakerAlgorithmContinuousParameterRangeSpecification is a nested property type of AWS::SageMaker::Algorithm.
`trainingSpecification.supportedHyperParameters[].range.continuousParameterRangeSpecification.maxValue`	string	yes
`trainingSpecification.supportedHyperParameters[].range.continuousParameterRangeSpecification.minValue`	string	yes
`trainingSpecification.supportedHyperParameters[].range.integerParameterRangeSpecification`	object	—	SageMakerAlgorithmIntegerParameterRangeSpecification is a nested property type of AWS::SageMaker::Algorithm.
`trainingSpecification.supportedHyperParameters[].range.integerParameterRangeSpecification.maxValue`	string	yes
`trainingSpecification.supportedHyperParameters[].range.integerParameterRangeSpecification.minValue`	string	yes
`trainingSpecification.supportedHyperParameters[].type`	enum	yes	Allowed: `Integer, Continuous, Categorical, FreeText`
`trainingSpecification.supportedTrainingInstanceTypes`	array<string>	yes
`trainingSpecification.supportedTuningJobObjectiveMetrics`	array<object>	—
`trainingSpecification.supportedTuningJobObjectiveMetrics[].metricName`	string	yes
`trainingSpecification.supportedTuningJobObjectiveMetrics[].type`	enum	yes	Allowed: `Maximize, Minimize`
`trainingSpecification.supportsDistributedTraining`	boolean	—
`trainingSpecification.trainingChannels`	array<object>	yes
`trainingSpecification.trainingChannels[].description`	string	—
`trainingSpecification.trainingChannels[].isRequired`	boolean	—
`trainingSpecification.trainingChannels[].name`	string	yes
`trainingSpecification.trainingChannels[].supportedCompressionTypes`	array<string>	—
`trainingSpecification.trainingChannels[].supportedContentTypes`	array<string>	yes
`trainingSpecification.trainingChannels[].supportedInputModes`	array<string>	yes
`trainingSpecification.trainingImage`	string	yes
`trainingSpecification.trainingImageDigest`	string	—

Status

Field	Type	Required	Description
`algorithmArn`	string	—	The Amazon Resource Name (ARN) of the algorithm.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	A timestamp specifying when the algorithm was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerApp # CLOUD CONTROL · AWS::SageMaker::App

SageMakerApp manages AWS::SageMaker::App through the AWS Cloud Control API.

kubectl get sagemakerapps · example on GitHub

Example

# SageMakerApp — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerapp
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerApp
metadata:
  name: example-sagemakerapp
  namespace: default
spec:
  appName: example-sagemakerapp
  appType: JupyterServer
  domainId: example-id
  providerRef:
    name: example-sagemakerapp
    region: us-east-1
  recoveryMode: false
  resourceSpec:
    instanceType: example
    lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
    sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
    sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
  tags:
  - key: example
    value: example
  userProfileName: example-sagemakerapp

Spec

Field	Type	Required	Description
`appName`	string	yes	The name of the app.
`appType`	enum	yes	The type of app. Allowed: `JupyterServer, KernelGateway, RStudioServerPro, RSessionGateway, Canvas`
`domainId`	string	yes	The domain ID.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`recoveryMode`	boolean	—	Indicates whether the application is launched in recovery mode.
`resourceSpec`	object	—	The instance type and the Amazon Resource Name (ARN) of the SageMaker image created on the instance.
`resourceSpec.instanceType`	string	—	The instance type that the image version runs on.
`resourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`resourceSpec.sageMakerImageArn`	string	—	The ARN of the SageMaker image that the image version belongs to.
`resourceSpec.sageMakerImageVersionArn`	string	—	The ARN of the image version created on the instance.
`tags`	array<object>	—	A list of tags to apply to the app.
`tags[].key`	string	yes
`tags[].value`	string	yes
`userProfileName`	string	yes	The user profile name.

Status

Field	Type	Required	Description
`appArn`	string	—	The Amazon Resource Name (ARN) of the app.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`builtInLifecycleConfigArn`	string	—	The lifecycle configuration that runs before the default lifecycle configuration.
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerAppImageConfig # CLOUD CONTROL · AWS::SageMaker::AppImageConfig

SageMakerAppImageConfig manages AWS::SageMaker::AppImageConfig through the AWS Cloud Control API.

kubectl get sagemakerappimageconfigs · example on GitHub

Example

# SageMakerAppImageConfig — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerappimageconfig
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerAppImageConfig
metadata:
  name: example-sagemakerappimageconfig
  namespace: default
spec:
  appImageConfigName: example-sagemakerappimageconfig
  codeEditorAppImageConfig:
    containerConfig:
      containerArguments:
      - example
      containerEntrypoint:
      - example
      containerEnvironmentVariables:
      - key: example
        value: example
  jupyterLabAppImageConfig:
    containerConfig:
      containerArguments:
      - example
      containerEntrypoint:
      - example
      containerEnvironmentVariables:
      - key: example
        value: example
  kernelGatewayImageConfig:
    fileSystemConfig:
      defaultGid: 1
      defaultUid: 1
      mountPath: example
    kernelSpecs:
    - displayName: example-sagemakerappimageconfig
      name: example-sagemakerappimageconfig
  providerRef:
    name: example-sagemakerappimageconfig
    region: us-east-1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`appImageConfigName`	string	yes	The Name of the AppImageConfig.
`codeEditorAppImageConfig`	object	—	The CodeEditorAppImageConfig.
`codeEditorAppImageConfig.containerConfig`	object	—	The container configuration for a SageMaker image.
`codeEditorAppImageConfig.containerConfig.containerArguments`	array<string>	—	A list of arguments to apply to the container.
`codeEditorAppImageConfig.containerConfig.containerEntrypoint`	array<string>	—	The custom entry point to use on container.
`codeEditorAppImageConfig.containerConfig.containerEnvironmentVariables`	array<object>	—	A list of variables to apply to the custom container.
`codeEditorAppImageConfig.containerConfig.containerEnvironmentVariables[].key`	string	yes
`codeEditorAppImageConfig.containerConfig.containerEnvironmentVariables[].value`	string	yes
`jupyterLabAppImageConfig`	object	—	The JupyterLabAppImageConfig.
`jupyterLabAppImageConfig.containerConfig`	object	—	The container configuration for a SageMaker image.
`jupyterLabAppImageConfig.containerConfig.containerArguments`	array<string>	—	A list of arguments to apply to the container.
`jupyterLabAppImageConfig.containerConfig.containerEntrypoint`	array<string>	—	The custom entry point to use on container.
`jupyterLabAppImageConfig.containerConfig.containerEnvironmentVariables`	array<object>	—	A list of variables to apply to the custom container.
`jupyterLabAppImageConfig.containerConfig.containerEnvironmentVariables[].key`	string	yes
`jupyterLabAppImageConfig.containerConfig.containerEnvironmentVariables[].value`	string	yes
`kernelGatewayImageConfig`	object	—	The KernelGatewayImageConfig.
`kernelGatewayImageConfig.fileSystemConfig`	object	—	The Amazon Elastic File System (EFS) storage configuration for a SageMaker image.
`kernelGatewayImageConfig.fileSystemConfig.defaultGid`	integer	—	The default POSIX group ID (GID).
`kernelGatewayImageConfig.fileSystemConfig.defaultUid`	integer	—	The default POSIX user ID (UID).
`kernelGatewayImageConfig.fileSystemConfig.mountPath`	string	—	The path within the image to mount the user's EFS home directory.
`kernelGatewayImageConfig.kernelSpecs`	array<object>	yes	The specification of the Jupyter kernels in the image.
`kernelGatewayImageConfig.kernelSpecs[].displayName`	string	—	The display name of the kernel.
`kernelGatewayImageConfig.kernelSpecs[].name`	string	yes	The name of the kernel.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	A list of tags to apply to the AppImageConfig.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`appImageConfigArn`	string	—	The Amazon Resource Name (ARN) of the AppImageConfig.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerArtifact # CLOUD CONTROL · AWS::SageMaker::Artifact

SageMakerArtifact manages AWS::SageMaker::Artifact through the AWS Cloud Control API.

kubectl get sagemakerartifacts · example on GitHub

Example

# SageMakerArtifact — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerartifact
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerArtifact
metadata:
  name: example-sagemakerartifact
  namespace: default
spec:
  artifactName: example-sagemakerartifact
  artifactType: example
  metadataProperties:
    generatedBy: example
    repository: example
  properties:
    key: value
  providerRef:
    name: example-sagemakerartifact
    region: us-east-1
  source:
    sourceTypes:
    - sourceIdType: MD5Hash
      value: example
    sourceUri: example
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`artifactName`	string	—	The name of the artifact.
`artifactType`	string	yes	The artifact type.
`metadataProperties`	object	—	Metadata properties of the tracking entity, trial, or trial component.
`metadataProperties.commitId`	string	—	The commit ID.
`metadataProperties.generatedBy`	string	—	The entity this entity was generated by.
`metadataProperties.projectId`	string	—	The project ID.
`metadataProperties.repository`	string	—	The repository.
`properties`	map<string>	—	A list of properties to add to the artifact.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`source`	object	yes	The source of the artifact.
`source.sourceTypes`	array<object>	—	A list of source types.
`source.sourceTypes[].sourceIdType`	enum	yes	The type of ID. Allowed: `MD5Hash, S3ETag, S3Version, Custom`
`source.sourceTypes[].value`	string	yes	The ID.
`source.sourceUri`	string	yes	The URI of the source.
`tags`	array<object>	—	A list of tags to apply to the artifact.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`arn`	string	—	The Amazon Resource Name (ARN) of the artifact.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	When the artifact was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	When the artifact was last modified.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerCluster # CLOUD CONTROL · AWS::SageMaker::Cluster

SageMakerCluster manages AWS::SageMaker::Cluster through the AWS Cloud Control API.

kubectl get sagemakerclusters · example on GitHub

Example

# SageMakerCluster — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakercluster
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerCluster
metadata:
  name: example-sagemakercluster
  namespace: default
spec:
  autoScaling:
    autoScalerType: Karpenter
    mode: Enable
  clusterName: example-sagemakercluster
  clusterRole: example
  instanceGroups:
  - autoPatchConfig:
      patchingStrategy: WhenIdle
    currentCount: 1
    executionRole: example-xxxxxxxxxxxx
    instanceCount: 1
    instanceGroupName: example-sagemakercluster
    instanceRequirements:
      instanceTypes:
      - example
    instanceType: example
    kubernetesConfig:
      taints:
      - effect: NoSchedule
        key: example
        value: example
    lifeCycleConfig:
      onCreate: example
      onInitComplete: example
      sourceS3Uri: example
    minInstanceCount: 1
    networkInterface:
      interfaceType: efa
    onStartDeepHealthChecks:
    - example
    overrideVpcConfig:
      securityGroupIds:
      - example
      subnets:
      - example
    scheduledUpdateConfig:
      scheduleExpression: example
    slurmConfig:
      nodeType: Controller
      partitionNames:
      - example
    threadsPerCore: 1
    trainingPlanArn: arn:aws:iam::123456789012:role/example-role
  nodeProvisioningMode: Continuous
  nodeRecovery: Automatic
  providerRef:
    name: example-sagemakercluster
    region: us-east-1
  restrictedInstanceGroups:
  - currentCount: 1
    executionRole: example-xxxxxxxxxxxx
    instanceCount: 1
    instanceGroupName: example-sagemakercluster
    instanceType: example
    onStartDeepHealthChecks:
    - example
    overrideVpcConfig:
      securityGroupIds:
      - example
      subnets:
      - example
    threadsPerCore: 1
    trainingPlanArn: arn:aws:iam::123456789012:role/example-role
  restrictedInstanceGroupsConfig:
    sharedEnvironmentConfig:
      fSxLustreConfig:
        perUnitStorageThroughput: 125
        sizeInGiB: 1200
      fSxLustreDeletionPolicy: Keep
  tags:
  - key: example
    value: example
  tieredStorageConfig:
    instanceMemoryAllocationPercentage: 1
    mode: Enable
  vpcConfig:
    securityGroupIds:
    - example
    subnets:
    - example

Spec

Field	Type	Required	Description
`autoScaling`	object	—	SageMakerClusterClusterAutoScalingConfig is a nested property type of AWS::SageMaker::Cluster.
`autoScaling.autoScalerType`	enum	—	The type of auto-scaler to use Allowed: `Karpenter`
`autoScaling.mode`	enum	yes	The auto-scaling mode for the cluster Allowed: `Enable, Disable`
`clusterName`	string	—	The name of the HyperPod Cluster.
`clusterRole`	string	—	The cluster role for the autoscaler to assume.
`instanceGroups`	array<object>	—
`instanceGroups[].autoPatchConfig`	object	—	SageMakerClusterAutoPatchConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].autoPatchConfig.deploymentConfig`	object	—	SageMakerClusterDeploymentConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].autoPatchConfig.deploymentConfig.autoRollbackConfiguration`	array<object>	—
`instanceGroups[].autoPatchConfig.deploymentConfig.autoRollbackConfiguration[].alarmName`	string	yes	The name of the alarm.
`instanceGroups[].autoPatchConfig.deploymentConfig.rollingUpdatePolicy`	object	—	SageMakerClusterRollingUpdatePolicy is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].autoPatchConfig.deploymentConfig.rollingUpdatePolicy.maximumBatchSize`	object	yes	SageMakerClusterCapacitySizeConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].autoPatchConfig.deploymentConfig.rollingUpdatePolicy.maximumBatchSize.type`	string	yes	Specifies whether SageMaker should process the update by amount or percentage of instances.
`instanceGroups[].autoPatchConfig.deploymentConfig.rollingUpdatePolicy.maximumBatchSize.value`	integer	yes	Specifies the amount or percentage of instances SageMaker updates at a time.
`instanceGroups[].autoPatchConfig.deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize`	object	—	SageMakerClusterCapacitySizeConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].autoPatchConfig.deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize.type`	string	yes	Specifies whether SageMaker should process the update by amount or percentage of instances.
`instanceGroups[].autoPatchConfig.deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize.value`	integer	yes	Specifies the amount or percentage of instances SageMaker updates at a time.
`instanceGroups[].autoPatchConfig.deploymentConfig.waitIntervalInSeconds`	integer	—	The duration in seconds that SageMaker waits before updating more instances in the cluster.
`instanceGroups[].autoPatchConfig.patchSchedule`	object	—	SageMakerClusterPatchSchedule is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].autoPatchConfig.patchSchedule.nextPatchDate`	string	—	The date and time of the next scheduled patch, set by the system when a patch AMI is detected.
`instanceGroups[].autoPatchConfig.patchingStrategy`	enum	yes	The patching strategy that determines when and how instances are patched. Allowed: `WhenIdle, WhenAllIdle`
`instanceGroups[].capacityRequirements`	object	—	SageMakerClusterClusterCapacityRequirements is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].capacityRequirements.onDemand`	object	—
`instanceGroups[].capacityRequirements.spot`	object	—
`instanceGroups[].currentCount`	integer	—	The number of instances that are currently in the instance group of a SageMaker HyperPod cluster.
`instanceGroups[].executionRole`	string	yes
`instanceGroups[].imageId`	string	—
`instanceGroups[].instanceCount`	integer	yes	The number of instances you specified to add to the instance group of a SageMaker HyperPod cluster.
`instanceGroups[].instanceGroupName`	string	yes
`instanceGroups[].instanceRequirements`	object	—	SageMakerClusterInstanceRequirements is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].instanceRequirements.instanceTypes`	array<string>	yes	A list of instance types that can be used for this instance group.
`instanceGroups[].instanceStorageConfigs`	array<object>	—
`instanceGroups[].instanceType`	string	—
`instanceGroups[].kubernetesConfig`	object	—	SageMakerClusterClusterKubernetesConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].kubernetesConfig.labels`	map<string>	—
`instanceGroups[].kubernetesConfig.taints`	array<object>	—
`instanceGroups[].kubernetesConfig.taints[].effect`	enum	yes	The effect of the taint. Allowed: `NoSchedule, PreferNoSchedule, NoExecute`
`instanceGroups[].kubernetesConfig.taints[].key`	string	yes	The key of the taint.
`instanceGroups[].kubernetesConfig.taints[].value`	string	—	The value of the taint.
`instanceGroups[].lifeCycleConfig`	object	—	SageMakerClusterClusterLifeCycleConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].lifeCycleConfig.onCreate`	string	—	The file name of the entrypoint script of lifecycle scripts under SourceS3Uri.
`instanceGroups[].lifeCycleConfig.onInitComplete`	string	—	The file name of the extension script under SourceS3Uri.
`instanceGroups[].lifeCycleConfig.sourceS3Uri`	string	—	An Amazon S3 bucket path where your lifecycle scripts are stored.
`instanceGroups[].minInstanceCount`	integer	—	The minimum number of instances required for the instance group to be InService.
`instanceGroups[].networkInterface`	object	—	SageMakerClusterClusterNetworkInterface is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].networkInterface.interfaceType`	enum	yes	The type of network interface. Allowed: `efa, efa-only`
`instanceGroups[].onStartDeepHealthChecks`	array<string>	—
`instanceGroups[].overrideVpcConfig`	object	—	SageMakerClusterVpcConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].overrideVpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form sg-xxxxxxxx.
`instanceGroups[].overrideVpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect your training job or model.
`instanceGroups[].scheduledUpdateConfig`	object	—	SageMakerClusterScheduledUpdateConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig`	object	—	SageMakerClusterDeploymentConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.autoRollbackConfiguration`	array<object>	—
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.autoRollbackConfiguration[].alarmName`	string	yes	The name of the alarm.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.rollingUpdatePolicy`	object	—	SageMakerClusterRollingUpdatePolicy is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.rollingUpdatePolicy.maximumBatchSize`	object	yes	SageMakerClusterCapacitySizeConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.rollingUpdatePolicy.maximumBatchSize.type`	string	yes	Specifies whether SageMaker should process the update by amount or percentage of instances.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.rollingUpdatePolicy.maximumBatchSize.value`	integer	yes	Specifies the amount or percentage of instances SageMaker updates at a time.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize`	object	—	SageMakerClusterCapacitySizeConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize.type`	string	yes	Specifies whether SageMaker should process the update by amount or percentage of instances.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize.value`	integer	yes	Specifies the amount or percentage of instances SageMaker updates at a time.
`instanceGroups[].scheduledUpdateConfig.deploymentConfig.waitIntervalInSeconds`	integer	—	The duration in seconds that SageMaker waits before updating more instances in the cluster.
`instanceGroups[].scheduledUpdateConfig.scheduleExpression`	string	yes	A cron expression that specifies the schedule that SageMaker follows when updating the AMI.
`instanceGroups[].slurmConfig`	object	—	SageMakerClusterClusterSlurmConfig is a nested property type of AWS::SageMaker::Cluster.
`instanceGroups[].slurmConfig.nodeType`	enum	yes	The type of Slurm node for this instance group. Allowed: `Controller, Login, Compute`
`instanceGroups[].slurmConfig.partitionNames`	array<string>	—	The Slurm partitions that this instance group belongs to.
`instanceGroups[].threadsPerCore`	integer	—	The number you specified to TreadsPerCore in CreateCluster for enabling or disabling multithreading.
`instanceGroups[].trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for this cluster instance group.
`nodeProvisioningMode`	enum	—	Determines the scaling strategy for the SageMaker HyperPod cluster. Allowed: `Continuous`
`nodeRecovery`	enum	—	If node auto-recovery is set to true, faulty nodes will be replaced or rebooted when a failure is detected. Allowed: `Automatic, None`
`orchestrator`	object	—
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`restrictedInstanceGroups`	array<object>	—
`restrictedInstanceGroups[].currentCount`	integer	—	The number of instances that are currently in the restricted instance group of a SageMaker HyperPod cluster.
`restrictedInstanceGroups[].environmentConfig`	object	—	SageMakerClusterEnvironmentConfig is a nested property type of AWS::SageMaker::Cluster.
`restrictedInstanceGroups[].environmentConfig.fSxLustreConfig`	object	—	SageMakerClusterFSxLustreConfig is a nested property type of AWS::SageMaker::Cluster.
`restrictedInstanceGroups[].environmentConfig.fSxLustreConfig.perUnitStorageThroughput`	integer	yes	The throughput capacity of the FSx for Lustre file system, measured in MB/s per TiB of storage.
`restrictedInstanceGroups[].environmentConfig.fSxLustreConfig.sizeInGiB`	integer	yes	The storage capacity of the FSx for Lustre file system, specified in gibibytes (GiB).
`restrictedInstanceGroups[].executionRole`	string	yes
`restrictedInstanceGroups[].instanceCount`	integer	yes	The number of instances you specified to add to the restricted instance group of a SageMaker HyperPod cluster.
`restrictedInstanceGroups[].instanceGroupName`	string	yes
`restrictedInstanceGroups[].instanceStorageConfigs`	array<object>	—
`restrictedInstanceGroups[].instanceType`	string	yes
`restrictedInstanceGroups[].onStartDeepHealthChecks`	array<string>	—
`restrictedInstanceGroups[].overrideVpcConfig`	object	—	SageMakerClusterVpcConfig is a nested property type of AWS::SageMaker::Cluster.
`restrictedInstanceGroups[].overrideVpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form sg-xxxxxxxx.
`restrictedInstanceGroups[].overrideVpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect your training job or model.
`restrictedInstanceGroups[].threadsPerCore`	integer	—	The number you specified to TreadsPerCore in CreateCluster for enabling or disabling multithreading.
`restrictedInstanceGroups[].trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for this cluster restricted instance group.
`restrictedInstanceGroupsConfig`	object	—	SageMakerClusterRestrictedInstanceGroupsConfig is a nested property type of AWS::SageMaker::Cluster.
`restrictedInstanceGroupsConfig.sharedEnvironmentConfig`	object	yes	SageMakerClusterSharedEnvironmentConfig is a nested property type of AWS::SageMaker::Cluster.
`restrictedInstanceGroupsConfig.sharedEnvironmentConfig.fSxLustreConfig`	object	—	SageMakerClusterFSxLustreConfig is a nested property type of AWS::SageMaker::Cluster.
`restrictedInstanceGroupsConfig.sharedEnvironmentConfig.fSxLustreConfig.perUnitStorageThroughput`	integer	yes	The throughput capacity of the FSx for Lustre file system, measured in MB/s per TiB of storage.
`restrictedInstanceGroupsConfig.sharedEnvironmentConfig.fSxLustreConfig.sizeInGiB`	integer	yes	The storage capacity of the FSx for Lustre file system, specified in gibibytes (GiB).
`restrictedInstanceGroupsConfig.sharedEnvironmentConfig.fSxLustreDeletionPolicy`	enum	yes	The deletion policy for the shared FSx Lustre file system. Allowed: `Keep, DeleteIfNotUsed`
`tags`	array<object>	—	Custom tags for managing the SageMaker HyperPod cluster as an AWS resource.
`tags[].key`	string	yes
`tags[].value`	string	yes
`tieredStorageConfig`	object	—	SageMakerClusterTieredStorageConfig is a nested property type of AWS::SageMaker::Cluster.
`tieredStorageConfig.instanceMemoryAllocationPercentage`	integer	—	The percentage of instance memory to allocate for tiered storage.
`tieredStorageConfig.mode`	enum	yes	The mode of tiered storage. Allowed: `Enable, Disable`
`vpcConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`vpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form sg-xxxxxxxx.
`vpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect your training job or model.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`clusterArn`	string	—	The Amazon Resource Name (ARN) of the HyperPod Cluster.
`clusterStatus`	enum	—	The status of the HyperPod Cluster. Allowed: `Creating, Deleting, Failed, InService, RollingBack, SystemUpdating, Updating`
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The time at which the HyperPod cluster was created.
`failureMessage`	string	—	The failure message of the HyperPod Cluster.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerContext # CLOUD CONTROL · AWS::SageMaker::Context

SageMakerContext manages AWS::SageMaker::Context through the AWS Cloud Control API.

kubectl get sagemakercontexts · example on GitHub

Example

# SageMakerContext — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakercontext
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerContext
metadata:
  name: example-sagemakercontext
  namespace: default
spec:
  contextName: example-sagemakercontext
  contextType: example
  description: Managed by konfig-konector
  properties:
    key: value
  providerRef:
    name: example-sagemakercontext
    region: us-east-1
  source:
    sourceType: example
    sourceUri: example
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`contextName`	string	yes	The name of the context.
`contextType`	string	yes	The context type.
`description`	string	—	The description of the context.
`properties`	map<string>	—	A list of properties to add to the context.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`source`	object	yes	The source type, ID, and URI.
`source.sourceId`	string	—	The ID of the source.
`source.sourceType`	string	—	The type of the source.
`source.sourceUri`	string	yes	The URI of the source.
`tags`	array<object>	—	A list of tags to apply to the context.
`tags[].key`	string	yes	The tag key.
`tags[].value`	string	yes	The tag value.

Status

Field	Type	Required	Description
`arn`	string	—	The Amazon Resource Name (ARN) of the context.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	When the context was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	When the context was last modified.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerDataQualityJobDefinition # CLOUD CONTROL · AWS::SageMaker::DataQualityJobDefinition

SageMakerDataQualityJobDefinition manages AWS::SageMaker::DataQualityJobDefinition through the AWS Cloud Control API.

kubectl get sagemakerdataqualityjobdefinitions · example on GitHub

Example

# SageMakerDataQualityJobDefinition — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerdataqualityjobdefinition
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerDataQualityJobDefinition
metadata:
  name: example-sagemakerdataqualityjobdefinition
  namespace: default
spec:
  dataQualityAppSpecification:
    containerArguments:
    - example
    containerEntrypoint:
    - example
    environment:
      key: value
    imageUri: example
    postAnalyticsProcessorSourceUri: example
    recordPreprocessorSourceUri: example
  dataQualityBaselineConfig:
    baseliningJobName: example-sagemakerdataqualityjobdefinition
    constraintsResource:
      s3Uri: example
    statisticsResource:
      s3Uri: example
  dataQualityJobInput:
    batchTransformInput:
      dataCapturedDestinationS3Uri: example
      datasetFormat:
        parquet: false
      excludeFeaturesAttribute: example
      localPath: example
      s3DataDistributionType: FullyReplicated
      s3InputMode: Pipe
    endpointInput:
      endpointName: example-sagemakerdataqualityjobdefinition
      excludeFeaturesAttribute: example
      localPath: example
      s3DataDistributionType: FullyReplicated
      s3InputMode: Pipe
  dataQualityJobOutputConfig:
    kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    monitoringOutputs:
    - s3Output:
        localPath: example
        s3UploadMode: Continuous
        s3Uri: example
  endpointName: example-sagemakerdataqualityjobdefinition
  jobDefinitionName: example-sagemakerdataqualityjobdefinition
  jobResources:
    clusterConfig:
      instanceCount: 1
      instanceType: example
      volumeKmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
      volumeSizeInGB: 1
  networkConfig:
    enableInterContainerTrafficEncryption: true
    enableNetworkIsolation: true
    vpcConfig:
      securityGroupIds:
      - example
      subnets:
      - example
  providerRef:
    name: example-sagemakerdataqualityjobdefinition
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  stoppingCondition:
    maxRuntimeInSeconds: 1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`dataQualityAppSpecification`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`dataQualityAppSpecification.containerArguments`	array<string>	—	An array of arguments for the container used to run the monitoring job.
`dataQualityAppSpecification.containerEntrypoint`	array<string>	—	Specifies the entrypoint for a container used to run the monitoring job.
`dataQualityAppSpecification.environment`	map<string>	—	Sets the environment variables in the Docker container
`dataQualityAppSpecification.imageUri`	string	yes	The container image to be run by the monitoring job.
`dataQualityAppSpecification.postAnalyticsProcessorSourceUri`	string	—	An Amazon S3 URI to a script that is called after analysis has been performed.
`dataQualityAppSpecification.recordPreprocessorSourceUri`	string	—	An Amazon S3 URI to a script that is called per row prior to running analysis.
`dataQualityBaselineConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`dataQualityBaselineConfig.baseliningJobName`	string	—
`dataQualityBaselineConfig.constraintsResource`	object	—	SageMakerDataQualityJobDefinitionConstraintsResource is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`dataQualityBaselineConfig.constraintsResource.s3Uri`	string	—	The Amazon S3 URI for baseline constraint file in Amazon S3 that the current monitoring job should validated against.
`dataQualityBaselineConfig.statisticsResource`	object	—	SageMakerDataQualityJobDefinitionStatisticsResource is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`dataQualityBaselineConfig.statisticsResource.s3Uri`	string	—	The Amazon S3 URI for the baseline statistics file in Amazon S3 that the current monitoring job should be validated against.
`dataQualityJobInput`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`dataQualityJobInput.batchTransformInput`	object	—	SageMakerDataQualityJobDefinitionBatchTransformInput is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`dataQualityJobInput.batchTransformInput.dataCapturedDestinationS3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Batch Transform Job captures data.
`dataQualityJobInput.batchTransformInput.datasetFormat`	object	yes	SageMakerDataQualityJobDefinitionDatasetFormat is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`dataQualityJobInput.batchTransformInput.datasetFormat.csv`	object	—	SageMakerDataQualityJobDefinitionCsv is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`dataQualityJobInput.batchTransformInput.datasetFormat.csv.header`	boolean	—	A boolean flag indicating if given CSV has header
`dataQualityJobInput.batchTransformInput.datasetFormat.json`	object	—	SageMakerDataQualityJobDefinitionJson is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`dataQualityJobInput.batchTransformInput.datasetFormat.json.line`	boolean	—	A boolean flag indicating if it is JSON line format
`dataQualityJobInput.batchTransformInput.datasetFormat.parquet`	boolean	—
`dataQualityJobInput.batchTransformInput.excludeFeaturesAttribute`	string	—	Indexes or names of the features to be excluded from analysis
`dataQualityJobInput.batchTransformInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`dataQualityJobInput.batchTransformInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`dataQualityJobInput.batchTransformInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`dataQualityJobInput.endpointInput`	object	—	SageMakerDataQualityJobDefinitionEndpointInput is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`dataQualityJobInput.endpointInput.endpointName`	string	yes
`dataQualityJobInput.endpointInput.excludeFeaturesAttribute`	string	—	Indexes or names of the features to be excluded from analysis
`dataQualityJobInput.endpointInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`dataQualityJobInput.endpointInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`dataQualityJobInput.endpointInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`dataQualityJobOutputConfig`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`dataQualityJobOutputConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt the model artifacts at rest using Amazon S3 server-side encryption.
`dataQualityJobOutputConfig.monitoringOutputs`	array<object>	yes	Monitoring outputs for monitoring jobs.
`dataQualityJobOutputConfig.monitoringOutputs[].s3Output`	object	yes	SageMakerDataQualityJobDefinitionS3Output is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`dataQualityJobOutputConfig.monitoringOutputs[].s3Output.localPath`	string	yes	The local path to the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`dataQualityJobOutputConfig.monitoringOutputs[].s3Output.s3UploadMode`	enum	—	Whether to upload the results of the monitoring job continuously or after the job completes. Allowed: `Continuous, EndOfJob`
`dataQualityJobOutputConfig.monitoringOutputs[].s3Output.s3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`endpointName`	string	—	Create-only in AWS: changing it requires replacing the resource.
`jobDefinitionName`	string	—	Create-only in AWS: changing it requires replacing the resource.
`jobResources`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`jobResources.clusterConfig`	object	yes	SageMakerDataQualityJobDefinitionClusterConfig is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`jobResources.clusterConfig.instanceCount`	integer	yes	The number of ML compute instances to use in the model monitoring job.
`jobResources.clusterConfig.instanceType`	string	yes	The ML compute instance type for the processing job.
`jobResources.clusterConfig.volumeKmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt data on the storage volume attached to the ML compute instance(s) that run the model monitoring job.
`jobResources.clusterConfig.volumeSizeInGB`	integer	yes	The size of the ML storage volume, in gigabytes, that you want to provision.
`networkConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`networkConfig.enableInterContainerTrafficEncryption`	boolean	—	Whether to encrypt all communications between distributed processing jobs.
`networkConfig.enableNetworkIsolation`	boolean	—	Whether to allow inbound and outbound network calls to and from the containers used for the processing job.
`networkConfig.vpcConfig`	object	—	SageMakerDataQualityJobDefinitionVpcConfig is a nested property type of AWS::SageMaker::DataQualityJobDefinition.
`networkConfig.vpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form sg-xxxxxxxx.
`networkConfig.vpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect to your monitoring jobs.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker can assume to perform tasks on your behalf.
`stoppingCondition`	object	—	Create-only in AWS: changing it requires replacing the resource.
`stoppingCondition.maxRuntimeInSeconds`	integer	yes	The maximum runtime allowed in seconds.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The time at which the job definition was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`jobDefinitionArn`	string	—	The Amazon Resource Name (ARN) of job definition.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerDevice # CLOUD CONTROL · AWS::SageMaker::Device

SageMakerDevice manages AWS::SageMaker::Device through the AWS Cloud Control API.

kubectl get sagemakerdevices · example on GitHub

Example

# SageMakerDevice — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerdevice
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerDevice
metadata:
  name: example-sagemakerdevice
  namespace: default
spec:
  device:
    description: Managed by konfig-konector
    deviceName: example-sagemakerdevice
    iotThingName: example-sagemakerdevice
  deviceFleetName: example-sagemakerdevice
  providerRef:
    name: example-sagemakerdevice
    region: us-east-1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`device`	object	—	The Edge Device you want to register against a device fleet
`device.description`	string	—	Description of the device
`device.deviceName`	string	yes	The name of the device
`device.iotThingName`	string	—	AWS Internet of Things (IoT) object name.
`deviceFleetName`	string	yes	The name of the edge device fleet
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	Associate tags with the resource
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerDeviceFleet # CLOUD CONTROL · AWS::SageMaker::DeviceFleet

SageMakerDeviceFleet manages AWS::SageMaker::DeviceFleet through the AWS Cloud Control API.

kubectl get sagemakerdevicefleets · example on GitHub

Example

# SageMakerDeviceFleet — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerdevicefleet
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerDeviceFleet
metadata:
  name: example-sagemakerdevicefleet
  namespace: default
spec:
  description: Managed by konfig-konector
  deviceFleetName: example-sagemakerdevicefleet
  outputConfig:
    kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    s3OutputLocation: example
  providerRef:
    name: example-sagemakerdevicefleet
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`description`	string	—	Description for the edge device fleet
`deviceFleetName`	string	yes	The name of the edge device fleet Create-only in AWS: changing it requires replacing the resource.
`outputConfig`	object	yes	S3 bucket and an ecryption key id (if available) to store outputs for the fleet
`outputConfig.kmsKeyId`	string	—	The KMS key id used for encryption on the S3 bucket
`outputConfig.s3OutputLocation`	string	yes	The Amazon Simple Storage (S3) bucket URI
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	Role associated with the device fleet
`tags`	array<object>	—	Associate tags with the resource
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerDomain # CLOUD CONTROL · AWS::SageMaker::Domain

SageMakerDomain manages AWS::SageMaker::Domain through the AWS Cloud Control API.

kubectl get sagemakerdomains · example on GitHub

Example

# SageMakerDomain — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerdomain
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerDomain
metadata:
  name: example-sagemakerdomain
  namespace: default
spec:
  appNetworkAccessType: PublicInternetOnly
  appSecurityGroupManagement: Service
  authMode: SSO
  defaultSpaceSettings:
    customPosixUserConfig:
      gid: 1001
      uid: 10000
    executionRole: example-xxxxxxxxxxxx
    jupyterLabAppSettings:
      builtInLifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
      codeRepositories:
      - repositoryUrl: example
      customImages:
      - appImageConfigName: example-sagemakerdomain
        imageName: example-sagemakerdomain
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    jupyterServerAppSettings:
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    kernelGatewayAppSettings:
      customImages:
      - appImageConfigName: example-sagemakerdomain
        imageName: example-sagemakerdomain
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    securityGroups:
    - example
    spaceStorageSettings:
      defaultEbsStorageSettings:
        defaultEbsVolumeSizeInGb: 5
        maximumEbsVolumeSizeInGb: 5
  defaultUserSettings:
    autoMountHomeEFS: Enabled
    codeEditorAppSettings:
      builtInLifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
      customImages:
      - appImageConfigName: example-sagemakerdomain
        imageName: example-sagemakerdomain
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    customPosixUserConfig:
      gid: 1001
      uid: 10000
    defaultLandingUri: example
    executionRole: example-xxxxxxxxxxxx
    jupyterLabAppSettings:
      builtInLifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
      codeRepositories:
      - repositoryUrl: example
      customImages:
      - appImageConfigName: example-sagemakerdomain
        imageName: example-sagemakerdomain
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    jupyterServerAppSettings:
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    kernelGatewayAppSettings:
      customImages:
      - appImageConfigName: example-sagemakerdomain
        imageName: example-sagemakerdomain
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    rSessionAppSettings:
      customImages:
      - appImageConfigName: example-sagemakerdomain
        imageName: example-sagemakerdomain
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
    rStudioServerProAppSettings:
      accessStatus: ENABLED
      userGroup: R_STUDIO_ADMIN
    securityGroups:
    - example
    sharingSettings:
      notebookOutputOption: Allowed
      s3KmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
      s3OutputPath: example
    spaceStorageSettings:
      defaultEbsStorageSettings:
        defaultEbsVolumeSizeInGb: 5
        maximumEbsVolumeSizeInGb: 5
    studioWebPortal: ENABLED
    studioWebPortalSettings:
      hiddenAppTypes:
      - example
      hiddenInstanceTypes:
      - example
      hiddenMlTools:
      - example
      hiddenSageMakerImageVersionAliases:
      - sageMakerImageName: sagemaker_distribution
        versionAliases:
        - example
  domainName: example-sagemakerdomain
  domainSettings:
    dockerSettings:
      enableDockerAccess: ENABLED
      vpcOnlyTrustedAccounts:
      - example
    executionRoleIdentityConfig: USER_PROFILE_NAME
    ipAddressType: IPV4
    rStudioServerProDomainSettings:
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      domainExecutionRoleArn: arn:aws:iam::123456789012:role/example-role
      rStudioConnectUrl: example
      rStudioPackageManagerUrl: example
    securityGroupIds:
    - example
    unifiedStudioSettings:
      domainAccountId: '123456789012'
      domainRegion: example
      projectS3Path: example
      studioWebPortalAccess: ENABLED
  homeEfsFileSystemCreation: Enabled
  kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
  providerRef:
    name: example-sagemakerdomain
    region: us-east-1
  subnetIds:
  - example
  tagPropagation: ENABLED
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`appNetworkAccessType`	enum	—	Specifies the VPC used for non-EFS traffic. Allowed: `PublicInternetOnly, VpcOnly`
`appSecurityGroupManagement`	enum	—	The entity that creates and manages the required security groups for inter-app communication in VPCOnly mode. Allowed: `Service, Customer`
`authMode`	enum	yes	The mode of authentication that members use to access the domain. Allowed: `SSO, IAM`
`defaultSpaceSettings`	object	—	The default space settings.
`defaultSpaceSettings.customFileSystemConfigs`	array<object>	—
`defaultSpaceSettings.customFileSystemConfigs[].efsFileSystemConfig`	object	—	SageMakerDomainEFSFileSystemConfig is a nested property type of AWS::SageMaker::Domain.
`defaultSpaceSettings.customFileSystemConfigs[].efsFileSystemConfig.fileSystemId`	string	yes
`defaultSpaceSettings.customFileSystemConfigs[].efsFileSystemConfig.fileSystemPath`	string	—
`defaultSpaceSettings.customFileSystemConfigs[].fSxLustreFileSystemConfig`	object	—	SageMakerDomainFSxLustreFileSystemConfig is a nested property type of AWS::SageMaker::Domain.
`defaultSpaceSettings.customFileSystemConfigs[].fSxLustreFileSystemConfig.fileSystemId`	string	yes
`defaultSpaceSettings.customFileSystemConfigs[].fSxLustreFileSystemConfig.fileSystemPath`	string	—
`defaultSpaceSettings.customFileSystemConfigs[].s3FileSystemConfig`	object	—	SageMakerDomainS3FileSystemConfig is a nested property type of AWS::SageMaker::Domain.
`defaultSpaceSettings.customFileSystemConfigs[].s3FileSystemConfig.mountPath`	string	—
`defaultSpaceSettings.customFileSystemConfigs[].s3FileSystemConfig.s3Uri`	string	—
`defaultSpaceSettings.customPosixUserConfig`	object	—	The Jupyter lab's custom posix user configurations.
`defaultSpaceSettings.customPosixUserConfig.gid`	integer	yes
`defaultSpaceSettings.customPosixUserConfig.uid`	integer	yes
`defaultSpaceSettings.executionRole`	string	yes	The execution role for the space.
`defaultSpaceSettings.jupyterLabAppSettings`	object	—	The Jupyter lab's app settings.
`defaultSpaceSettings.jupyterLabAppSettings.appLifecycleManagement`	object	—	SageMakerDomainAppLifecycleManagement is a nested property type of AWS::SageMaker::Domain.
`defaultSpaceSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings`	object	—	SageMakerDomainIdleSettings is a nested property type of AWS::SageMaker::Domain.
`defaultSpaceSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.idleTimeoutInMinutes`	integer	—
`defaultSpaceSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.lifecycleManagement`	enum	—	Allowed: `ENABLED, DISABLED`
`defaultSpaceSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.maxIdleTimeoutInMinutes`	integer	—
`defaultSpaceSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.minIdleTimeoutInMinutes`	integer	—
`defaultSpaceSettings.jupyterLabAppSettings.builtInLifecycleConfigArn`	string	—	The lifecycle configuration that runs before the default lifecycle configuration.
`defaultSpaceSettings.jupyterLabAppSettings.codeRepositories`	array<object>	—	A list of CodeRepositories available for use with JupyterLab apps.
`defaultSpaceSettings.jupyterLabAppSettings.codeRepositories[].repositoryUrl`	string	yes	A CodeRepository (valid URL) to be used within Jupyter's Git extension.
`defaultSpaceSettings.jupyterLabAppSettings.customImages`	array<object>	—	A list of custom images for use for JupyterLab apps.
`defaultSpaceSettings.jupyterLabAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`defaultSpaceSettings.jupyterLabAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`defaultSpaceSettings.jupyterLabAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`defaultSpaceSettings.jupyterLabAppSettings.defaultResourceSpec`	object	—	The default instance type and the Amazon Resource Name (ARN) of the default SageMaker image used by the JupyterLab app.
`defaultSpaceSettings.jupyterLabAppSettings.defaultResourceSpec.instanceType`	string	—
`defaultSpaceSettings.jupyterLabAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`defaultSpaceSettings.jupyterLabAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The Amazon Resource Name (ARN) of the SageMaker image that the image version belongs to.
`defaultSpaceSettings.jupyterLabAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The Amazon Resource Name (ARN) of the image version created on the instance.
`defaultSpaceSettings.jupyterLabAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`defaultSpaceSettings.jupyterLabAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with JupyterLab apps.
`defaultSpaceSettings.jupyterServerAppSettings`	object	—	The Jupyter server's app settings.
`defaultSpaceSettings.jupyterServerAppSettings.defaultResourceSpec`	object	—	SageMakerDomainResourceSpec is a nested property type of AWS::SageMaker::Domain.
`defaultSpaceSettings.jupyterServerAppSettings.defaultResourceSpec.instanceType`	string	—
`defaultSpaceSettings.jupyterServerAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`defaultSpaceSettings.jupyterServerAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The Amazon Resource Name (ARN) of the SageMaker image that the image version belongs to.
`defaultSpaceSettings.jupyterServerAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The Amazon Resource Name (ARN) of the image version created on the instance.
`defaultSpaceSettings.jupyterServerAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`defaultSpaceSettings.jupyterServerAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with JupyterServer apps.
`defaultSpaceSettings.kernelGatewayAppSettings`	object	—	The kernel gateway app settings.
`defaultSpaceSettings.kernelGatewayAppSettings.customImages`	array<object>	—	A list of custom SageMaker images that are configured to run as a KernelGateway app.
`defaultSpaceSettings.kernelGatewayAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`defaultSpaceSettings.kernelGatewayAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`defaultSpaceSettings.kernelGatewayAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`defaultSpaceSettings.kernelGatewayAppSettings.defaultResourceSpec`	object	—	The default instance type and the Amazon Resource Name (ARN) of the default SageMaker image used by the KernelGateway app.
`defaultSpaceSettings.kernelGatewayAppSettings.defaultResourceSpec.instanceType`	string	—
`defaultSpaceSettings.kernelGatewayAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`defaultSpaceSettings.kernelGatewayAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The Amazon Resource Name (ARN) of the SageMaker image that the image version belongs to.
`defaultSpaceSettings.kernelGatewayAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The Amazon Resource Name (ARN) of the image version created on the instance.
`defaultSpaceSettings.kernelGatewayAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`defaultSpaceSettings.kernelGatewayAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with KernelGateway apps.
`defaultSpaceSettings.securityGroups`	array<string>	—	The security groups for the Amazon Virtual Private Cloud (VPC) that Studio uses for communication.
`defaultSpaceSettings.spaceStorageSettings`	object	—	The Jupyter lab's space storage settings.
`defaultSpaceSettings.spaceStorageSettings.defaultEbsStorageSettings`	object	—	SageMakerDomainDefaultEbsStorageSettings is a nested property type of AWS::SageMaker::Domain.
`defaultSpaceSettings.spaceStorageSettings.defaultEbsStorageSettings.defaultEbsVolumeSizeInGb`	integer	yes	Default size of the Amazon EBS volume in Gb
`defaultSpaceSettings.spaceStorageSettings.defaultEbsStorageSettings.maximumEbsVolumeSizeInGb`	integer	yes	Maximum size of the Amazon EBS volume in Gb.
`defaultUserSettings`	object	yes	The default user settings.
`defaultUserSettings.autoMountHomeEFS`	enum	—	Indicates whether auto-mounting of an EFS volume is supported for the user profile. Allowed: `Enabled, Disabled, DefaultAsDomain`
`defaultUserSettings.codeEditorAppSettings`	object	—	SageMakerDomainCodeEditorAppSettings is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.codeEditorAppSettings.appLifecycleManagement`	object	—	SageMakerDomainAppLifecycleManagement is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings`	object	—	SageMakerDomainIdleSettings is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings.idleTimeoutInMinutes`	integer	—
`defaultUserSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings.lifecycleManagement`	enum	—	Allowed: `ENABLED, DISABLED`
`defaultUserSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings.maxIdleTimeoutInMinutes`	integer	—
`defaultUserSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings.minIdleTimeoutInMinutes`	integer	—
`defaultUserSettings.codeEditorAppSettings.builtInLifecycleConfigArn`	string	—	The lifecycle configuration that runs before the default lifecycle configuration.
`defaultUserSettings.codeEditorAppSettings.customImages`	array<object>	—	A list of custom images for use for CodeEditor apps.
`defaultUserSettings.codeEditorAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`defaultUserSettings.codeEditorAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`defaultUserSettings.codeEditorAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`defaultUserSettings.codeEditorAppSettings.defaultResourceSpec`	object	—	The default instance type and the Amazon Resource Name (ARN) of the default SageMaker image used by the CodeEditor app.
`defaultUserSettings.codeEditorAppSettings.defaultResourceSpec.instanceType`	string	—
`defaultUserSettings.codeEditorAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`defaultUserSettings.codeEditorAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The Amazon Resource Name (ARN) of the SageMaker image that the image version belongs to.
`defaultUserSettings.codeEditorAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The Amazon Resource Name (ARN) of the image version created on the instance.
`defaultUserSettings.codeEditorAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`defaultUserSettings.codeEditorAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with CodeEditor apps.
`defaultUserSettings.customFileSystemConfigs`	array<object>	—
`defaultUserSettings.customFileSystemConfigs[].efsFileSystemConfig`	object	—	SageMakerDomainEFSFileSystemConfig is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.customFileSystemConfigs[].efsFileSystemConfig.fileSystemId`	string	yes
`defaultUserSettings.customFileSystemConfigs[].efsFileSystemConfig.fileSystemPath`	string	—
`defaultUserSettings.customFileSystemConfigs[].fSxLustreFileSystemConfig`	object	—	SageMakerDomainFSxLustreFileSystemConfig is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.customFileSystemConfigs[].fSxLustreFileSystemConfig.fileSystemId`	string	yes
`defaultUserSettings.customFileSystemConfigs[].fSxLustreFileSystemConfig.fileSystemPath`	string	—
`defaultUserSettings.customFileSystemConfigs[].s3FileSystemConfig`	object	—	SageMakerDomainS3FileSystemConfig is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.customFileSystemConfigs[].s3FileSystemConfig.mountPath`	string	—
`defaultUserSettings.customFileSystemConfigs[].s3FileSystemConfig.s3Uri`	string	—
`defaultUserSettings.customPosixUserConfig`	object	—	SageMakerDomainCustomPosixUserConfig is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.customPosixUserConfig.gid`	integer	yes
`defaultUserSettings.customPosixUserConfig.uid`	integer	yes
`defaultUserSettings.defaultLandingUri`	string	—	Defines which Amazon SageMaker application users are directed to by default.
`defaultUserSettings.executionRole`	string	yes	The execution role for the user.
`defaultUserSettings.jupyterLabAppSettings`	object	—	SageMakerDomainJupyterLabAppSettings is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.jupyterLabAppSettings.appLifecycleManagement`	object	—	SageMakerDomainAppLifecycleManagement is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings`	object	—	SageMakerDomainIdleSettings is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.idleTimeoutInMinutes`	integer	—
`defaultUserSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.lifecycleManagement`	enum	—	Allowed: `ENABLED, DISABLED`
`defaultUserSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.maxIdleTimeoutInMinutes`	integer	—
`defaultUserSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.minIdleTimeoutInMinutes`	integer	—
`defaultUserSettings.jupyterLabAppSettings.builtInLifecycleConfigArn`	string	—	The lifecycle configuration that runs before the default lifecycle configuration.
`defaultUserSettings.jupyterLabAppSettings.codeRepositories`	array<object>	—	A list of CodeRepositories available for use with JupyterLab apps.
`defaultUserSettings.jupyterLabAppSettings.codeRepositories[].repositoryUrl`	string	yes	A CodeRepository (valid URL) to be used within Jupyter's Git extension.
`defaultUserSettings.jupyterLabAppSettings.customImages`	array<object>	—	A list of custom images for use for JupyterLab apps.
`defaultUserSettings.jupyterLabAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`defaultUserSettings.jupyterLabAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`defaultUserSettings.jupyterLabAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`defaultUserSettings.jupyterLabAppSettings.defaultResourceSpec`	object	—	The default instance type and the Amazon Resource Name (ARN) of the default SageMaker image used by the JupyterLab app.
`defaultUserSettings.jupyterLabAppSettings.defaultResourceSpec.instanceType`	string	—
`defaultUserSettings.jupyterLabAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`defaultUserSettings.jupyterLabAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The Amazon Resource Name (ARN) of the SageMaker image that the image version belongs to.
`defaultUserSettings.jupyterLabAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The Amazon Resource Name (ARN) of the image version created on the instance.
`defaultUserSettings.jupyterLabAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`defaultUserSettings.jupyterLabAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with JupyterLab apps.
`defaultUserSettings.jupyterServerAppSettings`	object	—	The Jupyter server's app settings.
`defaultUserSettings.jupyterServerAppSettings.defaultResourceSpec`	object	—	SageMakerDomainResourceSpec is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.jupyterServerAppSettings.defaultResourceSpec.instanceType`	string	—
`defaultUserSettings.jupyterServerAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`defaultUserSettings.jupyterServerAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The Amazon Resource Name (ARN) of the SageMaker image that the image version belongs to.
`defaultUserSettings.jupyterServerAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The Amazon Resource Name (ARN) of the image version created on the instance.
`defaultUserSettings.jupyterServerAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`defaultUserSettings.jupyterServerAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with JupyterServer apps.
`defaultUserSettings.kernelGatewayAppSettings`	object	—	The kernel gateway app settings.
`defaultUserSettings.kernelGatewayAppSettings.customImages`	array<object>	—	A list of custom SageMaker images that are configured to run as a KernelGateway app.
`defaultUserSettings.kernelGatewayAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`defaultUserSettings.kernelGatewayAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`defaultUserSettings.kernelGatewayAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`defaultUserSettings.kernelGatewayAppSettings.defaultResourceSpec`	object	—	The default instance type and the Amazon Resource Name (ARN) of the default SageMaker image used by the KernelGateway app.
`defaultUserSettings.kernelGatewayAppSettings.defaultResourceSpec.instanceType`	string	—
`defaultUserSettings.kernelGatewayAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`defaultUserSettings.kernelGatewayAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The Amazon Resource Name (ARN) of the SageMaker image that the image version belongs to.
`defaultUserSettings.kernelGatewayAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The Amazon Resource Name (ARN) of the image version created on the instance.
`defaultUserSettings.kernelGatewayAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`defaultUserSettings.kernelGatewayAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with KernelGateway apps.
`defaultUserSettings.rSessionAppSettings`	object	—	SageMakerDomainRSessionAppSettings is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.rSessionAppSettings.customImages`	array<object>	—	A list of custom SageMaker images that are configured to run as a KernelGateway app.
`defaultUserSettings.rSessionAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`defaultUserSettings.rSessionAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`defaultUserSettings.rSessionAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`defaultUserSettings.rSessionAppSettings.defaultResourceSpec`	object	—	SageMakerDomainResourceSpec is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.rSessionAppSettings.defaultResourceSpec.instanceType`	string	—
`defaultUserSettings.rSessionAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`defaultUserSettings.rSessionAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The Amazon Resource Name (ARN) of the SageMaker image that the image version belongs to.
`defaultUserSettings.rSessionAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The Amazon Resource Name (ARN) of the image version created on the instance.
`defaultUserSettings.rSessionAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`defaultUserSettings.rStudioServerProAppSettings`	object	—	SageMakerDomainRStudioServerProAppSettings is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.rStudioServerProAppSettings.accessStatus`	enum	—	Indicates whether the current user has access to the RStudioServerPro app. Allowed: `ENABLED, DISABLED`
`defaultUserSettings.rStudioServerProAppSettings.userGroup`	enum	—	The level of permissions that the user has within the RStudioServerPro app. Allowed: `R_STUDIO_ADMIN, R_STUDIO_USER`
`defaultUserSettings.securityGroups`	array<string>	—	The security groups for the Amazon Virtual Private Cloud (VPC) that Studio uses for communication.
`defaultUserSettings.sharingSettings`	object	—	The sharing settings.
`defaultUserSettings.sharingSettings.notebookOutputOption`	enum	—	Whether to include the notebook cell output when sharing the notebook. Allowed: `Allowed, Disabled`
`defaultUserSettings.sharingSettings.s3KmsKeyId`	string	—	When NotebookOutputOption is Allowed, the AWS Key Management Service (KMS) encryption key ID used to encrypt the notebook cell output in the Amazon S3 bucket.
`defaultUserSettings.sharingSettings.s3OutputPath`	string	—	When NotebookOutputOption is Allowed, the Amazon S3 bucket used to store the shared notebook snapshots.
`defaultUserSettings.spaceStorageSettings`	object	—	SageMakerDomainDefaultSpaceStorageSettings is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.spaceStorageSettings.defaultEbsStorageSettings`	object	—	SageMakerDomainDefaultEbsStorageSettings is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.spaceStorageSettings.defaultEbsStorageSettings.defaultEbsVolumeSizeInGb`	integer	yes	Default size of the Amazon EBS volume in Gb
`defaultUserSettings.spaceStorageSettings.defaultEbsStorageSettings.maximumEbsVolumeSizeInGb`	integer	yes	Maximum size of the Amazon EBS volume in Gb.
`defaultUserSettings.studioWebPortal`	enum	—	Indicates whether the Studio experience is available to users. Allowed: `ENABLED, DISABLED`
`defaultUserSettings.studioWebPortalSettings`	object	—	SageMakerDomainStudioWebPortalSettings is a nested property type of AWS::SageMaker::Domain.
`defaultUserSettings.studioWebPortalSettings.hiddenAppTypes`	array<string>	—	Applications supported in Studio that are hidden from the Studio left navigation pane.
`defaultUserSettings.studioWebPortalSettings.hiddenInstanceTypes`	array<string>	—	The instance types you are hiding from the Studio user interface.
`defaultUserSettings.studioWebPortalSettings.hiddenMlTools`	array<string>	—	The machine learning tools that are hidden from the Studio left navigation pane.
`defaultUserSettings.studioWebPortalSettings.hiddenSageMakerImageVersionAliases`	array<object>	—	The version aliases you are hiding from the Studio user interface.
`defaultUserSettings.studioWebPortalSettings.hiddenSageMakerImageVersionAliases[].sageMakerImageName`	enum	—	The SageMaker image name that you are hiding from the Studio user interface. Allowed: `sagemaker_distribution`
`defaultUserSettings.studioWebPortalSettings.hiddenSageMakerImageVersionAliases[].versionAliases`	array<string>	—
`domainName`	string	yes	A name for the domain.
`domainSettings`	object	—	SageMakerDomainDomainSettings is a nested property type of AWS::SageMaker::Domain.
`domainSettings.dockerSettings`	object	—	SageMakerDomainDockerSettings is a nested property type of AWS::SageMaker::Domain.
`domainSettings.dockerSettings.enableDockerAccess`	enum	—	The flag to enable/disable docker-proxy server Allowed: `ENABLED, DISABLED`
`domainSettings.dockerSettings.vpcOnlyTrustedAccounts`	array<string>	—	A list of account id's that would be used to pull images from in VpcOnly mode
`domainSettings.executionRoleIdentityConfig`	enum	—	The configuration for attaching a SageMaker user profile name to the execution role as a sts:SourceIdentity key. Allowed: `USER_PROFILE_NAME, DISABLED`
`domainSettings.ipAddressType`	enum	—	Allowed: `IPV4, DUALSTACK`
`domainSettings.rStudioServerProDomainSettings`	object	—	SageMakerDomainRStudioServerProDomainSettings is a nested property type of AWS::SageMaker::Domain.
`domainSettings.rStudioServerProDomainSettings.defaultResourceSpec`	object	—	SageMakerDomainResourceSpec is a nested property type of AWS::SageMaker::Domain.
`domainSettings.rStudioServerProDomainSettings.defaultResourceSpec.instanceType`	string	—
`domainSettings.rStudioServerProDomainSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`domainSettings.rStudioServerProDomainSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The Amazon Resource Name (ARN) of the SageMaker image that the image version belongs to.
`domainSettings.rStudioServerProDomainSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The Amazon Resource Name (ARN) of the image version created on the instance.
`domainSettings.rStudioServerProDomainSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`domainSettings.rStudioServerProDomainSettings.domainExecutionRoleArn`	string	yes	The ARN of the execution role for the RStudioServerPro Domain-level app.
`domainSettings.rStudioServerProDomainSettings.rStudioConnectUrl`	string	—	A URL pointing to an RStudio Connect server.
`domainSettings.rStudioServerProDomainSettings.rStudioPackageManagerUrl`	string	—	A URL pointing to an RStudio Package Manager server.
`domainSettings.securityGroupIds`	array<string>	—	The security groups for the Amazon Virtual Private Cloud that the Domain uses for communication between Domain-level apps and user apps.
`domainSettings.unifiedStudioSettings`	object	—	SageMakerDomainUnifiedStudioSettings is a nested property type of AWS::SageMaker::Domain.
`domainSettings.unifiedStudioSettings.domainAccountId`	string	—	The ID of the AWS account that has the Amazon SageMaker Unified Studio domain.
`domainSettings.unifiedStudioSettings.domainId`	string	—	The ID of the Amazon SageMaker Unified Studio domain associated with this domain.
`domainSettings.unifiedStudioSettings.domainRegion`	string	—	The AWS Region where the domain is located in Amazon SageMaker Unified Studio.
`domainSettings.unifiedStudioSettings.environmentId`	string	—	The ID of the environment that Amazon SageMaker Unified Studio associates with the domain.
`domainSettings.unifiedStudioSettings.projectId`	string	—	The ID of the Amazon SageMaker Unified Studio project that corresponds to the domain.
`domainSettings.unifiedStudioSettings.projectS3Path`	string	—	The location where Amazon S3 stores temporary execution data and other artifacts for the project that corresponds to the domain.
`domainSettings.unifiedStudioSettings.studioWebPortalAccess`	enum	—	Sets whether you can access the domain in Amazon SageMaker Studio: ENABLED You can access the domain in Amazon SageMaker Studio. Allowed: `ENABLED, DISABLED`
`homeEfsFileSystemCreation`	enum	—	Indicates whether a home EFS file system is created for the domain. Allowed: `Enabled, Disabled`
`kmsKeyId`	string	—	SageMaker uses AWS KMS to encrypt the EFS volume attached to the domain with an AWS managed customer master key (CMK) by default.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`subnetIds`	array<string>	—	The VPC subnets that Studio uses for communication.
`tagPropagation`	enum	—	Indicates whether the tags added to Domain, User Profile and Space entity is propagated to all SageMaker resources. Allowed: `ENABLED, DISABLED`
`tags`	array<object>	—	A list of tags to apply to the user profile.
`tags[].key`	string	yes
`tags[].value`	string	yes
`vpcId`	string	—	The ID of the Amazon Virtual Private Cloud (VPC) that Studio uses for communication.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`domainArn`	string	—	The Amazon Resource Name (ARN) of the created domain.
`domainId`	string	—	The domain name.
`homeEfsFileSystemId`	string	—	The ID of the Amazon Elastic File System (EFS) managed by this Domain.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`securityGroupIdForDomainBoundary`	string	—	The ID of the security group that authorizes traffic between the RSessionGateway apps and the RStudioServerPro app.
`singleSignOnApplicationArn`	string	—	The ARN of the application managed by SageMaker in IAM Identity Center.
`singleSignOnManagedApplicationInstanceId`	string	—	The SSO managed application instance ID.
`url`	string	—	The URL to the created domain.

SageMakerEndpoint # CLOUD CONTROL · AWS::SageMaker::Endpoint

SageMakerEndpoint manages AWS::SageMaker::Endpoint through the AWS Cloud Control API.

kubectl get sagemakerendpoints · example on GitHub

Example

# SageMakerEndpoint — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerendpoint
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerEndpoint
metadata:
  name: example-sagemakerendpoint
  namespace: default
spec:
  deploymentConfig:
    autoRollbackConfiguration:
      alarms:
      - alarmName: example-sagemakerendpoint
    blueGreenUpdatePolicy:
      maximumExecutionTimeoutInSeconds: 600
      terminationWaitInSeconds: 1
      trafficRoutingConfiguration:
        type: example
        waitIntervalInSeconds: 1
    rollingUpdatePolicy:
      maximumBatchSize:
        type: example
        value: 1
      maximumExecutionTimeoutInSeconds: 600
      rollbackMaximumBatchSize:
        type: example
        value: 1
      waitIntervalInSeconds: 1
  endpointConfigName: example-sagemakerendpoint
  endpointName: example-sagemakerendpoint
  excludeRetainedVariantProperties:
  - variantPropertyType: example
  providerRef:
    name: example-sagemakerendpoint
    region: us-east-1
  retainAllVariantProperties: false
  retainDeploymentConfig: false
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`deploymentConfig`	object	—	Specifies deployment configuration for updating the SageMaker endpoint.
`deploymentConfig.autoRollbackConfiguration`	object	—	Configuration for automatic rollback if an error occurs during deployment.
`deploymentConfig.autoRollbackConfiguration.alarms`	array<object>	yes	List of CloudWatch alarms to monitor during the deployment.
`deploymentConfig.autoRollbackConfiguration.alarms[].alarmName`	string	yes	The name of the CloudWatch alarm.
`deploymentConfig.blueGreenUpdatePolicy`	object	—	Configuration for blue-green update deployment policies.
`deploymentConfig.blueGreenUpdatePolicy.maximumExecutionTimeoutInSeconds`	integer	—	The maximum time allowed for the blue/green update, in seconds.
`deploymentConfig.blueGreenUpdatePolicy.terminationWaitInSeconds`	integer	—	The wait time before terminating the old endpoint during a blue/green deployment.
`deploymentConfig.blueGreenUpdatePolicy.trafficRoutingConfiguration`	object	yes	The traffic routing configuration for the blue/green deployment.
`deploymentConfig.blueGreenUpdatePolicy.trafficRoutingConfiguration.canarySize`	object	—	Specifies the size of the canary traffic in a canary deployment.
`deploymentConfig.blueGreenUpdatePolicy.trafficRoutingConfiguration.canarySize.type`	string	yes	Specifies whether the `Value` is an instance count or a capacity unit.
`deploymentConfig.blueGreenUpdatePolicy.trafficRoutingConfiguration.canarySize.value`	integer	yes	The value representing either the number of instances or the number of capacity units.
`deploymentConfig.blueGreenUpdatePolicy.trafficRoutingConfiguration.linearStepSize`	object	—	Specifies the step size for linear traffic routing.
`deploymentConfig.blueGreenUpdatePolicy.trafficRoutingConfiguration.linearStepSize.type`	string	yes	Specifies whether the `Value` is an instance count or a capacity unit.
`deploymentConfig.blueGreenUpdatePolicy.trafficRoutingConfiguration.linearStepSize.value`	integer	yes	The value representing either the number of instances or the number of capacity units.
`deploymentConfig.blueGreenUpdatePolicy.trafficRoutingConfiguration.type`	string	yes	Specifies the type of traffic routing (e.g., 'AllAtOnce', 'Canary', 'Linear').
`deploymentConfig.blueGreenUpdatePolicy.trafficRoutingConfiguration.waitIntervalInSeconds`	integer	—	Specifies the wait interval between traffic shifts, in seconds.
`deploymentConfig.rollingUpdatePolicy`	object	—	Configuration for rolling update deployment policies.
`deploymentConfig.rollingUpdatePolicy.maximumBatchSize`	object	yes	Specifies the maximum batch size for each rolling update.
`deploymentConfig.rollingUpdatePolicy.maximumBatchSize.type`	string	yes	Specifies whether the `Value` is an instance count or a capacity unit.
`deploymentConfig.rollingUpdatePolicy.maximumBatchSize.value`	integer	yes	The value representing either the number of instances or the number of capacity units.
`deploymentConfig.rollingUpdatePolicy.maximumExecutionTimeoutInSeconds`	integer	—	The maximum time allowed for the rolling update, in seconds.
`deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize`	object	—	The maximum batch size for rollback during an update failure.
`deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize.type`	string	yes	Specifies whether the `Value` is an instance count or a capacity unit.
`deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize.value`	integer	yes	The value representing either the number of instances or the number of capacity units.
`deploymentConfig.rollingUpdatePolicy.waitIntervalInSeconds`	integer	yes	The time to wait between steps during the rolling update, in seconds.
`endpointConfigName`	string	yes	The name of the endpoint configuration for the SageMaker endpoint.
`endpointName`	string	—	The name of the SageMaker endpoint.
`excludeRetainedVariantProperties`	array<object>	—	Specifies a list of variant properties that you want to exclude when updating an endpoint.
`excludeRetainedVariantProperties[].variantPropertyType`	string	—	The type of variant property (e.g., 'DesiredInstanceCount', 'DesiredWeight', 'DataCaptureConfig').
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`retainAllVariantProperties`	boolean	—	When set to true, retains all variant properties for an endpoint when it is updated.
`retainDeploymentConfig`	boolean	—	When set to true, retains the deployment configuration during endpoint updates.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`endpointArn`	string	—	The Amazon Resource Name (ARN) of the endpoint.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerEndpointConfig # CLOUD CONTROL · AWS::SageMaker::EndpointConfig

SageMakerEndpointConfig manages AWS::SageMaker::EndpointConfig through the AWS Cloud Control API.

kubectl get sagemakerendpointconfigs · example on GitHub

Example

# SageMakerEndpointConfig — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerendpointconfig
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerEndpointConfig
metadata:
  name: example-sagemakerendpointconfig
  namespace: default
spec:
  asyncInferenceConfig:
    clientConfig:
      maxConcurrentInvocationsPerInstance: 1
    outputConfig:
      kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
      notificationConfig:
        errorTopic: example
        includeInferenceResponseIn:
        - example
        successTopic: example
      s3FailurePath: example
      s3OutputPath: example
  dataCaptureConfig:
    captureContentTypeHeader:
      csvContentTypes:
      - example
      jsonContentTypes:
      - example
    captureOptions:
    - captureMode: example
    destinationS3Uri: example
    enableCapture: true
    initialSamplingPercentage: 1
    kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
  enableNetworkIsolation: true
  endpointConfigName: example-sagemakerendpointconfig
  executionRoleArn: arn:aws:iam::123456789012:role/example-role
  explainerConfig:
    clarifyExplainerConfig:
      enableExplanations: example
      inferenceConfig:
        contentTemplate: example
        featureHeaders:
        - example
        featureTypes:
        - example
        featuresAttribute: example
        labelAttribute: example
        labelHeaders:
        - example
        labelIndex: 1
        maxPayloadInMB: 1
        maxRecordCount: 1
        probabilityAttribute: example
        probabilityIndex: 1
      shapConfig:
        numberOfSamples: 1
        seed: 1
        shapBaselineConfig:
          mimeType: example
          shapBaseline: example
          shapBaselineUri: example
        useLogit: false
  kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
  metricsConfig:
    enableDetailedObservability: true
    enableEnhancedMetrics: true
    metricPublishFrequencyInSeconds: 300
  productionVariants:
  - capacityReservationConfig:
      capacityReservationPreference: example
      mlReservationArn: arn:aws:iam::123456789012:role/example-role
    containerStartupHealthCheckTimeoutInSeconds: 60
    coreDumpConfig:
      destinationS3Uri: example
      kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    enableSSMAccess: true
    inferenceAmiVersion: '1.0'
    initialInstanceCount: 1
    initialVariantWeight: 1
    instancePools:
    - instanceType: example
      modelNameOverride: example
      priority: 1
    instanceType: example
    managedInstanceScaling:
      maxInstanceCount: 2
      minInstanceCount: 2
      status: example
    modelDataDownloadTimeoutInSeconds: 60
    modelName: example-sagemakerendpointconfig
    routingConfig:
      routingStrategy: example
    serverlessConfig:
      maxConcurrency: 1
      memorySizeInMB: 2
      provisionedConcurrency: 1
    variantInstanceProvisionTimeoutInSeconds: 300
    variantName: example-sagemakerendpointconfig
    volumeSizeInGB: 1
  providerRef:
    name: example-sagemakerendpointconfig
    region: us-east-1
  shadowProductionVariants:
  - capacityReservationConfig:
      capacityReservationPreference: example
      mlReservationArn: arn:aws:iam::123456789012:role/example-role
    containerStartupHealthCheckTimeoutInSeconds: 60
    coreDumpConfig:
      destinationS3Uri: example
      kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    enableSSMAccess: true
    inferenceAmiVersion: '1.0'
    initialInstanceCount: 1
    initialVariantWeight: 1
    instancePools:
    - instanceType: example
      modelNameOverride: example
      priority: 1
    instanceType: example
    managedInstanceScaling:
      maxInstanceCount: 2
      minInstanceCount: 2
      status: example
    modelDataDownloadTimeoutInSeconds: 60
    modelName: example-sagemakerendpointconfig
    routingConfig:
      routingStrategy: example
    serverlessConfig:
      maxConcurrency: 1
      memorySizeInMB: 2
      provisionedConcurrency: 1
    variantInstanceProvisionTimeoutInSeconds: 300
    variantName: example-sagemakerendpointconfig
    volumeSizeInGB: 1
  tags:
  - key: example
    value: example
  vpcConfig:
    securityGroupIds:
    - example
    subnets:
    - example

Spec

Field	Type	Required	Description
`asyncInferenceConfig`	object	—	Specifies configuration for how an endpoint performs asynchronous inference.
`asyncInferenceConfig.clientConfig`	object	—	Configures the behavior of the client used by SageMaker to interact with the model container during asynchronous inference.
`asyncInferenceConfig.clientConfig.maxConcurrentInvocationsPerInstance`	integer	—	The maximum number of concurrent requests sent by the SageMaker client to the model container.
`asyncInferenceConfig.outputConfig`	object	yes	Specifies the configuration for asynchronous inference invocation outputs.
`asyncInferenceConfig.outputConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt the asynchronous inference output in Amazon S3.
`asyncInferenceConfig.outputConfig.notificationConfig`	object	—	Specifies the configuration for notifications of inference results for asynchronous inference.
`asyncInferenceConfig.outputConfig.notificationConfig.errorTopic`	string	—	Amazon SNS topic to post a notification to when an inference fails.
`asyncInferenceConfig.outputConfig.notificationConfig.includeInferenceResponseIn`	array<string>	—	The Amazon SNS topics where you want the inference response to be included.
`asyncInferenceConfig.outputConfig.notificationConfig.successTopic`	string	—	Amazon SNS topic to post a notification to when an inference completes successfully.
`asyncInferenceConfig.outputConfig.s3FailurePath`	string	—	The Amazon S3 location to upload failure inference responses to.
`asyncInferenceConfig.outputConfig.s3OutputPath`	string	—	The Amazon S3 location to upload inference responses to.
`dataCaptureConfig`	object	—	Specifies how to capture endpoint data for model monitor.
`dataCaptureConfig.captureContentTypeHeader`	object	—	A list of the JSON and CSV content type that the endpoint captures.
`dataCaptureConfig.captureContentTypeHeader.csvContentTypes`	array<string>	—	A list of the CSV content types of the data that the endpoint captures.
`dataCaptureConfig.captureContentTypeHeader.jsonContentTypes`	array<string>	—	A list of the JSON content types of the data that the endpoint captures.
`dataCaptureConfig.captureOptions`	array<object>	yes	Specifies whether the endpoint captures input data to your model, output data from your model, or both.
`dataCaptureConfig.captureOptions[].captureMode`	string	yes	Specifies whether the endpoint captures input data or output data.
`dataCaptureConfig.destinationS3Uri`	string	yes	The S3 bucket where model monitor stores captured data.
`dataCaptureConfig.enableCapture`	boolean	—	Set to True to enable data capture.
`dataCaptureConfig.initialSamplingPercentage`	integer	yes	The percentage of data to capture.
`dataCaptureConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt the captured data at rest using Amazon S3 server-side encryption.
`enableNetworkIsolation`	boolean	—	Sets whether all model containers deployed to the endpoint are isolated.
`endpointConfigName`	string	—	The name of the endpoint configuration.
`executionRoleArn`	string	—	The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker AI can assume to perform actions on your behalf.
`explainerConfig`	object	—	A parameter to activate explainers.
`explainerConfig.clarifyExplainerConfig`	object	—	A member of ExplainerConfig that contains configuration parameters for the SageMaker Clarify explainer.
`explainerConfig.clarifyExplainerConfig.enableExplanations`	string	—	A JMESPath boolean expression used to filter which records to explain.
`explainerConfig.clarifyExplainerConfig.inferenceConfig`	object	—	The inference configuration parameter for the model container.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.contentTemplate`	string	—	A template string used to format a JSON record into an acceptable model container input.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.featureHeaders`	array<string>	—	The names of the features.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.featureTypes`	array<string>	—	A list of data types of the features (optional).
`explainerConfig.clarifyExplainerConfig.inferenceConfig.featuresAttribute`	string	—	Provides the JMESPath expression to extract the features from a model container input in JSON Lines format.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.labelAttribute`	string	—	A JMESPath expression used to locate the list of label headers in the model container output.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.labelHeaders`	array<string>	—	For multiclass classification problems, the label headers are the names of the classes.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.labelIndex`	integer	—	A zero-based index used to extract a label header or list of label headers from model container output in CSV format.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.maxPayloadInMB`	integer	—	The maximum payload size (MB) allowed of a request from the explainer to the model container.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.maxRecordCount`	integer	—	The maximum number of records in a request that the model container can process when querying the model container for the predictions of a synthetic dataset.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.probabilityAttribute`	string	—	A JMESPath expression used to extract the probability (or score) from the model container output if the model container is in JSON Lines format.
`explainerConfig.clarifyExplainerConfig.inferenceConfig.probabilityIndex`	integer	—	A zero-based index used to extract a probability value (score) or list from model container output in CSV format.
`explainerConfig.clarifyExplainerConfig.shapConfig`	object	yes	The configuration for SHAP analysis.
`explainerConfig.clarifyExplainerConfig.shapConfig.numberOfSamples`	integer	—	The number of samples to be used for analysis by the Kernal SHAP algorithm.
`explainerConfig.clarifyExplainerConfig.shapConfig.seed`	integer	—	The starting value used to initialize the random number generator in the explainer.
`explainerConfig.clarifyExplainerConfig.shapConfig.shapBaselineConfig`	object	yes	The configuration for the SHAP baseline of the Kernal SHAP algorithm.
`explainerConfig.clarifyExplainerConfig.shapConfig.shapBaselineConfig.mimeType`	string	—	The MIME type of the baseline data.
`explainerConfig.clarifyExplainerConfig.shapConfig.shapBaselineConfig.shapBaseline`	string	—	The inline SHAP baseline data in string format.
`explainerConfig.clarifyExplainerConfig.shapConfig.shapBaselineConfig.shapBaselineUri`	string	—	The uniform resource identifier (URI) of the S3 bucket where the SHAP baseline file is stored.
`explainerConfig.clarifyExplainerConfig.shapConfig.textConfig`	object	—	A parameter that indicates if text features are treated as text and explanations are provided for individual units of text.
`explainerConfig.clarifyExplainerConfig.shapConfig.textConfig.granularity`	string	yes	The unit of granularity for the analysis of text features.
`explainerConfig.clarifyExplainerConfig.shapConfig.textConfig.language`	string	yes	Specifies the language of the text features in ISO 639-1 or ISO 639-3 code of a supported language.
`explainerConfig.clarifyExplainerConfig.shapConfig.useLogit`	boolean	—	A Boolean toggle to indicate if you want to use the logit function (true) or log-odds units (false) for model predictions.
`kmsKeyId`	string	—	The Amazon Resource Name (ARN) of an AWS Key Management Service key that Amazon SageMaker uses to encrypt data on the storage volume attached to the ML compute instance that hosts the endpoint.
`metricsConfig`	object	—	Specifies the metrics that the endpoint publishes to Amazon CloudWatch, the frequency of publication, and whether to enable enhanced or detailed observability metrics.
`metricsConfig.enableDetailedObservability`	boolean	—	Specifies whether to enable detailed observability for the endpoint.
`metricsConfig.enableEnhancedMetrics`	boolean	—	Specifies whether to enable enhanced metrics for the endpoint.
`metricsConfig.metricPublishFrequencyInSeconds`	integer	—	The interval, in seconds, at which the endpoint publishes metrics to Amazon CloudWatch.
`productionVariants`	array<object>	yes	A list of ProductionVariant objects, one for each model that you want to host at this endpoint.
`productionVariants[].capacityReservationConfig`	object	—	Settings for the capacity reservation for the compute instances that SageMaker AI reserves for an endpoint.
`productionVariants[].capacityReservationConfig.capacityReservationPreference`	string	—	Options that you can choose for the capacity reservation.
`productionVariants[].capacityReservationConfig.mlReservationArn`	string	—	The Amazon Resource Name (ARN) that uniquely identifies the ML capacity reservation that SageMaker AI applies when it deploys the endpoint.
`productionVariants[].containerStartupHealthCheckTimeoutInSeconds`	integer	—	The timeout value, in seconds, for your inference container to pass health check by SageMaker Hosting.
`productionVariants[].coreDumpConfig`	object	—	Specifies configuration for a core dump from the model container when the process crashes.
`productionVariants[].coreDumpConfig.destinationS3Uri`	string	yes	The Amazon S3 bucket to send the core dump to.
`productionVariants[].coreDumpConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that SageMaker uses to encrypt the core dump data at rest using Amazon S3 server-side encryption.
`productionVariants[].enableSSMAccess`	boolean	—	You can use this parameter to turn on native AWS Systems Manager (SSM) access for a production variant behind an endpoint.
`productionVariants[].inferenceAmiVersion`	string	—	Specifies an option from a collection of preconfigured Amazon Machine Image (AMI) images.
`productionVariants[].initialInstanceCount`	integer	—	Number of instances to launch initially.
`productionVariants[].initialVariantWeight`	number	—	Determines initial traffic distribution among all of the models that you specify in the endpoint configuration.
`productionVariants[].instancePools`	array<object>	—	A list of instance pools for the production variant.
`productionVariants[].instancePools[].instanceType`	string	yes	The ML compute instance type for the instance pool.
`productionVariants[].instancePools[].modelNameOverride`	string	—	The name of a SageMaker model to use for this instance pool instead of the model specified for the production variant.
`productionVariants[].instancePools[].priority`	integer	yes	The priority for the instance pool.
`productionVariants[].instanceType`	string	—	The ML compute instance type.
`productionVariants[].managedInstanceScaling`	object	—	SageMakerEndpointConfigManagedInstanceScaling is a nested property type of AWS::SageMaker::EndpointConfig.
`productionVariants[].managedInstanceScaling.maxInstanceCount`	integer	—	The maximum number of instances that the endpoint can provision when it scales up to accommodate an increase in traffic.
`productionVariants[].managedInstanceScaling.minInstanceCount`	integer	—	The minimum number of instances that the endpoint must retain when it scales down to accommodate a decrease in traffic.
`productionVariants[].managedInstanceScaling.scaleInPolicy`	object	—	Configures the scale-in behavior for managed instance scaling.
`productionVariants[].managedInstanceScaling.scaleInPolicy.cooldownInMinutes`	integer	—	The cooldown period, in minutes, after the last endpoint operation before the endpoint evaluates consolidation scale-in opportunities.
`productionVariants[].managedInstanceScaling.scaleInPolicy.maximumStepSize`	integer	—	The maximum number of instances that the endpoint can terminate at a time during a consolidation scale-in operation.
`productionVariants[].managedInstanceScaling.scaleInPolicy.strategy`	string	yes	The strategy for scaling in instances.
`productionVariants[].managedInstanceScaling.status`	string	—	Indicates whether managed instance scaling is enabled.
`productionVariants[].modelDataDownloadTimeoutInSeconds`	integer	—	The timeout value, in seconds, to download and extract the model that you want to host from Amazon S3 to the individual inference instance associated with this production variant.
`productionVariants[].modelName`	string	—	The name of the model that you want to host.
`productionVariants[].routingConfig`	object	—	Settings that control how the endpoint routes incoming traffic to the instances that the endpoint hosts.
`productionVariants[].routingConfig.prefixAwareRoutingConfig`	object	—	The configuration for prefix-aware routing.
`productionVariants[].routingConfig.prefixAwareRoutingConfig.concurrencyThreshold`	integer	—	The maximum number of in-flight requests on the target instance before the endpoint routes to another instance.
`productionVariants[].routingConfig.prefixAwareRoutingConfig.prefixLength`	integer	—	The maximum length of the prefix used for routing decisions.
`productionVariants[].routingConfig.routingStrategy`	string	—	Sets how the endpoint routes incoming traffic.
`productionVariants[].serverlessConfig`	object	—	The serverless configuration for an endpoint.
`productionVariants[].serverlessConfig.maxConcurrency`	integer	yes	The maximum number of concurrent invocations your serverless endpoint can process.
`productionVariants[].serverlessConfig.memorySizeInMB`	integer	yes	The memory size of your serverless endpoint.
`productionVariants[].serverlessConfig.provisionedConcurrency`	integer	—	The amount of provisioned concurrency to allocate for the serverless endpoint.
`productionVariants[].variantInstanceProvisionTimeoutInSeconds`	integer	—	The timeout value, in seconds, for provisioning instances for the production variant.
`productionVariants[].variantName`	string	yes	The name of the production variant.
`productionVariants[].volumeSizeInGB`	integer	—	The size, in GB, of the ML storage volume attached to individual inference instance associated with the production variant.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`shadowProductionVariants`	array<object>	—	Array of ProductionVariant objects.
`shadowProductionVariants[].capacityReservationConfig`	object	—	Settings for the capacity reservation for the compute instances that SageMaker AI reserves for an endpoint.
`shadowProductionVariants[].capacityReservationConfig.capacityReservationPreference`	string	—	Options that you can choose for the capacity reservation.
`shadowProductionVariants[].capacityReservationConfig.mlReservationArn`	string	—	The Amazon Resource Name (ARN) that uniquely identifies the ML capacity reservation that SageMaker AI applies when it deploys the endpoint.
`shadowProductionVariants[].containerStartupHealthCheckTimeoutInSeconds`	integer	—	The timeout value, in seconds, for your inference container to pass health check by SageMaker Hosting.
`shadowProductionVariants[].coreDumpConfig`	object	—	Specifies configuration for a core dump from the model container when the process crashes.
`shadowProductionVariants[].coreDumpConfig.destinationS3Uri`	string	yes	The Amazon S3 bucket to send the core dump to.
`shadowProductionVariants[].coreDumpConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that SageMaker uses to encrypt the core dump data at rest using Amazon S3 server-side encryption.
`shadowProductionVariants[].enableSSMAccess`	boolean	—	You can use this parameter to turn on native AWS Systems Manager (SSM) access for a production variant behind an endpoint.
`shadowProductionVariants[].inferenceAmiVersion`	string	—	Specifies an option from a collection of preconfigured Amazon Machine Image (AMI) images.
`shadowProductionVariants[].initialInstanceCount`	integer	—	Number of instances to launch initially.
`shadowProductionVariants[].initialVariantWeight`	number	—	Determines initial traffic distribution among all of the models that you specify in the endpoint configuration.
`shadowProductionVariants[].instancePools`	array<object>	—	A list of instance pools for the production variant.
`shadowProductionVariants[].instancePools[].instanceType`	string	yes	The ML compute instance type for the instance pool.
`shadowProductionVariants[].instancePools[].modelNameOverride`	string	—	The name of a SageMaker model to use for this instance pool instead of the model specified for the production variant.
`shadowProductionVariants[].instancePools[].priority`	integer	yes	The priority for the instance pool.
`shadowProductionVariants[].instanceType`	string	—	The ML compute instance type.
`shadowProductionVariants[].managedInstanceScaling`	object	—	SageMakerEndpointConfigManagedInstanceScaling is a nested property type of AWS::SageMaker::EndpointConfig.
`shadowProductionVariants[].managedInstanceScaling.maxInstanceCount`	integer	—	The maximum number of instances that the endpoint can provision when it scales up to accommodate an increase in traffic.
`shadowProductionVariants[].managedInstanceScaling.minInstanceCount`	integer	—	The minimum number of instances that the endpoint must retain when it scales down to accommodate a decrease in traffic.
`shadowProductionVariants[].managedInstanceScaling.scaleInPolicy`	object	—	Configures the scale-in behavior for managed instance scaling.
`shadowProductionVariants[].managedInstanceScaling.scaleInPolicy.cooldownInMinutes`	integer	—	The cooldown period, in minutes, after the last endpoint operation before the endpoint evaluates consolidation scale-in opportunities.
`shadowProductionVariants[].managedInstanceScaling.scaleInPolicy.maximumStepSize`	integer	—	The maximum number of instances that the endpoint can terminate at a time during a consolidation scale-in operation.
`shadowProductionVariants[].managedInstanceScaling.scaleInPolicy.strategy`	string	yes	The strategy for scaling in instances.
`shadowProductionVariants[].managedInstanceScaling.status`	string	—	Indicates whether managed instance scaling is enabled.
`shadowProductionVariants[].modelDataDownloadTimeoutInSeconds`	integer	—	The timeout value, in seconds, to download and extract the model that you want to host from Amazon S3 to the individual inference instance associated with this production variant.
`shadowProductionVariants[].modelName`	string	—	The name of the model that you want to host.
`shadowProductionVariants[].routingConfig`	object	—	Settings that control how the endpoint routes incoming traffic to the instances that the endpoint hosts.
`shadowProductionVariants[].routingConfig.prefixAwareRoutingConfig`	object	—	The configuration for prefix-aware routing.
`shadowProductionVariants[].routingConfig.prefixAwareRoutingConfig.concurrencyThreshold`	integer	—	The maximum number of in-flight requests on the target instance before the endpoint routes to another instance.
`shadowProductionVariants[].routingConfig.prefixAwareRoutingConfig.prefixLength`	integer	—	The maximum length of the prefix used for routing decisions.
`shadowProductionVariants[].routingConfig.routingStrategy`	string	—	Sets how the endpoint routes incoming traffic.
`shadowProductionVariants[].serverlessConfig`	object	—	The serverless configuration for an endpoint.
`shadowProductionVariants[].serverlessConfig.maxConcurrency`	integer	yes	The maximum number of concurrent invocations your serverless endpoint can process.
`shadowProductionVariants[].serverlessConfig.memorySizeInMB`	integer	yes	The memory size of your serverless endpoint.
`shadowProductionVariants[].serverlessConfig.provisionedConcurrency`	integer	—	The amount of provisioned concurrency to allocate for the serverless endpoint.
`shadowProductionVariants[].variantInstanceProvisionTimeoutInSeconds`	integer	—	The timeout value, in seconds, for provisioning instances for the production variant.
`shadowProductionVariants[].variantName`	string	yes	The name of the production variant.
`shadowProductionVariants[].volumeSizeInGB`	integer	—	The size, in GB, of the ML storage volume attached to individual inference instance associated with the production variant.
`tags`	array<object>	—	A list of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes
`vpcConfig`	object	—	Specifies an Amazon Virtual Private Cloud (VPC) that your SageMaker jobs, hosted models, and compute resources have access to.
`vpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form sg-xxxxxxxx.
`vpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect your training job or model.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`endpointConfigArn`	string	—	The Amazon Resource Name (ARN) of the endpoint configuration.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerExperiment # CLOUD CONTROL · AWS::SageMaker::Experiment

SageMakerExperiment manages AWS::SageMaker::Experiment through the AWS Cloud Control API.

kubectl get sagemakerexperiments · example on GitHub

Example

# SageMakerExperiment — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerexperiment
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerExperiment
metadata:
  name: example-sagemakerexperiment
  namespace: default
spec:
  description: Managed by konfig-konector
  displayName: example-sagemakerexperiment
  experimentName: example-sagemakerexperiment
  providerRef:
    name: example-sagemakerexperiment
    region: us-east-1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`description`	string	—	The description of the experiment.
`displayName`	string	—	The name of the experiment as displayed.
`experimentName`	string	yes	The name of the experiment.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	A list of tags to associate with the experiment.
`tags[].key`	string	yes	The tag key.
`tags[].value`	string	yes	The tag value.

Status

Field	Type	Required	Description
`arn`	string	—	The Amazon Resource Name (ARN) of the experiment.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	When the experiment was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	When the experiment was last modified.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerFeatureGroup # CLOUD CONTROL · AWS::SageMaker::FeatureGroup

SageMakerFeatureGroup manages AWS::SageMaker::FeatureGroup through the AWS Cloud Control API.

kubectl get sagemakerfeaturegroups · example on GitHub

Example

# SageMakerFeatureGroup — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerfeaturegroup
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerFeatureGroup
metadata:
  name: example-sagemakerfeaturegroup
  namespace: default
spec:
  description: Managed by konfig-konector
  eventTimeFeatureName: example-sagemakerfeaturegroup
  featureDefinitions:
  - featureName: example-sagemakerfeaturegroup
    featureType: Integral
  featureGroupName: example-sagemakerfeaturegroup
  offlineStoreConfig:
    dataCatalogConfig:
      catalog: example
      database: example
      tableName: example-sagemakerfeaturegroup
    disableGlueTableCreation: false
    s3StorageConfig:
      kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
      s3Uri: example
    tableFormat: Iceberg
  onlineStoreConfig:
    enableOnlineStore: true
    securityConfig:
      kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    storageType: Standard
    ttlDuration:
      unit: Seconds
      value: 1
  providerRef:
    name: example-sagemakerfeaturegroup
    region: us-east-1
  recordIdentifierFeatureName: example-sagemakerfeaturegroup
  roleArn: arn:aws:iam::123456789012:role/example-role
  tags:
  - key: example
    value: example
  throughputConfig:
    provisionedReadCapacityUnits: 2
    provisionedWriteCapacityUnits: 2
    throughputMode: OnDemand

Spec

Field	Type	Required	Description
`description`	string	—	Description about the FeatureGroup.
`eventTimeFeatureName`	string	yes	The Event Time Feature Name.
`featureDefinitions`	array<object>	yes	An Array of Feature Definition
`featureDefinitions[].featureName`	string	yes
`featureDefinitions[].featureType`	enum	yes	Allowed: `Integral, Fractional, String`
`featureGroupName`	string	yes	The Name of the FeatureGroup.
`offlineStoreConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`offlineStoreConfig.dataCatalogConfig`	object	—	SageMakerFeatureGroupDataCatalogConfig is a nested property type of AWS::SageMaker::FeatureGroup.
`offlineStoreConfig.dataCatalogConfig.catalog`	string	yes
`offlineStoreConfig.dataCatalogConfig.database`	string	yes
`offlineStoreConfig.dataCatalogConfig.tableName`	string	yes
`offlineStoreConfig.disableGlueTableCreation`	boolean	—
`offlineStoreConfig.s3StorageConfig`	object	yes	SageMakerFeatureGroupS3StorageConfig is a nested property type of AWS::SageMaker::FeatureGroup.
`offlineStoreConfig.s3StorageConfig.kmsKeyId`	string	—
`offlineStoreConfig.s3StorageConfig.s3Uri`	string	yes
`offlineStoreConfig.tableFormat`	enum	—	Allowed: `Iceberg, Glue`
`onlineStoreConfig`	object	—	SageMakerFeatureGroupOnlineStoreConfig is a nested property type of AWS::SageMaker::FeatureGroup.
`onlineStoreConfig.enableOnlineStore`	boolean	—
`onlineStoreConfig.securityConfig`	object	—	SageMakerFeatureGroupOnlineStoreSecurityConfig is a nested property type of AWS::SageMaker::FeatureGroup.
`onlineStoreConfig.securityConfig.kmsKeyId`	string	—
`onlineStoreConfig.storageType`	enum	—	Allowed: `Standard, InMemory`
`onlineStoreConfig.ttlDuration`	object	—	SageMakerFeatureGroupTtlDuration is a nested property type of AWS::SageMaker::FeatureGroup.
`onlineStoreConfig.ttlDuration.unit`	enum	—	Allowed: `Seconds, Minutes, Hours, Days, Weeks`
`onlineStoreConfig.ttlDuration.value`	integer	—
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`recordIdentifierFeatureName`	string	yes	The Record Identifier Feature Name.
`roleArn`	string	—	Role Arn Create-only in AWS: changing it requires replacing the resource.
`tags`	array<object>	—	An array of key-value pair to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes
`throughputConfig`	object	—	SageMakerFeatureGroupThroughputConfig is a nested property type of AWS::SageMaker::FeatureGroup.
`throughputConfig.provisionedReadCapacityUnits`	integer	—	For provisioned feature groups with online store enabled, this indicates the read throughput you are billed for and can consume without throttling.
`throughputConfig.provisionedWriteCapacityUnits`	integer	—	For provisioned feature groups, this indicates the write throughput you are billed for and can consume without throttling.
`throughputConfig.throughputMode`	enum	yes	Allowed: `OnDemand, Provisioned`

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	A timestamp of FeatureGroup creation time.
`featureGroupStatus`	string	—	The status of the feature group.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerHub # CLOUD CONTROL · AWS::SageMaker::Hub

SageMakerHub manages AWS::SageMaker::Hub through the AWS Cloud Control API.

kubectl get sagemakerhubs · example on GitHub

Example

# SageMakerHub — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerhub
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerHub
metadata:
  name: example-sagemakerhub
  namespace: default
spec:
  hubDescription: Managed by konfig-konector
  hubDisplayName: example-sagemakerhub
  hubName: example-sagemakerhub
  hubSearchKeywords:
  - example
  providerRef:
    name: example-sagemakerhub
    region: us-east-1
  s3StorageConfig:
    s3OutputPath: example
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`hubDescription`	string	yes	A description of the hub.
`hubDisplayName`	string	—	The display name of the hub.
`hubName`	string	yes	The name of the hub.
`hubSearchKeywords`	array<string>	—	The searchable keywords for the hub.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`s3StorageConfig`	object	—	The Amazon S3 storage configuration for the hub.
`s3StorageConfig.s3OutputPath`	string	—	The Amazon S3 bucket prefix for hosting hub content.
`tags`	array<object>	—	Tags to associate with the hub.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The date and time that the hub was created.
`hubArn`	string	—	The Amazon Resource Name (ARN) of the hub.
`hubStatus`	enum	—	The status of the hub. Allowed: `InService, Creating, Updating, Deleting, CreateFailed, UpdateFailed, DeleteFailed`
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	The date and time that the hub was last modified.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerHumanTaskUi # CLOUD CONTROL · AWS::SageMaker::HumanTaskUi

SageMakerHumanTaskUi manages AWS::SageMaker::HumanTaskUi through the AWS Cloud Control API.

kubectl get sagemakerhumantaskuis · example on GitHub

Example

# SageMakerHumanTaskUi — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerhumantaskui
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerHumanTaskUi
metadata:
  name: example-sagemakerhumantaskui
  namespace: default
spec:
  humanTaskUiName: example-sagemakerhumantaskui
  providerRef:
    name: example-sagemakerhumantaskui
    region: us-east-1
  tags:
  - key: example
    value: example
  uiTemplate:
    content: example

Spec

Field	Type	Required	Description
`humanTaskUiName`	string	yes	The name of the human task user interface.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	An array of key-value pairs that contain metadata to help you categorize and organize a human review workflow user interface.
`tags[].key`	string	yes
`tags[].value`	string	yes
`uiTemplate`	object	—	The Liquid template for the worker user interface.
`uiTemplate.content`	string	yes	The content of the Liquid template for the worker user interface.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The timestamp when the human task user interface was created.
`humanTaskUiArn`	string	—	The Amazon Resource Name (ARN) of the human task user interface.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerImage # CLOUD CONTROL · AWS::SageMaker::Image

SageMakerImage manages AWS::SageMaker::Image through the AWS Cloud Control API.

kubectl get sagemakerimages · example on GitHub

Example

# SageMakerImage — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerimage
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerImage
metadata:
  name: example-sagemakerimage
  namespace: default
spec:
  imageDescription: Managed by konfig-konector
  imageDisplayName: example-sagemakerimage
  imageName: example-sagemakerimage
  imageRoleArn: arn:aws:iam::123456789012:role/example-role
  providerRef:
    name: example-sagemakerimage
    region: us-east-1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`imageDescription`	string	—
`imageDisplayName`	string	—
`imageName`	string	yes	Create-only in AWS: changing it requires replacing the resource.
`imageRoleArn`	string	yes
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`imageArn`	string	—	Read-only: reported by AWS after creation.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerImageVersion # CLOUD CONTROL · AWS::SageMaker::ImageVersion

SageMakerImageVersion manages AWS::SageMaker::ImageVersion through the AWS Cloud Control API.

kubectl get sagemakerimageversions · example on GitHub

Example

# SageMakerImageVersion — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerimageversion
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerImageVersion
metadata:
  name: example-sagemakerimageversion
  namespace: default
spec:
  alias: example
  aliases:
  - example
  baseImage: example
  horovod: false
  imageName: example-sagemakerimageversion
  jobType: TRAINING
  mlFramework: example
  processor: CPU
  programmingLang: example
  providerRef:
    name: example-sagemakerimageversion
    region: us-east-1
  releaseNotes: example
  vendorGuidance: NOT_PROVIDED

Spec

Field	Type	Required	Description
`alias`	string	—
`aliases`	array<string>	—
`baseImage`	string	yes	Create-only in AWS: changing it requires replacing the resource.
`horovod`	boolean	—
`imageName`	string	yes	Create-only in AWS: changing it requires replacing the resource.
`jobType`	enum	—	Allowed: `TRAINING, INFERENCE, NOTEBOOK_KERNEL`
`mlFramework`	string	—
`processor`	enum	—	Allowed: `CPU, GPU`
`programmingLang`	string	—
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`releaseNotes`	string	—
`vendorGuidance`	enum	—	Allowed: `NOT_PROVIDED, STABLE, TO_BE_ARCHIVED, ARCHIVED`

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`containerImage`	string	—	Read-only: reported by AWS after creation.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`imageArn`	string	—	Read-only: reported by AWS after creation.
`imageVersionArn`	string	—	Read-only: reported by AWS after creation.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`version`	integer	—	Read-only: reported by AWS after creation.

SageMakerInferenceComponent # CLOUD CONTROL · AWS::SageMaker::InferenceComponent

SageMakerInferenceComponent manages AWS::SageMaker::InferenceComponent through the AWS Cloud Control API.

kubectl get sagemakerinferencecomponents · example on GitHub

Example

# SageMakerInferenceComponent — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerinferencecomponent
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerInferenceComponent
metadata:
  name: example-sagemakerinferencecomponent
  namespace: default
spec:
  deploymentConfig:
    autoRollbackConfiguration:
      alarms:
      - alarmName: example-sagemakerinferencecomponent
    rollingUpdatePolicy:
      maximumBatchSize:
        type: COPY_COUNT
        value: 1
      maximumExecutionTimeoutInSeconds: 600
      rollbackMaximumBatchSize:
        type: COPY_COUNT
        value: 1
      waitIntervalInSeconds: 1
  endpointArn: arn:aws:iam::123456789012:role/example-role
  endpointName: example-sagemakerinferencecomponent
  inferenceComponentName: example-sagemakerinferencecomponent
  providerRef:
    name: example-sagemakerinferencecomponent
    region: us-east-1
  runtimeConfig:
    copyCount: 1
    currentCopyCount: 1
    desiredCopyCount: 1
    placementStatus:
    - currentCopyCount: 1
      instanceType: example
  specification:
    baseInferenceComponentName: example-sagemakerinferencecomponent
    computeResourceRequirements:
      maxMemoryRequiredInMb: 128
      minMemoryRequiredInMb: 128
      numberOfAcceleratorDevicesRequired: 1
      numberOfCpuCoresRequired: 1
    container:
      artifactUrl: example
      containerMetricsConfig:
        metricsEndpoints:
        - metricPublishFrequencyInSeconds: 10
          metricsEndpointPath: example
      deployedImage:
        resolutionTime: '2026-01-01T00:00:00Z'
        resolvedImage: example
        specifiedImage: example
      environment:
        key: value
      image: example
    currentDataCacheConfig:
      enableCaching: true
    dataCacheConfig:
      enableCaching: true
    modelName: example-sagemakerinferencecomponent
    schedulingConfig:
      availabilityZoneBalance:
        enforcementMode: PERMISSIVE
        maxImbalance: 1
      placementStrategy: SPREAD
    startupParameters:
      containerStartupHealthCheckTimeoutInSeconds: 60
      modelDataDownloadTimeoutInSeconds: 60
  specifications:
  - computeResourceRequirements:
      maxMemoryRequiredInMb: 128
      minMemoryRequiredInMb: 128
      numberOfAcceleratorDevicesRequired: 1
      numberOfCpuCoresRequired: 1
    container:
      artifactUrl: example
      image: example
    currentDataCacheConfig:
      enableCaching: true
    dataCacheConfig:
      enableCaching: true
    instanceType: example
    modelName: example-sagemakerinferencecomponent
    schedulingConfig:
      availabilityZoneBalance:
        enforcementMode: PERMISSIVE
        maxImbalance: 1
      placementStrategy: SPREAD
    startupParameters:
      containerStartupHealthCheckTimeoutInSeconds: 60
      modelDataDownloadTimeoutInSeconds: 60
  tags:
  - key: example
    value: example
  variantName: example-sagemakerinferencecomponent

Spec

Field	Type	Required	Description
`deploymentConfig`	object	—	SageMakerInferenceComponentInferenceComponentDeploymentConfig is a nested property type of AWS::SageMaker::InferenceComponent.
`deploymentConfig.autoRollbackConfiguration`	object	—	SageMakerInferenceComponentAutoRollbackConfiguration is a nested property type of AWS::SageMaker::InferenceComponent.
`deploymentConfig.autoRollbackConfiguration.alarms`	array<object>	yes
`deploymentConfig.autoRollbackConfiguration.alarms[].alarmName`	string	yes
`deploymentConfig.rollingUpdatePolicy`	object	—	SageMakerInferenceComponentInferenceComponentRollingUpdatePolicy is a nested property type of AWS::SageMaker::InferenceComponent.
`deploymentConfig.rollingUpdatePolicy.maximumBatchSize`	object	—	SageMakerInferenceComponentInferenceComponentCapacitySize is a nested property type of AWS::SageMaker::InferenceComponent.
`deploymentConfig.rollingUpdatePolicy.maximumBatchSize.type`	enum	yes	Allowed: `COPY_COUNT, CAPACITY_PERCENT`
`deploymentConfig.rollingUpdatePolicy.maximumBatchSize.value`	integer	yes
`deploymentConfig.rollingUpdatePolicy.maximumExecutionTimeoutInSeconds`	integer	—
`deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize`	object	—	SageMakerInferenceComponentInferenceComponentCapacitySize is a nested property type of AWS::SageMaker::InferenceComponent.
`deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize.type`	enum	yes	Allowed: `COPY_COUNT, CAPACITY_PERCENT`
`deploymentConfig.rollingUpdatePolicy.rollbackMaximumBatchSize.value`	integer	yes
`deploymentConfig.rollingUpdatePolicy.waitIntervalInSeconds`	integer	—
`endpointArn`	string	—
`endpointName`	string	yes
`inferenceComponentName`	string	—
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`runtimeConfig`	object	—	SageMakerInferenceComponentInferenceComponentRuntimeConfig is a nested property type of AWS::SageMaker::InferenceComponent.
`runtimeConfig.copyCount`	integer	—
`runtimeConfig.currentCopyCount`	integer	—
`runtimeConfig.desiredCopyCount`	integer	—
`runtimeConfig.placementStatus`	array<object>	—	The placement status of the inference component across instance types
`runtimeConfig.placementStatus[].currentCopyCount`	integer	yes
`runtimeConfig.placementStatus[].instanceType`	string	yes
`specification`	object	—	SageMakerInferenceComponentInferenceComponentSpecification is a nested property type of AWS::SageMaker::InferenceComponent.
`specification.baseInferenceComponentName`	string	—
`specification.computeResourceRequirements`	object	—	SageMakerInferenceComponentInferenceComponentComputeResourceRequirements is a nested property type of AWS::SageMaker::InferenceComponent.
`specification.computeResourceRequirements.maxMemoryRequiredInMb`	integer	—
`specification.computeResourceRequirements.minMemoryRequiredInMb`	integer	—
`specification.computeResourceRequirements.numberOfAcceleratorDevicesRequired`	number	—
`specification.computeResourceRequirements.numberOfCpuCoresRequired`	number	—
`specification.container`	object	—	SageMakerInferenceComponentInferenceComponentContainerSpecification is a nested property type of AWS::SageMaker::InferenceComponent.
`specification.container.artifactUrl`	string	—
`specification.container.containerMetricsConfig`	object	—	SageMakerInferenceComponentContainerMetricsConfig is a nested property type of AWS::SageMaker::InferenceComponent.
`specification.container.containerMetricsConfig.metricsEndpoints`	array<object>	yes
`specification.container.containerMetricsConfig.metricsEndpoints[].metricPublishFrequencyInSeconds`	integer	—	The interval, in seconds, at which container metrics scraped from the endpoint are published to Amazon CloudWatch.
`specification.container.containerMetricsConfig.metricsEndpoints[].metricsEndpointPath`	string	yes	The path to the Prometheus formatted metrics endpoint exposed by the container
`specification.container.deployedImage`	object	—	SageMakerInferenceComponentDeployedImage is a nested property type of AWS::SageMaker::InferenceComponent.
`specification.container.deployedImage.resolutionTime`	string	—
`specification.container.deployedImage.resolvedImage`	string	—
`specification.container.deployedImage.specifiedImage`	string	—
`specification.container.environment`	map<string>	—
`specification.container.image`	string	—
`specification.currentDataCacheConfig`	object	—	The data caching configuration actually in effect, including a value the service chose rather than the template: SageMaker enables caching automatically on instance types with more than 232 GiB of local NVMe storage, whether or not DataCacheConfig was set.
`specification.currentDataCacheConfig.enableCaching`	boolean	yes	Whether the endpoint caches the model artifacts and container image on each instance it provisions for the inference component
`specification.dataCacheConfig`	object	—	SageMakerInferenceComponentInferenceComponentDataCacheConfig is a nested property type of AWS::SageMaker::InferenceComponent.
`specification.dataCacheConfig.enableCaching`	boolean	yes	Whether the endpoint caches the model artifacts and container image on each instance it provisions for the inference component
`specification.modelName`	string	—
`specification.schedulingConfig`	object	—	SageMakerInferenceComponentInferenceComponentSchedulingConfig is a nested property type of AWS::SageMaker::InferenceComponent.
`specification.schedulingConfig.availabilityZoneBalance`	object	yes	SageMakerInferenceComponentInferenceComponentAvailabilityZoneBalance is a nested property type of AWS::SageMaker::InferenceComponent.
`specification.schedulingConfig.availabilityZoneBalance.enforcementMode`	enum	yes	Allowed: `PERMISSIVE`
`specification.schedulingConfig.availabilityZoneBalance.maxImbalance`	integer	—	The maximum allowed difference in the number of inference component copies between any two Availability Zones
`specification.schedulingConfig.placementStrategy`	enum	yes	Allowed: `SPREAD, BINPACK`
`specification.startupParameters`	object	—	SageMakerInferenceComponentInferenceComponentStartupParameters is a nested property type of AWS::SageMaker::InferenceComponent.
`specification.startupParameters.containerStartupHealthCheckTimeoutInSeconds`	integer	—
`specification.startupParameters.modelDataDownloadTimeoutInSeconds`	integer	—
`specifications`	array<object>	—
`specifications[].computeResourceRequirements`	object	—	SageMakerInferenceComponentInferenceComponentComputeResourceRequirements is a nested property type of AWS::SageMaker::InferenceComponent.
`specifications[].computeResourceRequirements.maxMemoryRequiredInMb`	integer	—
`specifications[].computeResourceRequirements.minMemoryRequiredInMb`	integer	—
`specifications[].computeResourceRequirements.numberOfAcceleratorDevicesRequired`	number	—
`specifications[].computeResourceRequirements.numberOfCpuCoresRequired`	number	—
`specifications[].container`	object	—	SageMakerInferenceComponentInferenceComponentContainerSpecificationForInstanceType is a nested property type of AWS::SageMaker::InferenceComponent.
`specifications[].container.artifactUrl`	string	—
`specifications[].container.containerMetricsConfig`	object	—	SageMakerInferenceComponentContainerMetricsConfig is a nested property type of AWS::SageMaker::InferenceComponent.
`specifications[].container.containerMetricsConfig.metricsEndpoints`	array<object>	yes
`specifications[].container.containerMetricsConfig.metricsEndpoints[].metricPublishFrequencyInSeconds`	integer	—	The interval, in seconds, at which container metrics scraped from the endpoint are published to Amazon CloudWatch.
`specifications[].container.containerMetricsConfig.metricsEndpoints[].metricsEndpointPath`	string	yes	The path to the Prometheus formatted metrics endpoint exposed by the container
`specifications[].container.environment`	map<string>	—
`specifications[].container.image`	string	—
`specifications[].currentDataCacheConfig`	object	—	The data caching configuration actually in effect for this instance type, including a value the service chose rather than the template: SageMaker enables caching automatically on instance types with more than 232 GiB of local NVMe storage, whether or not DataCacheConfig was set.
`specifications[].currentDataCacheConfig.enableCaching`	boolean	yes	Whether the endpoint caches the model artifacts and container image on each instance it provisions for the inference component
`specifications[].dataCacheConfig`	object	—	SageMakerInferenceComponentInferenceComponentDataCacheConfig is a nested property type of AWS::SageMaker::InferenceComponent.
`specifications[].dataCacheConfig.enableCaching`	boolean	yes	Whether the endpoint caches the model artifacts and container image on each instance it provisions for the inference component
`specifications[].instanceType`	string	yes
`specifications[].modelName`	string	—
`specifications[].schedulingConfig`	object	—	SageMakerInferenceComponentInferenceComponentSchedulingConfig is a nested property type of AWS::SageMaker::InferenceComponent.
`specifications[].schedulingConfig.availabilityZoneBalance`	object	yes	SageMakerInferenceComponentInferenceComponentAvailabilityZoneBalance is a nested property type of AWS::SageMaker::InferenceComponent.
`specifications[].schedulingConfig.availabilityZoneBalance.enforcementMode`	enum	yes	Allowed: `PERMISSIVE`
`specifications[].schedulingConfig.availabilityZoneBalance.maxImbalance`	integer	—	The maximum allowed difference in the number of inference component copies between any two Availability Zones
`specifications[].schedulingConfig.placementStrategy`	enum	yes	Allowed: `SPREAD, BINPACK`
`specifications[].startupParameters`	object	—	SageMakerInferenceComponentInferenceComponentStartupParameters is a nested property type of AWS::SageMaker::InferenceComponent.
`specifications[].startupParameters.containerStartupHealthCheckTimeoutInSeconds`	integer	—
`specifications[].startupParameters.modelDataDownloadTimeoutInSeconds`	integer	—
`tags`	array<object>	—
`tags[].key`	string	yes
`tags[].value`	string	yes
`variantName`	string	—

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	Read-only: reported by AWS after creation.
`failureReason`	string	—	Read-only: reported by AWS after creation.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`inferenceComponentArn`	string	—	Read-only: reported by AWS after creation.
`inferenceComponentStatus`	enum	—	Read-only: reported by AWS after creation. Allowed: `InService, Creating, Updating, Failed, Deleting`
`lastModifiedTime`	string	—	Read-only: reported by AWS after creation.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerInferenceExperiment # CLOUD CONTROL · AWS::SageMaker::InferenceExperiment

SageMakerInferenceExperiment manages AWS::SageMaker::InferenceExperiment through the AWS Cloud Control API.

kubectl get sagemakerinferenceexperiments · example on GitHub

Example

# SageMakerInferenceExperiment — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerinferenceexperiment
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerInferenceExperiment
metadata:
  name: example-sagemakerinferenceexperiment
  namespace: default
spec:
  dataStorageConfig:
    contentType:
      csvContentTypes:
      - example
      jsonContentTypes:
      - example
    destination: example
    kmsKey: example
  description: Managed by konfig-konector
  desiredState: Running
  endpointName: example-sagemakerinferenceexperiment
  kmsKey: example
  modelVariants:
  - infrastructureConfig:
      infrastructureType: RealTimeInference
      realTimeInferenceConfig:
        instanceCount: 2
        instanceType: example
    modelName: example-sagemakerinferenceexperiment
    variantName: example-sagemakerinferenceexperiment
  name: example-sagemakerinferenceexperiment
  providerRef:
    name: example-sagemakerinferenceexperiment
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  schedule:
    endTime: '2026-01-01T00:00:00Z'
    startTime: '2026-01-01T00:00:00Z'
  shadowModeConfig:
    shadowModelVariants:
    - samplingPercentage: 1
      shadowModelVariantName: example-sagemakerinferenceexperiment
    sourceModelVariantName: example-sagemakerinferenceexperiment
  statusReason: example
  tags:
  - key: example
    value: example
  type: ShadowMode

Spec

Field	Type	Required	Description
`dataStorageConfig`	object	—	SageMakerInferenceExperimentDataStorageConfig is a nested property type of AWS::SageMaker::InferenceExperiment.
`dataStorageConfig.contentType`	object	—	SageMakerInferenceExperimentCaptureContentTypeHeader is a nested property type of AWS::SageMaker::InferenceExperiment.
`dataStorageConfig.contentType.csvContentTypes`	array<string>	—	The list of all content type headers that SageMaker will treat as CSV and capture accordingly.
`dataStorageConfig.contentType.jsonContentTypes`	array<string>	—	The list of all content type headers that SageMaker will treat as JSON and capture accordingly.
`dataStorageConfig.destination`	string	yes	The Amazon S3 bucket where the inference request and response data is stored.
`dataStorageConfig.kmsKey`	string	—	The AWS Key Management Service key that Amazon SageMaker uses to encrypt captured data at rest using Amazon S3 server-side encryption.
`description`	string	—	The description of the inference experiment.
`desiredState`	enum	—	The desired state of the experiment after starting or stopping operation. Allowed: `Running, Completed, Cancelled`
`endpointName`	string	yes	Create-only in AWS: changing it requires replacing the resource.
`kmsKey`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt data on the storage volume attached to the ML compute instance that hosts the endpoint.
`modelVariants`	array<object>	yes	An array of ModelVariantConfig objects.
`modelVariants[].infrastructureConfig`	object	yes	SageMakerInferenceExperimentModelInfrastructureConfig is a nested property type of AWS::SageMaker::InferenceExperiment.
`modelVariants[].infrastructureConfig.infrastructureType`	enum	yes	The type of the inference experiment that you want to run. Allowed: `RealTimeInference`
`modelVariants[].infrastructureConfig.realTimeInferenceConfig`	object	yes	SageMakerInferenceExperimentRealTimeInferenceConfig is a nested property type of AWS::SageMaker::InferenceExperiment.
`modelVariants[].infrastructureConfig.realTimeInferenceConfig.instanceCount`	integer	yes	The number of instances of the type specified by InstanceType.
`modelVariants[].infrastructureConfig.realTimeInferenceConfig.instanceType`	string	yes	The instance type the model is deployed to.
`modelVariants[].modelName`	string	yes	The name of the Amazon SageMaker Model entity.
`modelVariants[].variantName`	string	yes	The name of the variant.
`name`	string	yes	The name for the inference experiment.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker can assume to access model artifacts and container images, and manage Amazon SageMaker Inference endpoints for model deployment.
`schedule`	object	—	SageMakerInferenceExperimentInferenceExperimentSchedule is a nested property type of AWS::SageMaker::InferenceExperiment.
`schedule.endTime`	string	—	The timestamp at which the inference experiment ended or will end.
`schedule.startTime`	string	—	The timestamp at which the inference experiment started or will start.
`shadowModeConfig`	object	—	SageMakerInferenceExperimentShadowModeConfig is a nested property type of AWS::SageMaker::InferenceExperiment.
`shadowModeConfig.shadowModelVariants`	array<object>	yes	List of shadow variant configurations.
`shadowModeConfig.shadowModelVariants[].samplingPercentage`	integer	yes	The percentage of inference requests that Amazon SageMaker replicates from the production variant to the shadow variant.
`shadowModeConfig.shadowModelVariants[].shadowModelVariantName`	string	yes	The name of the shadow variant.
`shadowModeConfig.sourceModelVariantName`	string	yes	The name of the production variant, which takes all the inference requests.
`statusReason`	string	—	The error message or client-specified reason from the StopInferenceExperiment API, that explains the status of the inference experiment.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes
`type`	enum	yes	The type of the inference experiment that you want to run. Allowed: `ShadowMode`

Status

Field	Type	Required	Description
`arn`	string	—	The Amazon Resource Name (ARN) of the inference experiment.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The timestamp at which you created the inference experiment.
`endpointMetadata`	object	—	Read-only: reported by AWS after creation.
`endpointMetadata.endpointConfigName`	string	—	The name of the endpoint configuration.
`endpointMetadata.endpointName`	string	yes
`endpointMetadata.endpointStatus`	enum	—	The status of the endpoint. Allowed: `Creating, Updating, SystemUpdating, RollingBack, InService, OutOfService, Deleting, Failed`
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	The timestamp at which you last modified the inference experiment.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`status`	enum	—	The status of the inference experiment. Allowed: `Creating, Created, Updating, Starting, Stopping, Running, Completed, Cancelled`

SageMakerMlflowApp # CLOUD CONTROL · AWS::SageMaker::MlflowApp

SageMakerMlflowApp manages AWS::SageMaker::MlflowApp through the AWS Cloud Control API.

kubectl get sagemakermlflowapps · example on GitHub

Example

# SageMakerMlflowApp — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermlflowapp
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerMlflowApp
metadata:
  name: example-sagemakermlflowapp
  namespace: default
spec:
  artifactStoreUri: example
  modelRegistrationMode: AutoModelRegistrationEnabled
  name: example-sagemakermlflowapp
  providerRef:
    name: example-sagemakermlflowapp
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  tags:
  - key: example
    value: example
  weeklyMaintenanceWindowStart: example

Spec

Field	Type	Required	Description
`artifactStoreUri`	string	yes	The S3 URI for a general purpose bucket to use as the MLflow App artifact store.
`modelRegistrationMode`	enum	—	Whether to enable or disable automatic registration of new MLflow models to the SageMaker Model Registry. Allowed: `AutoModelRegistrationEnabled, AutoModelRegistrationDisabled`
`name`	string	yes	The name of the MLflow App.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	The Amazon Resource Name (ARN) for an IAM role in your account that the MLflow App uses to access the artifact store in Amazon S3.
`tags`	array<object>	—	Tags to associate with the MLflow App.
`tags[].key`	string	yes
`tags[].value`	string	yes
`weeklyMaintenanceWindowStart`	string	—	The day and time of the week in Coordinated Universal Time (UTC) 24-hour standard time that weekly maintenance updates are scheduled.

Status

Field	Type	Required	Description
`arn`	string	—	The Amazon Resource Name (ARN) of the MLflow App.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The date and time that the MLflow App was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	The date and time that the MLflow App was last modified.
`lastSyncTime`	string	—
`mlflowAppId`	string	—	The server-generated identifier of the MLflow App.
`mlflowVersion`	string	—	The MLflow version used by the MLflow App.
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`status`	enum	—	The status of the MLflow App. Allowed: `Creating, Created, CreateFailed, Updating, Updated, UpdateFailed, Deleting, DeleteFailed…`

SageMakerMlflowTrackingServer # CLOUD CONTROL · AWS::SageMaker::MlflowTrackingServer

SageMakerMlflowTrackingServer manages AWS::SageMaker::MlflowTrackingServer through the AWS Cloud Control API.

kubectl get sagemakermlflowtrackingservers · example on GitHub

Example

# SageMakerMlflowTrackingServer — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermlflowtrackingserver
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerMlflowTrackingServer
metadata:
  name: example-sagemakermlflowtrackingserver
  namespace: default
spec:
  artifactStoreUri: example
  automaticModelRegistration: false
  mlflowVersion: '1.0'
  providerRef:
    name: example-sagemakermlflowtrackingserver
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  tags:
  - key: example
    value: example
  trackingServerName: example-sagemakermlflowtrackingserver
  trackingServerSize: Small
  weeklyMaintenanceWindowStart: example

Spec

Field	Type	Required	Description
`artifactStoreUri`	string	yes	The Amazon S3 URI for MLFlow Tracking Server artifacts.
`automaticModelRegistration`	boolean	—	A flag to enable Automatic SageMaker Model Registration.
`mlflowVersion`	string	—	The MLFlow Version used on the MLFlow Tracking Server.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	The Amazon Resource Name (ARN) of an IAM role that enables Amazon SageMaker to perform tasks on behalf of the customer.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes
`trackingServerName`	string	yes	The name of the MLFlow Tracking Server.
`trackingServerSize`	enum	—	The size of the MLFlow Tracking Server. Allowed: `Small, Medium, Large`
`weeklyMaintenanceWindowStart`	string	—	The start of the time window for maintenance of the MLFlow Tracking Server in UTC time.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`trackingServerArn`	string	—	The Amazon Resource Name (ARN) of the MLFlow Tracking Server.

SageMakerModel # CLOUD CONTROL · AWS::SageMaker::Model

SageMakerModel manages AWS::SageMaker::Model through the AWS Cloud Control API.

kubectl get sagemakermodels · example on GitHub

Example

# SageMakerModel — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermodel
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerModel
metadata:
  name: example-sagemakermodel
  namespace: default
spec:
  containers:
  - containerHostname: example-sagemakermodel
    image: example
    imageConfig:
      repositoryAccessMode: Platform
    inferenceSpecificationName: example-sagemakermodel
    mode: SingleModel
    modelDataSource:
      s3DataSource:
        compressionType: None
        s3DataType: S3Prefix
        s3Uri: example
    modelDataUrl: example
    modelPackageName: example-sagemakermodel
    multiModelConfig:
      modelCacheSetting: Enabled
  enableNetworkIsolation: true
  executionRoleArn: arn:aws:iam::123456789012:role/example-role
  inferenceExecutionConfig:
    mode: Serial
  modelName: example-sagemakermodel
  primaryContainer:
    containerHostname: example-sagemakermodel
    image: example
    imageConfig:
      repositoryAccessMode: Platform
      repositoryAuthConfig:
        repositoryCredentialsProviderArn: arn:aws:iam::123456789012:role/example-role
    inferenceSpecificationName: example-sagemakermodel
    mode: SingleModel
    modelDataSource:
      s3DataSource:
        compressionType: None
        s3DataType: S3Prefix
        s3Uri: example
    modelDataUrl: example
    modelPackageName: example-sagemakermodel
    multiModelConfig:
      modelCacheSetting: Enabled
  providerRef:
    name: example-sagemakermodel
    region: us-east-1
  tags:
  - key: example
    value: example
  vpcConfig:
    securityGroupIds:
    - example
    subnets:
    - example

Spec

Field	Type	Required	Description
`containers`	array<object>	—	Specifies the containers in the inference pipeline.
`containers[].containerHostname`	string	—	This parameter is ignored for models that contain only a PrimaryContainer.
`containers[].environment`	object	—	The environment variables to set in the Docker container.
`containers[].image`	string	—	The path where inference code is stored.
`containers[].imageConfig`	object	—	SageMakerModelImageConfig is a nested property type of AWS::SageMaker::Model.
`containers[].imageConfig.repositoryAccessMode`	enum	yes	Set this to one of the following values: Platform - The model image is hosted in Amazon ECR. Allowed: `Platform, Vpc`
`containers[].imageConfig.repositoryAuthConfig`	object	—	SageMakerModelRepositoryAuthConfig is a nested property type of AWS::SageMaker::Model.
`containers[].imageConfig.repositoryAuthConfig.repositoryCredentialsProviderArn`	string	yes	The Amazon Resource Name (ARN) of an AWS Lambda function that provides credentials to authenticate to the private Docker registry where your model image is hosted.
`containers[].inferenceSpecificationName`	string	—	The inference specification name in the model package version.
`containers[].mode`	enum	—	Whether the container hosts a single model or multiple models. Allowed: `SingleModel, MultiModel`
`containers[].modelDataSource`	object	—	SageMakerModelModelDataSource is a nested property type of AWS::SageMaker::Model.
`containers[].modelDataSource.s3DataSource`	object	yes	SageMakerModelS3DataSource is a nested property type of AWS::SageMaker::Model.
`containers[].modelDataSource.s3DataSource.compressionType`	enum	yes	Specifies how the ML model data is prepared. Allowed: `None, Gzip`
`containers[].modelDataSource.s3DataSource.hubAccessConfig`	object	—	SageMakerModelHubAccessConfig is a nested property type of AWS::SageMaker::Model.
`containers[].modelDataSource.s3DataSource.hubAccessConfig.hubContentArn`	string	yes	The ARN of the hub content for which deployment access is allowed.
`containers[].modelDataSource.s3DataSource.modelAccessConfig`	object	—	SageMakerModelModelAccessConfig is a nested property type of AWS::SageMaker::Model.
`containers[].modelDataSource.s3DataSource.modelAccessConfig.acceptEula`	boolean	yes	Specifies agreement to the model end-user license agreement (EULA).
`containers[].modelDataSource.s3DataSource.s3DataType`	enum	yes	Specifies the type of ML model data to deploy. Allowed: `S3Prefix, S3Object`
`containers[].modelDataSource.s3DataSource.s3Uri`	string	yes	Specifies the S3 path of ML model data to deploy.
`containers[].modelDataUrl`	string	—	The S3 path where the model artifacts, which result from model training, are stored.
`containers[].modelPackageName`	string	—	The name or Amazon Resource Name (ARN) of the model package to use to create the model.
`containers[].multiModelConfig`	object	—	SageMakerModelMultiModelConfig is a nested property type of AWS::SageMaker::Model.
`containers[].multiModelConfig.modelCacheSetting`	enum	—	Whether to cache models for a multi-model endpoint. Allowed: `Enabled, Disabled`
`enableNetworkIsolation`	boolean	—	Isolates the model container.
`executionRoleArn`	string	—	The Amazon Resource Name (ARN) of the IAM role that you specified for the model.
`inferenceExecutionConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`inferenceExecutionConfig.mode`	enum	yes	How containers in a multi-container are run. Allowed: `Serial, Direct`
`modelName`	string	—	The name of the new model.
`primaryContainer`	object	—	Create-only in AWS: changing it requires replacing the resource.
`primaryContainer.containerHostname`	string	—	This parameter is ignored for models that contain only a PrimaryContainer.
`primaryContainer.environment`	object	—	The environment variables to set in the Docker container.
`primaryContainer.image`	string	—	The path where inference code is stored.
`primaryContainer.imageConfig`	object	—	SageMakerModelImageConfig is a nested property type of AWS::SageMaker::Model.
`primaryContainer.imageConfig.repositoryAccessMode`	enum	yes	Set this to one of the following values: Platform - The model image is hosted in Amazon ECR. Allowed: `Platform, Vpc`
`primaryContainer.imageConfig.repositoryAuthConfig`	object	—	SageMakerModelRepositoryAuthConfig is a nested property type of AWS::SageMaker::Model.
`primaryContainer.imageConfig.repositoryAuthConfig.repositoryCredentialsProviderArn`	string	yes	The Amazon Resource Name (ARN) of an AWS Lambda function that provides credentials to authenticate to the private Docker registry where your model image is hosted.
`primaryContainer.inferenceSpecificationName`	string	—	The inference specification name in the model package version.
`primaryContainer.mode`	enum	—	Whether the container hosts a single model or multiple models. Allowed: `SingleModel, MultiModel`
`primaryContainer.modelDataSource`	object	—	SageMakerModelModelDataSource is a nested property type of AWS::SageMaker::Model.
`primaryContainer.modelDataSource.s3DataSource`	object	yes	SageMakerModelS3DataSource is a nested property type of AWS::SageMaker::Model.
`primaryContainer.modelDataSource.s3DataSource.compressionType`	enum	yes	Specifies how the ML model data is prepared. Allowed: `None, Gzip`
`primaryContainer.modelDataSource.s3DataSource.hubAccessConfig`	object	—	SageMakerModelHubAccessConfig is a nested property type of AWS::SageMaker::Model.
`primaryContainer.modelDataSource.s3DataSource.hubAccessConfig.hubContentArn`	string	yes	The ARN of the hub content for which deployment access is allowed.
`primaryContainer.modelDataSource.s3DataSource.modelAccessConfig`	object	—	SageMakerModelModelAccessConfig is a nested property type of AWS::SageMaker::Model.
`primaryContainer.modelDataSource.s3DataSource.modelAccessConfig.acceptEula`	boolean	yes	Specifies agreement to the model end-user license agreement (EULA).
`primaryContainer.modelDataSource.s3DataSource.s3DataType`	enum	yes	Specifies the type of ML model data to deploy. Allowed: `S3Prefix, S3Object`
`primaryContainer.modelDataSource.s3DataSource.s3Uri`	string	yes	Specifies the S3 path of ML model data to deploy.
`primaryContainer.modelDataUrl`	string	—	The S3 path where the model artifacts, which result from model training, are stored.
`primaryContainer.modelPackageName`	string	—	The name or Amazon Resource Name (ARN) of the model package to use to create the model.
`primaryContainer.multiModelConfig`	object	—	SageMakerModelMultiModelConfig is a nested property type of AWS::SageMaker::Model.
`primaryContainer.multiModelConfig.modelCacheSetting`	enum	—	Whether to cache models for a multi-model endpoint. Allowed: `Enabled, Disabled`
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	An array of key-value pairs.
`tags[].key`	string	yes
`tags[].value`	string	yes
`vpcConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`vpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form `sg-xxxxxxxx`.
`vpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect your training job or model.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`modelArn`	string	—	The Amazon Resource Name (ARN) of the model.
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerModelBiasJobDefinition # CLOUD CONTROL · AWS::SageMaker::ModelBiasJobDefinition

SageMakerModelBiasJobDefinition manages AWS::SageMaker::ModelBiasJobDefinition through the AWS Cloud Control API.

kubectl get sagemakermodelbiasjobdefinitions · example on GitHub

Example

# SageMakerModelBiasJobDefinition — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermodelbiasjobdefinition
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerModelBiasJobDefinition
metadata:
  name: example-sagemakermodelbiasjobdefinition
  namespace: default
spec:
  endpointName: example-sagemakermodelbiasjobdefinition
  jobDefinitionName: example-sagemakermodelbiasjobdefinition
  jobResources:
    clusterConfig:
      instanceCount: 1
      instanceType: example
      volumeKmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
      volumeSizeInGB: 1
  modelBiasAppSpecification:
    configUri: example
    environment:
      key: value
    imageUri: example
  modelBiasBaselineConfig:
    baseliningJobName: example-sagemakermodelbiasjobdefinition
    constraintsResource:
      s3Uri: example
  modelBiasJobInput:
    batchTransformInput:
      dataCapturedDestinationS3Uri: example
      datasetFormat:
        parquet: false
      endTimeOffset: 2026-01-01T00:0
      featuresAttribute: example
      inferenceAttribute: example
      localPath: example
      probabilityAttribute: example
      probabilityThresholdAttribute: 1
      s3DataDistributionType: FullyReplicated
      s3InputMode: Pipe
      startTimeOffset: 2026-01-01T00:0
    endpointInput:
      endTimeOffset: 2026-01-01T00:0
      endpointName: example-sagemakermodelbiasjobdefinition
      featuresAttribute: example
      inferenceAttribute: example
      localPath: example
      probabilityAttribute: example
      probabilityThresholdAttribute: 1
      s3DataDistributionType: FullyReplicated
      s3InputMode: Pipe
      startTimeOffset: 2026-01-01T00:0
    groundTruthS3Input:
      s3Uri: example
  modelBiasJobOutputConfig:
    kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    monitoringOutputs:
    - s3Output:
        localPath: example
        s3UploadMode: Continuous
        s3Uri: example
  networkConfig:
    enableInterContainerTrafficEncryption: true
    enableNetworkIsolation: true
    vpcConfig:
      securityGroupIds:
      - example
      subnets:
      - example
  providerRef:
    name: example-sagemakermodelbiasjobdefinition
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  stoppingCondition:
    maxRuntimeInSeconds: 1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`endpointName`	string	—	Create-only in AWS: changing it requires replacing the resource.
`jobDefinitionName`	string	—	Create-only in AWS: changing it requires replacing the resource.
`jobResources`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`jobResources.clusterConfig`	object	yes	SageMakerModelBiasJobDefinitionClusterConfig is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`jobResources.clusterConfig.instanceCount`	integer	yes	The number of ML compute instances to use in the model monitoring job.
`jobResources.clusterConfig.instanceType`	string	yes	The ML compute instance type for the processing job.
`jobResources.clusterConfig.volumeKmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt data on the storage volume attached to the ML compute instance(s) that run the model monitoring job.
`jobResources.clusterConfig.volumeSizeInGB`	integer	yes	The size of the ML storage volume, in gigabytes, that you want to provision.
`modelBiasAppSpecification`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`modelBiasAppSpecification.configUri`	string	yes	The S3 URI to an analysis configuration file
`modelBiasAppSpecification.environment`	map<string>	—	Sets the environment variables in the Docker container
`modelBiasAppSpecification.imageUri`	string	yes	The container image to be run by the monitoring job.
`modelBiasBaselineConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`modelBiasBaselineConfig.baseliningJobName`	string	—
`modelBiasBaselineConfig.constraintsResource`	object	—	SageMakerModelBiasJobDefinitionConstraintsResource is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`modelBiasBaselineConfig.constraintsResource.s3Uri`	string	—	The Amazon S3 URI for baseline constraint file in Amazon S3 that the current monitoring job should validated against.
`modelBiasJobInput`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`modelBiasJobInput.batchTransformInput`	object	—	SageMakerModelBiasJobDefinitionBatchTransformInput is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`modelBiasJobInput.batchTransformInput.dataCapturedDestinationS3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Batch Transform Job captures data.
`modelBiasJobInput.batchTransformInput.datasetFormat`	object	yes	SageMakerModelBiasJobDefinitionDatasetFormat is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`modelBiasJobInput.batchTransformInput.datasetFormat.csv`	object	—	SageMakerModelBiasJobDefinitionCsv is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`modelBiasJobInput.batchTransformInput.datasetFormat.csv.header`	boolean	—	A boolean flag indicating if given CSV has header
`modelBiasJobInput.batchTransformInput.datasetFormat.json`	object	—	SageMakerModelBiasJobDefinitionJson is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`modelBiasJobInput.batchTransformInput.datasetFormat.json.line`	boolean	—	A boolean flag indicating if it is JSON line format
`modelBiasJobInput.batchTransformInput.datasetFormat.parquet`	boolean	—
`modelBiasJobInput.batchTransformInput.endTimeOffset`	string	—	Monitoring end time offset, e.g.
`modelBiasJobInput.batchTransformInput.featuresAttribute`	string	—	JSONpath to locate features in JSONlines dataset
`modelBiasJobInput.batchTransformInput.inferenceAttribute`	string	—	Index or JSONpath to locate predicted label(s)
`modelBiasJobInput.batchTransformInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`modelBiasJobInput.batchTransformInput.probabilityAttribute`	string	—	Index or JSONpath to locate probabilities
`modelBiasJobInput.batchTransformInput.probabilityThresholdAttribute`	number	—
`modelBiasJobInput.batchTransformInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`modelBiasJobInput.batchTransformInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`modelBiasJobInput.batchTransformInput.startTimeOffset`	string	—	Monitoring start time offset, e.g.
`modelBiasJobInput.endpointInput`	object	—	SageMakerModelBiasJobDefinitionEndpointInput is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`modelBiasJobInput.endpointInput.endTimeOffset`	string	—	Monitoring end time offset, e.g.
`modelBiasJobInput.endpointInput.endpointName`	string	yes
`modelBiasJobInput.endpointInput.featuresAttribute`	string	—	JSONpath to locate features in JSONlines dataset
`modelBiasJobInput.endpointInput.inferenceAttribute`	string	—	Index or JSONpath to locate predicted label(s)
`modelBiasJobInput.endpointInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`modelBiasJobInput.endpointInput.probabilityAttribute`	string	—	Index or JSONpath to locate probabilities
`modelBiasJobInput.endpointInput.probabilityThresholdAttribute`	number	—
`modelBiasJobInput.endpointInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`modelBiasJobInput.endpointInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`modelBiasJobInput.endpointInput.startTimeOffset`	string	—	Monitoring start time offset, e.g.
`modelBiasJobInput.groundTruthS3Input`	object	yes	SageMakerModelBiasJobDefinitionMonitoringGroundTruthS3Input is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`modelBiasJobInput.groundTruthS3Input.s3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`modelBiasJobOutputConfig`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`modelBiasJobOutputConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt the model artifacts at rest using Amazon S3 server-side encryption.
`modelBiasJobOutputConfig.monitoringOutputs`	array<object>	yes	Monitoring outputs for monitoring jobs.
`modelBiasJobOutputConfig.monitoringOutputs[].s3Output`	object	yes	SageMakerModelBiasJobDefinitionS3Output is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`modelBiasJobOutputConfig.monitoringOutputs[].s3Output.localPath`	string	yes	The local path to the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`modelBiasJobOutputConfig.monitoringOutputs[].s3Output.s3UploadMode`	enum	—	Whether to upload the results of the monitoring job continuously or after the job completes. Allowed: `Continuous, EndOfJob`
`modelBiasJobOutputConfig.monitoringOutputs[].s3Output.s3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`networkConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`networkConfig.enableInterContainerTrafficEncryption`	boolean	—	Whether to encrypt all communications between distributed processing jobs.
`networkConfig.enableNetworkIsolation`	boolean	—	Whether to allow inbound and outbound network calls to and from the containers used for the processing job.
`networkConfig.vpcConfig`	object	—	SageMakerModelBiasJobDefinitionVpcConfig is a nested property type of AWS::SageMaker::ModelBiasJobDefinition.
`networkConfig.vpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form sg-xxxxxxxx.
`networkConfig.vpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect to your monitoring jobs.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker can assume to perform tasks on your behalf.
`stoppingCondition`	object	—	Create-only in AWS: changing it requires replacing the resource.
`stoppingCondition.maxRuntimeInSeconds`	integer	yes	The maximum runtime allowed in seconds.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The time at which the job definition was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`jobDefinitionArn`	string	—	The Amazon Resource Name (ARN) of job definition.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerModelCard # CLOUD CONTROL · AWS::SageMaker::ModelCard

SageMakerModelCard manages AWS::SageMaker::ModelCard through the AWS Cloud Control API.

kubectl get sagemakermodelcards · example on GitHub

Example

# SageMakerModelCard — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermodelcard
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerModelCard
metadata:
  name: example-sagemakermodelcard
  namespace: default
spec:
  content:
    additionalInformation:
      caveatsAndRecommendations: example
      customDetails:
        key: value
      ethicalConsiderations: example
    businessDetails:
      businessProblem: example
      businessStakeholders: example
      lineOfBusiness: example
    evaluationDetails:
    - datasets:
      - example
      evaluationJobArn: arn:aws:iam::123456789012:role/example-role
      evaluationObservation: example
      metricGroups:
      - name: example-sagemakermodelcard
      name: example-sagemakermodelcard
    intendedUses:
      explanationsForRiskRating: example
      factorsAffectingModelEfficiency: example
      intendedUses: example
      purposeOfModel: example
      riskRating: High
    modelOverview:
      algorithmType: example
      inferenceEnvironment:
        containerImage:
        - example
      modelArtifact:
      - example
      modelCreator: example
      modelDescription: Managed by konfig-konector
      modelName: example-sagemakermodelcard
      modelOwner: example
      modelVersion: 1
      problemType: example
    modelPackageDetails:
      approvalDescription: Managed by konfig-konector
      createdBy:
        userProfileName: example-sagemakermodelcard
      domain: example
      inferenceSpecification:
        containers:
        - image: example
          modelDataUrl: example
          nearestModelName: example-sagemakermodelcard
      modelApprovalStatus: Approved
      modelPackageArn: arn:aws:iam::123456789012:role/example-role
      modelPackageDescription: Managed by konfig-konector
      modelPackageGroupName: example-sagemakermodelcard
      modelPackageName: example-sagemakermodelcard
      modelPackageStatus: Pending
      modelPackageVersion: 1
      sourceAlgorithms:
      - algorithmName: example-sagemakermodelcard
        modelDataUrl: example
      task: example
    trainingDetails:
      objectiveFunction:
        notes: example
      trainingJobDetails:
        hyperParameters:
        - name: example-sagemakermodelcard
          value: example
        trainingArn: arn:aws:iam::123456789012:role/example-role
        trainingDatasets:
        - example
        trainingMetrics:
        - name: example-sagemakermodelcard
          notes: example
          value: 1
        userProvidedHyperParameters:
        - name: example-sagemakermodelcard
          value: example
        userProvidedTrainingMetrics:
        - name: example-sagemakermodelcard
          notes: example
          value: 1
      trainingObservations: example
  createdBy:
    userProfileArn: arn:aws:iam::123456789012:role/example-role
    userProfileName: example-sagemakermodelcard
  lastModifiedBy:
    userProfileArn: arn:aws:iam::123456789012:role/example-role
    userProfileName: example-sagemakermodelcard
  modelCardName: example-sagemakermodelcard
  modelCardStatus: Draft
  providerRef:
    name: example-sagemakermodelcard
    region: us-east-1
  securityConfig:
    kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`content`	object	yes	SageMakerModelCardContent is a nested property type of AWS::SageMaker::ModelCard.
`content.additionalInformation`	object	—	SageMakerModelCardAdditionalInformation is a nested property type of AWS::SageMaker::ModelCard.
`content.additionalInformation.caveatsAndRecommendations`	string	—	Caveats and recommendations for people who might use this model in their applications.
`content.additionalInformation.customDetails`	map<string>	—	customer details.
`content.additionalInformation.ethicalConsiderations`	string	—	Any ethical considerations that the author wants to provide.
`content.businessDetails`	object	—	SageMakerModelCardBusinessDetails is a nested property type of AWS::SageMaker::ModelCard.
`content.businessDetails.businessProblem`	string	—	What business problem does the model solve?
`content.businessDetails.businessStakeholders`	string	—	Business stakeholders.
`content.businessDetails.lineOfBusiness`	string	—	Line of business.
`content.evaluationDetails`	array<object>	—
`content.evaluationDetails[].datasets`	array<string>	—
`content.evaluationDetails[].evaluationJobArn`	string	—
`content.evaluationDetails[].evaluationObservation`	string	—
`content.evaluationDetails[].metadata`	map<string>	—	additional attributes associated with the evaluation results.
`content.evaluationDetails[].metricGroups`	array<object>	—
`content.evaluationDetails[].metricGroups[].metricData`	array<object>	yes
`content.evaluationDetails[].metricGroups[].name`	string	yes
`content.evaluationDetails[].name`	string	yes
`content.intendedUses`	object	—	SageMakerModelCardIntendedUses is a nested property type of AWS::SageMaker::ModelCard.
`content.intendedUses.explanationsForRiskRating`	string	—
`content.intendedUses.factorsAffectingModelEfficiency`	string	—
`content.intendedUses.intendedUses`	string	—	intended use cases.
`content.intendedUses.purposeOfModel`	string	—	Why the model was developed?
`content.intendedUses.riskRating`	enum	—	Allowed: `High, Medium, Low, Unknown`
`content.modelOverview`	object	—	SageMakerModelCardModelOverview is a nested property type of AWS::SageMaker::ModelCard.
`content.modelOverview.algorithmType`	string	—	Algorithm used to solve the problem.
`content.modelOverview.inferenceEnvironment`	object	—	Overview about the inference.
`content.modelOverview.inferenceEnvironment.containerImage`	array<string>	—	SageMaker inference image uri.
`content.modelOverview.modelArtifact`	array<string>	—	Location of the model artifact.
`content.modelOverview.modelCreator`	string	—	Creator of model.
`content.modelOverview.modelDescription`	string	—	description of model.
`content.modelOverview.modelId`	string	—	SageMaker Model Arn or Non SageMaker Model id.
`content.modelOverview.modelName`	string	—	Name of the model.
`content.modelOverview.modelOwner`	string	—	Owner of model.
`content.modelOverview.modelVersion`	number	—	Version of the model.
`content.modelOverview.problemType`	string	—	Problem being solved with the model.
`content.modelPackageDetails`	object	—	SageMakerModelCardModelPackageDetails is a nested property type of AWS::SageMaker::ModelCard.
`content.modelPackageDetails.approvalDescription`	string	—	A description provided for the model approval
`content.modelPackageDetails.createdBy`	object	—	Information about the user who created model package.
`content.modelPackageDetails.createdBy.userProfileName`	string	—	The name of the user's profile in Studio
`content.modelPackageDetails.domain`	string	—	The machine learning domain of the model package you specified.
`content.modelPackageDetails.inferenceSpecification`	object	—	Details about inference jobs that can be run with models based on this model package.
`content.modelPackageDetails.inferenceSpecification.containers`	array<object>	yes	Contains inference related information which were used to create model package.
`content.modelPackageDetails.inferenceSpecification.containers[].image`	string	yes	Inference environment path.
`content.modelPackageDetails.inferenceSpecification.containers[].modelDataUrl`	string	—	The Amazon S3 path where the model artifacts, which result from model training, are stored.
`content.modelPackageDetails.inferenceSpecification.containers[].nearestModelName`	string	—	The name of a pre-trained machine learning benchmarked by Amazon SageMaker Inference Recommender model that matches your model.
`content.modelPackageDetails.modelApprovalStatus`	enum	—	Current approval status of model package Allowed: `Approved, Rejected, PendingManualApproval`
`content.modelPackageDetails.modelPackageArn`	string	—	The Amazon Resource Name (ARN) of the model package
`content.modelPackageDetails.modelPackageDescription`	string	—	A brief summary of the model package
`content.modelPackageDetails.modelPackageGroupName`	string	—	If the model is a versioned model, the name of the model group that the versioned model belongs to.
`content.modelPackageDetails.modelPackageName`	string	—	Name of the model package
`content.modelPackageDetails.modelPackageStatus`	enum	—	Current status of model package Allowed: `Pending, InProgress, Completed, Failed, Deleting`
`content.modelPackageDetails.modelPackageVersion`	number	—	Version of the model package
`content.modelPackageDetails.sourceAlgorithms`	array<object>	—	A list of algorithms that were used to create a model package.
`content.modelPackageDetails.sourceAlgorithms[].algorithmName`	string	yes	The name of an algorithm that was used to create the model package.
`content.modelPackageDetails.sourceAlgorithms[].modelDataUrl`	string	—	The Amazon S3 path where the model artifacts, which result from model training, are stored.
`content.modelPackageDetails.task`	string	—	The machine learning task you specified that your model package accomplishes.
`content.trainingDetails`	object	—	SageMakerModelCardTrainingDetails is a nested property type of AWS::SageMaker::ModelCard.
`content.trainingDetails.objectiveFunction`	object	—	SageMakerModelCardObjectiveFunction is a nested property type of AWS::SageMaker::ModelCard.
`content.trainingDetails.objectiveFunction.function`	object	—	objective function that training job is optimized for.
`content.trainingDetails.objectiveFunction.function.condition`	string	—
`content.trainingDetails.objectiveFunction.function.facet`	string	—
`content.trainingDetails.objectiveFunction.function.function`	enum	—	Allowed: `Maximize, Minimize`
`content.trainingDetails.objectiveFunction.notes`	string	—
`content.trainingDetails.trainingJobDetails`	object	—	SageMakerModelCardTrainingDetailsTrainingJobDetails is a nested property type of AWS::SageMaker::ModelCard.
`content.trainingDetails.trainingJobDetails.hyperParameters`	array<object>	—
`content.trainingDetails.trainingJobDetails.hyperParameters[].name`	string	yes
`content.trainingDetails.trainingJobDetails.hyperParameters[].value`	string	yes
`content.trainingDetails.trainingJobDetails.trainingArn`	string	—	SageMaker Training job arn.
`content.trainingDetails.trainingJobDetails.trainingDatasets`	array<string>	—	Location of the model datasets.
`content.trainingDetails.trainingJobDetails.trainingEnvironment`	object	—	SageMakerModelCardTrainingDetailsTrainingJobDetailsTrainingEnvironment is a nested property type of AWS::SageMaker::ModelCard.
`content.trainingDetails.trainingJobDetails.trainingEnvironment.containerImage`	array<string>	—	SageMaker training image uri.
`content.trainingDetails.trainingJobDetails.trainingMetrics`	array<object>	—
`content.trainingDetails.trainingJobDetails.trainingMetrics[].name`	string	yes
`content.trainingDetails.trainingJobDetails.trainingMetrics[].notes`	string	—
`content.trainingDetails.trainingJobDetails.trainingMetrics[].value`	number	yes
`content.trainingDetails.trainingJobDetails.userProvidedHyperParameters`	array<object>	—
`content.trainingDetails.trainingJobDetails.userProvidedHyperParameters[].name`	string	yes
`content.trainingDetails.trainingJobDetails.userProvidedHyperParameters[].value`	string	yes
`content.trainingDetails.trainingJobDetails.userProvidedTrainingMetrics`	array<object>	—
`content.trainingDetails.trainingJobDetails.userProvidedTrainingMetrics[].name`	string	yes
`content.trainingDetails.trainingJobDetails.userProvidedTrainingMetrics[].notes`	string	—
`content.trainingDetails.trainingJobDetails.userProvidedTrainingMetrics[].value`	number	yes
`content.trainingDetails.trainingObservations`	string	—
`createdBy`	object	—	Information about the user who created or modified an experiment, trial, trial component, lineage group, project, or model card.
`createdBy.domainId`	string	—	The domain associated with the user.
`createdBy.userProfileArn`	string	—	The Amazon Resource Name (ARN) of the user's profile.
`createdBy.userProfileName`	string	—	The name of the user's profile.
`lastModifiedBy`	object	—	Information about the user who created or modified an experiment, trial, trial component, lineage group, project, or model card.
`lastModifiedBy.domainId`	string	—	The domain associated with the user.
`lastModifiedBy.userProfileArn`	string	—	The Amazon Resource Name (ARN) of the user's profile.
`lastModifiedBy.userProfileName`	string	—	The name of the user's profile.
`modelCardName`	string	yes	The unique name of the model card.
`modelCardStatus`	enum	yes	The approval status of the model card within your organization. Allowed: `Draft, PendingReview, Approved, Archived`
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`securityConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`securityConfig.kmsKeyId`	string	—	A Key Management Service key ID to use for encrypting a model card.
`tags`	array<object>	—	Key-value pairs used to manage metadata for model cards.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The date and time the model card was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	The date and time the model card was last modified.
`lastSyncTime`	string	—
`modelCardArn`	string	—	The Amazon Resource Name (ARN) of the successfully created model card.
`modelCardProcessingStatus`	enum	—	The processing status of model card deletion. Allowed: `UnsetValue, DeleteInProgress, DeletePending, ContentDeleted, ExportJobsDeleted, DeleteCompleted, DeleteFailed`
`modelCardVersion`	integer	—	A version of the model card.
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerModelExplainabilityJobDefinition # CLOUD CONTROL · AWS::SageMaker::ModelExplainabilityJobDefinition

SageMakerModelExplainabilityJobDefinition manages AWS::SageMaker::ModelExplainabilityJobDefinition through the AWS Cloud Control API.

kubectl get sagemakermodelexplainabilityjobdefinitions · example on GitHub

Example

# SageMakerModelExplainabilityJobDefinition — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermodelexplainabilityjobdefinition
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerModelExplainabilityJobDefinition
metadata:
  name: example-sagemakermodelexplainabilityjobdefinition
  namespace: default
spec:
  endpointName: example-sagemakermodelexplainabilityjobdefinition
  jobDefinitionName: example-sagemakermodelexplainabilityjobdefinition
  jobResources:
    clusterConfig:
      instanceCount: 1
      instanceType: example
      volumeKmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
      volumeSizeInGB: 1
  modelExplainabilityAppSpecification:
    configUri: example
    environment:
      key: value
    imageUri: example
  modelExplainabilityBaselineConfig:
    baseliningJobName: example-sagemakermodelexplainabilityjobdefinition
    constraintsResource:
      s3Uri: example
  modelExplainabilityJobInput:
    batchTransformInput:
      dataCapturedDestinationS3Uri: example
      datasetFormat:
        parquet: false
      featuresAttribute: example
      inferenceAttribute: example
      localPath: example
      probabilityAttribute: example
      s3DataDistributionType: FullyReplicated
      s3InputMode: Pipe
    endpointInput:
      endpointName: example-sagemakermodelexplainabilityjobdefinition
      featuresAttribute: example
      inferenceAttribute: example
      localPath: example
      probabilityAttribute: example
      s3DataDistributionType: FullyReplicated
      s3InputMode: Pipe
  modelExplainabilityJobOutputConfig:
    kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    monitoringOutputs:
    - s3Output:
        localPath: example
        s3UploadMode: Continuous
        s3Uri: example
  networkConfig:
    enableInterContainerTrafficEncryption: true
    enableNetworkIsolation: true
    vpcConfig:
      securityGroupIds:
      - example
      subnets:
      - example
  providerRef:
    name: example-sagemakermodelexplainabilityjobdefinition
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  stoppingCondition:
    maxRuntimeInSeconds: 1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`endpointName`	string	—	Create-only in AWS: changing it requires replacing the resource.
`jobDefinitionName`	string	—	Create-only in AWS: changing it requires replacing the resource.
`jobResources`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`jobResources.clusterConfig`	object	yes	SageMakerModelExplainabilityJobDefinitionClusterConfig is a nested property type of AWS::SageMaker::ModelExplainabilityJobDefinition.
`jobResources.clusterConfig.instanceCount`	integer	yes	The number of ML compute instances to use in the model monitoring job.
`jobResources.clusterConfig.instanceType`	string	yes	The ML compute instance type for the processing job.
`jobResources.clusterConfig.volumeKmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt data on the storage volume attached to the ML compute instance(s) that run the model monitoring job.
`jobResources.clusterConfig.volumeSizeInGB`	integer	yes	The size of the ML storage volume, in gigabytes, that you want to provision.
`modelExplainabilityAppSpecification`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`modelExplainabilityAppSpecification.configUri`	string	yes	The S3 URI to an analysis configuration file
`modelExplainabilityAppSpecification.environment`	map<string>	—	Sets the environment variables in the Docker container
`modelExplainabilityAppSpecification.imageUri`	string	yes	The container image to be run by the monitoring job.
`modelExplainabilityBaselineConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`modelExplainabilityBaselineConfig.baseliningJobName`	string	—
`modelExplainabilityBaselineConfig.constraintsResource`	object	—	SageMakerModelExplainabilityJobDefinitionConstraintsResource is a nested property type of AWS::SageMaker::ModelExplainabilityJobDefinition.
`modelExplainabilityBaselineConfig.constraintsResource.s3Uri`	string	—	The Amazon S3 URI for baseline constraint file in Amazon S3 that the current monitoring job should validated against.
`modelExplainabilityJobInput`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`modelExplainabilityJobInput.batchTransformInput`	object	—	SageMakerModelExplainabilityJobDefinitionBatchTransformInput is a nested property type of AWS::SageMaker::ModelExplainabilityJobDefinition.
`modelExplainabilityJobInput.batchTransformInput.dataCapturedDestinationS3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Batch Transform Job captures data.
`modelExplainabilityJobInput.batchTransformInput.datasetFormat`	object	yes	SageMakerModelExplainabilityJobDefinitionDatasetFormat is a nested property type of AWS::SageMaker::ModelExplainabilityJobDefinition.
`modelExplainabilityJobInput.batchTransformInput.datasetFormat.csv`	object	—	SageMakerModelExplainabilityJobDefinitionCsv is a nested property type of AWS::SageMaker::ModelExplainabilityJobDefinition.
`modelExplainabilityJobInput.batchTransformInput.datasetFormat.csv.header`	boolean	—	A boolean flag indicating if given CSV has header
`modelExplainabilityJobInput.batchTransformInput.datasetFormat.json`	object	—	SageMakerModelExplainabilityJobDefinitionJson is a nested property type of AWS::SageMaker::ModelExplainabilityJobDefinition.
`modelExplainabilityJobInput.batchTransformInput.datasetFormat.json.line`	boolean	—	A boolean flag indicating if it is JSON line format
`modelExplainabilityJobInput.batchTransformInput.datasetFormat.parquet`	boolean	—
`modelExplainabilityJobInput.batchTransformInput.featuresAttribute`	string	—	JSONpath to locate features in JSONlines dataset
`modelExplainabilityJobInput.batchTransformInput.inferenceAttribute`	string	—	Index or JSONpath to locate predicted label(s)
`modelExplainabilityJobInput.batchTransformInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`modelExplainabilityJobInput.batchTransformInput.probabilityAttribute`	string	—	Index or JSONpath to locate probabilities
`modelExplainabilityJobInput.batchTransformInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`modelExplainabilityJobInput.batchTransformInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`modelExplainabilityJobInput.endpointInput`	object	—	SageMakerModelExplainabilityJobDefinitionEndpointInput is a nested property type of AWS::SageMaker::ModelExplainabilityJobDefinition.
`modelExplainabilityJobInput.endpointInput.endpointName`	string	yes
`modelExplainabilityJobInput.endpointInput.featuresAttribute`	string	—	JSONpath to locate features in JSONlines dataset
`modelExplainabilityJobInput.endpointInput.inferenceAttribute`	string	—	Index or JSONpath to locate predicted label(s)
`modelExplainabilityJobInput.endpointInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`modelExplainabilityJobInput.endpointInput.probabilityAttribute`	string	—	Index or JSONpath to locate probabilities
`modelExplainabilityJobInput.endpointInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`modelExplainabilityJobInput.endpointInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`modelExplainabilityJobOutputConfig`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`modelExplainabilityJobOutputConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt the model artifacts at rest using Amazon S3 server-side encryption.
`modelExplainabilityJobOutputConfig.monitoringOutputs`	array<object>	yes	Monitoring outputs for monitoring jobs.
`modelExplainabilityJobOutputConfig.monitoringOutputs[].s3Output`	object	yes	SageMakerModelExplainabilityJobDefinitionS3Output is a nested property type of AWS::SageMaker::ModelExplainabilityJobDefinition.
`modelExplainabilityJobOutputConfig.monitoringOutputs[].s3Output.localPath`	string	yes	The local path to the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`modelExplainabilityJobOutputConfig.monitoringOutputs[].s3Output.s3UploadMode`	enum	—	Whether to upload the results of the monitoring job continuously or after the job completes. Allowed: `Continuous, EndOfJob`
`modelExplainabilityJobOutputConfig.monitoringOutputs[].s3Output.s3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`networkConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`networkConfig.enableInterContainerTrafficEncryption`	boolean	—	Whether to encrypt all communications between distributed processing jobs.
`networkConfig.enableNetworkIsolation`	boolean	—	Whether to allow inbound and outbound network calls to and from the containers used for the processing job.
`networkConfig.vpcConfig`	object	—	SageMakerModelExplainabilityJobDefinitionVpcConfig is a nested property type of AWS::SageMaker::ModelExplainabilityJobDefinition.
`networkConfig.vpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form sg-xxxxxxxx.
`networkConfig.vpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect to your monitoring jobs.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker can assume to perform tasks on your behalf.
`stoppingCondition`	object	—	Create-only in AWS: changing it requires replacing the resource.
`stoppingCondition.maxRuntimeInSeconds`	integer	yes	The maximum runtime allowed in seconds.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The time at which the job definition was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`jobDefinitionArn`	string	—	The Amazon Resource Name (ARN) of job definition.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerModelPackage # CLOUD CONTROL · AWS::SageMaker::ModelPackage

SageMakerModelPackage manages AWS::SageMaker::ModelPackage through the AWS Cloud Control API.

kubectl get sagemakermodelpackages · example on GitHub

Example

# SageMakerModelPackage — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermodelpackage
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerModelPackage
metadata:
  name: example-sagemakermodelpackage
  namespace: default
spec:
  additionalInferenceSpecifications:
  - containers:
    - containerHostname: example-sagemakermodelpackage
      framework: example
      frameworkVersion: '1.0'
      image: example
      imageDigest: example
      modelDataUrl: example
      nearestModelName: example-sagemakermodelpackage
    description: Managed by konfig-konector
    name: example-sagemakermodelpackage
    supportedContentTypes:
    - example
    supportedRealtimeInferenceInstanceTypes:
    - '2026-01-01T00:00:00Z'
    supportedResponseMIMETypes:
    - example
    supportedTransformInstanceTypes:
    - example
  additionalInferenceSpecificationsToAdd:
  - containers:
    - containerHostname: example-sagemakermodelpackage
      framework: example
      frameworkVersion: '1.0'
      image: example
      imageDigest: example
      modelDataUrl: example
      nearestModelName: example-sagemakermodelpackage
    description: Managed by konfig-konector
    name: example-sagemakermodelpackage
    supportedContentTypes:
    - example
    supportedRealtimeInferenceInstanceTypes:
    - '2026-01-01T00:00:00Z'
    supportedResponseMIMETypes:
    - example
    supportedTransformInstanceTypes:
    - example
  approvalDescription: Managed by konfig-konector
  certifyForMarketplace: false
  clientToken: example
  customerMetadataProperties:
    key: value
  domain: example
  driftCheckBaselines:
    bias:
      configFile:
        contentDigest: example
        contentType: example
        s3Uri: example
      postTrainingConstraints:
        contentDigest: example
        contentType: example
        s3Uri: example
      preTrainingConstraints:
        contentDigest: example
        contentType: example
        s3Uri: example
    explainability:
      configFile:
        contentDigest: example
        contentType: example
        s3Uri: example
      constraints:
        contentDigest: example
        contentType: example
        s3Uri: example
    modelDataQuality:
      constraints:
        contentDigest: example
        contentType: example
        s3Uri: example
      statistics:
        contentDigest: example
        contentType: example
        s3Uri: example
    modelQuality:
      constraints:
        contentDigest: example
        contentType: example
        s3Uri: example
      statistics:
        contentDigest: example
        contentType: example
        s3Uri: example
  inferenceSpecification:
    containers:
    - containerHostname: example-sagemakermodelpackage
      framework: example
      frameworkVersion: '1.0'
      image: example
      imageDigest: example
      modelDataUrl: example
      nearestModelName: example-sagemakermodelpackage
    supportedContentTypes:
    - example
    supportedRealtimeInferenceInstanceTypes:
    - '2026-01-01T00:00:00Z'
    supportedResponseMIMETypes:
    - example
    supportedTransformInstanceTypes:
    - example
  metadataProperties:
    generatedBy: example
    repository: example
  modelApprovalStatus: Approved
  modelCard:
    modelCardContent: example
    modelCardStatus: Draft
  modelMetrics:
    bias:
      postTrainingReport:
        contentDigest: example
        contentType: example
        s3Uri: example
      preTrainingReport:
        contentDigest: example
        contentType: example
        s3Uri: example
      report:
        contentDigest: example
        contentType: example
        s3Uri: example
    explainability:
      report:
        contentDigest: example
        contentType: example
        s3Uri: example
    modelDataQuality:
      constraints:
        contentDigest: example
        contentType: example
        s3Uri: example
      statistics:
        contentDigest: example
        contentType: example
        s3Uri: example
    modelQuality:
      constraints:
        contentDigest: example
        contentType: example
        s3Uri: example
      statistics:
        contentDigest: example
        contentType: example
        s3Uri: example
  modelPackageDescription: Managed by konfig-konector
  modelPackageGroupName: example-sagemakermodelpackage
  modelPackageName: example-sagemakermodelpackage
  modelPackageStatusDetails:
    validationStatuses:
    - failureReason: example
      name: example-sagemakermodelpackage
      status: NotStarted
  modelPackageVersion: 1
  providerRef:
    name: example-sagemakermodelpackage
    region: us-east-1
  samplePayloadUrl: example
  securityConfig:
    kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
  skipModelValidation: None
  sourceAlgorithmSpecification:
    sourceAlgorithms:
    - algorithmName: example-sagemakermodelpackage
      modelDataUrl: example
  sourceUri: example
  tags:
  - key: example
    value: example
  task: example
  validationSpecification:
    validationProfiles:
    - profileName: example-sagemakermodelpackage
      transformJobDefinition:
        batchStrategy: MultiRecord
        maxConcurrentTransforms: 1
        maxPayloadInMB: 1
        transformInput:
          compressionType: None
          contentType: example
          dataSource:
            s3DataSource:
              s3DataType: ManifestFile
              s3Uri: example
          splitType: None
        transformOutput:
          accept: example
          assembleWith: None
          kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
          s3OutputPath: example
        transformResources:
          instanceCount: 1
          instanceType: example
          volumeKmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    validationRole: example-xxxxxxxxxxxx

Spec

Field	Type	Required	Description
`additionalInferenceSpecifications`	array<object>	—
`additionalInferenceSpecifications[].containers`	array<object>	yes	The Amazon ECR registry path of the Docker image that contains the inference code.
`additionalInferenceSpecifications[].containers[].containerHostname`	string	—	The DNS host name for the Docker container.
`additionalInferenceSpecifications[].containers[].environment`	map<string>	—
`additionalInferenceSpecifications[].containers[].framework`	string	—	The machine learning framework of the model package container image.
`additionalInferenceSpecifications[].containers[].frameworkVersion`	string	—	The framework version of the Model Package Container Image.
`additionalInferenceSpecifications[].containers[].image`	string	yes	The Amazon EC2 Container Registry (Amazon ECR) path where inference code is stored.
`additionalInferenceSpecifications[].containers[].imageDigest`	string	—	An MD5 hash of the training algorithm that identifies the Docker image used for training.
`additionalInferenceSpecifications[].containers[].modelDataSource`	object	—	SageMakerModelPackageModelDataSource is a nested property type of AWS::SageMaker::ModelPackage.
`additionalInferenceSpecifications[].containers[].modelDataSource.s3DataSource`	object	—	SageMakerModelPackageS3ModelDataSource is a nested property type of AWS::SageMaker::ModelPackage.
`additionalInferenceSpecifications[].containers[].modelDataSource.s3DataSource.compressionType`	enum	yes	Specifies how the ML model data is prepared. Allowed: `None, Gzip`
`additionalInferenceSpecifications[].containers[].modelDataSource.s3DataSource.modelAccessConfig`	object	—	SageMakerModelPackageModelAccessConfig is a nested property type of AWS::SageMaker::ModelPackage.
`additionalInferenceSpecifications[].containers[].modelDataSource.s3DataSource.modelAccessConfig.acceptEula`	boolean	yes	Specifies agreement to the model end-user license agreement (EULA).
`additionalInferenceSpecifications[].containers[].modelDataSource.s3DataSource.s3DataType`	enum	yes	Specifies the type of ML model data to deploy. Allowed: `S3Prefix, S3Object`
`additionalInferenceSpecifications[].containers[].modelDataSource.s3DataSource.s3Uri`	string	yes	Specifies the S3 path of ML model data to deploy.
`additionalInferenceSpecifications[].containers[].modelDataUrl`	string	—	A structure with Model Input details.
`additionalInferenceSpecifications[].containers[].modelInput`	object	—	SageMakerModelPackageModelPackageContainerDefinitionModelInput is a nested property type of AWS::SageMaker::ModelPackage.
`additionalInferenceSpecifications[].containers[].modelInput.dataInputConfig`	string	yes	The input configuration object for the model.
`additionalInferenceSpecifications[].containers[].nearestModelName`	string	—	The name of a pre-trained machine learning benchmarked by Amazon SageMaker Inference Recommender model that matches your model.
`additionalInferenceSpecifications[].description`	string	—	A description of the additional Inference specification.
`additionalInferenceSpecifications[].name`	string	yes	A unique name to identify the additional inference specification.
`additionalInferenceSpecifications[].supportedContentTypes`	array<string>	—	The supported MIME types for the input data.
`additionalInferenceSpecifications[].supportedRealtimeInferenceInstanceTypes`	array<string>	—	A list of the instance types that are used to generate inferences in real-time
`additionalInferenceSpecifications[].supportedResponseMIMETypes`	array<string>	—	The supported MIME types for the output data.
`additionalInferenceSpecifications[].supportedTransformInstanceTypes`	array<string>	—	A list of the instance types on which a transformation job can be run or on which an endpoint can be deployed.
`additionalInferenceSpecificationsToAdd`	array<object>	—
`additionalInferenceSpecificationsToAdd[].containers`	array<object>	yes	The Amazon ECR registry path of the Docker image that contains the inference code.
`additionalInferenceSpecificationsToAdd[].containers[].containerHostname`	string	—	The DNS host name for the Docker container.
`additionalInferenceSpecificationsToAdd[].containers[].environment`	map<string>	—
`additionalInferenceSpecificationsToAdd[].containers[].framework`	string	—	The machine learning framework of the model package container image.
`additionalInferenceSpecificationsToAdd[].containers[].frameworkVersion`	string	—	The framework version of the Model Package Container Image.
`additionalInferenceSpecificationsToAdd[].containers[].image`	string	yes	The Amazon EC2 Container Registry (Amazon ECR) path where inference code is stored.
`additionalInferenceSpecificationsToAdd[].containers[].imageDigest`	string	—	An MD5 hash of the training algorithm that identifies the Docker image used for training.
`additionalInferenceSpecificationsToAdd[].containers[].modelDataSource`	object	—	SageMakerModelPackageModelDataSource is a nested property type of AWS::SageMaker::ModelPackage.
`additionalInferenceSpecificationsToAdd[].containers[].modelDataSource.s3DataSource`	object	—	SageMakerModelPackageS3ModelDataSource is a nested property type of AWS::SageMaker::ModelPackage.
`additionalInferenceSpecificationsToAdd[].containers[].modelDataSource.s3DataSource.compressionType`	enum	yes	Specifies how the ML model data is prepared. Allowed: `None, Gzip`
`additionalInferenceSpecificationsToAdd[].containers[].modelDataSource.s3DataSource.modelAccessConfig`	object	—	SageMakerModelPackageModelAccessConfig is a nested property type of AWS::SageMaker::ModelPackage.
`additionalInferenceSpecificationsToAdd[].containers[].modelDataSource.s3DataSource.modelAccessConfig.acceptEula`	boolean	yes	Specifies agreement to the model end-user license agreement (EULA).
`additionalInferenceSpecificationsToAdd[].containers[].modelDataSource.s3DataSource.s3DataType`	enum	yes	Specifies the type of ML model data to deploy. Allowed: `S3Prefix, S3Object`
`additionalInferenceSpecificationsToAdd[].containers[].modelDataSource.s3DataSource.s3Uri`	string	yes	Specifies the S3 path of ML model data to deploy.
`additionalInferenceSpecificationsToAdd[].containers[].modelDataUrl`	string	—	A structure with Model Input details.
`additionalInferenceSpecificationsToAdd[].containers[].modelInput`	object	—	SageMakerModelPackageModelPackageContainerDefinitionModelInput is a nested property type of AWS::SageMaker::ModelPackage.
`additionalInferenceSpecificationsToAdd[].containers[].modelInput.dataInputConfig`	string	yes	The input configuration object for the model.
`additionalInferenceSpecificationsToAdd[].containers[].nearestModelName`	string	—	The name of a pre-trained machine learning benchmarked by Amazon SageMaker Inference Recommender model that matches your model.
`additionalInferenceSpecificationsToAdd[].description`	string	—	A description of the additional Inference specification.
`additionalInferenceSpecificationsToAdd[].name`	string	yes	A unique name to identify the additional inference specification.
`additionalInferenceSpecificationsToAdd[].supportedContentTypes`	array<string>	—	The supported MIME types for the input data.
`additionalInferenceSpecificationsToAdd[].supportedRealtimeInferenceInstanceTypes`	array<string>	—	A list of the instance types that are used to generate inferences in real-time
`additionalInferenceSpecificationsToAdd[].supportedResponseMIMETypes`	array<string>	—	The supported MIME types for the output data.
`additionalInferenceSpecificationsToAdd[].supportedTransformInstanceTypes`	array<string>	—	A list of the instance types on which a transformation job can be run or on which an endpoint can be deployed.
`approvalDescription`	string	—
`certifyForMarketplace`	boolean	—
`clientToken`	string	—	Create-only in AWS: changing it requires replacing the resource.
`customerMetadataProperties`	map<string>	—
`domain`	string	—	Create-only in AWS: changing it requires replacing the resource.
`driftCheckBaselines`	object	—	Create-only in AWS: changing it requires replacing the resource.
`driftCheckBaselines.bias`	object	—	SageMakerModelPackageDriftCheckBias is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.bias.configFile`	object	—	SageMakerModelPackageFileSource is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.bias.configFile.contentDigest`	string	—	The digest of the file source.
`driftCheckBaselines.bias.configFile.contentType`	string	—	The type of content stored in the file source.
`driftCheckBaselines.bias.configFile.s3Uri`	string	yes	The Amazon S3 URI for the file source.
`driftCheckBaselines.bias.postTrainingConstraints`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.bias.postTrainingConstraints.contentDigest`	string	—	The digest of the metric source.
`driftCheckBaselines.bias.postTrainingConstraints.contentType`	string	yes	The type of content stored in the metric source.
`driftCheckBaselines.bias.postTrainingConstraints.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`driftCheckBaselines.bias.preTrainingConstraints`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.bias.preTrainingConstraints.contentDigest`	string	—	The digest of the metric source.
`driftCheckBaselines.bias.preTrainingConstraints.contentType`	string	yes	The type of content stored in the metric source.
`driftCheckBaselines.bias.preTrainingConstraints.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`driftCheckBaselines.explainability`	object	—	SageMakerModelPackageDriftCheckExplainability is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.explainability.configFile`	object	—	SageMakerModelPackageFileSource is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.explainability.configFile.contentDigest`	string	—	The digest of the file source.
`driftCheckBaselines.explainability.configFile.contentType`	string	—	The type of content stored in the file source.
`driftCheckBaselines.explainability.configFile.s3Uri`	string	yes	The Amazon S3 URI for the file source.
`driftCheckBaselines.explainability.constraints`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.explainability.constraints.contentDigest`	string	—	The digest of the metric source.
`driftCheckBaselines.explainability.constraints.contentType`	string	yes	The type of content stored in the metric source.
`driftCheckBaselines.explainability.constraints.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`driftCheckBaselines.modelDataQuality`	object	—	SageMakerModelPackageDriftCheckModelDataQuality is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.modelDataQuality.constraints`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.modelDataQuality.constraints.contentDigest`	string	—	The digest of the metric source.
`driftCheckBaselines.modelDataQuality.constraints.contentType`	string	yes	The type of content stored in the metric source.
`driftCheckBaselines.modelDataQuality.constraints.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`driftCheckBaselines.modelDataQuality.statistics`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.modelDataQuality.statistics.contentDigest`	string	—	The digest of the metric source.
`driftCheckBaselines.modelDataQuality.statistics.contentType`	string	yes	The type of content stored in the metric source.
`driftCheckBaselines.modelDataQuality.statistics.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`driftCheckBaselines.modelQuality`	object	—	SageMakerModelPackageDriftCheckModelQuality is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.modelQuality.constraints`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.modelQuality.constraints.contentDigest`	string	—	The digest of the metric source.
`driftCheckBaselines.modelQuality.constraints.contentType`	string	yes	The type of content stored in the metric source.
`driftCheckBaselines.modelQuality.constraints.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`driftCheckBaselines.modelQuality.statistics`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`driftCheckBaselines.modelQuality.statistics.contentDigest`	string	—	The digest of the metric source.
`driftCheckBaselines.modelQuality.statistics.contentType`	string	yes	The type of content stored in the metric source.
`driftCheckBaselines.modelQuality.statistics.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`inferenceSpecification`	object	—	Create-only in AWS: changing it requires replacing the resource.
`inferenceSpecification.containers`	array<object>	yes	The Amazon ECR registry path of the Docker image that contains the inference code.
`inferenceSpecification.containers[].containerHostname`	string	—	The DNS host name for the Docker container.
`inferenceSpecification.containers[].environment`	map<string>	—
`inferenceSpecification.containers[].framework`	string	—	The machine learning framework of the model package container image.
`inferenceSpecification.containers[].frameworkVersion`	string	—	The framework version of the Model Package Container Image.
`inferenceSpecification.containers[].image`	string	yes	The Amazon EC2 Container Registry (Amazon ECR) path where inference code is stored.
`inferenceSpecification.containers[].imageDigest`	string	—	An MD5 hash of the training algorithm that identifies the Docker image used for training.
`inferenceSpecification.containers[].modelDataSource`	object	—	SageMakerModelPackageModelDataSource is a nested property type of AWS::SageMaker::ModelPackage.
`inferenceSpecification.containers[].modelDataSource.s3DataSource`	object	—	SageMakerModelPackageS3ModelDataSource is a nested property type of AWS::SageMaker::ModelPackage.
`inferenceSpecification.containers[].modelDataSource.s3DataSource.compressionType`	enum	yes	Specifies how the ML model data is prepared. Allowed: `None, Gzip`
`inferenceSpecification.containers[].modelDataSource.s3DataSource.modelAccessConfig`	object	—	SageMakerModelPackageModelAccessConfig is a nested property type of AWS::SageMaker::ModelPackage.
`inferenceSpecification.containers[].modelDataSource.s3DataSource.modelAccessConfig.acceptEula`	boolean	yes	Specifies agreement to the model end-user license agreement (EULA).
`inferenceSpecification.containers[].modelDataSource.s3DataSource.s3DataType`	enum	yes	Specifies the type of ML model data to deploy. Allowed: `S3Prefix, S3Object`
`inferenceSpecification.containers[].modelDataSource.s3DataSource.s3Uri`	string	yes	Specifies the S3 path of ML model data to deploy.
`inferenceSpecification.containers[].modelDataUrl`	string	—	A structure with Model Input details.
`inferenceSpecification.containers[].modelInput`	object	—	SageMakerModelPackageModelPackageContainerDefinitionModelInput is a nested property type of AWS::SageMaker::ModelPackage.
`inferenceSpecification.containers[].modelInput.dataInputConfig`	string	yes	The input configuration object for the model.
`inferenceSpecification.containers[].nearestModelName`	string	—	The name of a pre-trained machine learning benchmarked by Amazon SageMaker Inference Recommender model that matches your model.
`inferenceSpecification.supportedContentTypes`	array<string>	yes	The supported MIME types for the input data.
`inferenceSpecification.supportedRealtimeInferenceInstanceTypes`	array<string>	—	A list of the instance types that are used to generate inferences in real-time
`inferenceSpecification.supportedResponseMIMETypes`	array<string>	yes	The supported MIME types for the output data.
`inferenceSpecification.supportedTransformInstanceTypes`	array<string>	—	A list of the instance types on which a transformation job can be run or on which an endpoint can be deployed.
`metadataProperties`	object	—	Create-only in AWS: changing it requires replacing the resource.
`metadataProperties.commitId`	string	—	The commit ID.
`metadataProperties.generatedBy`	string	—	The entity this entity was generated by.
`metadataProperties.projectId`	string	—	The project ID metadata.
`metadataProperties.repository`	string	—	The repository metadata.
`modelApprovalStatus`	enum	—	Allowed: `Approved, Rejected, PendingManualApproval`
`modelCard`	object	—	SageMakerModelPackageModelCard is a nested property type of AWS::SageMaker::ModelPackage.
`modelCard.modelCardContent`	string	yes	The content of the model card.
`modelCard.modelCardStatus`	enum	yes	The approval status of the model card within your organization. Allowed: `Draft, PendingReview, Approved, Archived`
`modelMetrics`	object	—	Create-only in AWS: changing it requires replacing the resource.
`modelMetrics.bias`	object	—	SageMakerModelPackageBias is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.bias.postTrainingReport`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.bias.postTrainingReport.contentDigest`	string	—	The digest of the metric source.
`modelMetrics.bias.postTrainingReport.contentType`	string	yes	The type of content stored in the metric source.
`modelMetrics.bias.postTrainingReport.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`modelMetrics.bias.preTrainingReport`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.bias.preTrainingReport.contentDigest`	string	—	The digest of the metric source.
`modelMetrics.bias.preTrainingReport.contentType`	string	yes	The type of content stored in the metric source.
`modelMetrics.bias.preTrainingReport.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`modelMetrics.bias.report`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.bias.report.contentDigest`	string	—	The digest of the metric source.
`modelMetrics.bias.report.contentType`	string	yes	The type of content stored in the metric source.
`modelMetrics.bias.report.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`modelMetrics.explainability`	object	—	SageMakerModelPackageExplainability is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.explainability.report`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.explainability.report.contentDigest`	string	—	The digest of the metric source.
`modelMetrics.explainability.report.contentType`	string	yes	The type of content stored in the metric source.
`modelMetrics.explainability.report.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`modelMetrics.modelDataQuality`	object	—	SageMakerModelPackageModelDataQuality is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.modelDataQuality.constraints`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.modelDataQuality.constraints.contentDigest`	string	—	The digest of the metric source.
`modelMetrics.modelDataQuality.constraints.contentType`	string	yes	The type of content stored in the metric source.
`modelMetrics.modelDataQuality.constraints.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`modelMetrics.modelDataQuality.statistics`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.modelDataQuality.statistics.contentDigest`	string	—	The digest of the metric source.
`modelMetrics.modelDataQuality.statistics.contentType`	string	yes	The type of content stored in the metric source.
`modelMetrics.modelDataQuality.statistics.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`modelMetrics.modelQuality`	object	—	SageMakerModelPackageModelQuality is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.modelQuality.constraints`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.modelQuality.constraints.contentDigest`	string	—	The digest of the metric source.
`modelMetrics.modelQuality.constraints.contentType`	string	yes	The type of content stored in the metric source.
`modelMetrics.modelQuality.constraints.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`modelMetrics.modelQuality.statistics`	object	—	SageMakerModelPackageMetricsSource is a nested property type of AWS::SageMaker::ModelPackage.
`modelMetrics.modelQuality.statistics.contentDigest`	string	—	The digest of the metric source.
`modelMetrics.modelQuality.statistics.contentType`	string	yes	The type of content stored in the metric source.
`modelMetrics.modelQuality.statistics.s3Uri`	string	yes	The Amazon S3 URI for the metric source.
`modelPackageDescription`	string	—	Create-only in AWS: changing it requires replacing the resource.
`modelPackageGroupName`	string	—	Create-only in AWS: changing it requires replacing the resource.
`modelPackageName`	string	—
`modelPackageStatusDetails`	object	—	SageMakerModelPackageModelPackageStatusDetails is a nested property type of AWS::SageMaker::ModelPackage.
`modelPackageStatusDetails.validationStatuses`	array<object>	—
`modelPackageStatusDetails.validationStatuses[].failureReason`	string	—	If the overall status is Failed, the reason for the failure.
`modelPackageStatusDetails.validationStatuses[].name`	string	yes	The name of the model package for which the overall status is being reported.
`modelPackageStatusDetails.validationStatuses[].status`	enum	yes	The current status. Allowed: `NotStarted, Failed, InProgress, Completed`
`modelPackageVersion`	integer	—
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`samplePayloadUrl`	string	—	Create-only in AWS: changing it requires replacing the resource.
`securityConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`securityConfig.kmsKeyId`	string	yes	The AWS KMS Key ID (KMSKeyId) used for encryption of model package information.
`skipModelValidation`	enum	—	Allowed: `None, All`
`sourceAlgorithmSpecification`	object	—	Create-only in AWS: changing it requires replacing the resource.
`sourceAlgorithmSpecification.sourceAlgorithms`	array<object>	yes	A list of algorithms that were used to create a model package.
`sourceAlgorithmSpecification.sourceAlgorithms[].algorithmName`	string	yes	The name of an algorithm that was used to create the model package.
`sourceAlgorithmSpecification.sourceAlgorithms[].modelDataUrl`	string	—	The Amazon S3 path where the model artifacts, which result from model training, are stored.
`sourceUri`	string	—
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes
`task`	string	—	Create-only in AWS: changing it requires replacing the resource.
`validationSpecification`	object	—	Create-only in AWS: changing it requires replacing the resource.
`validationSpecification.validationProfiles`	array<object>	yes
`validationSpecification.validationProfiles[].profileName`	string	yes	The name of the profile for the model package.
`validationSpecification.validationProfiles[].transformJobDefinition`	object	yes	SageMakerModelPackageTransformJobDefinition is a nested property type of AWS::SageMaker::ModelPackage.
`validationSpecification.validationProfiles[].transformJobDefinition.batchStrategy`	enum	—	A string that determines the number of records included in a single mini-batch. Allowed: `MultiRecord, SingleRecord`
`validationSpecification.validationProfiles[].transformJobDefinition.environment`	map<string>	—
`validationSpecification.validationProfiles[].transformJobDefinition.maxConcurrentTransforms`	integer	—	The maximum number of parallel requests that can be sent to each instance in a transform job.
`validationSpecification.validationProfiles[].transformJobDefinition.maxPayloadInMB`	integer	—	The maximum payload size allowed, in MB.
`validationSpecification.validationProfiles[].transformJobDefinition.transformInput`	object	yes	SageMakerModelPackageTransformInput is a nested property type of AWS::SageMaker::ModelPackage.
`validationSpecification.validationProfiles[].transformJobDefinition.transformInput.compressionType`	enum	—	If your transform data is compressed, specify the compression type. Allowed: `None, Gzip`
`validationSpecification.validationProfiles[].transformJobDefinition.transformInput.contentType`	string	—	The multipurpose internet mail extension (MIME) type of the data.
`validationSpecification.validationProfiles[].transformJobDefinition.transformInput.dataSource`	object	yes	SageMakerModelPackageDataSource is a nested property type of AWS::SageMaker::ModelPackage.
`validationSpecification.validationProfiles[].transformJobDefinition.transformInput.dataSource.s3DataSource`	object	yes	SageMakerModelPackageS3DataSource is a nested property type of AWS::SageMaker::ModelPackage.
`validationSpecification.validationProfiles[].transformJobDefinition.transformInput.splitType`	enum	—	The method to use to split the transform job's data files into smaller batches. Allowed: `None, TFRecord, Line, RecordIO`
`validationSpecification.validationProfiles[].transformJobDefinition.transformOutput`	object	yes	SageMakerModelPackageTransformOutput is a nested property type of AWS::SageMaker::ModelPackage.
`validationSpecification.validationProfiles[].transformJobDefinition.transformOutput.accept`	string	—	The MIME type used to specify the output data.
`validationSpecification.validationProfiles[].transformJobDefinition.transformOutput.assembleWith`	enum	—	Defines how to assemble the results of the transform job as a single S3 object. Allowed: `None, Line`
`validationSpecification.validationProfiles[].transformJobDefinition.transformOutput.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt the model artifacts at rest using Amazon S3 server-side encryption.
`validationSpecification.validationProfiles[].transformJobDefinition.transformOutput.s3OutputPath`	string	yes	The Amazon S3 path where you want Amazon SageMaker to store the results of the transform job.
`validationSpecification.validationProfiles[].transformJobDefinition.transformResources`	object	yes	SageMakerModelPackageTransformResources is a nested property type of AWS::SageMaker::ModelPackage.
`validationSpecification.validationProfiles[].transformJobDefinition.transformResources.instanceCount`	integer	yes	The number of ML compute instances to use in the transform job.
`validationSpecification.validationProfiles[].transformJobDefinition.transformResources.instanceType`	string	yes	The ML compute instance type for the transform job.
`validationSpecification.validationProfiles[].transformJobDefinition.transformResources.volumeKmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt model data on the storage volume attached to the ML compute instance(s) that run the batch transform job.
`validationSpecification.validationRole`	string	yes	The IAM roles to be used for the validation of the model package.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	Read-only: reported by AWS after creation.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	Read-only: reported by AWS after creation.
`lastSyncTime`	string	—
`modelPackageArn`	string	—	Read-only: reported by AWS after creation.
`modelPackageStatus`	enum	—	Read-only: reported by AWS after creation. Allowed: `Pending, Deleting, InProgress, Completed, Failed`
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerModelPackageGroup # CLOUD CONTROL · AWS::SageMaker::ModelPackageGroup

SageMakerModelPackageGroup manages AWS::SageMaker::ModelPackageGroup through the AWS Cloud Control API.

kubectl get sagemakermodelpackagegroups · example on GitHub

Example

# SageMakerModelPackageGroup — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermodelpackagegroup
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerModelPackageGroup
metadata:
  name: example-sagemakermodelpackagegroup
  namespace: default
spec:
  modelPackageGroupDescription: Managed by konfig-konector
  modelPackageGroupName: example-sagemakermodelpackagegroup
  providerRef:
    name: example-sagemakermodelpackagegroup
    region: us-east-1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`modelPackageGroupDescription`	string	—	Create-only in AWS: changing it requires replacing the resource.
`modelPackageGroupName`	string	yes	Create-only in AWS: changing it requires replacing the resource.
`modelPackageGroupPolicy`	object	—
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The time at which the model package group was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`modelPackageGroupArn`	string	—	Read-only: reported by AWS after creation.
`modelPackageGroupStatus`	enum	—	The status of a modelpackage group job. Allowed: `Pending, InProgress, Completed, Failed, Deleting, DeleteFailed`
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerModelQualityJobDefinition # CLOUD CONTROL · AWS::SageMaker::ModelQualityJobDefinition

SageMakerModelQualityJobDefinition manages AWS::SageMaker::ModelQualityJobDefinition through the AWS Cloud Control API.

kubectl get sagemakermodelqualityjobdefinitions · example on GitHub

Example

# SageMakerModelQualityJobDefinition — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermodelqualityjobdefinition
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerModelQualityJobDefinition
metadata:
  name: example-sagemakermodelqualityjobdefinition
  namespace: default
spec:
  endpointName: example-sagemakermodelqualityjobdefinition
  jobDefinitionName: example-sagemakermodelqualityjobdefinition
  jobResources:
    clusterConfig:
      instanceCount: 1
      instanceType: example
      volumeKmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
      volumeSizeInGB: 1
  modelQualityAppSpecification:
    containerArguments:
    - example
    containerEntrypoint:
    - example
    environment:
      key: value
    imageUri: example
    postAnalyticsProcessorSourceUri: example
    problemType: BinaryClassification
    recordPreprocessorSourceUri: example
  modelQualityBaselineConfig:
    baseliningJobName: example-sagemakermodelqualityjobdefinition
    constraintsResource:
      s3Uri: example
  modelQualityJobInput:
    batchTransformInput:
      dataCapturedDestinationS3Uri: example
      datasetFormat:
        parquet: false
      endTimeOffset: 2026-01-01T00:0
      inferenceAttribute: example
      localPath: example
      probabilityAttribute: example
      probabilityThresholdAttribute: 1
      s3DataDistributionType: FullyReplicated
      s3InputMode: Pipe
      startTimeOffset: 2026-01-01T00:0
    endpointInput:
      endTimeOffset: 2026-01-01T00:0
      endpointName: example-sagemakermodelqualityjobdefinition
      inferenceAttribute: example
      localPath: example
      probabilityAttribute: example
      probabilityThresholdAttribute: 1
      s3DataDistributionType: FullyReplicated
      s3InputMode: Pipe
      startTimeOffset: 2026-01-01T00:0
    groundTruthS3Input:
      s3Uri: example
  modelQualityJobOutputConfig:
    kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    monitoringOutputs:
    - s3Output:
        localPath: example
        s3UploadMode: Continuous
        s3Uri: example
  networkConfig:
    enableInterContainerTrafficEncryption: true
    enableNetworkIsolation: true
    vpcConfig:
      securityGroupIds:
      - example
      subnets:
      - example
  providerRef:
    name: example-sagemakermodelqualityjobdefinition
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  stoppingCondition:
    maxRuntimeInSeconds: 1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`endpointName`	string	—	Create-only in AWS: changing it requires replacing the resource.
`jobDefinitionName`	string	—	Create-only in AWS: changing it requires replacing the resource.
`jobResources`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`jobResources.clusterConfig`	object	yes	SageMakerModelQualityJobDefinitionClusterConfig is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`jobResources.clusterConfig.instanceCount`	integer	yes	The number of ML compute instances to use in the model monitoring job.
`jobResources.clusterConfig.instanceType`	string	yes	The ML compute instance type for the processing job.
`jobResources.clusterConfig.volumeKmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt data on the storage volume attached to the ML compute instance(s) that run the model monitoring job.
`jobResources.clusterConfig.volumeSizeInGB`	integer	yes	The size of the ML storage volume, in gigabytes, that you want to provision.
`modelQualityAppSpecification`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`modelQualityAppSpecification.containerArguments`	array<string>	—	An array of arguments for the container used to run the monitoring job.
`modelQualityAppSpecification.containerEntrypoint`	array<string>	—	Specifies the entrypoint for a container used to run the monitoring job.
`modelQualityAppSpecification.environment`	map<string>	—	Sets the environment variables in the Docker container
`modelQualityAppSpecification.imageUri`	string	yes	The container image to be run by the monitoring job.
`modelQualityAppSpecification.postAnalyticsProcessorSourceUri`	string	—	An Amazon S3 URI to a script that is called after analysis has been performed.
`modelQualityAppSpecification.problemType`	enum	yes	Allowed: `BinaryClassification, MulticlassClassification, Regression`
`modelQualityAppSpecification.recordPreprocessorSourceUri`	string	—	An Amazon S3 URI to a script that is called per row prior to running analysis.
`modelQualityBaselineConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`modelQualityBaselineConfig.baseliningJobName`	string	—
`modelQualityBaselineConfig.constraintsResource`	object	—	SageMakerModelQualityJobDefinitionConstraintsResource is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`modelQualityBaselineConfig.constraintsResource.s3Uri`	string	—	The Amazon S3 URI for baseline constraint file in Amazon S3 that the current monitoring job should validated against.
`modelQualityJobInput`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`modelQualityJobInput.batchTransformInput`	object	—	SageMakerModelQualityJobDefinitionBatchTransformInput is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`modelQualityJobInput.batchTransformInput.dataCapturedDestinationS3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Batch Transform Job captures data.
`modelQualityJobInput.batchTransformInput.datasetFormat`	object	yes	SageMakerModelQualityJobDefinitionDatasetFormat is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`modelQualityJobInput.batchTransformInput.datasetFormat.csv`	object	—	SageMakerModelQualityJobDefinitionCsv is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`modelQualityJobInput.batchTransformInput.datasetFormat.csv.header`	boolean	—	A boolean flag indicating if given CSV has header
`modelQualityJobInput.batchTransformInput.datasetFormat.json`	object	—	SageMakerModelQualityJobDefinitionJson is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`modelQualityJobInput.batchTransformInput.datasetFormat.json.line`	boolean	—	A boolean flag indicating if it is JSON line format
`modelQualityJobInput.batchTransformInput.datasetFormat.parquet`	boolean	—
`modelQualityJobInput.batchTransformInput.endTimeOffset`	string	—	Monitoring end time offset, e.g.
`modelQualityJobInput.batchTransformInput.inferenceAttribute`	string	—	Index or JSONpath to locate predicted label(s)
`modelQualityJobInput.batchTransformInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`modelQualityJobInput.batchTransformInput.probabilityAttribute`	string	—	Index or JSONpath to locate probabilities
`modelQualityJobInput.batchTransformInput.probabilityThresholdAttribute`	number	—
`modelQualityJobInput.batchTransformInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`modelQualityJobInput.batchTransformInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`modelQualityJobInput.batchTransformInput.startTimeOffset`	string	—	Monitoring start time offset, e.g.
`modelQualityJobInput.endpointInput`	object	—	SageMakerModelQualityJobDefinitionEndpointInput is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`modelQualityJobInput.endpointInput.endTimeOffset`	string	—	Monitoring end time offset, e.g.
`modelQualityJobInput.endpointInput.endpointName`	string	yes
`modelQualityJobInput.endpointInput.inferenceAttribute`	string	—	Index or JSONpath to locate predicted label(s)
`modelQualityJobInput.endpointInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`modelQualityJobInput.endpointInput.probabilityAttribute`	string	—	Index or JSONpath to locate probabilities
`modelQualityJobInput.endpointInput.probabilityThresholdAttribute`	number	—
`modelQualityJobInput.endpointInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`modelQualityJobInput.endpointInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`modelQualityJobInput.endpointInput.startTimeOffset`	string	—	Monitoring start time offset, e.g.
`modelQualityJobInput.groundTruthS3Input`	object	yes	SageMakerModelQualityJobDefinitionMonitoringGroundTruthS3Input is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`modelQualityJobInput.groundTruthS3Input.s3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`modelQualityJobOutputConfig`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`modelQualityJobOutputConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt the model artifacts at rest using Amazon S3 server-side encryption.
`modelQualityJobOutputConfig.monitoringOutputs`	array<object>	yes	Monitoring outputs for monitoring jobs.
`modelQualityJobOutputConfig.monitoringOutputs[].s3Output`	object	yes	SageMakerModelQualityJobDefinitionS3Output is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`modelQualityJobOutputConfig.monitoringOutputs[].s3Output.localPath`	string	yes	The local path to the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`modelQualityJobOutputConfig.monitoringOutputs[].s3Output.s3UploadMode`	enum	—	Whether to upload the results of the monitoring job continuously or after the job completes. Allowed: `Continuous, EndOfJob`
`modelQualityJobOutputConfig.monitoringOutputs[].s3Output.s3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`networkConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`networkConfig.enableInterContainerTrafficEncryption`	boolean	—	Whether to encrypt all communications between distributed processing jobs.
`networkConfig.enableNetworkIsolation`	boolean	—	Whether to allow inbound and outbound network calls to and from the containers used for the processing job.
`networkConfig.vpcConfig`	object	—	SageMakerModelQualityJobDefinitionVpcConfig is a nested property type of AWS::SageMaker::ModelQualityJobDefinition.
`networkConfig.vpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form sg-xxxxxxxx.
`networkConfig.vpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect to your monitoring jobs.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker can assume to perform tasks on your behalf.
`stoppingCondition`	object	—	Create-only in AWS: changing it requires replacing the resource.
`stoppingCondition.maxRuntimeInSeconds`	integer	yes	The maximum runtime allowed in seconds.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The time at which the job definition was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`jobDefinitionArn`	string	—	The Amazon Resource Name (ARN) of job definition.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerMonitoringSchedule # CLOUD CONTROL · AWS::SageMaker::MonitoringSchedule

SageMakerMonitoringSchedule manages AWS::SageMaker::MonitoringSchedule through the AWS Cloud Control API.

kubectl get sagemakermonitoringschedules · example on GitHub

Example

# SageMakerMonitoringSchedule — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakermonitoringschedule
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerMonitoringSchedule
metadata:
  name: example-sagemakermonitoringschedule
  namespace: default
spec:
  endpointName: example-sagemakermonitoringschedule
  failureReason: example
  lastMonitoringExecutionSummary:
    creationTime: '2026-01-01T00:00:00Z'
    endpointName: example-sagemakermonitoringschedule
    failureReason: example
    lastModifiedTime: '2026-01-01T00:00:00Z'
    monitoringExecutionStatus: Pending
    monitoringScheduleName: example-sagemakermonitoringschedule
    processingJobArn: arn:aws:iam::123456789012:role/example-role
    scheduledTime: '2026-01-01T00:00:00Z'
  monitoringScheduleConfig:
    monitoringJobDefinition:
      environment:
        key: value
      monitoringAppSpecification:
        containerArguments:
        - example
        containerEntrypoint:
        - example
        imageUri: example
        postAnalyticsProcessorSourceUri: example
        recordPreprocessorSourceUri: example
      monitoringOutputConfig:
        kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
        monitoringOutputs:
        - s3Output:
            localPath: example
            s3UploadMode: Continuous
            s3Uri: example
      monitoringResources:
        clusterConfig:
          instanceCount: 1
          instanceType: example
          volumeKmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
          volumeSizeInGB: 1
      networkConfig:
        enableInterContainerTrafficEncryption: true
        enableNetworkIsolation: true
      roleArn: arn:aws:iam::123456789012:role/example-role
      stoppingCondition:
        maxRuntimeInSeconds: 1
    monitoringJobDefinitionName: example-sagemakermonitoringschedule
    monitoringType: DataQuality
    scheduleConfig:
      dataAnalysisEndTime: 2026-01-01T00:0
      dataAnalysisStartTime: 2026-01-01T00:0
      scheduleExpression: example
  monitoringScheduleName: example-sagemakermonitoringschedule
  monitoringScheduleStatus: Pending
  providerRef:
    name: example-sagemakermonitoringschedule
    region: us-east-1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`endpointName`	string	—
`failureReason`	string	—	Contains the reason a monitoring job failed, if it failed.
`lastMonitoringExecutionSummary`	object	—	Describes metadata on the last execution to run, if there was one.
`lastMonitoringExecutionSummary.creationTime`	string	yes	The time at which the monitoring job was created.
`lastMonitoringExecutionSummary.endpointName`	string	—
`lastMonitoringExecutionSummary.failureReason`	string	—	Contains the reason a monitoring job failed, if it failed.
`lastMonitoringExecutionSummary.lastModifiedTime`	string	yes	A timestamp that indicates the last time the monitoring job was modified.
`lastMonitoringExecutionSummary.monitoringExecutionStatus`	enum	yes	The status of the monitoring job. Allowed: `Pending, Completed, CompletedWithViolations, InProgress, Failed, Stopping, Stopped`
`lastMonitoringExecutionSummary.monitoringScheduleName`	string	yes
`lastMonitoringExecutionSummary.processingJobArn`	string	—	The Amazon Resource Name (ARN) of the monitoring job.
`lastMonitoringExecutionSummary.scheduledTime`	string	yes	The time the monitoring job was scheduled.
`monitoringScheduleConfig`	object	yes	SageMakerMonitoringScheduleMonitoringScheduleConfig is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition`	object	—	SageMakerMonitoringScheduleMonitoringJobDefinition is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.baselineConfig`	object	—	SageMakerMonitoringScheduleBaselineConfig is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.baselineConfig.constraintsResource`	object	—	SageMakerMonitoringScheduleConstraintsResource is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.baselineConfig.constraintsResource.s3Uri`	string	—	The Amazon S3 URI for baseline constraint file in Amazon S3 that the current monitoring job should validated against.
`monitoringScheduleConfig.monitoringJobDefinition.baselineConfig.statisticsResource`	object	—	SageMakerMonitoringScheduleStatisticsResource is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.baselineConfig.statisticsResource.s3Uri`	string	—	The Amazon S3 URI for the baseline statistics file in Amazon S3 that the current monitoring job should be validated against.
`monitoringScheduleConfig.monitoringJobDefinition.environment`	map<string>	—	Sets the environment variables in the Docker container
`monitoringScheduleConfig.monitoringJobDefinition.monitoringAppSpecification`	object	yes	SageMakerMonitoringScheduleMonitoringAppSpecification is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringAppSpecification.containerArguments`	array<string>	—	An array of arguments for the container used to run the monitoring job.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringAppSpecification.containerEntrypoint`	array<string>	—	Specifies the entrypoint for a container used to run the monitoring job.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringAppSpecification.imageUri`	string	yes	The container image to be run by the monitoring job.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringAppSpecification.postAnalyticsProcessorSourceUri`	string	—	An Amazon S3 URI to a script that is called after analysis has been performed.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringAppSpecification.recordPreprocessorSourceUri`	string	—	An Amazon S3 URI to a script that is called per row prior to running analysis.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs`	array<object>	yes
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput`	object	—	SageMakerMonitoringScheduleBatchTransformInput is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput.dataCapturedDestinationS3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Batch Transform Job captures data.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput.datasetFormat`	object	yes	SageMakerMonitoringScheduleDatasetFormat is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput.datasetFormat.csv`	object	—	SageMakerMonitoringScheduleCsv is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput.datasetFormat.json`	object	—	SageMakerMonitoringScheduleJson is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput.datasetFormat.parquet`	boolean	—
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput.excludeFeaturesAttribute`	string	—	Indexes or names of the features to be excluded from analysis
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].batchTransformInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].endpointInput`	object	—	SageMakerMonitoringScheduleEndpointInput is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].endpointInput.endpointName`	string	yes
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].endpointInput.excludeFeaturesAttribute`	string	—	Indexes or names of the features to be excluded from analysis
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].endpointInput.localPath`	string	yes	Path to the filesystem where the endpoint data is available to the container.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].endpointInput.s3DataDistributionType`	enum	—	Whether input data distributed in Amazon S3 is fully replicated or sharded by an S3 key. Allowed: `FullyReplicated, ShardedByS3Key`
`monitoringScheduleConfig.monitoringJobDefinition.monitoringInputs[].endpointInput.s3InputMode`	enum	—	Whether the Pipe or File is used as the input mode for transfering data for the monitoring job. Allowed: `Pipe, File`
`monitoringScheduleConfig.monitoringJobDefinition.monitoringOutputConfig`	object	yes	SageMakerMonitoringScheduleMonitoringOutputConfig is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringOutputConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt the model artifacts at rest using Amazon S3 server-side encryption.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringOutputConfig.monitoringOutputs`	array<object>	yes	Monitoring outputs for monitoring jobs.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringOutputConfig.monitoringOutputs[].s3Output`	object	yes	SageMakerMonitoringScheduleS3Output is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringOutputConfig.monitoringOutputs[].s3Output.localPath`	string	yes	The local path to the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringOutputConfig.monitoringOutputs[].s3Output.s3UploadMode`	enum	—	Whether to upload the results of the monitoring job continuously or after the job completes. Allowed: `Continuous, EndOfJob`
`monitoringScheduleConfig.monitoringJobDefinition.monitoringOutputConfig.monitoringOutputs[].s3Output.s3Uri`	string	yes	A URI that identifies the Amazon S3 storage location where Amazon SageMaker saves the results of a monitoring job.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringResources`	object	yes	SageMakerMonitoringScheduleMonitoringResources is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringResources.clusterConfig`	object	yes	SageMakerMonitoringScheduleClusterConfig is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringResources.clusterConfig.instanceCount`	integer	yes	The number of ML compute instances to use in the model monitoring job.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringResources.clusterConfig.instanceType`	string	yes	The ML compute instance type for the processing job.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringResources.clusterConfig.volumeKmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt data on the storage volume attached to the ML compute instance(s) that run the model monitoring job.
`monitoringScheduleConfig.monitoringJobDefinition.monitoringResources.clusterConfig.volumeSizeInGB`	integer	yes	The size of the ML storage volume, in gigabytes, that you want to provision.
`monitoringScheduleConfig.monitoringJobDefinition.networkConfig`	object	—	SageMakerMonitoringScheduleNetworkConfig is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.networkConfig.enableInterContainerTrafficEncryption`	boolean	—	Whether to encrypt all communications between distributed processing jobs.
`monitoringScheduleConfig.monitoringJobDefinition.networkConfig.enableNetworkIsolation`	boolean	—	Whether to allow inbound and outbound network calls to and from the containers used for the processing job.
`monitoringScheduleConfig.monitoringJobDefinition.networkConfig.vpcConfig`	object	—	SageMakerMonitoringScheduleVpcConfig is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.networkConfig.vpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form sg-xxxxxxxx.
`monitoringScheduleConfig.monitoringJobDefinition.networkConfig.vpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect to your monitoring jobs.
`monitoringScheduleConfig.monitoringJobDefinition.roleArn`	string	yes	The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker can assume to perform tasks on your behalf.
`monitoringScheduleConfig.monitoringJobDefinition.stoppingCondition`	object	—	SageMakerMonitoringScheduleStoppingCondition is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.monitoringJobDefinition.stoppingCondition.maxRuntimeInSeconds`	integer	yes	The maximum runtime allowed in seconds.
`monitoringScheduleConfig.monitoringJobDefinitionName`	string	—	Name of the job definition
`monitoringScheduleConfig.monitoringType`	enum	—	Allowed: `DataQuality, ModelQuality, ModelBias, ModelExplainability`
`monitoringScheduleConfig.scheduleConfig`	object	—	SageMakerMonitoringScheduleScheduleConfig is a nested property type of AWS::SageMaker::MonitoringSchedule.
`monitoringScheduleConfig.scheduleConfig.dataAnalysisEndTime`	string	—	Data Analysis end time, e.g.
`monitoringScheduleConfig.scheduleConfig.dataAnalysisStartTime`	string	—	Data Analysis start time, e.g.
`monitoringScheduleConfig.scheduleConfig.scheduleExpression`	string	yes	A cron expression or 'NOW' that describes details about the monitoring schedule.
`monitoringScheduleName`	string	yes	Create-only in AWS: changing it requires replacing the resource.
`monitoringScheduleStatus`	enum	—	The status of a schedule job. Allowed: `Pending, Failed, Scheduled, Stopped`
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The time at which the schedule was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	A timestamp that indicates the last time the monitoring job was modified.
`lastSyncTime`	string	—
`monitoringScheduleArn`	string	—	The Amazon Resource Name (ARN) of the monitoring schedule.
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerPartnerApp # CLOUD CONTROL · AWS::SageMaker::PartnerApp

SageMakerPartnerApp manages AWS::SageMaker::PartnerApp through the AWS Cloud Control API.

kubectl get sagemakerpartnerapps · example on GitHub

Example

# SageMakerPartnerApp — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerpartnerapp
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerPartnerApp
metadata:
  name: example-sagemakerpartnerapp
  namespace: default
spec:
  appVersion: '1.0'
  applicationConfig:
    adminUsers:
    - example
    arguments:
      key: value
  authType: IAM
  clientToken: example
  enableAutoMinorVersionUpgrade: true
  enableIamSessionBasedIdentity: true
  executionRoleArn: arn:aws:iam::123456789012:role/example-role
  kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
  maintenanceConfig:
    maintenanceWindowStart: example
  name: example-sagemakerpartnerapp
  providerRef:
    name: example-sagemakerpartnerapp
    region: us-east-1
  tags:
  - key: example
    value: example
  tier: example
  type: lakera-guard

Spec

Field	Type	Required	Description
`appVersion`	string	—	The version of the PartnerApp.
`applicationConfig`	object	—	A collection of settings that specify the maintenance schedule for the PartnerApp.
`applicationConfig.adminUsers`	array<string>	—	A list of users with administrator privileges for the PartnerApp.
`applicationConfig.arguments`	map<string>	—	A list of arguments to pass to the PartnerApp.
`authType`	enum	yes	The Auth type of PartnerApp. Allowed: `IAM`
`clientToken`	string	—	The client token for the PartnerApp.
`enableAutoMinorVersionUpgrade`	boolean	—	Enables automatic minor version upgrades for the PartnerApp.
`enableIamSessionBasedIdentity`	boolean	—	Enables IAM Session based Identity for PartnerApp.
`executionRoleArn`	string	yes	The execution role for the user.
`kmsKeyId`	string	—	The AWS KMS customer managed key used to encrypt the data associated with the PartnerApp.
`maintenanceConfig`	object	—	A collection of settings that specify the maintenance schedule for the PartnerApp.
`maintenanceConfig.maintenanceWindowStart`	string	yes	The maintenance window start day and time for the PartnerApp.
`name`	string	yes	A name for the PartnerApp.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`tags`	array<object>	—	A list of tags to apply to the PartnerApp.
`tags[].key`	string	yes
`tags[].value`	string	yes
`tier`	string	yes	The tier of the PartnerApp.
`type`	enum	yes	The type of PartnerApp. Allowed: `lakera-guard, comet, deepchecks-llm-evaluation, fiddler`

Status

Field	Type	Required	Description
`arn`	string	—	The Amazon Resource Name (ARN) of the created PartnerApp.
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`baseUrl`	string	—	The AppServerUrl based on app and account-info.
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`currentVersionEolDate`	string	—	The end-of-life date for the current version of the PartnerApp.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerPipeline # CLOUD CONTROL · AWS::SageMaker::Pipeline

SageMakerPipeline manages AWS::SageMaker::Pipeline through the AWS Cloud Control API.

kubectl get sagemakerpipelines · example on GitHub

Example

# SageMakerPipeline — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerpipeline
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerPipeline
metadata:
  name: example-sagemakerpipeline
  namespace: default
spec:
  parallelismConfiguration:
    maxParallelExecutionSteps: 1
  pipelineDescription: Managed by konfig-konector
  pipelineDisplayName: example-sagemakerpipeline
  pipelineName: example-sagemakerpipeline
  providerRef:
    name: example-sagemakerpipeline
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`parallelismConfiguration`	object	—	SageMakerPipelineParallelismConfiguration is a nested property type of AWS::SageMaker::Pipeline.
`parallelismConfiguration.maxParallelExecutionSteps`	integer	yes	Maximum parallel execution steps
`pipelineDefinition`	object	yes
`pipelineDescription`	string	—	The description of the Pipeline.
`pipelineDisplayName`	string	—	The display name of the Pipeline.
`pipelineName`	string	yes	The name of the Pipeline.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	Role Arn
`tags`	array<object>	—
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerProcessingJob # CLOUD CONTROL · AWS::SageMaker::ProcessingJob

SageMakerProcessingJob manages AWS::SageMaker::ProcessingJob through the AWS Cloud Control API.

kubectl get sagemakerprocessingjobs · example on GitHub

Example

# SageMakerProcessingJob — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerprocessingjob
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerProcessingJob
metadata:
  name: example-sagemakerprocessingjob
  namespace: default
spec:
  appSpecification:
    containerArguments:
    - example
    containerEntrypoint:
    - example
    imageUri: example
  environment:
    key: value
  experimentConfig:
    experimentName: example-sagemakerprocessingjob
    runName: example-sagemakerprocessingjob
    trialComponentDisplayName: example-sagemakerprocessingjob
    trialName: example-sagemakerprocessingjob
  networkConfig:
    enableInterContainerTrafficEncryption: true
    enableNetworkIsolation: true
    vpcConfig:
      securityGroupIds:
      - example
      subnets:
      - example
  processingInputs:
  - appManaged: false
    datasetDefinition:
      dataDistributionType: FullyReplicated
      inputMode: File
      localPath: example
    inputName: example-sagemakerprocessingjob
    s3Input:
      localPath: example
      s3CompressionType: None
      s3DataDistributionType: FullyReplicated
      s3DataType: ManifestFile
      s3InputMode: File
      s3Uri: example
  processingJobName: example-sagemakerprocessingjob
  processingOutputConfig:
    kmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
    outputs:
    - appManaged: false
      outputName: example-sagemakerprocessingjob
  processingResources:
    clusterConfig:
      instanceCount: 1
      instanceType: example
      volumeKmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
      volumeSizeInGB: 1
  providerRef:
    name: example-sagemakerprocessingjob
    region: us-east-1
  roleArn: arn:aws:iam::123456789012:role/example-role
  stoppingCondition:
    maxRuntimeInSeconds: 1
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`appSpecification`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`appSpecification.containerArguments`	array<string>	—	The arguments for a container used to run a processing job.
`appSpecification.containerEntrypoint`	array<string>	—	The entrypoint for a container used to run a processing job.
`appSpecification.imageUri`	string	yes	The container image to be run by the processing job.
`environment`	map<string>	—	Create-only in AWS: changing it requires replacing the resource.
`experimentConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`experimentConfig.experimentName`	string	—	The name of an existing experiment to associate with the trial component.
`experimentConfig.runName`	string	—	The name of the experiment run to associate with the trial component.
`experimentConfig.trialComponentDisplayName`	string	—	The display name for the trial component.
`experimentConfig.trialName`	string	—	The name of an existing trial to associate the trial component with.
`networkConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`networkConfig.enableInterContainerTrafficEncryption`	boolean	—	Whether to encrypt all communications between distributed processing jobs.
`networkConfig.enableNetworkIsolation`	boolean	—	Whether to allow inbound and outbound network calls to and from the containers used for the processing job.
`networkConfig.vpcConfig`	object	—	SageMakerProcessingJobVpcConfig is a nested property type of AWS::SageMaker::ProcessingJob.
`networkConfig.vpcConfig.securityGroupIds`	array<string>	yes	The VPC security group IDs, in the form 'sg-xxxxxxxx'.
`networkConfig.vpcConfig.subnets`	array<string>	yes	The ID of the subnets in the VPC to which you want to connect your training job or model.
`processingInputs`	array<object>	—	An array of inputs configuring the data to download into the processing container.
`processingInputs[].appManaged`	boolean	—	When True, input operations such as data download are managed natively by the processing job application.
`processingInputs[].datasetDefinition`	object	—	SageMakerProcessingJobDatasetDefinition is a nested property type of AWS::SageMaker::ProcessingJob.
`processingInputs[].datasetDefinition.athenaDatasetDefinition`	object	—	SageMakerProcessingJobAthenaDatasetDefinition is a nested property type of AWS::SageMaker::ProcessingJob.
`processingInputs[].datasetDefinition.athenaDatasetDefinition.catalog`	string	yes	The name of the data catalog used in Athena query execution.
`processingInputs[].datasetDefinition.athenaDatasetDefinition.database`	string	yes	The name of the database used in the Athena query execution.
`processingInputs[].datasetDefinition.athenaDatasetDefinition.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt data generated from an Athena query execution.
`processingInputs[].datasetDefinition.athenaDatasetDefinition.outputCompression`	enum	—	The compression used for Athena query results. Allowed: `GZIP, SNAPPY, ZLIB`
`processingInputs[].datasetDefinition.athenaDatasetDefinition.outputFormat`	enum	yes	The data storage format for Athena query results. Allowed: `PARQUET, AVRO, ORC, JSON, TEXTFILE`
`processingInputs[].datasetDefinition.athenaDatasetDefinition.outputS3Uri`	string	yes	The location in Amazon S3 where Athena query results are stored.
`processingInputs[].datasetDefinition.athenaDatasetDefinition.queryString`	string	yes	The SQL query statements, to be executed.
`processingInputs[].datasetDefinition.athenaDatasetDefinition.workGroup`	string	—	The name of the workgroup in which the Athena query is being started.
`processingInputs[].datasetDefinition.dataDistributionType`	enum	—	Whether the generated dataset is FullyReplicated or ShardedByS3Key (default). Allowed: `FullyReplicated, ShardedByS3Key`
`processingInputs[].datasetDefinition.inputMode`	enum	—	Whether to use File or Pipe input mode. Allowed: `File, Pipe`
`processingInputs[].datasetDefinition.localPath`	string	—	The local path where you want Amazon SageMaker to download the Dataset Definition inputs to run a processing job.
`processingInputs[].datasetDefinition.redshiftDatasetDefinition`	object	—	SageMakerProcessingJobRedshiftDatasetDefinition is a nested property type of AWS::SageMaker::ProcessingJob.
`processingInputs[].datasetDefinition.redshiftDatasetDefinition.clusterId`	string	yes	The Redshift cluster Identifier.
`processingInputs[].datasetDefinition.redshiftDatasetDefinition.clusterRoleArn`	string	yes	The IAM role attached to your Redshift cluster that Amazon SageMaker uses to generate datasets.
`processingInputs[].datasetDefinition.redshiftDatasetDefinition.database`	string	yes	The name of the Redshift database used in Redshift query execution.
`processingInputs[].datasetDefinition.redshiftDatasetDefinition.dbUser`	string	yes	The database user name used in Redshift query execution.
`processingInputs[].datasetDefinition.redshiftDatasetDefinition.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt data from a Redshift execution.
`processingInputs[].datasetDefinition.redshiftDatasetDefinition.outputCompression`	enum	—	The compression used for Redshift query results. Allowed: `None, GZIP, SNAPPY, ZSTD, BZIP2`
`processingInputs[].datasetDefinition.redshiftDatasetDefinition.outputFormat`	enum	yes	The data storage format for Redshift query results. Allowed: `PARQUET, CSV`
`processingInputs[].datasetDefinition.redshiftDatasetDefinition.outputS3Uri`	string	yes	The location in Amazon S3 where the Redshift query results are stored.
`processingInputs[].datasetDefinition.redshiftDatasetDefinition.queryString`	string	yes	The SQL query statements to be executed.
`processingInputs[].inputName`	string	yes	The name for the processing job input.
`processingInputs[].s3Input`	object	—	SageMakerProcessingJobS3Input is a nested property type of AWS::SageMaker::ProcessingJob.
`processingInputs[].s3Input.localPath`	string	—	The local path in your container where you want Amazon SageMaker to write input data to.
`processingInputs[].s3Input.s3CompressionType`	enum	—	Whether to GZIP-decompress the data in Amazon S3 as it is streamed into the processing container. Allowed: `None, Gzip`
`processingInputs[].s3Input.s3DataDistributionType`	enum	—	Whether to distribute the data from Amazon S3 to all processing instances with `FullyReplicated`, or whether the data from Amazon S3 is shared by Amazon S3 key, downloading one shard of data to each processing instance. Allowed: `FullyReplicated, ShardedByS3Key`
`processingInputs[].s3Input.s3DataType`	enum	yes	Whether you use an S3Prefix or a ManifestFile for the data type. Allowed: `ManifestFile, S3Prefix`
`processingInputs[].s3Input.s3InputMode`	enum	—	Whether to use File or Pipe input mode. Allowed: `File, Pipe`
`processingInputs[].s3Input.s3Uri`	string	yes	The URI of the Amazon S3 prefix Amazon SageMaker downloads data required to run a processing job.
`processingJobName`	string	—	The name of the processing job.
`processingOutputConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`processingOutputConfig.kmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt the processing job output.
`processingOutputConfig.outputs`	array<object>	yes	An array of outputs configuring the data to upload from the processing container.
`processingOutputConfig.outputs[].appManaged`	boolean	—	When True, output operations such as data upload are managed natively by the processing job application.
`processingOutputConfig.outputs[].featureStoreOutput`	object	—	SageMakerProcessingJobFeatureStoreOutput is a nested property type of AWS::SageMaker::ProcessingJob.
`processingOutputConfig.outputs[].featureStoreOutput.featureGroupName`	string	yes	The name of the Amazon SageMaker FeatureGroup to use as the destination for processing job output.
`processingOutputConfig.outputs[].outputName`	string	yes	The name for the processing job output.
`processingOutputConfig.outputs[].s3Output`	object	—	SageMakerProcessingJobS3Output is a nested property type of AWS::SageMaker::ProcessingJob.
`processingOutputConfig.outputs[].s3Output.localPath`	string	—	The local path of a directory where you want Amazon SageMaker to upload its contents to Amazon S3.
`processingOutputConfig.outputs[].s3Output.s3UploadMode`	enum	yes	Whether to upload the results of the processing job continuously or after the job completes. Allowed: `Continuous, EndOfJob`
`processingOutputConfig.outputs[].s3Output.s3Uri`	string	yes	A URI that identifies the Amazon S3 bucket where you want Amazon SageMaker to save the results of a processing job.
`processingResources`	object	yes	Create-only in AWS: changing it requires replacing the resource.
`processingResources.clusterConfig`	object	yes	SageMakerProcessingJobClusterConfig is a nested property type of AWS::SageMaker::ProcessingJob.
`processingResources.clusterConfig.instanceCount`	integer	yes	The number of ML compute instances to use in the processing job.
`processingResources.clusterConfig.instanceType`	string	yes	The ML compute instance type for the processing job.
`processingResources.clusterConfig.volumeKmsKeyId`	string	—	The AWS Key Management Service (AWS KMS) key that Amazon SageMaker uses to encrypt data on the storage volume attached to the ML compute instance(s) that run the processing job.
`processingResources.clusterConfig.volumeSizeInGB`	integer	yes	The size of the ML storage volume in gigabytes that you want to provision.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`roleArn`	string	yes	The Amazon Resource Name (ARN) of an IAM role that Amazon SageMaker can assume to perform tasks on your behalf.
`stoppingCondition`	object	—	Create-only in AWS: changing it requires replacing the resource.
`stoppingCondition.maxRuntimeInSeconds`	integer	yes	Specifies the maximum runtime in seconds.
`tags`	array<object>	—	(Optional) An array of key-value pairs.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`autoMLJobArn`	string	—	The ARN of an AutoML job associated with this processing job.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The time at which the processing job was created.
`exitMessage`	string	—	An optional string, up to one KB in size, that contains metadata from the processing container when the processing job exits.
`failureReason`	string	—	A string, up to one KB in size, that contains the reason a processing job failed, if it failed.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	The time at which the processing job was last modified.
`lastSyncTime`	string	—
`monitoringScheduleArn`	string	—	The ARN of a monitoring schedule for an endpoint associated with this processing job.
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`processingEndTime`	string	—	The time at which the processing job completed.
`processingJobArn`	string	—	The Amazon Resource Name (ARN) of the processing job.
`processingJobStatus`	enum	—	Provides the status of a processing job. Allowed: `Completed, InProgress, Stopping, Stopped, Failed`
`processingStartTime`	string	—	The time at which the processing job started.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`trainingJobArn`	string	—	The ARN of a training job associated with this processing job Read-only: reported by AWS after creation.

SageMakerProject # CLOUD CONTROL · AWS::SageMaker::Project

SageMakerProject manages AWS::SageMaker::Project through the AWS Cloud Control API.

kubectl get sagemakerprojects · example on GitHub

Example

# SageMakerProject — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerproject
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerProject
metadata:
  name: example-sagemakerproject
  namespace: default
spec:
  projectDescription: Managed by konfig-konector
  projectName: example-sagemakerproject
  providerRef:
    name: example-sagemakerproject
    region: us-east-1
  serviceCatalogProvisionedProductDetails:
    provisionedProductStatusMessage: example
  serviceCatalogProvisioningDetails:
    productId: example-id
    provisioningParameters:
    - key: example
      value: example
  tags:
  - key: example
    value: example
  templateProviderDetails:
  - cfnTemplateProviderDetail:
      parameters:
      - key: example
        value: example
      roleARN: arn:aws:iam::123456789012:role/example-role
      templateName: example-sagemakerproject
      templateURL: example

Spec

Field	Type	Required	Description
`projectDescription`	string	—	Create-only in AWS: changing it requires replacing the resource.
`projectName`	string	yes	Create-only in AWS: changing it requires replacing the resource.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`serviceCatalogProvisionedProductDetails`	object	—	Provisioned ServiceCatalog Details
`serviceCatalogProvisionedProductDetails.provisionedProductId`	string	—
`serviceCatalogProvisionedProductDetails.provisionedProductStatusMessage`	string	—
`serviceCatalogProvisioningDetails`	object	—	Input ServiceCatalog Provisioning Details Create-only in AWS: changing it requires replacing the resource.
`serviceCatalogProvisioningDetails.pathId`	string	—
`serviceCatalogProvisioningDetails.productId`	string	yes
`serviceCatalogProvisioningDetails.provisioningArtifactId`	string	—
`serviceCatalogProvisioningDetails.provisioningParameters`	array<object>	—	Parameters specified by the administrator that are required for provisioning the product.
`serviceCatalogProvisioningDetails.provisioningParameters[].key`	string	yes
`serviceCatalogProvisioningDetails.provisioningParameters[].value`	string	yes
`tags`	array<object>	—	An array of key-value pairs to apply to this resource.
`tags[].key`	string	yes
`tags[].value`	string	yes
`templateProviderDetails`	array<object>	—	An array of template providers associated with the project.
`templateProviderDetails[].cfnTemplateProviderDetail`	object	—	SageMakerProjectCfnTemplateProviderDetail is a nested property type of AWS::SageMaker::Project.
`templateProviderDetails[].cfnTemplateProviderDetail.parameters`	array<object>	—
`templateProviderDetails[].cfnTemplateProviderDetail.parameters[].key`	string	yes
`templateProviderDetails[].cfnTemplateProviderDetail.parameters[].value`	string	yes
`templateProviderDetails[].cfnTemplateProviderDetail.roleARN`	string	—	The Amazon Resource Name (ARN) of the IAM role used by the template provider.
`templateProviderDetails[].cfnTemplateProviderDetail.templateName`	string	yes	The name of the template used for the project.
`templateProviderDetails[].cfnTemplateProviderDetail.templateURL`	string	yes	The URL of the CloudFormation template.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	The time at which the project was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`projectArn`	string	—	Read-only: reported by AWS after creation.
`projectId`	string	—	Read-only: reported by AWS after creation.
`projectStatus`	enum	—	The status of a project. Allowed: `Pending, CreateInProgress, CreateCompleted, CreateFailed, DeleteInProgress, DeleteFailed, DeleteCompleted`
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.

SageMakerSpace # CLOUD CONTROL · AWS::SageMaker::Space

SageMakerSpace manages AWS::SageMaker::Space through the AWS Cloud Control API.

kubectl get sagemakerspaces · example on GitHub

Example

# SageMakerSpace — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerspace
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerSpace
metadata:
  name: example-sagemakerspace
  namespace: default
spec:
  domainId: example-id
  ownershipSettings:
    ownerUserProfileName: example-sagemakerspace
  providerRef:
    name: example-sagemakerspace
    region: us-east-1
  spaceDisplayName: example-sagemakerspace
  spaceName: example-sagemakerspace
  spaceSettings:
    appType: JupyterServer
    codeEditorAppSettings:
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
    jupyterLabAppSettings:
      codeRepositories:
      - repositoryUrl: example
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
    jupyterServerAppSettings:
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    kernelGatewayAppSettings:
      customImages:
      - appImageConfigName: example-sagemakerspace
        imageName: example-sagemakerspace
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    remoteAccess: ENABLED
    spaceManagedResources: ENABLED
    spaceStorageSettings:
      ebsStorageSettings:
        ebsVolumeSizeInGb: 5
  spaceSharingSettings:
    sharingType: Private
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`domainId`	string	yes	The ID of the associated Domain.
`ownershipSettings`	object	—	Create-only in AWS: changing it requires replacing the resource.
`ownershipSettings.ownerUserProfileName`	string	yes
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`spaceDisplayName`	string	—
`spaceName`	string	yes	A name for the Space.
`spaceSettings`	object	—	A collection of settings.
`spaceSettings.appType`	enum	—	Allowed: `JupyterServer, KernelGateway, TensorBoard, RStudioServerPro, RSessionGateway, JupyterLab, CodeEditor`
`spaceSettings.codeEditorAppSettings`	object	—	The CodeEditor app settings.
`spaceSettings.codeEditorAppSettings.appLifecycleManagement`	object	—	SageMakerSpaceSpaceAppLifecycleManagement is a nested property type of AWS::SageMaker::Space.
`spaceSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings`	object	—	SageMakerSpaceSpaceIdleSettings is a nested property type of AWS::SageMaker::Space.
`spaceSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings.idleTimeoutInMinutes`	integer	—	The space idle timeout value set in minutes
`spaceSettings.codeEditorAppSettings.defaultResourceSpec`	object	—	SageMakerSpaceResourceSpec is a nested property type of AWS::SageMaker::Space.
`spaceSettings.codeEditorAppSettings.defaultResourceSpec.instanceType`	string	—	The instance type that the image version runs on.
`spaceSettings.codeEditorAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`spaceSettings.codeEditorAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The ARN of the SageMaker image that the image version belongs to.
`spaceSettings.codeEditorAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The ARN of the image version created on the instance.
`spaceSettings.customFileSystems`	array<object>	—
`spaceSettings.customFileSystems[].efsFileSystem`	object	—	SageMakerSpaceEFSFileSystem is a nested property type of AWS::SageMaker::Space.
`spaceSettings.customFileSystems[].efsFileSystem.fileSystemId`	string	yes
`spaceSettings.customFileSystems[].fSxLustreFileSystem`	object	—	SageMakerSpaceFSxLustreFileSystem is a nested property type of AWS::SageMaker::Space.
`spaceSettings.customFileSystems[].fSxLustreFileSystem.fileSystemId`	string	yes
`spaceSettings.customFileSystems[].s3FileSystem`	object	—	SageMakerSpaceS3FileSystem is a nested property type of AWS::SageMaker::Space.
`spaceSettings.customFileSystems[].s3FileSystem.s3Uri`	string	—
`spaceSettings.jupyterLabAppSettings`	object	—	The JupyterLab app settings.
`spaceSettings.jupyterLabAppSettings.appLifecycleManagement`	object	—	SageMakerSpaceSpaceAppLifecycleManagement is a nested property type of AWS::SageMaker::Space.
`spaceSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings`	object	—	SageMakerSpaceSpaceIdleSettings is a nested property type of AWS::SageMaker::Space.
`spaceSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.idleTimeoutInMinutes`	integer	—	The space idle timeout value set in minutes
`spaceSettings.jupyterLabAppSettings.codeRepositories`	array<object>	—	A list of CodeRepositories available for use with JupyterLab apps.
`spaceSettings.jupyterLabAppSettings.codeRepositories[].repositoryUrl`	string	yes	A CodeRepository (valid URL) to be used within Jupyter's Git extension.
`spaceSettings.jupyterLabAppSettings.defaultResourceSpec`	object	—	SageMakerSpaceResourceSpec is a nested property type of AWS::SageMaker::Space.
`spaceSettings.jupyterLabAppSettings.defaultResourceSpec.instanceType`	string	—	The instance type that the image version runs on.
`spaceSettings.jupyterLabAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`spaceSettings.jupyterLabAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The ARN of the SageMaker image that the image version belongs to.
`spaceSettings.jupyterLabAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The ARN of the image version created on the instance.
`spaceSettings.jupyterServerAppSettings`	object	—	The Jupyter server's app settings.
`spaceSettings.jupyterServerAppSettings.defaultResourceSpec`	object	—	SageMakerSpaceResourceSpec is a nested property type of AWS::SageMaker::Space.
`spaceSettings.jupyterServerAppSettings.defaultResourceSpec.instanceType`	string	—	The instance type that the image version runs on.
`spaceSettings.jupyterServerAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`spaceSettings.jupyterServerAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The ARN of the SageMaker image that the image version belongs to.
`spaceSettings.jupyterServerAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The ARN of the image version created on the instance.
`spaceSettings.jupyterServerAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with JupyterServer apps.
`spaceSettings.kernelGatewayAppSettings`	object	—	The kernel gateway app settings.
`spaceSettings.kernelGatewayAppSettings.customImages`	array<object>	—	A list of custom SageMaker images that are configured to run as a KernelGateway app.
`spaceSettings.kernelGatewayAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`spaceSettings.kernelGatewayAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`spaceSettings.kernelGatewayAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`spaceSettings.kernelGatewayAppSettings.defaultResourceSpec`	object	—	The default instance type and the Amazon Resource Name (ARN) of the default SageMaker image used by the KernelGateway app.
`spaceSettings.kernelGatewayAppSettings.defaultResourceSpec.instanceType`	string	—	The instance type that the image version runs on.
`spaceSettings.kernelGatewayAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`spaceSettings.kernelGatewayAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The ARN of the SageMaker image that the image version belongs to.
`spaceSettings.kernelGatewayAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The ARN of the image version created on the instance.
`spaceSettings.kernelGatewayAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with KernelGateway apps.
`spaceSettings.remoteAccess`	enum	—	This is a flag used to indicate if remote access is enabled. Allowed: `ENABLED, DISABLED`
`spaceSettings.spaceManagedResources`	enum	—	This is a flag used to indicate if space managed resources needs to be created. Allowed: `ENABLED, DISABLED`
`spaceSettings.spaceStorageSettings`	object	—	Default storage settings for a space.
`spaceSettings.spaceStorageSettings.ebsStorageSettings`	object	—	SageMakerSpaceEbsStorageSettings is a nested property type of AWS::SageMaker::Space.
`spaceSettings.spaceStorageSettings.ebsStorageSettings.ebsVolumeSizeInGb`	integer	yes	Size of the Amazon EBS volume in Gb
`spaceSharingSettings`	object	—	Create-only in AWS: changing it requires replacing the resource.
`spaceSharingSettings.sharingType`	enum	yes	Allowed: `Private, Shared`
`tags`	array<object>	—	A list of tags to apply to the space.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`spaceArn`	string	—	The space Amazon Resource Name (ARN).
`url`	string	—	Read-only: reported by AWS after creation.

SageMakerStudioLifecycleConfig # CLOUD CONTROL · AWS::SageMaker::StudioLifecycleConfig

SageMakerStudioLifecycleConfig manages AWS::SageMaker::StudioLifecycleConfig through the AWS Cloud Control API.

kubectl get sagemakerstudiolifecycleconfigs · example on GitHub

Example

# SageMakerStudioLifecycleConfig — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerstudiolifecycleconfig
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerStudioLifecycleConfig
metadata:
  name: example-sagemakerstudiolifecycleconfig
  namespace: default
spec:
  providerRef:
    name: example-sagemakerstudiolifecycleconfig
    region: us-east-1
  studioLifecycleConfigAppType: JupyterServer
  studioLifecycleConfigContent: example
  studioLifecycleConfigName: example-sagemakerstudiolifecycleconfig
  tags:
  - key: example
    value: example

Spec

Field	Type	Required	Description
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`studioLifecycleConfigAppType`	enum	yes	The App type that the Lifecycle Configuration is attached to. Allowed: `JupyterServer, KernelGateway, CodeEditor, JupyterLab`
`studioLifecycleConfigContent`	string	yes	The content of your Amazon SageMaker Studio Lifecycle Configuration script.
`studioLifecycleConfigName`	string	yes	The name of the Amazon SageMaker Studio Lifecycle Configuration.
`tags`	array<object>	—	Tags to be associated with the Lifecycle Configuration.
`tags[].key`	string	yes
`tags[].value`	string	yes

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`studioLifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration.

SageMakerTrialComponent # CLOUD CONTROL · AWS::SageMaker::TrialComponent

SageMakerTrialComponent manages AWS::SageMaker::TrialComponent through the AWS Cloud Control API.

kubectl get sagemakertrialcomponents · example on GitHub

Example

# SageMakerTrialComponent — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakertrialcomponent
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerTrialComponent
metadata:
  name: example-sagemakertrialcomponent
  namespace: default
spec:
  displayName: example-sagemakertrialcomponent
  inputArtifacts: {}
  metadataProperties:
    generatedBy: example
    repository: example
  outputArtifacts: {}
  parameters: {}
  providerRef:
    name: example-sagemakertrialcomponent
    region: us-east-1
  status:
    message: example
    primaryStatus: InProgress
  tags:
  - key: example
    value: example
  trialComponentName: example-sagemakertrialcomponent

Spec

Field	Type	Required	Description
`displayName`	string	—	The name of the component as displayed.
`inputArtifacts`	map<string>	—	The input artifacts for the component.
`metadataProperties`	object	—	Create-only in AWS: changing it requires replacing the resource.
`metadataProperties.commitId`	string	—	The commit ID.
`metadataProperties.generatedBy`	string	—	The entity this entity was generated by.
`metadataProperties.projectId`	string	—	The project ID.
`metadataProperties.repository`	string	—	The repository.
`outputArtifacts`	map<string>	—	The output artifacts for the component.
`parameters`	map<string>	—	The hyperparameters for the component.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`status`	object	—	SageMakerTrialComponentTrialComponentStatus is a nested property type of AWS::SageMaker::TrialComponent.
`status.message`	string	—	If the component failed, a message describing why.
`status.primaryStatus`	enum	—	The status of the trial component. Allowed: `InProgress, Completed, Failed, Stopping, Stopped`
`tags`	array<object>	—	A list of tags to associate with the trial component.
`tags[].key`	string	yes	The tag key.
`tags[].value`	string	yes	The tag value.
`trialComponentName`	string	yes	The name of the trial component.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`creationTime`	string	—	When the trial component was created.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastModifiedTime`	string	—	When the trial component was last modified.
`lastSyncTime`	string	—
`lineageGroupArn`	string	—	The Amazon Resource Name (ARN) of the lineage group.
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`trialComponentArn`	string	—	The Amazon Resource Name (ARN) of the trial component.

SageMakerUserProfile # CLOUD CONTROL · AWS::SageMaker::UserProfile

SageMakerUserProfile manages AWS::SageMaker::UserProfile through the AWS Cloud Control API.

kubectl get sagemakeruserprofiles · example on GitHub

Example

# SageMakerUserProfile — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakeruserprofile
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerUserProfile
metadata:
  name: example-sagemakeruserprofile
  namespace: default
spec:
  domainId: example-id
  providerRef:
    name: example-sagemakeruserprofile
    region: us-east-1
  singleSignOnUserIdentifier: example
  singleSignOnUserValue: example
  tags:
  - key: example
    value: example
  userProfileName: example-sagemakeruserprofile
  userSettings:
    autoMountHomeEFS: Enabled
    codeEditorAppSettings:
      builtInLifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
      customImages:
      - appImageConfigName: example-sagemakeruserprofile
        imageName: example-sagemakeruserprofile
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    customPosixUserConfig:
      gid: 1001
      uid: 10000
    defaultLandingUri: example
    executionRole: example
    jupyterLabAppSettings:
      builtInLifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
      codeRepositories:
      - repositoryUrl: example
      customImages:
      - appImageConfigName: example-sagemakeruserprofile
        imageName: example-sagemakeruserprofile
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      emrSettings:
        assumableRoleArns:
        - arn:aws:iam::123456789012:role/example-role
        executionRoleArns:
        - arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    jupyterServerAppSettings:
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    kernelGatewayAppSettings:
      customImages:
      - appImageConfigName: example-sagemakeruserprofile
        imageName: example-sagemakeruserprofile
        imageVersionNumber: 1
      defaultResourceSpec:
        instanceType: example
        lifecycleConfigArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageArn: arn:aws:iam::123456789012:role/example-role
        sageMakerImageVersionArn: arn:aws:iam::123456789012:role/example-role
        trainingPlanArn: arn:aws:iam::123456789012:role/example-role
      lifecycleConfigArns:
      - example
    rStudioServerProAppSettings:
      accessStatus: ENABLED
      userGroup: R_STUDIO_ADMIN
    securityGroups:
    - example
    sharingSettings:
      notebookOutputOption: Allowed
      s3KmsKeyId: arn:aws:kms:us-east-1:123456789012:key/1234abcd-12ab-34cd-56ef-1234567890ab
      s3OutputPath: example
    spaceStorageSettings:
      defaultEbsStorageSettings:
        defaultEbsVolumeSizeInGb: 5
        maximumEbsVolumeSizeInGb: 5
    studioWebPortal: ENABLED
    studioWebPortalSettings:
      hiddenAppTypes:
      - example
      hiddenInstanceTypes:
      - example
      hiddenMlTools:
      - example
      hiddenSageMakerImageVersionAliases:
      - sageMakerImageName: sagemaker_distribution
        versionAliases:
        - example

Spec

Field	Type	Required	Description
`domainId`	string	yes	The ID of the associated Domain.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`singleSignOnUserIdentifier`	string	—	A specifier for the type of value specified in SingleSignOnUserValue.
`singleSignOnUserValue`	string	—	The username of the associated AWS Single Sign-On User for this UserProfile.
`tags`	array<object>	—	A list of tags to apply to the user profile.
`tags[].key`	string	yes
`tags[].value`	string	yes
`userProfileName`	string	yes	A name for the UserProfile.
`userSettings`	object	—	A collection of settings.
`userSettings.autoMountHomeEFS`	enum	—	Indicates whether auto-mounting of an EFS volume is supported for the user profile. Allowed: `Enabled, Disabled, DefaultAsDomain`
`userSettings.codeEditorAppSettings`	object	—	SageMakerUserProfileCodeEditorAppSettings is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.codeEditorAppSettings.appLifecycleManagement`	object	—	SageMakerUserProfileAppLifecycleManagement is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings`	object	—	SageMakerUserProfileIdleSettings is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings.idleTimeoutInMinutes`	integer	—
`userSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings.lifecycleManagement`	enum	—	Allowed: `ENABLED, DISABLED`
`userSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings.maxIdleTimeoutInMinutes`	integer	—
`userSettings.codeEditorAppSettings.appLifecycleManagement.idleSettings.minIdleTimeoutInMinutes`	integer	—
`userSettings.codeEditorAppSettings.builtInLifecycleConfigArn`	string	—	The lifecycle configuration that runs before the default lifecycle configuration.
`userSettings.codeEditorAppSettings.customImages`	array<object>	—	A list of custom images for use for CodeEditor apps.
`userSettings.codeEditorAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`userSettings.codeEditorAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`userSettings.codeEditorAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`userSettings.codeEditorAppSettings.defaultResourceSpec`	object	—	The default instance type and the Amazon Resource Name (ARN) of the default SageMaker image used by the CodeEditor app.
`userSettings.codeEditorAppSettings.defaultResourceSpec.instanceType`	string	—
`userSettings.codeEditorAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`userSettings.codeEditorAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The ARN of the SageMaker image that the image version belongs to.
`userSettings.codeEditorAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The ARN of the image version created on the instance.
`userSettings.codeEditorAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`userSettings.codeEditorAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with CodeEditor apps.
`userSettings.customFileSystemConfigs`	array<object>	—
`userSettings.customFileSystemConfigs[].efsFileSystemConfig`	object	—	SageMakerUserProfileEFSFileSystemConfig is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.customFileSystemConfigs[].efsFileSystemConfig.fileSystemId`	string	yes
`userSettings.customFileSystemConfigs[].efsFileSystemConfig.fileSystemPath`	string	—
`userSettings.customFileSystemConfigs[].fSxLustreFileSystemConfig`	object	—	SageMakerUserProfileFSxLustreFileSystemConfig is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.customFileSystemConfigs[].fSxLustreFileSystemConfig.fileSystemId`	string	yes
`userSettings.customFileSystemConfigs[].fSxLustreFileSystemConfig.fileSystemPath`	string	—
`userSettings.customFileSystemConfigs[].s3FileSystemConfig`	object	—	SageMakerUserProfileS3FileSystemConfig is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.customFileSystemConfigs[].s3FileSystemConfig.mountPath`	string	—
`userSettings.customFileSystemConfigs[].s3FileSystemConfig.s3Uri`	string	—
`userSettings.customPosixUserConfig`	object	—	SageMakerUserProfileCustomPosixUserConfig is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.customPosixUserConfig.gid`	integer	yes
`userSettings.customPosixUserConfig.uid`	integer	yes
`userSettings.defaultLandingUri`	string	—	Defines which Amazon SageMaker application users are directed to by default.
`userSettings.executionRole`	string	—	The user profile Amazon Resource Name (ARN).
`userSettings.jupyterLabAppSettings`	object	—	SageMakerUserProfileJupyterLabAppSettings is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.jupyterLabAppSettings.appLifecycleManagement`	object	—	SageMakerUserProfileAppLifecycleManagement is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings`	object	—	SageMakerUserProfileIdleSettings is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.idleTimeoutInMinutes`	integer	—
`userSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.lifecycleManagement`	enum	—	Allowed: `ENABLED, DISABLED`
`userSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.maxIdleTimeoutInMinutes`	integer	—
`userSettings.jupyterLabAppSettings.appLifecycleManagement.idleSettings.minIdleTimeoutInMinutes`	integer	—
`userSettings.jupyterLabAppSettings.builtInLifecycleConfigArn`	string	—	The lifecycle configuration that runs before the default lifecycle configuration.
`userSettings.jupyterLabAppSettings.codeRepositories`	array<object>	—	A list of CodeRepositories available for use with JupyterLab apps.
`userSettings.jupyterLabAppSettings.codeRepositories[].repositoryUrl`	string	yes	A CodeRepository (valid URL) to be used within Jupyter's Git extension.
`userSettings.jupyterLabAppSettings.customImages`	array<object>	—	A list of custom images available for use for JupyterLab apps
`userSettings.jupyterLabAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`userSettings.jupyterLabAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`userSettings.jupyterLabAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`userSettings.jupyterLabAppSettings.defaultResourceSpec`	object	—	The default instance type and the Amazon Resource Name (ARN) of the default SageMaker image used by the JupyterLab app.
`userSettings.jupyterLabAppSettings.defaultResourceSpec.instanceType`	string	—
`userSettings.jupyterLabAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`userSettings.jupyterLabAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The ARN of the SageMaker image that the image version belongs to.
`userSettings.jupyterLabAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The ARN of the image version created on the instance.
`userSettings.jupyterLabAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`userSettings.jupyterLabAppSettings.emrSettings`	object	—	The configuration parameters for EMR settings.
`userSettings.jupyterLabAppSettings.emrSettings.assumableRoleArns`	array<string>	—	An array of Amazon Resource Names (ARNs) of the IAM roles that the execution role of SageMaker can assume.
`userSettings.jupyterLabAppSettings.emrSettings.executionRoleArns`	array<string>	—	An array of ARNs of IAM roles used by EMR cluster instances or job execution environments.
`userSettings.jupyterLabAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with JupyterLab apps.
`userSettings.jupyterServerAppSettings`	object	—	The Jupyter server's app settings.
`userSettings.jupyterServerAppSettings.defaultResourceSpec`	object	—	SageMakerUserProfileResourceSpec is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.jupyterServerAppSettings.defaultResourceSpec.instanceType`	string	—
`userSettings.jupyterServerAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`userSettings.jupyterServerAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The ARN of the SageMaker image that the image version belongs to.
`userSettings.jupyterServerAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The ARN of the image version created on the instance.
`userSettings.jupyterServerAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`userSettings.jupyterServerAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with JupyterServer apps.
`userSettings.kernelGatewayAppSettings`	object	—	The kernel gateway app settings.
`userSettings.kernelGatewayAppSettings.customImages`	array<object>	—	A list of custom SageMaker images that are configured to run as a KernelGateway app.
`userSettings.kernelGatewayAppSettings.customImages[].appImageConfigName`	string	yes	The Name of the AppImageConfig.
`userSettings.kernelGatewayAppSettings.customImages[].imageName`	string	yes	The name of the CustomImage.
`userSettings.kernelGatewayAppSettings.customImages[].imageVersionNumber`	integer	—	The version number of the CustomImage.
`userSettings.kernelGatewayAppSettings.defaultResourceSpec`	object	—	The default instance type and the Amazon Resource Name (ARN) of the default SageMaker image used by the KernelGateway app.
`userSettings.kernelGatewayAppSettings.defaultResourceSpec.instanceType`	string	—
`userSettings.kernelGatewayAppSettings.defaultResourceSpec.lifecycleConfigArn`	string	—	The Amazon Resource Name (ARN) of the Lifecycle Configuration to attach to the Resource.
`userSettings.kernelGatewayAppSettings.defaultResourceSpec.sageMakerImageArn`	string	—	The ARN of the SageMaker image that the image version belongs to.
`userSettings.kernelGatewayAppSettings.defaultResourceSpec.sageMakerImageVersionArn`	string	—	The ARN of the image version created on the instance.
`userSettings.kernelGatewayAppSettings.defaultResourceSpec.trainingPlanArn`	string	—	The Amazon Resource Name (ARN) of the training plan to use for the ResourceSpec.
`userSettings.kernelGatewayAppSettings.lifecycleConfigArns`	array<string>	—	A list of LifecycleConfigArns available for use with KernelGateway apps.
`userSettings.rStudioServerProAppSettings`	object	—	SageMakerUserProfileRStudioServerProAppSettings is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.rStudioServerProAppSettings.accessStatus`	enum	—	Indicates whether the current user has access to the RStudioServerPro app. Allowed: `ENABLED, DISABLED`
`userSettings.rStudioServerProAppSettings.userGroup`	enum	—	The level of permissions that the user has within the RStudioServerPro app. Allowed: `R_STUDIO_ADMIN, R_STUDIO_USER`
`userSettings.securityGroups`	array<string>	—	The security groups for the Amazon Virtual Private Cloud (VPC) that Studio uses for communication.
`userSettings.sharingSettings`	object	—	The sharing settings.
`userSettings.sharingSettings.notebookOutputOption`	enum	—	Whether to include the notebook cell output when sharing the notebook. Allowed: `Allowed, Disabled`
`userSettings.sharingSettings.s3KmsKeyId`	string	—	When NotebookOutputOption is Allowed, the AWS Key Management Service (KMS) encryption key ID used to encrypt the notebook cell output in the Amazon S3 bucket.
`userSettings.sharingSettings.s3OutputPath`	string	—	When NotebookOutputOption is Allowed, the Amazon S3 bucket used to store the shared notebook snapshots.
`userSettings.spaceStorageSettings`	object	—	SageMakerUserProfileDefaultSpaceStorageSettings is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.spaceStorageSettings.defaultEbsStorageSettings`	object	—	SageMakerUserProfileDefaultEbsStorageSettings is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.spaceStorageSettings.defaultEbsStorageSettings.defaultEbsVolumeSizeInGb`	integer	yes	Default size of the Amazon EBS volume in Gb
`userSettings.spaceStorageSettings.defaultEbsStorageSettings.maximumEbsVolumeSizeInGb`	integer	yes	Maximum size of the Amazon EBS volume in Gb.
`userSettings.studioWebPortal`	enum	—	Indicates whether the Studio experience is available to users. Allowed: `ENABLED, DISABLED`
`userSettings.studioWebPortalSettings`	object	—	SageMakerUserProfileStudioWebPortalSettings is a nested property type of AWS::SageMaker::UserProfile.
`userSettings.studioWebPortalSettings.hiddenAppTypes`	array<string>	—	Applications supported in Studio that are hidden from the Studio left navigation pane.
`userSettings.studioWebPortalSettings.hiddenInstanceTypes`	array<string>	—	The instance types you are hiding from the Studio user interface.
`userSettings.studioWebPortalSettings.hiddenMlTools`	array<string>	—	The machine learning tools that are hidden from the Studio left navigation pane.
`userSettings.studioWebPortalSettings.hiddenSageMakerImageVersionAliases`	array<object>	—	The version aliases you are hiding from the Studio user interface.
`userSettings.studioWebPortalSettings.hiddenSageMakerImageVersionAliases[].sageMakerImageName`	enum	—	The SageMaker image name that you are hiding from the Studio user interface. Allowed: `sagemaker_distribution`
`userSettings.studioWebPortalSettings.hiddenSageMakerImageVersionAliases[].versionAliases`	array<string>	—

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`userProfileArn`	string	—	The user profile Amazon Resource Name (ARN).

SageMakerWorkforce # CLOUD CONTROL · AWS::SageMaker::Workforce

SageMakerWorkforce manages AWS::SageMaker::Workforce through the AWS Cloud Control API.

kubectl get sagemakerworkforces · example on GitHub

Example

# SageMakerWorkforce — SageMaker
# Full-options example generated from the CRD schema. Optional fields are
# included with sample values; delete what you don't need. Field reference:
# https://konfig-konector.io/docs/sagemaker.html#sagemakerworkforce
apiVersion: aws.konfig.io/v1alpha1
kind: SageMakerWorkforce
metadata:
  name: example-sagemakerworkforce
  namespace: default
spec:
  cognitoConfig:
    clientId: example-id
    userPool: example
  ipAddressType: ipv4
  oidcConfig:
    authenticationRequestExtraParams:
      key: value
    authorizationEndpoint: example
    clientId: example-id
    clientSecret: example
    issuer: example
    jwksUri: example
    logoutEndpoint: example
    scope: example
    tokenEndpoint: example
    userInfoEndpoint: example
  providerRef:
    name: example-sagemakerworkforce
    region: us-east-1
  sourceIpConfig:
    cidrs:
    - 10.0.0.0/16
  tags:
  - key: example
    value: example
  workforceName: example-sagemakerworkforce
  workforceVpcConfig:
    securityGroupIds:
    - example
    subnets:
    - example

Spec

Field	Type	Required	Description
`cognitoConfig`	object	—	Create-only in AWS: changing it requires replacing the resource.
`cognitoConfig.clientId`	string	yes	The client ID for your Amazon Cognito user pool.
`cognitoConfig.userPool`	string	yes	The ID for your Amazon Cognito user pool.
`ipAddressType`	enum	—	The IP address type for the workforce. Allowed: `ipv4, dualstack`
`oidcConfig`	object	—	SageMakerWorkforceOidcConfig is a nested property type of AWS::SageMaker::Workforce.
`oidcConfig.authenticationRequestExtraParams`	map<string>	—	A string to string map of identifiers specific to the custom identity provider (IdP) being used.
`oidcConfig.authorizationEndpoint`	string	yes	The OIDC IdP authorization endpoint used to configure your private workforce.
`oidcConfig.clientId`	string	yes	The OIDC IdP client ID used to configure your private workforce.
`oidcConfig.clientSecret`	string	—	The OIDC IdP client secret used to configure your private workforce.
`oidcConfig.issuer`	string	yes	The OIDC IdP issuer used to configure your private workforce.
`oidcConfig.jwksUri`	string	yes	The OIDC IdP JSON Web Key Set (Jwks) URI used to configure your private workforce.
`oidcConfig.logoutEndpoint`	string	yes	The OIDC IdP logout endpoint used to configure your private workforce.
`oidcConfig.scope`	string	—	An array of string identifiers used to refer to the specific pieces of user data or claims that the client application wants to access.
`oidcConfig.tokenEndpoint`	string	yes	The OIDC IdP token endpoint used to configure your private workforce.
`oidcConfig.userInfoEndpoint`	string	yes	The OIDC IdP user info endpoint used to configure your private workforce.
`providerRef`	object	—	ProviderRef selects the AWSProvider (account/region) this resource is reconciled against.
`providerRef.name`	string	yes	Name of a cluster-scoped AWSProvider.
`providerRef.region`	string	—	Region overrides the provider's region for this resource only.
`sourceIpConfig`	object	—	SageMakerWorkforceSourceIpConfig is a nested property type of AWS::SageMaker::Workforce.
`sourceIpConfig.cidrs`	array<string>	yes	A list of one to ten Classless Inter-Domain Routing (CIDR) values.
`tags`	array<object>	—	An array of key-value pairs.
`tags[].key`	string	yes
`tags[].value`	string	yes
`workforceName`	string	yes	The name of the private workforce.
`workforceVpcConfig`	object	—	SageMakerWorkforceWorkforceVpcConfigRequest is a nested property type of AWS::SageMaker::Workforce.
`workforceVpcConfig.securityGroupIds`	array<string>	—	The VPC security group IDs.
`workforceVpcConfig.subnets`	array<string>	—	The VPC subnets.
`workforceVpcConfig.vpcId`	string	—	The ID of the VPC.

Status

Field	Type	Required	Description
`attempt`	integer	—	Attempt counts failed create/delete requests; it salts the Cloud Control idempotency token so a retry is not served the cached failure.
`awsProvider`	object	—	AWSProvider confirms the account and region this resource was reconciled against.
`awsProvider.accountId`	string	—	AccountID the resource lives in.
`awsProvider.name`	string	—	Name of the AWSProvider used; empty when the operator's own credentials were used.
`awsProvider.region`	string	—	Region the resource lives in (empty for global services).
`conditions`	array<object>	—	Conditions describe the current state of the resource.
`conditions[].lastTransitionTime`	string	yes	lastTransitionTime is the last time the condition transitioned from one status to another.
`conditions[].message`	string	yes	message is a human readable message indicating details about the transition.
`conditions[].observedGeneration`	integer	—	observedGeneration represents the .metadata.generation that the condition was set based upon.
`conditions[].reason`	string	yes	reason contains a programmatic identifier indicating the reason for the condition's last transition.
`conditions[].status`	enum	yes	status of the condition, one of True, False, Unknown. Allowed: `True, False, Unknown`
`conditions[].type`	string	yes	type of condition in CamelCase or in foo.example.com/CamelCase.
`identifier`	string	—	Identifier is the Cloud Control primary identifier of the resource.
`lastSyncTime`	string	—
`observedGeneration`	integer	—
`operation`	string	—	Operation is the in-flight operation (CREATE, UPDATE, DELETE).
`operationStatus`	string	—	OperationStatus is the last reported operation status.
`requestToken`	string	—	RequestToken tracks an in-flight asynchronous operation.
`subDomain`	string	—	The subdomain for your OIDC Identity Provider.
`workforceArn`	string	—	The Amazon Resource Name (ARN) of the private workforce.