This function should be synchronized with the function `is_terminal()` in Ray Job.

Custom Reason for RayClusterCondition.

Failed is deprecated, but we keep it to avoid compilation errors in projects that import the KubeRay Golang module.

Custom Reason for RayClusterCondition.

HeadPodReady indicates whether RayCluster's head Pod is ready for requests.

Custom Reason for RayClusterCondition.

Submit job via HTTP request.

https://docs.ray.io/en/latest/cluster/running-applications/job-submission/jobs-package-ref.html#jobstatus.

https://docs.ray.io/en/latest/cluster/running-applications/job-submission/jobs-package-ref.html#jobstatus.

https://docs.ray.io/en/latest/cluster/running-applications/job-submission/jobs-package-ref.html#jobstatus.

https://docs.ray.io/en/latest/cluster/running-applications/job-submission/jobs-package-ref.html#jobstatus.

https://docs.ray.io/en/latest/cluster/running-applications/job-submission/jobs-package-ref.html#jobstatus.

https://docs.ray.io/en/latest/cluster/running-applications/job-submission/jobs-package-ref.html#jobstatus.

Submit job via Kubernetes Job.

Custom Reason for RayClusterCondition.

RayClusterProvisioned indicates whether all Ray Pods are ready for the first time.

RayClusterReplicaFailure is added in a RayCluster when one of its pods fails to be created or deleted.

RedisCleanupNode is a Pod managed by a Kubernetes Job that cleans up Redis data after a RayCluster with GCS fault tolerance enabled is deleted.

UnknownReason says that the reason for the condition is unknown.

Don't submit job in KubeRay.

AddToScheme adds the types in this group-version to the given scheme.

These statuses should match Ray Serve's application statuses See `enum ApplicationStatus` in https://sourcegraph.com/github.com/ray-project/ray/-/blob/src/ray/protobuf/serve.proto for more details.

These statuses should match Ray Serve's deployment statuses.

GroupVersion is group version used to register these objects.

SchemeBuilder is used to add go types to the GroupVersionKind scheme.

SchemeGroupVersion is group version used to register these objects.

AutoscalerOptions specifies optional configuration for the Ray autoscaler.

HeadGroupSpec are the spec for the head pod.

HeadInfo gives info about head.

RayCluster is the Schema for the RayClusters API +kubebuilder:object:root=true +kubebuilder:resource:categories=all +kubebuilder:subresource:status +kubebuilder:storageversion +kubebuilder:printcolumn:name="desired workers",type=integer,JSONPath=".status.desiredWorkerReplicas",priority=0 +kubebuilder:printcolumn:name="available workers",type=integer,JSONPath=".status.availableWorkerReplicas",priority=0 +kubebuilder:printcolumn:name="cpus",type=string,JSONPath=".status.desiredCPU",priority=0 +kubebuilder:printcolumn:name="memory",type=string,JSONPath=".status.desiredMemory",priority=0 +kubebuilder:printcolumn:name="gpus",type=string,JSONPath=".status.desiredGPU",priority=0 +kubebuilder:printcolumn:name="tpus",type=string,JSONPath=".status.desiredTPU",priority=1 +kubebuilder:printcolumn:name="status",type="string",JSONPath=".status.state",priority=0 +kubebuilder:printcolumn:name="age",type="date",JSONPath=".metadata.creationTimestamp",priority=0 +kubebuilder:printcolumn:name="head pod IP",type="string",JSONPath=".status.head.podIP",priority=1 +kubebuilder:printcolumn:name="head service IP",type="string",JSONPath=".status.head.serviceIP",priority=1 +genclient.

RayClusterList contains a list of RayCluster.

RayClusterSpec defines the desired state of RayCluster.

RayClusterStatus defines the observed state of RayCluster.

+kubebuilder:object:root=true +kubebuilder:resource:categories=all +kubebuilder:subresource:status +kubebuilder:storageversion +kubebuilder:printcolumn:name="job status",type=string,JSONPath=".status.jobStatus",priority=0 +kubebuilder:printcolumn:name="deployment status",type=string,JSONPath=".status.jobDeploymentStatus",priority=0 +kubebuilder:printcolumn:name="ray cluster name",type="string",JSONPath=".status.rayClusterName",priority=0 +kubebuilder:printcolumn:name="start time",type=string,JSONPath=".status.startTime",priority=0 +kubebuilder:printcolumn:name="end time",type=string,JSONPath=".status.endTime",priority=0 +kubebuilder:printcolumn:name="age",type="date",JSONPath=".metadata.creationTimestamp",priority=0 +genclient RayJob is the Schema for the rayjobs API.

RayJobList contains a list of RayJob.

RayJobSpec defines the desired state of RayJob.

RayJobStatus defines the observed state of RayJob.

+kubebuilder:object:root=true +kubebuilder:resource:categories=all +kubebuilder:subresource:status +kubebuilder:storageversion +kubebuilder:printcolumn:name="service status",type=string,JSONPath=".status.serviceStatus" +kubebuilder:printcolumn:name="num serve endpoints",type=string,JSONPath=".status.numServeEndpoints" +genclient RayService is the Schema for the rayservices API.

RayServiceList contains a list of RayService.

RayServiceSpec defines the desired state of RayService.

RayServiceStatuses defines the observed state of RayService.

ScaleStrategy to remove workers.

ServeDeploymentStatus defines the current state of a Serve deployment.

WorkerGroupSpec are the specs for the worker pods.

The overall state of the Ray cluster.

JobDeploymentStatus indicates RayJob status including RayCluster lifecycle management and Job submission.

JobFailedReason indicates the reason the RayJob changes its JobDeploymentStatus to 'Failed'.

JobStatus is the Ray Job Status.

RayNodeType the type of a ray node: head/worker.

+kubebuilder:validation:Enum=Default;Aggressive;Conservative.