Kubernetes1.8.4中如何安装Heapster

这篇文章给大家分享的是有关Kubernetes 1.8.4中如何安装Heapster的内容。小编觉得挺实用的，因此分享给大家做个参考，一起跟随小编过来看看吧。

站在用户的角度思考问题，与客户深入沟通，找到河曲网站设计与河曲网站推广的解决方案，凭借多年的经验，让设计与互联网技术结合，创造个性化、用户体验好的作品，建站类型包括：成都网站设计、网站制作、外贸营销网站建设、企业官网、英文网站、手机端网站、网站推广、空间域名、网络空间、企业邮箱。业务覆盖河曲地区。

Heapster

Heapster 是 Kubernetes 社区维护的容器集群监控分析工具。Heapster 会从 Kubernetes apiserver 获得所有 Node 信息，然后再通过这些 Node 来获得 kubelet 上的数据，最后再将所有收集到数据送到 Heapster 的后台储存 InfluxDB，最后利用 Grafana 来抓取 InfluxDB 的数据源来进行可视化。

在master通过 kubectl 来创建 kubernetes monitor 即可
生成kube-heapster.yml

cat < kube-heapster.yml
apiVersion: v1
kind: ServiceAccount
metadata:
  name: heapster-sa
  namespace: kube-system
  labels:
    kubernetes.io/cluster-service: "true"
    addonmanager.kubernetes.io/mode: Reconcile
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: heapster-binding
  labels:
    kubernetes.io/cluster-service: "true"
    addonmanager.kubernetes.io/mode: Reconcile
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: system:heapster
subjects:
- kind: ServiceAccount
  name: heapster
  namespace: kube-system
---
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  name: system:pod-nanny
  namespace: kube-system
  labels:
    kubernetes.io/cluster-service: "true"
    addonmanager.kubernetes.io/mode: Reconcile
rules:
- apiGroups:
  - ""
  resources:
  - pods
  verbs:
  - get
- apiGroups:
  - "extensions"
  resources:
  - deployments
  verbs:
  - get
  - update
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: heapster-binding
  namespace: kube-system
  labels:
    kubernetes.io/cluster-service: "true"
    addonmanager.kubernetes.io/mode: Reconcile
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: Role
  name: system:pod-nanny
subjects:
- kind: ServiceAccount
  name: heapster
  namespace: kube-system
---
kind: Service
apiVersion: v1
metadata:
  name: heapster
  namespace: kube-system
  labels:
    kubernetes.io/cluster-service: "true"
    addonmanager.kubernetes.io/mode: Reconcile
spec:
  ports:
    - port: 80
      targetPort: 8082
  selector:
    k8s-app: heapster
---
apiVersion: v1
kind: Service
metadata:
  name: monitoring-grafana
  namespace: kube-system
  labels:
    kubernetes.io/cluster-service: "true"
    addonmanager.kubernetes.io/mode: Reconcile
spec:
  ports:
    - port: 80
      protocol: TCP
      targetPort: ui
  selector:
    k8s-app: influxGrafana
---
apiVersion: v1
kind: Service
metadata:
  name: monitoring-influxdb
  namespace: kube-system
  labels:
    kubernetes.io/cluster-service: "true"
    addonmanager.kubernetes.io/mode: Reconcile
spec:
  ports:
    - name: http
      port: 8083
      targetPort: 8083
    - name: api
      port: 8086
      targetPort: 8086
  selector:
    k8s-app: influxGrafana
---
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: heapster
  labels:
    k8s-app: heapster
    kubernetes.io/cluster-service: "true"
    addonmanager.kubernetes.io/mode: Reconcile
  namespace: kube-system
spec:
  replicas: 1
  selector:
    matchLabels:
      k8s-app: heapster
  template:
    metadata:
      labels:
        k8s-app: heapster
      annotations:
        scheduler.alpha.kubernetes.io/critical-pod: ''
    spec:
      serviceAccountName: heapster-sa
      tolerations:
      - key: node-role.kubernetes.io/master
        effect: NoSchedule
      - key: "CriticalAddonsOnly"
        operator: "Exists"
      containers:
        - image: registry.cn-shenzhen.aliyuncs.com/rancher_cn/heapster-amd64:v1.5.0-beta.0
          name: heapster
          livenessProbe:
            httpGet:
              path: /healthz
              port: 8082
              scheme: HTTP
            initialDelaySeconds: 180
            timeoutSeconds: 5
          command:
            - /heapster
            - --source=kubernetes.summary_api:''
            - --sink=influxdb:http://monitoring-influxdb:8086
        - image: registry.cn-shenzhen.aliyuncs.com/rancher_cn/heapster-amd64:v1.5.0-beta.0
          name: eventer
          command:
            - /eventer
            - --source=kubernetes:''
            - --sink=influxdb:http://monitoring-influxdb:8086
        - image: registry.cn-hangzhou.aliyuncs.com/google_containers/addon-resizer:1.7
          name: heapster-nanny
          resources:
            limits:
              cpu: 100m
              memory: 128Mi
            requests:
              cpu: 50m
              memory: 90Mi
          env:
            - name: MY_POD_NAME
              valueFrom:
                fieldRef:
                  fieldPath: metadata.name
            - name: MY_POD_NAMESPACE
              valueFrom:
                fieldRef:
                  fieldPath: metadata.namespace
          command:
            - /pod_nanny
            - --cpu=80m
            - --extra-cpu=0.5m
            - --memory=140Mi
            - --extra-memory=4Mi
            - --threshold=5
            - --deployment=heapster
            - --container=heapster
            - --poll-period=300000
            - --estimator=exponential
        - image: registry.cn-hangzhou.aliyuncs.com/google_containers/addon-resizer:1.7
          name: eventer-nanny
          resources:
            limits:
              cpu: 100m
              memory: 128Mi
            requests:
              cpu: 50m
              memory: 90Mi
          env:
            - name: MY_POD_NAME
              valueFrom:
                fieldRef:
                  fieldPath: metadata.name
            - name: MY_POD_NAMESPACE
              valueFrom:
                fieldRef:
                  fieldPath: metadata.namespace
          command:
            - /pod_nanny
            - --cpu=100m
            - --extra-cpu=0m
            - --memory=190Mi
            - --extra-memory=500Ki
            - --threshold=5
            - --deployment=heapster
            - --container=eventer
            - --poll-period=300000
            - --estimator=exponential
---
kind: Deployment
apiVersion: extensions/v1beta1
metadata:
  name: influxdb-grafana
  namespace: kube-system
  labels:
    k8s-app: influxGrafana
    kubernetes.io/cluster-service: "true"
    addonmanager.kubernetes.io/mode: Reconcile
spec:
  replicas: 1
  selector:
    matchLabels:
      k8s-app: influxGrafana
  template:
    metadata:
      labels:
        k8s-app: influxGrafana
      annotations:
        scheduler.alpha.kubernetes.io/critical-pod: ''
    spec:
      tolerations:
      - key: node-role.kubernetes.io/master
        effect: NoSchedule
      - key: "CriticalAddonsOnly"
        operator: "Exists"
      containers:
        - name: influxdb
          image: registry.cn-shenzhen.aliyuncs.com/rancher_cn/heapster-influxdb-amd64:v1.3.3
          resources:
            limits:
              cpu: 100m
              memory: 500Mi
            requests:
              cpu: 100m
              memory: 500Mi
          ports:
            - name: http
              containerPort: 8083
            - name: api
              containerPort: 8086
          volumeMounts:
          - name: influxdb-persistent-storage
            mountPath: /data
        - name: grafana
          image: registry.cn-shenzhen.aliyuncs.com/rancher_cn/heapster-grafana-amd64:v4.4.3
          env:
          resources:
            limits:
              cpu: 100m
              memory: 100Mi
            requests:
              cpu: 100m
              memory: 100Mi
          env:
            - name: INFLUXDB_SERVICE_URL
              value: http://monitoring-influxdb:8086
            - name: GF_AUTH_BASIC_ENABLED
              value: "false"
            - name: GF_AUTH_ANONYMOUS_ENABLED
              value: "true"
            - name: GF_AUTH_ANONYMOUS_ORG_ROLE
              value: Admin
            - name: GF_SERVER_ROOT_URL
              value: /api/v1/proxy/namespaces/kube-system/services/monitoring-grafana/
          ports:
          - name: ui
            containerPort: 3000
            protocol: TCP
          volumeMounts:
          - name: grafana-persistent-storage
            mountPath: /var
      volumes:
      - name: influxdb-persistent-storage
        emptyDir: {}
      - name: grafana-persistent-storage
        emptyDir: {}
EOF

kubectl apply -f kube-heapster.yml

查看状态

kubectl -n kube-system get po,svc

完成后，就可以通过浏览器访问Grafana Dashboard

https://10.0.0.162:6443/api/v1/proxy/namespaces/kube-system/services/monitoring-grafana

感谢各位的阅读！关于“Kubernetes 1.8.4中如何安装Heapster”这篇文章就分享到这里了，希望以上内容可以对大家有一定的帮助，让大家可以学到更多知识，如果觉得文章不错，可以把它分享出去让更多的人看到吧！

标题名称：Kubernetes1.8.4中如何安装Heapster
地址分享：http://hbruida.cn/article/jepgip.html

Kubernetes1.8.4中如何安装Heapster

Heapster

其他资讯