From b2d1393ecd523d2f6b65662b81adabe8f6036547 Mon Sep 17 00:00:00 2001
From: Luke <feixiang.li@smartnews.com>
Date: Mon, 20 May 2024 14:45:14 +0900
Subject: [PATCH] handle deleting nodes

---
 pkg/controllers/provisioning/provisioner.go | 26 +++++++++++++--------
 1 file changed, 16 insertions(+), 10 deletions(-)

diff --git a/pkg/controllers/provisioning/provisioner.go b/pkg/controllers/provisioning/provisioner.go
index 4a507b2224..6f63eeda4d 100644
--- a/pkg/controllers/provisioning/provisioner.go
+++ b/pkg/controllers/provisioning/provisioner.go
@@ -160,7 +160,7 @@ func (p *Provisioner) GetPendingPods(ctx context.Context) ([]*v1.Pod, error) {
 	}
 	return lo.Reject(pods, func(po *v1.Pod, _ int) bool {
 		if err := p.Validate(ctx, po); err != nil {
-			logging.FromContext(ctx).With("pod", client.ObjectKeyFromObject(po)).Infof("ignoring pod, %s", err)
+			logging.FromContext(ctx).With("pod", client.ObjectKeyFromObject(po)).Debugf("ignoring pod, %s", err)
 			return true
 		}
 		p.consolidationWarnings(ctx, po)
@@ -314,7 +314,13 @@ func (p *Provisioner) Schedule(ctx context.Context) (scheduler.Results, error) {
 	if len(pods) == 0 {
 		return scheduler.Results{}, nil
 	}
-	s, err := p.NewScheduler(ctx, pods, nodes.Active())
+	targetPods := lo.FilterMap(pods, func(pod *v1.Pod, _ int) (*v1.Pod, bool) {
+		if !p.isPodHandled(ctx, pod) {
+			return pod, true
+		}
+		return nil, false
+	})
+	s, err := p.NewScheduler(ctx, targetPods, nodes.Active())
 	if err != nil {
 		if errors.Is(err, ErrNodePoolsNotFound) {
 			logging.FromContext(ctx).Info(ErrNodePoolsNotFound)
@@ -322,8 +328,8 @@ func (p *Provisioner) Schedule(ctx context.Context) (scheduler.Results, error) {
 		}
 		return scheduler.Results{}, fmt.Errorf("creating scheduler, %w", err)
 	}
-	results := s.Solve(ctx, pods).TruncateInstanceTypes(scheduler.MaxInstanceTypes)
-	logging.FromContext(ctx).With("pods", pretty.Slice(lo.Map(pods, func(p *v1.Pod, _ int) string { return client.ObjectKeyFromObject(p).String() }), 5)).
+	results := s.Solve(ctx, targetPods).TruncateInstanceTypes(scheduler.MaxInstanceTypes)
+	logging.FromContext(ctx).With("pods", pretty.Slice(lo.Map(targetPods, func(p *v1.Pod, _ int) string { return client.ObjectKeyFromObject(p).String() }), 5)).
 		With("duration", time.Since(start)).
 		Infof("found provisionable pod(s)")
 	results.Record(ctx, p.recorder, p.cluster)
@@ -416,11 +422,10 @@ func (p *Provisioner) Validate(ctx context.Context, pod *v1.Pod) error {
 		validateNodeSelector(pod),
 		validateAffinity(pod),
 		p.volumeTopology.ValidatePersistentVolumeClaims(ctx, pod),
-		p.isPodHandled(ctx, pod),
 	)
 }
 
-func (p *Provisioner) isPodHandled(ctx context.Context, pod *v1.Pod) (err error) {
+func (p *Provisioner) isPodHandled(ctx context.Context, pod *v1.Pod) bool {
 	var events v1.EventList
 	filter := client.MatchingFields{
 		"namespace":           pod.Namespace,
@@ -428,20 +433,21 @@ func (p *Provisioner) isPodHandled(ctx context.Context, pod *v1.Pod) (err error)
 		"involvedObject.name": pod.Name,
 		"reason":              "HandledByKarpenter",
 	}
-	logging.FromContext(ctx).Debugf("get event for %s/%s", pod.Namespace, pod.Name)
+	logging.FromContext(ctx).Infof("get event for %s/%s", pod.Namespace, pod.Name)
 	if err := p.kubeClient.List(ctx, &events, filter); err == nil {
 		for _, event := range events.Items {
-			logging.FromContext(ctx).Debugf("process event %s", event.Name)
+			logging.FromContext(ctx).Infof("process event %s", event.Name)
 			// ignore the pod if it's already handled in 3 minute
 			if !time.Now().Add(3 * time.Minute).After(event.LastTimestamp.Time) {
-				return fmt.Errorf("pod %s/%s is handled", pod.Namespace, pod.Name)
+				logging.FromContext(ctx).Infof("pod %s/%s is handled", pod.Namespace, pod.Name)
+				return true
 			}
 		}
 	} else {
 		logging.FromContext(ctx).Errorf("failed to get event for %s/%s: %w", pod.Namespace, pod.Name, err)
 	}
 	p.recorder.Publish(scheduler.PodHandledEvent(pod))
-	return nil
+	return false
 }
 
 // validateKarpenterManagedLabelCanExist provides a more clear error message in the event of scheduling a pod that specifically doesn't