Skip to content

fix(cluster-autoscaler): prevent panic in SimulateNodeRemoval by handling missing node info #8449

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
+25 −0
Open

fix(cluster-autoscaler): prevent panic in SimulateNodeRemoval by handling missing node info #8449

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions cluster-autoscaler/simulator/cluster.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,10 +17,12 @@ limitations under the License.
package simulator

import (
"errors"
"fmt"
"time"

apiv1 "k8s.io/api/core/v1"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
"k8s.io/autoscaler/cluster-autoscaler/core/scaledown/pdb"
"k8s.io/autoscaler/cluster-autoscaler/simulator/clustersnapshot"
"k8s.io/autoscaler/cluster-autoscaler/simulator/drainability/rules"
Expand Down Expand Up @@ -91,6 +93,8 @@ const (
BlockedByPod
// UnexpectedError - node can't be removed because of an unexpected error.
UnexpectedError
// NoNodeInfo - node can't be removed because it doesn't have any node info in the cluster snapshot.
NoNodeInfo
)

// RemovalSimulator is a helper object for simulating node removal scenarios.
Expand Down Expand Up @@ -151,6 +155,12 @@ func (r *RemovalSimulator) SimulateNodeRemoval(
nodeInfo, err := r.clusterSnapshot.GetNodeInfo(nodeName)
if err != nil {
klog.Errorf("Can't retrieve node %s from snapshot, err: %v", nodeName, err)
unremovableReason := UnexpectedError
if errors.Is(err, clustersnapshot.ErrNodeNotFound) {
unremovableReason = NoNodeInfo
}
unremovableNode := &UnremovableNode{Node: &apiv1.Node{ObjectMeta: metav1.ObjectMeta{Name: nodeName}}, Reason: unremovableReason}
return nil, unremovableNode
}
klog.V(2).Infof("Simulating node %s removal", nodeName)

Expand Down
15 changes: 15 additions & 0 deletions cluster-autoscaler/simulator/cluster_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,6 +88,9 @@ func TestFindNodesToRemove(t *testing.T) {
fullNode := BuildTestNode("n4", 1000, 2000000)
fullNodeInfo := framework.NewTestNodeInfo(fullNode)

// node with no info in cluster snapshot
nodeWithoutInfo := &apiv1.Node{ObjectMeta: metav1.ObjectMeta{Name: "n5"}}

SetNodeReadyState(emptyNode, true, time.Time{})
SetNodeReadyState(drainableNode, true, time.Time{})
SetNodeReadyState(nonDrainableNode, true, time.Time{})
Expand Down Expand Up @@ -137,6 +140,10 @@ func TestFindNodesToRemove(t *testing.T) {
Node: drainableNode,
PodsToReschedule: []*apiv1.Pod{pod1, pod2},
}
nodeWithoutInfoUnremovable := UnremovableNode{
Node: nodeWithoutInfo,
Reason: NoNodeInfo,
}

clusterSnapshot := testsnapshot.NewTestSnapshotOrDie(t)

Expand Down Expand Up @@ -240,6 +247,14 @@ func TestFindNodesToRemove(t *testing.T) {
{Node: topoNode3, Reason: BlockedByPod, BlockingPod: &drain.BlockingPod{Pod: blocker2, Reason: drain.NotReplicated}},
},
},
{
name: "candidate not in clusterSnapshot should be marked unremovable",
candidates: []string{nodeWithoutInfo.Name},
allNodes: []*apiv1.Node{},
pods: []*apiv1.Pod{},
toRemove: nil,
unremovable: []*UnremovableNode{&nodeWithoutInfoUnremovable},
},
}

for _, test := range tests {
Expand Down
Loading