配置Grafana用于prometheus

安装Grafana
$ docker pull grafana/grafana:5.3.4
1、准备pvc

apiVersion: v1kind: PersistentVolumemetadata:  name: grafanaspec:  capacity:    storage: 1Gi  accessModes:  - ReadWriteOnce  persistentVolumeReclaimPolicy: Recycle  nfs:    server: 192.168.1.244    path: /data/k8s---apiVersion: v1kind: PersistentVolumeClaimmetadata:  name: grafana  namespace: kube-opsspec:  accessModes:  - ReadWriteOnce  resources:    requests:      storage: 1Gi

2、修改/var/lib/grafana目录的权限

apiVersion: batch/v1kind: Jobmetadata:  name: grafana-chown  namespace: kube-opsspec:  template:    spec:      restartPolicy: Never      containers:      - name: grafana-chown        command: ["chown", "-R", "472:472", "/var/lib/grafana"]        image: busybox        imagePullPolicy: IfNotPresent        volumeMounts:        - name: storage          subPath: grafana          mountPath: /var/lib/grafana      volumes:      - name: storage        persistentVolumeClaim:          claimName: grafana

3、创建grafana pod

apiVersion: extensions/v1beta1kind: Deploymentmetadata:  name: grafana  namespace: kube-ops  labels:    app: grafanaspec:  revisionHistoryLimit: 10  template:    metadata:      labels:        app: grafana    spec:      containers:      - name: grafana        image: grafana/grafana:5.3.4        imagePullPolicy: IfNotPresent        ports:        - containerPort: 3000          name: grafana        env:        - name: GF_SECURITY_ADMIN_USER          value: admin         #登录用户名        - name: GF_SECURITY_ADMIN_PASSWORD          value: admin321        #登录密码        readinessProbe:          failureThreshold: 10          httpGet:            path: /api/health            port: 3000            scheme: HTTP          initialDelaySeconds: 60          periodSeconds: 10          successThreshold: 1          timeoutSeconds: 30        livenessProbe:          failureThreshold: 3          httpGet:            path: /api/health            port: 3000            scheme: HTTP          periodSeconds: 10          successThreshold: 1          timeoutSeconds: 1        resources:          limits:            cpu: 100m            memory: 256Mi          requests:            cpu: 100m            memory: 256Mi        volumeMounts:        - mountPath: /var/lib/grafana          subPath: grafana          name: storage      securityContext:        fsGroup: 472        runAsUser: 472      volumes:      - name: storage        persistentVolumeClaim:          claimName: grafana

$ kubectl get pod -n kube-ops |grep grafana
grafana-bbfd776d6-w9xf7 1/1 Running 1 111m
grafana-chown-26wmr 0/1 Completed 0 111m
$ kubectl logs grafana-bbfd776d6-w9xf7 -n kube-ops
4、创建svc

apiVersion: v1kind: Servicemetadata:  name: grafana  namespace: kube-ops  labels:    app: grafanaspec:  type: NodePort  ports:    - port: 3000  selector:    app: grafana

$ kubectl get svc -n kube-ops |grep grafana
grafana NodePort 10.107.208.176 3000:31171/TCP

使用grafana
1、基本配置
http://192.168.1.243:31171
使用grafana pod环境变量中设置的用户名和密码登录
点击Add data source进入添加数据源界面----数据源名字----数据源类型： Prometheus----url：http://prometheus:9090 ----访问模式：server
访问模式是用来控制如何处理对数据源的请求的，分两种
服务器(Server)访问模式（默认）
浏览器(Browser)访问模式
导入dashboard：点击左侧加号----import----Grafana.com Dashboard：输入162
导入配置好的dashboard：点击左侧加号----import----upload .json file----选择本地文件mydashboard.json
2、安装k8s插件grafana-kubernetes-app
要安装这个插件，需要到 grafana 的 Pod 里面去执行安装命令
$ kubectl get pods -n kube-ops |grep grafana
grafana-bbfd776d6-w9xf7 1/1 Running 1
$ kubectl exec -it grafana-bbfd776d6-w9xf7 bash
grafana-kubernetes-app6d6-w9xf7:/usr/share/grafana$ grafana-cli plugins install grafana-kubernetes-app
$ kubectl delete -f grafana-deploy.yaml
$ kubectl apply -f grafana-deploy.yaml
重新打开garfana
http://192.168.1.243:31171
点击Installed Apps下面的----kubernetes----enable----点击链接Connect to your Kubernetes Cluster and deploy metric exporters，进入kubernetes / Cluster Config配置界面

未完待续