scheduler

Documentation

Index

• Variables
• type ScheduleType
• type Scheduler
  • func NewScheduler(dynamicClient dynamic.Interface, karmadaClient karmadaclientset.Interface, ...) *Scheduler
  • func (s *Scheduler) Run(ctx context.Context)

Variables

var Failover bool

Failover indicates if the scheduler should performs re-scheduler in case of cluster failure. TODO(RainbowMango): Remove the temporary solution by introducing feature flag

Types

type ScheduleType

type ScheduleType string

ScheduleType defines the schedule type of a binding object should be performed.

const (
	// FirstSchedule means the binding object hasn't been scheduled.
	FirstSchedule ScheduleType = "FirstSchedule"

	// ReconcileSchedule means the binding object associated policy has been changed.
	ReconcileSchedule ScheduleType = "ReconcileSchedule"

	// FailoverSchedule means one of the cluster a binding object associated with becomes failure.
	FailoverSchedule ScheduleType = "FailoverSchedule"

	// AvoidSchedule means don't need to trigger scheduler.
	AvoidSchedule ScheduleType = "AvoidSchedule"

	// Unknown means can't detect the schedule type
	Unknown ScheduleType = "Unknown"
)

type Scheduler

type Scheduler struct {
	DynamicClient dynamic.Interface
	KarmadaClient karmadaclientset.Interface
	KubeClient    kubernetes.Interface

	Algorithm core.ScheduleAlgorithm
	// contains filtered or unexported fields
}

Scheduler is the scheduler schema, which is used to schedule a specific resource to specific clusters

func NewScheduler

func NewScheduler(dynamicClient dynamic.Interface, karmadaClient karmadaclientset.Interface, kubeClient kubernetes.Interface) *Scheduler

NewScheduler instantiates a scheduler

func (*Scheduler) Run

func (s *Scheduler) Run(ctx context.Context)

Run runs the scheduler