Pod stuck in Pending state

1
2
3
4
5

# kubectl describe pod my-pod
Events:
  Warning  FailedScheduling  default-scheduler  0/3 nodes are available:
  1 node(s) had taint {node.kubernetes.io/not-ready: }, that the pod didn't tolerate,
  2 Insufficient cpu, 2 Insufficient memory.

1
2

# Events
Warning  FailedScheduling  0/3 nodes are available: 3 Insufficient cpu.

1
2
3
4
5
6
7
8

# リソース要求を減らす
spec:
  containers:
    - name: app
      resources:
        requests:
          cpu: "100m"      # 1000mから100mに減らす
          memory: "128Mi"  # 1Giから128Miに減らす

1
2
3
4

# ノードのリソース確認
kubectl describe nodes | grep -A5 "Allocated resources"

# クラスターにノードを追加

1
2

# Events
Warning  FailedScheduling  0/3 nodes are available: 3 node(s) didn't match Pod's node affinity/selector.

1
2
3
4
5

# ノードのラベル確認
kubectl get nodes --show-labels

# ラベルを追加
kubectl label nodes my-node disktype=ssd

1
2
3
4

# またはnodeSelectorを削除/修正
spec:
  nodeSelector:
    disktype: ssd  # このラベルを持つノードが必要

1
2

# Events
Warning  FailedScheduling  0/3 nodes are available: 3 node(s) had taint {gpu: true}, that the pod didn't tolerate.

1
2
3
4
5
6

spec:
  tolerations:
    - key: "gpu"
      operator: "Equal"
      value: "true"
      effect: "NoSchedule"

1
2

# Events
Warning  FailedScheduling  persistentvolumeclaim "my-pvc" not found

1
2
3
4
5

# PVCの状態確認
kubectl get pvc

# PVが利用可能か確認
kubectl get pv

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

# PVCを作成
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: my-pvc
spec:
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 1Gi

1
2
3
4
5

# クォータ確認
kubectl describe resourcequota

# 使用量確認
kubectl get resourcequota -o yaml

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12

# 1. Podの詳細確認
kubectl describe pod my-pod

# 2. Eventsを確認
kubectl get events --sort-by=.metadata.creationTimestamp

# 3. ノードの状態確認
kubectl get nodes
kubectl describe node <node-name>

# 4. スケジューラのログ
kubectl logs -n kube-system -l component=kube-scheduler