mpi

package
v1.7.0 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: Nov 1, 2023 License: Apache-2.0 Imports: 42 Imported by: 0

Documentation

Index

Constants

View Source
const (
	// ErrResourceExists is used as part of the Event 'reason' when an MPIJob
	// fails to sync due to dependent resources of the same name already
	// existing.
	ErrResourceExists = "ErrResourceExists"

	// MessageResourceExists is the message used for Events when a resource
	// fails to sync due to dependent resources already existing.
	MessageResourceExists = "Resource %q of MPIJobKind %q already exists and is not managed by MPIJob"

	// ErrResourceDoesNotExist is used as part of the Event 'reason' when some
	// resource is missing in yaml
	ErrResourceDoesNotExist = "ErrResourceDoesNotExist"

	// MessageResourceDoesNotExist is used for Events when some
	// resource is missing in yaml
	MessageResourceDoesNotExist = "Resource %q is missing in yaml"
)
View Source
const (
	FailedDeleteJobReason     = "FailedDeleteJob"
	SuccessfulDeleteJobReason = "SuccessfulDeleteJob"
)

Variables

This section is empty.

Functions

This section is empty.

Types

type MPIJobReconciler

type MPIJobReconciler struct {
	common.JobController
	client.Client
	Scheme *runtime.Scheme

	Log logr.Logger
	// contains filtered or unexported fields
}

MPIJobReconciler reconciles a MPIJob object

func NewReconciler

func NewReconciler(mgr manager.Manager, gangSchedulingSetupFunc common.GangSchedulingSetupFunc) *MPIJobReconciler

func (*MPIJobReconciler) ControllerName

func (jc *MPIJobReconciler) ControllerName() string

func (*MPIJobReconciler) DeleteJob

func (jc *MPIJobReconciler) DeleteJob(job interface{}) error

func (*MPIJobReconciler) GetAPIGroupVersion

func (jc *MPIJobReconciler) GetAPIGroupVersion() schema.GroupVersion

func (*MPIJobReconciler) GetAPIGroupVersionKind

func (jc *MPIJobReconciler) GetAPIGroupVersionKind() schema.GroupVersionKind

func (*MPIJobReconciler) GetDefaultContainerName

func (jc *MPIJobReconciler) GetDefaultContainerName() string

func (*MPIJobReconciler) GetDefaultContainerPortName

func (jc *MPIJobReconciler) GetDefaultContainerPortName() string

func (*MPIJobReconciler) GetFrameworkName added in v1.7.0

func (jc *MPIJobReconciler) GetFrameworkName() string

func (*MPIJobReconciler) GetGroupNameLabelValue

func (jc *MPIJobReconciler) GetGroupNameLabelValue() string

func (*MPIJobReconciler) GetJobFromAPIClient

func (jc *MPIJobReconciler) GetJobFromAPIClient(namespace, name string) (metav1.Object, error)

func (*MPIJobReconciler) GetJobFromInformerCache

func (jc *MPIJobReconciler) GetJobFromInformerCache(namespace, name string) (metav1.Object, error)

func (*MPIJobReconciler) GetPodsForJob

func (jc *MPIJobReconciler) GetPodsForJob(jobObject interface{}) ([]*corev1.Pod, error)

GetPodsForJob returns the set of pods that this job should manage. It also reconciles ControllerRef by adopting/orphaning. Note that the returned Pods are pointers into the cache.

func (*MPIJobReconciler) GetServicesForJob

func (jc *MPIJobReconciler) GetServicesForJob(jobObject interface{}) ([]*corev1.Service, error)

GetServicesForJob returns the set of services that this job should manage. It also reconciles ControllerRef by adopting/orphaning. Note that the returned services are pointers into the cache.

func (*MPIJobReconciler) IsMasterRole

func (jc *MPIJobReconciler) IsMasterRole(replicas map[kubeflowv1.ReplicaType]*kubeflowv1.ReplicaSpec,
	rtype kubeflowv1.ReplicaType, index int) bool

func (*MPIJobReconciler) Reconcile

func (jc *MPIJobReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctrl.Result, error)

Reconcile is part of the main kubernetes reconciliation loop which aims to move the current state of the cluster closer to the desired state.

func (*MPIJobReconciler) ReconcilePods

func (jc *MPIJobReconciler) ReconcilePods(
	job interface{},
	jobStatus *kubeflowv1.JobStatus,
	pods []*corev1.Pod,
	rtype kubeflowv1.ReplicaType,
	spec *kubeflowv1.ReplicaSpec,
	replicas map[kubeflowv1.ReplicaType]*kubeflowv1.ReplicaSpec,
) error

func (*MPIJobReconciler) ReconcileServices

func (jc *MPIJobReconciler) ReconcileServices(
	job metav1.Object,
	services []*corev1.Service,
	rtype kubeflowv1.ReplicaType,
	spec *kubeflowv1.ReplicaSpec) error

ReconcileServices is overridden because mpi-reconciler.v1 does not need to reconcile services

func (*MPIJobReconciler) SetClusterSpec

func (jc *MPIJobReconciler) SetClusterSpec(job interface{}, podTemplate *corev1.PodTemplateSpec, rtype, index string) error

SetClusterSpec is overridden because no cluster spec is needed for MPIJob

func (*MPIJobReconciler) SetupWithManager

func (jc *MPIJobReconciler) SetupWithManager(mgr ctrl.Manager, controllerThreads int) error

SetupWithManager sets up the controller with the Manager.

func (*MPIJobReconciler) UpdateJobStatus

func (jc *MPIJobReconciler) UpdateJobStatus(job interface{}, replicas map[kubeflowv1.ReplicaType]*kubeflowv1.ReplicaSpec, jobStatus *kubeflowv1.JobStatus) error

func (*MPIJobReconciler) UpdateJobStatusInApiServer

func (jc *MPIJobReconciler) UpdateJobStatusInApiServer(job interface{}, jobStatus *kubeflowv1.JobStatus) error

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL