scaler

Documentation

Index

• Constants
• func DrainNode(node *apiv1.Node, pods []*apiv1.Pod, client kube_client.Interface, ...) error

Constants

const (
	// EvictionRetryTime is the time after CA retries failed pod eviction.
	EvictionRetryTime = 10 * time.Second
)

Functions

func DrainNode

func DrainNode(node *apiv1.Node, pods []*apiv1.Pod, client kube_client.Interface, recorder kube_record.EventRecorder,
	maxGracefulTerminationSec int, maxPodEvictionTime time.Duration, waitBetweenRetries time.Duration) error

DrainNode performs drain logic on the node. Marks the node as unschedulable and later removes all pods, giving them up to MaxGracefulTerminationTime to finish.

Originally from https://github.com/kubernetes/autoscaler/blob/bf59e3daa5922c0e44027fa211948b50cb6b7a12/cluster-autoscaler/core/scale_down.go#L725-L783