github
/
k3s
mirror of https://github.com/k3s-io/k3s
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Merge pull request #71118 from ping035627/k8s-181115 

Add some exception handling of Preempt
pull/564/head
Kubernetes Prow Robot 2018-12-17 12:10:31 -08:00 committed by GitHub
parent 9aeb7b07c0 98f852a441
commit 1d59ab05fa
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 15 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
pkg/scheduler/core/generic_scheduler.go
View File

@ -134,7 +134,7 @@ type genericScheduler struct {
percentageOfNodesToScore int32
}
// snapshot snapshots equivalane cache and node infos for all fit and priority
// snapshot snapshots equivalence cache and node infos for all fit and priority
// functions.
func (g *genericScheduler) snapshot() error {
// IMPORTANT NOTE: We must snapshot equivalence cache before snapshotting
@ -143,7 +143,7 @@ func (g *genericScheduler) snapshot() error {
// 1. snapshot cache
// 2. event arrives, updating cache and invalidating predicates or whole node cache
// 3. snapshot ecache
// 4. evaludate predicates
// 4. evaluate predicates
// 5. stale result will be written to ecache
if g.equivalenceCache != nil {
g.equivalenceCache.Snapshot()
@ -309,7 +309,7 @@ func (g *genericScheduler) Preempt(pod *v1.Pod, nodeLister algorithm.NodeLister,
candidateNode := pickOneNodeForPreemption(nodeToVictims)
if candidateNode == nil {
return nil, nil, nil, err
return nil, nil, nil, nil
}
// Lower priority pods nominated to run on this node, may no longer fit on
@ -318,7 +318,7 @@ func (g *genericScheduler) Preempt(pod *v1.Pod, nodeLister algorithm.NodeLister,
// lets scheduler find another place for them.
nominatedPods := g.getLowerPriorityNominatedPods(pod, candidateNode.Name)
if nodeInfo, ok := g.cachedNodeInfoMap[candidateNode.Name]; ok {
return nodeInfo.Node(), nodeToVictims[candidateNode].Pods, nominatedPods, err
return nodeInfo.Node(), nodeToVictims[candidateNode].Pods, nominatedPods, nil
}
return nil, nil, nil, fmt.Errorf(

22
pkg/scheduler/scheduler.go
View File

@ -318,11 +318,6 @@ func (sched *Scheduler) schedule(pod *v1.Pod) (string, error) {
// If it succeeds, it adds the name of the node where preemption has happened to the pod annotations.
// It returns the node name and an error if any.
func (sched *Scheduler) preempt(preemptor *v1.Pod, scheduleErr error) (string, error) {
if !util.PodPriorityEnabled() || sched.config.DisablePreemption {
klog.V(3).Infof("Pod priority feature is not enabled or preemption is disabled by scheduler configuration." +
" No preemption is performed.")
return "", nil
}
preemptor, err := sched.config.PodPreemptor.GetUpdatedPod(preemptor)
if err != nil {
klog.Errorf("Error getting the updated preemptor pod object: %v", err)
@ -330,7 +325,6 @@ func (sched *Scheduler) preempt(preemptor *v1.Pod, scheduleErr error) (string, e
}
node, victims, nominatedPodsToClear, err := sched.config.Algorithm.Preempt(preemptor, sched.config.NodeLister, scheduleErr)
metrics.PreemptionVictims.Set(float64(len(victims)))
if err != nil {
klog.Errorf("Error preempting victims to make room for %v/%v.", preemptor.Namespace, preemptor.Name)
return "", err
@ -350,6 +344,7 @@ func (sched *Scheduler) preempt(preemptor *v1.Pod, scheduleErr error) (string, e
}
sched.config.Recorder.Eventf(victim, v1.EventTypeNormal, "Preempted", "by %v/%v on node %v", preemptor.Namespace, preemptor.Name, nodeName)
}
metrics.PreemptionVictims.Set(float64(len(victims)))
}
// Clearing nominated pods should happen outside of "if node != nil". Node could
// be nil when a pod with nominated node name is eligible to preempt again,
@ -508,11 +503,16 @@ func (sched *Scheduler) scheduleOne() {
// will fit due to the preemption. It is also possible that a different pod will schedule
// into the resources that were preempted, but this is harmless.
if fitError, ok := err.(*core.FitError); ok {
preemptionStartTime := time.Now()
sched.preempt(pod, fitError)
metrics.PreemptionAttempts.Inc()
metrics.SchedulingAlgorithmPremptionEvaluationDuration.Observe(metrics.SinceInMicroseconds(preemptionStartTime))
metrics.SchedulingLatency.WithLabelValues(metrics.PreemptionEvaluation).Observe(metrics.SinceInSeconds(preemptionStartTime))
if !util.PodPriorityEnabled() || sched.config.DisablePreemption {
klog.V(3).Infof("Pod priority feature is not enabled or preemption is disabled by scheduler configuration." +
" No preemption is performed.")
} else {
preemptionStartTime := time.Now()
sched.preempt(pod, fitError)
metrics.PreemptionAttempts.Inc()
metrics.SchedulingAlgorithmPremptionEvaluationDuration.Observe(metrics.SinceInMicroseconds(preemptionStartTime))
metrics.SchedulingLatency.WithLabelValues(metrics.PreemptionEvaluation).Observe(metrics.SinceInSeconds(preemptionStartTime))
}
// Pod did not fit anywhere, so it is counted as a failure. If preemption
// succeeds, the pod should get counted as a success the next time we try to
// schedule it. (hopefully)