Kube State Metrics

本文說明如何設定 Google Kubernetes Engine 部署作業，以便使用 Google Cloud Managed Service for Prometheus 從 Kube State Metrics 收集指標。本文將說明如何執行下列操作：

• 設定 Kube State Metrics，以便回報指標。
• 設定 Managed Service for Prometheus 的 PodMonitoring 資源，以收集匯出的指標。
• 在 Cloud Monitoring 中存取資訊主頁，即可查看指標。
• 設定快訊規則來監控指標。

只有在使用 Managed Service for Prometheus 的 代管收集作業時，才適用這些操作說明。如果您使用自行部署的收集作業，請參閱 Kube State Metrics 的來源存放區，瞭解安裝資訊。

這些操作說明僅供參考，適用於大多數 Kubernetes 環境。 如果因安全或機構政策限制而無法安裝應用程式或匯出工具，建議您參閱開放原始碼文件尋求支援。

必要條件

如要使用 Managed Service for Prometheus 和代管收集功能，從 Kube State Metrics 收集指標，部署作業必須符合下列條件：

• 叢集必須執行 Google Kubernetes Engine 1.21.4-gke.300 以上版本。
• 您必須執行 Managed Service for Prometheus，並啟用代管收集作業。詳情請參閱「 開始使用代管集合」一文。

• 如要使用 Cloud Monitoring 提供的整合資訊主頁，請務必使用 kube-state-metrics 2.4.2 以上版本。

  如要進一步瞭解可用的資訊主頁，請參閱「安裝資訊主頁」。

受管理 Kube State Metrics

GKE 提供全代管的 kube 狀態指標部署作業，可做為設定這項整合功能的替代方案。可安裝的套件提供一組精選的 Kube 狀態指標，設定起來更輕鬆。這項整合功能可提供額外指標，但設定較為複雜。如要先查看精選套件再決定使用哪個，請參閱「 套件：Kube 狀態指標」。

安裝 Kube State Metrics

您可以使用下列設定安裝 Kube State Metrics：

# Copyright 2023 Google LLC
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

apiVersion: apps/v1
kind: StatefulSet
metadata:
  labels:
    app.kubernetes.io/name: kube-state-metrics
    app.kubernetes.io/version: 2.12.0
  namespace: gmp-public
  name: kube-state-metrics
spec:
  replicas: 1
  selector:
    matchLabels:
      app.kubernetes.io/name: kube-state-metrics
  serviceName: kube-state-metrics
  template:
    metadata:
      labels:
        app.kubernetes.io/name: kube-state-metrics
        app.kubernetes.io/version: 2.12.0
    spec:
      affinity:
        nodeAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            nodeSelectorTerms:
            - matchExpressions:
              - key: kubernetes.io/arch
                operator: In
                values:
                - arm64
                - amd64
              - key: kubernetes.io/os
                operator: In
                values:
                - linux
      containers:
      - name: kube-state-metric
        image: registry.k8s.io/kube-state-metrics/kube-state-metrics:v2.12.0
        env:
        - name: POD_NAME
          valueFrom:
            fieldRef:
              fieldPath: metadata.name
        - name: POD_NAMESPACE
          valueFrom:
            fieldRef:
              fieldPath: metadata.namespace
        args:
        - --pod=$(POD_NAME)
        - --pod-namespace=$(POD_NAMESPACE)
        - --port=8080
        - --telemetry-port=8081
        ports:
        - name: metrics
          containerPort: 8080
        - name: metrics-self
          containerPort: 8081
        resources:
          requests:
            cpu: 100m
            memory: 190Mi
          limits:
            memory: 250Mi
        securityContext:
          allowPrivilegeEscalation: false
          privileged: false
          capabilities:
            drop:
            - all
          runAsUser: 1000
          runAsGroup: 1000
        livenessProbe:
          httpGet:
            path: /healthz
            port: 8080
          initialDelaySeconds: 5
          timeoutSeconds: 5
        readinessProbe:
          httpGet:
            path: /
            port: 8081
          initialDelaySeconds: 5
          timeoutSeconds: 5
      serviceAccountName: kube-state-metrics
---
apiVersion: v1
kind: Service
metadata:
  labels:
    app.kubernetes.io/name: kube-state-metrics
    app.kubernetes.io/version: 2.12.0
  namespace: gmp-public
  name: kube-state-metrics
spec:
  clusterIP: None
  ports:
  - name: metrics
    port: 8080
    targetPort: metrics
  - name: metrics-self
    port: 8081
    targetPort: metrics-self
  selector:
    app.kubernetes.io/name: kube-state-metrics
---
apiVersion: v1
kind: ServiceAccount
metadata:
  namespace: gmp-public
  name: kube-state-metrics
  labels:
    app.kubernetes.io/name: kube-state-metrics
    app.kubernetes.io/version: 2.12.0
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: gmp-public:kube-state-metrics
  labels:
    app.kubernetes.io/name: kube-state-metrics
    app.kubernetes.io/version: 2.12.0
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: gmp-public:kube-state-metrics
subjects:
- kind: ServiceAccount
  namespace: gmp-public
  name: kube-state-metrics
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: gmp-public:kube-state-metrics
  labels:
    app.kubernetes.io/name: kube-state-metrics
    app.kubernetes.io/version: 2.12.0
rules:
- apiGroups:
  - ""
  resources:
  - configmaps
  - secrets
  - nodes
  - pods
  - services
  - resourcequotas
  - replicationcontrollers
  - limitranges
  - persistentvolumeclaims
  - persistentvolumes
  - namespaces
  - endpoints
  verbs:
  - list
  - watch
- apiGroups:
  - ""
  resources:
  - pods
  verbs:
  - get
- apiGroups:
  - extensions
  resources:
  - daemonsets
  - deployments
  - replicasets
  - ingresses
  verbs:
  - list
  - watch
- apiGroups:
  - apps
  resources:
  - statefulsets
  - daemonsets
  - deployments
  - replicasets
  verbs:
  - list
  - watch
- apiGroups:
  - apps
  resources:
  - statefulsets
  verbs:
  - get 
- apiGroups:
  - batch
  resources:
  - cronjobs
  - jobs
  verbs:
  - list
  - watch
- apiGroups:
  - autoscaling
  resources:
  - horizontalpodautoscalers
  verbs:
  - list
  - watch
- apiGroups:
  - authentication.k8s.io
  resources:
  - tokenreviews
  verbs:
  - create
- apiGroups:
  - authorization.k8s.io
  resources:
  - subjectaccessreviews
  verbs:
  - create
- apiGroups:
  - policy
  resources:
  - poddisruptionbudgets
  verbs:
  - list
  - watch
- apiGroups:
  - certificates.k8s.io
  resources:
  - certificatesigningrequests
  verbs:
  - list
  - watch
- apiGroups:
  - storage.k8s.io
  resources:
  - storageclasses
  - volumeattachments
  verbs:
  - list
  - watch
- apiGroups:
  - admissionregistration.k8s.io
  resources:
  - mutatingwebhookconfigurations
  - validatingwebhookconfigurations
  verbs:
  - list
  - watch
- apiGroups:
  - networking.k8s.io
  resources:
  - networkpolicies
  - ingresses
  verbs:
  - list
  - watch
- apiGroups:
  - coordination.k8s.io
  resources:
  - leases
  verbs:
  - list
  - watch
---
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: kube-state-metrics
  namespace: gmp-public
spec:
  maxReplicas: 10
  minReplicas: 1
  scaleTargetRef:
    apiVersion: apps/v1
    kind: StatefulSet
    name: kube-state-metrics
  metrics:
  - type: Resource
    resource:
      name: memory
      target:
        type: Utilization
        averageUtilization: 60
  behavior:
    scaleDown:
      policies:
      - type: Pods
        value: 1
        # Under-utilization needs to persist for `periodSeconds` before any action can be taken.
        # Current supported max from https://kubernetes.io/docs/reference/kubernetes-api/workload-resources/horizontal-pod-autoscaler-v2beta2/.
        periodSeconds: 1800
      # Current supported max from https://kubernetes.io/docs/reference/kubernetes-api/workload-resources/horizontal-pod-autoscaler-v2beta2/.
      stabilizationWindowSeconds: 3600
---
apiVersion: monitoring.googleapis.com/v1
kind: ClusterPodMonitoring
metadata:
  name: kube-state-metrics
  labels:
    app.kubernetes.io/name: kube-state-metrics
    app.kubernetes.io/part-of: google-cloud-managed-prometheus
spec:
  selector:
    matchLabels:
      app.kubernetes.io/name: kube-state-metrics
  endpoints:
  - port: metrics
    interval: 30s
    metricRelabeling:
    - action: keep
      # Curated subset of metrics to reduce costs while populating default set of sample dashboards at
      # https://github.com/GoogleCloudPlatform/monitoring-dashboard-samples/tree/master/dashboards/kubernetes
      # Change this regex to fit your needs for which objects you want to monitor    
      regex: kube_(daemonset|deployment|replicaset|pod|namespace|node|statefulset|persistentvolume|horizontalpodautoscaler|job_created)(_.+)?
      sourceLabels: [__name__]
  targetLabels:
    metadata: [] # explicitly empty so the metric labels are respected
---
apiVersion: monitoring.googleapis.com/v1
kind: PodMonitoring
metadata:
  namespace: gmp-public
  name: kube-state-metrics
  labels:
    app.kubernetes.io/name: kube-state-metrics
    app.kubernetes.io/part-of: google-cloud-managed-prometheus
spec:
  selector:
    matchLabels:
      app.kubernetes.io/name: kube-state-metrics
  endpoints:
  - port: metrics-self
    interval: 30s

如要套用本機檔案的設定變更，請執行下列指令：

kubectl apply -f FILE_NAME

您也可以使用 Terraform 管理設定。

定義規則和快訊

您可以使用下列 Rules 設定，為指標定義快訊：

# Copyright 2022 Google LLC
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

apiVersion: monitoring.googleapis.com/v1
kind: Rules
metadata:
  namespace: gmp-public
  name: kube-state-metrics-rules
  labels:
    app.kubernetes.io/component: rules
    app.kubernetes.io/name: kube-state-metrics
    app.kubernetes.io/part-of: google-cloud-managed-prometheus
spec:
  groups:
    - name: kube-state-metrics
      interval: 30s
      rules:
      - alert: KubeStateMetricsListErrors
        annotations:
          description: kube-state-metrics is experiencing errors at an elevated rate in list operations. This is likely causing it to not be able to expose metrics about Kubernetes objects correctly or at all.
          summary: kube-state-metrics is experiencing errors in list operations.
        expr: |
          (sum(rate(kube_state_metrics_list_total{job="kube-state-metrics",result="error"}[5m]))
            /
          sum(rate(kube_state_metrics_list_total{job="kube-state-metrics"}[5m])))
          > 0.01
        for: 15m
        labels:
          severity: critical
      - alert: KubeStateMetricsWatchErrors
        annotations:
          description: kube-state-metrics is experiencing errors at an elevated rate in watch operations. This is likely causing it to not be able to expose metrics about Kubernetes objects correctly or at all.
          summary: kube-state-metrics is experiencing errors in watch operations.
        expr: |
          (sum(rate(kube_state_metrics_watch_total{job="kube-state-metrics",result="error"}[5m]))
            /
          sum(rate(kube_state_metrics_watch_total{job="kube-state-metrics"}[5m])))
          > 0.01
        for: 15m
        labels:
          severity: critical
      - alert: KubeStateMetricsShardingMismatch
        annotations:
          description: kube-state-metrics pods are running with different --total-shards configuration, some Kubernetes objects may be exposed multiple times or not exposed at all.
          summary: kube-state-metrics sharding is misconfigured.
        expr: |
          stdvar (kube_state_metrics_total_shards{job="kube-state-metrics"}) != 0
        for: 15m
        labels:
          severity: critical
      - alert: KubeStateMetricsShardsMissing
        annotations:
          description: kube-state-metrics shards are missing, some Kubernetes objects are not being exposed.
          summary: kube-state-metrics shards are missing.
        expr: |
          2^max(kube_state_metrics_total_shards{job="kube-state-metrics"}) - 1
            -
          sum( 2 ^ max by (shard_ordinal) (kube_state_metrics_shard_ordinal{job="kube-state-metrics"}) )
          != 0
        for: 15m
        labels:
          severity: critical

如要套用本機檔案的設定變更，請執行下列指令：

kubectl apply -f FILE_NAME

您也可以使用 Terraform 管理設定。

如要進一步瞭解如何將規則套用至叢集，請參閱「受管理規則評估和快訊」。

這項Rules設定是根據kube-state-metrics 存放區提供的規則和快訊改編而成。

驗證設定

您可以使用 Metrics Explorer 確認 Kube State Metrics 設定正確無誤。Cloud Monitoring 可能需要一到兩分鐘才能擷取指標。

如要確認指標已擷取，請按照下列步驟操作：

1. 前往 Google Cloud 控制台的 「Metrics Explorer」頁面：

   前往 Metrics Explorer

   如果您是使用搜尋列尋找這個頁面，請選取子標題為「Monitoring」的結果。

2. 在查詢建構工具窗格的工具列中，選取名稱為  MQL 或  PromQL 的按鈕。
3. 確認已在「Language」(語言) 切換按鈕中選取「PromQL」。語言切換按鈕位於同一工具列，可供你設定查詢格式。
4. 輸入並執行下列查詢：
   

   up{job="kube-state-metrics", cluster="CLUSTER_NAME", namespace="gmp-public"}
   

安裝資訊主頁

Cloud Monitoring 提供整合的範例資訊主頁程式庫。範例程式庫包含「Prometheus」資訊主頁，您可以安裝這些資訊主頁，在 Google Cloud 控制台中查看資料。

請注意，Kubernetes 叢集 Prometheus 總覽資訊主頁需要安裝 Node Exporter。Kubernetes Pod Prometheus 總覽資訊主頁需要安裝 Node Exporter 和 cAdvisor/Kubelet。

如要從範本庫安裝資訊主頁，請按照下列步驟操作：

1. 在 Google Cloud 控制台中，前往「Dashboards」(資訊主頁) 頁面：

   前往「Dashboards」(資訊主頁)

   如果您是使用搜尋列尋找這個頁面，請選取子標題為「Monitoring」的結果。

2. 選取「範例庫」分頁標籤。
3. 選擇「其他」類別。
4. (選用) 如要查看資訊主頁的靜態預覽畫面，而不需安裝，請按一下「預覽」。
5. 選取要安裝的資訊主頁，然後按一下 「匯入」。

如要進一步瞭解如何安裝資訊主頁，請參閱「 安裝範例資訊主頁」。

疑難排解

如要瞭解如何排解指標擷取問題，請參閱「 排解擷取端問題」一文中的「 收集匯出工具資料時發生問題」一節。