api-overview.md

API Reference

Packages

• datasciencecluster.opendatahub.io/v1
• dscinitialization.opendatahub.io/v1

datasciencecluster.opendatahub.io/codeflare

Package codeflare provides utility functions to config CodeFlare as part of the stack which makes managing distributed compute infrastructure in the cloud easy and intuitive for Data Scientists

CodeFlare

CodeFlare struct holds the configuration for the CodeFlare component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component

datasciencecluster.opendatahub.io/components

Component

Component struct defines the basis for each OpenDataHub component configuration.

Appears in:

• CodeFlare
• Dashboard
• DataSciencePipelines
• Kserve
• Kueue
• ModelMeshServing
• ModelRegistry
• Ray
• TrainingOperator
• TrustyAI
• Workbenches

Field	Description	Default	Validation
managementState ManagementState	Set to one of the following values: - "Managed" : the operator is actively managing the component and trying to keep it active. It will only upgrade the component if it is safe to do so - "Removed" : the operator is actively managing the component and will not install it, or if it is installed, the operator will try to remove it		Enum: [Managed Removed]
devFlags DevFlags	Add developer fields

DevFlags

DevFlags defines list of fields that can be used by developers to test customizations. This is not recommended to be used in production environment.

Appears in:

• Component

Field	Description	Default	Validation
manifests ManifestsConfig array	List of custom manifests for the given component

ManifestsConfig

Appears in:

• DevFlags

Field	Description	Default	Validation
uri string	uri is the URI point to a git repo with tag/branch. e.g. https://github.com/org/repo/tarball/<tag/branch>
contextDir string	contextDir is the relative path to the folder containing manifests in a repository, default value "manifests"	manifests
sourcePath string	sourcePath is the subpath within contextDir where kustomize builds start. Examples include any sub-folder or path: base, overlays/dev, default, odh etc.

datasciencecluster.opendatahub.io/dashboard

Package dashboard provides utility functions to config Open Data Hub Dashboard: A web dashboard that displays installed Open Data Hub components with easy access to component UIs and documentation

Dashboard

Dashboard struct holds the configuration for the Dashboard component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component

datasciencecluster.opendatahub.io/datasciencepipelines

Package datasciencepipelines provides utility functions to config Data Science Pipelines: Pipeline solution for end to end MLOps workflows that support the Kubeflow Pipelines SDK and Argo Workflows.

DataSciencePipelines

DataSciencePipelines struct holds the configuration for the DataSciencePipelines component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component

datasciencecluster.opendatahub.io/kserve

Package kserve provides utility functions to config Kserve as the Controller for serving ML models on arbitrary frameworks

DefaultDeploymentMode

Underlying type: string

Validation:

• Pattern: ^(Serverless|RawDeployment)$

Appears in:

• Kserve

Kserve

Kserve struct holds the configuration for the Kserve component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component
serving ServingSpec	Serving configures the KNative-Serving stack used for model serving. A Service Mesh (Istio) is prerequisite, since it is used as networking layer.
defaultDeploymentMode DefaultDeploymentMode	Configures the default deployment mode for Kserve. This can be set to 'Serverless' or 'RawDeployment'. The value specified in this field will be used to set the default deployment mode in the 'inferenceservice-config' configmap for Kserve. This field is optional. If no default deployment mode is specified, Kserve will use Serverless mode.		Enum: [Serverless RawDeployment] Pattern: ^(Serverless|RawDeployment)$

datasciencecluster.opendatahub.io/kueue

Kueue

Kueue struct holds the configuration for the Kueue component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component

datasciencecluster.opendatahub.io/modelmeshserving

Package modelmeshserving provides utility functions to config MoModelMesh, a general-purpose model serving management/routing layer

ModelMeshServing

ModelMeshServing struct holds the configuration for the ModelMeshServing component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component

datasciencecluster.opendatahub.io/modelregistry

Package modelregistry provides utility functions to config ModelRegistry, an ML Model metadata repository service

ModelRegistry

Appears in:

• Components

Field	Description	Default	Validation
Component Component
registriesNamespace string	Namespace for model registries to be installed, configurable only once when model registry is enabled, defaults to "odh-model-registries"	odh-model-registries	MaxLength: 63 Pattern: ^([a-z0-9]([-a-z0-9]*[a-z0-9])?)?$

datasciencecluster.opendatahub.io/ray

Package ray provides utility functions to config Ray as part of the stack which makes managing distributed compute infrastructure in the cloud easy and intuitive for Data Scientists

Ray

Ray struct holds the configuration for the Ray component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component

datasciencecluster.opendatahub.io/trainingoperator

Package trainingoperator provides utility functions to config trainingoperator as part of the stack which makes managing distributed compute infrastructure in the cloud easy and intuitive for Data Scientists

TrainingOperator

TrainingOperator struct holds the configuration for the TrainingOperator component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component

datasciencecluster.opendatahub.io/trustyai

Package trustyai provides utility functions to config TrustyAI, a bias/fairness and explainability toolkit

TrustyAI

TrustyAI struct holds the configuration for the TrustyAI component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component

datasciencecluster.opendatahub.io/v1

Resource Types

• DataScienceCluster

AuthSpec

Appears in:

• ServiceMeshSpec

Field	Description	Default	Validation
namespace string	Namespace where it is deployed. If not provided, the default is to use '-auth-provider' suffix on the ApplicationsNamespace of the DSCI.		MaxLength: 63 Pattern: ^([a-z0-9]([-a-z0-9]*[a-z0-9])?)?$
audiences string	Audiences is a list of the identifiers that the resource server presented with the token identifies as. Audience-aware token authenticators will verify that the token was intended for at least one of the audiences in this list. If no audiences are provided, the audience will default to the audience of the Kubernetes apiserver (kubernetes.default.svc).	[https://kubernetes.default.svc]

CertType

Underlying type: string

Appears in:

• CertificateSpec

Field	Description
SelfSigned
Provided
OpenshiftDefaultIngress

CertificateSpec

CertificateSpec represents the specification of the certificate securing communications of an Istio Gateway.

Appears in:

• GatewaySpec

Field	Description	Default	Validation
secretName string	SecretName specifies the name of the Kubernetes Secret resource that contains a TLS certificate secure HTTP communications for the KNative network.
type CertType	Type specifies if the TLS certificate should be generated automatically, or if the certificate is provided by the user. Allowed values are: * SelfSigned: A certificate is going to be generated using an own private key. * Provided: Pre-existence of the TLS Secret (see SecretName) with a valid certificate is assumed. * OpenshiftDefaultIngress: Default ingress certificate configured for OpenShift	OpenshiftDefaultIngress	Enum: [SelfSigned Provided OpenshiftDefaultIngress]

Components

Appears in:

• DataScienceClusterSpec

Field	Description	Default	Validation
dashboard Dashboard	Dashboard component configuration.
workbenches Workbenches	Workbenches component configuration.
modelmeshserving ModelMeshServing	ModelMeshServing component configuration. Does not support enabled Kserve at the same time
datasciencepipelines DataSciencePipelines	DataServicePipeline component configuration. Require OpenShift Pipelines Operator to be installed before enable component
kserve Kserve	Kserve component configuration. Require OpenShift Serverless and OpenShift Service Mesh Operators to be installed before enable component Does not support enabled ModelMeshServing at the same time
kueue Kueue	Kueue component configuration.
codeflare CodeFlare	CodeFlare component configuration. If CodeFlare Operator has been installed in the cluster, it should be uninstalled first before enabled component.
ray Ray	Ray component configuration.
trustyai TrustyAI	TrustyAI component configuration.
modelregistry ModelRegistry	ModelRegistry component configuration.
trainingoperator TrainingOperator	Training Operator component configuration.

ComponentsStatus

ComponentsStatus defines the custom status of DataScienceCluster components.

Appears in:

• DataScienceClusterStatus

Field	Description	Default	Validation
modelregistry ModelRegistryStatus	ModelRegistry component status

ControlPlaneSpec

Appears in:

• ServiceMeshSpec

Field	Description	Default	Validation
name string	Name is a name Service Mesh Control Plane. Defaults to "data-science-smcp".	data-science-smcp
namespace string	Namespace is a namespace where Service Mesh is deployed. Defaults to "istio-system".	istio-system	MaxLength: 63 Pattern: ^([a-z0-9]([-a-z0-9]*[a-z0-9])?)?$
metricsCollection string	MetricsCollection specifies if metrics from components on the Mesh namespace should be collected. Setting the value to "Istio" will collect metrics from the control plane and any proxies on the Mesh namespace (like gateway pods). Setting to "None" will disable metrics collection.	Istio	Enum: [Istio None]

DataScienceCluster

DataScienceCluster is the Schema for the datascienceclusters API.

Field	Description	Default	Validation
apiVersion string	datasciencecluster.opendatahub.io/v1
kind string	DataScienceCluster
kind string	Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds
apiVersion string	APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources
metadata ObjectMeta	Refer to Kubernetes API documentation for fields of metadata.
spec DataScienceClusterSpec
status DataScienceClusterStatus

DataScienceClusterSpec

DataScienceClusterSpec defines the desired state of the cluster.

Appears in:

• DataScienceCluster

Field	Description	Default	Validation
components Components	Override and fine tune specific component configurations.

DataScienceClusterStatus

DataScienceClusterStatus defines the observed state of DataScienceCluster.

Appears in:

• DataScienceCluster

Field	Description	Default	Validation
phase string	Phase describes the Phase of DataScienceCluster reconciliation state This is used by OLM UI to provide status information to the user
conditions Condition array	Conditions describes the state of the DataScienceCluster resource.
relatedObjects ObjectReference array	RelatedObjects is a list of objects created and maintained by this operator. Object references will be added to this list after they have been created AND found in the cluster.
errorMessage string
installedComponents object (keys:string, values:boolean)	List of components with status if installed or not
components ComponentsStatus	Expose component's specific status
release Release	Version and release type

GatewaySpec

GatewaySpec represents the configuration of the Ingress Gateways.

Appears in:

• ServingSpec

Field	Description	Default	Validation
domain string	Domain specifies the host name for intercepting incoming requests. Most likely, you will want to use a wildcard name, like *.example.com. If not set, the domain of the OpenShift Ingress is used. If you choose to generate a certificate, this is the domain used for the certificate request.
certificate CertificateSpec	Certificate specifies configuration of the TLS certificate securing communication for the gateway.

ServiceMeshSpec

ServiceMeshSpec configures Service Mesh.

Appears in:

• DSCInitializationSpec

Field	Description	Default	Validation
managementState ManagementState		Removed	Enum: [Managed Unmanaged Removed]
controlPlane ControlPlaneSpec	ControlPlane holds configuration of Service Mesh used by Opendatahub.
auth AuthSpec	Auth holds configuration of authentication and authorization services used by Service Mesh in Opendatahub.

ServingSpec

ServingSpec specifies the configuration for the KNative Serving components and their bindings with the Service Mesh.

Appears in:

• Kserve

Field	Description	Default	Validation
managementState ManagementState		Managed	Enum: [Managed Unmanaged Removed]
name string	Name specifies the name of the KNativeServing resource that is going to be created to instruct the KNative Operator to deploy KNative serving components. This resource is created in the "knative-serving" namespace.	knative-serving
ingressGateway GatewaySpec	IngressGateway allows to customize some parameters for the Istio Ingress Gateway that is bound to KNative-Serving.

datasciencecluster.opendatahub.io/workbenches

Package workbenches provides utility functions to config Workbenches to secure Jupyter Notebook in Kubernetes environments with support for OAuth

Workbenches

Workbenches struct holds the configuration for the Workbenches component.

Appears in:

• Components

Field	Description	Default	Validation
Component Component

dscinitialization.opendatahub.io/v1

Package v1 contains API Schema definitions for the dscinitialization v1 API group

Resource Types

• DSCInitialization

DSCInitialization

DSCInitialization is the Schema for the dscinitializations API.

Field	Description	Default	Validation
apiVersion string	dscinitialization.opendatahub.io/v1
kind string	DSCInitialization
kind string	Kind is a string value representing the REST resource this object represents. Servers may infer this from the endpoint the client submits requests to. Cannot be updated. In CamelCase. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds
apiVersion string	APIVersion defines the versioned schema of this representation of an object. Servers should convert recognized schemas to the latest internal value, and may reject unrecognized values. More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources
metadata ObjectMeta	Refer to Kubernetes API documentation for fields of metadata.
spec DSCInitializationSpec
status DSCInitializationStatus

DSCInitializationSpec

DSCInitializationSpec defines the desired state of DSCInitialization.

Appears in:

• DSCInitialization

Field	Description	Default	Validation
applicationsNamespace string	Namespace for applications to be installed, non-configurable, default to "opendatahub"	opendatahub	MaxLength: 63 Pattern: ^([a-z0-9]([-a-z0-9]*[a-z0-9])?)?$
monitoring Monitoring	Enable monitoring on specified namespace
serviceMesh ServiceMeshSpec	Configures Service Mesh as networking layer for Data Science Clusters components. The Service Mesh is a mandatory prerequisite for single model serving (KServe) and you should review this configuration if you are planning to use KServe. For other components, it enhances user experience; e.g. it provides unified authentication giving a Single Sign On experience.
trustedCABundle TrustedCABundleSpec	When set to Managed, adds odh-trusted-ca-bundle Configmap to all namespaces that includes cluster-wide Trusted CA Bundle in .data["ca-bundle.crt"]. Additionally, this fields allows admins to add custom CA bundles to the configmap using the .CustomCABundle field.
devFlags DevFlags	Internal development useful field to test customizations. This is not recommended to be used in production environment.

DSCInitializationStatus

DSCInitializationStatus defines the observed state of DSCInitialization.

Appears in:

• DSCInitialization

Field	Description	Default	Validation
phase string	Phase describes the Phase of DSCInitializationStatus This is used by OLM UI to provide status information to the user
conditions Condition array	Conditions describes the state of the DSCInitializationStatus resource
relatedObjects ObjectReference array	RelatedObjects is a list of objects created and maintained by this operator. Object references will be added to this list after they have been created AND found in the cluster
errorMessage string
release Release	Version and release type

DevFlags

DevFlags defines list of fields that can be used by developers to test customizations. This is not recommended to be used in production environment.

Appears in:

• DSCInitializationSpec

Field	Description	Default	Validation
manifestsUri string	Custom manifests uri for odh-manifests
logmode string	## DEPRECATED ##: Ignored, use LogLevel instead	production	Enum: [devel development prod production default]
logLevel string	Override Zap log level. Can be "debug", "info", "error" or a number (more verbose).

Monitoring

Appears in:

• DSCInitializationSpec

Field	Description	Default	Validation
managementState ManagementState	Set to one of the following values: - "Managed" : the operator is actively managing the component and trying to keep it active. It will only upgrade the component if it is safe to do so. - "Removed" : the operator is actively managing the component and will not install it, or if it is installed, the operator will try to remove it.		Enum: [Managed Removed]
namespace string	Namespace for monitoring if it is enabled	opendatahub	MaxLength: 63 Pattern: ^([a-z0-9]([-a-z0-9]*[a-z0-9])?)?$

TrustedCABundleSpec

Appears in:

• DSCInitializationSpec

Field	Description	Default	Validation
managementState ManagementState	managementState indicates whether and how the operator should manage customized CA bundle	Removed	Enum: [Managed Removed Unmanaged]
customCABundle string	A custom CA bundle that will be available for all components in the Data Science Cluster(DSC). This bundle will be stored in odh-trusted-ca-bundle ConfigMap .data.odh-ca-bundle.crt .