參考自《kubernetes權威指南》

1.基本概念

作為服務發現機制的基本功能，在叢集內需要能夠通過服務名對服務進行訪問，這就需要一個叢集範圍的DNS服務來完成服務名到clusterIP的解析

kubernetes提供的虛擬DNS服務名為skydns，由4個元件組成

1. etcd：DNS儲存
2. kube2sky：將kubernetes master中的service註冊到etcd
3. skyDNS：提供dns域名解析服務
4. healthz：提供對skydns服務的健康檢查功能

2.實驗與操作

複製以下程式碼時，如果TAB長度不是八個空格，可能報檔案格式錯誤
1.建立skydns-rc.yaml的rc檔案
vim skydns-rc.yaml

apiVersion: v1
kind: ReplicationController
metadata:
  name: kube-dns-v11
  namespace: kube-system
  labels:
    k8s-app: kube-dns
    version: v11
    kubernetes.io/cluster-service: "true"
spec:
  replicas: 1
  selector:
    k8s-app: kube-dns
    version: v11
  template:
    metadata:
      labels:
        k8s-app: kube-dns
        version: v11
        kubernetes.io/cluster-service: "true"
    spec:
      containers:
      - name:
 
 etcd
        image: gcr.io/google_containers/etcd-amd64:2.2.1
        resources:
          limits:
            cpu: 100m
            memory: 50Mi
          requests:
            cpu: 100m
            memory: 50Mi
        command:
        - /usr/local/bin/etcd
        - -data-dir
        - /tmp/data
        - -listen-client-urls
 
        - http://127.0.0.1:2379,http://127.0.0.1:4001
        - -advertise-client-urls
        - http://127.0.0.1:2379,http://127.0.0.1:4001
        - -initial-cluster-token
        - skydns-etcd
        volumeMounts:
        - name: etcd-storage
          mountPath: /tmp/data
      - name: kube2sky
        image: gcr.io/google_containers/kube2sky-amd64:1.15
        resources:
          limits:
            cpu: 100m
            memory: 50Mi
          requests:
            cpu: 100m
            memory: 50Mi
        livenessProbe:
          httpGet:
            path: /healthz
            port: 8080
            scheme: HTTP
          initialDelaySeconds: 60
          timeoutSeconds: 5
          successThreshold: 1
          failureThreshold: 5
        readinessProbe:
          httpGet:
            path: /readiness
            port: 8081
            scheme: HTTP
          initialDelaySeconds: 30
          timeoutSeconds: 5
        args:
        - --kube-master-url=http://192.168.56.102:8080
        - --domain=cluster.local
      - name: skydns
        image: gcr.io/google_containers/skydns:2015-10-13-8c72f8c
        resources:
          limits:
            cpu: 100m
            memory: 50Mi
          requests:
            cpu: 100m
            memory: 50Mi
        args:
        - -machines=http://127.0.0.1:4001
        - -addr=0.0.0.0:53
        - -ns-rotate=false
        - -domain=cluster.local
        ports:
        - containerPort: 53
          name: dns
          protocol: UDP
        - containerPort: 53
          name: dns-tcp
          protocol: TCP
      - name: healthz
        image: gcr.io/google_containers/exechealthz:1.0
        resources:
          limits:
            cpu: 10m
            memory: 20Mi
          requests:
            cpu: 10m
            memory: 20Mi
        args:
        - -cmd=nslookup kubernetes.default.svc.cluster.local 127.0.0.1 > /de/null
        - -port=8080
        ports:
        - containerPort: 8080
          protocol: TCP
      volumes:
      - name: etcd-storage
        emptyDir: {}
      dnsPolicy: Default

2.建立skydns的svc檔案
vim skydns-svc.yaml

apiVersion: v1
kind: Service
metadata:
  name: kube-dns
  namespace: kube-system
  labels:
    k8s-app: kube-dns
    kubernetes.io/cluster-service: "true"
    kubernetes.io/name: "KubeDNS"
spec:
  selector:
    k8s-app: kube-dns
  clusterIP: 169.169.0.100
  ports:
  - name: dns
    port: 53
    protocol: UDP
  - name: dns-tcp
    port: 53
    protocol: TCP

如果無法訪問google , 需參考文章最後疑難解析修改映象名

3.修改每臺node上kubelet啟動引數,新增
–cluster_dns=169.169.0.100
–cluster_domain=cluser.local

4.建立資源
kubectl create -f skydns-rc.yaml
kubectl create -f skydns-svc.yaml

5.檢查資源是否正常
kubectl get rc –namespace=kube-system
kubectl get pods –namespace=kube-system
kubectl get svc –namespace=kube-system

6.為redis-master建立一個service
vim redis-master-service.yaml

apiVersion: v1
kind: Service
metadata:
  name: redis-master
  labels:
    name: redis-master
spec:
  ports:
  - port: 6379
    targetPort: 6379
  selector:
    name: redis-master

kubectl get services 檢查剛剛建立的redis-master服務是否正常

7.使用一個帶有nslookup工具的pod來驗證dns服務是否正常工作
vim busybox.yaml

apiVersion: v1
kind: Pod
metadata:
  name: busybox
  namespace: default
spec:
  containers:
  - name: busybox
    image: busybox
    command:
    - sleep
    - "3600"

kubectl create -f busybox.yaml

kubectl exec busybox – nslookup redis-master

實驗結果：無法解析redis-master
[[email protected] kubernetes]# kubectl exec kube-dns-v11-bh30c -c etcd –namespace=kube-system etcdctl ls /skydns/local/cluster
/skydns/local/cluster/svc
/skydns/local/cluster/pod
結果發現redis-master存在上述svc/default下，因此查詢需要修改為：

kubectl exec busybox -- nslookup redis-master.defualt.svc

[root@kube-master kubernetes]# kubectl exec kube-dns-v11-bh30c -c etcd --namespace=kube-system etcdctl -- ls --recursive  /skydns/local/cluster/
/skydns/local/cluster/svc
/skydns/local/cluster/svc/default
/skydns/local/cluster/svc/default/kubernetes
/skydns/local/cluster/svc/default/kubernetes/430884aa
/skydns/local/cluster/svc/default/redis-master
/skydns/local/cluster/svc/default/redis-master/3896faab
/skydns/local/cluster/svc/kube-system
/skydns/local/cluster/svc/kube-system/kube-dns
/skydns/local/cluster/svc/kube-system/kube-dns/1c613ef2
/skydns/local/cluster/pod
/skydns/local/cluster/pod/kube-system
/skydns/local/cluster/pod/kube-system/172-17-0-2
/skydns/local/cluster/pod/kube-system/172-17-0-2/c6db276b
/skydns/local/cluster/pod/default
/skydns/local/cluster/pod/default/172-17-0-3
/skydns/local/cluster/pod/default/172-17-0-3/edc9142

3.疑難解決

1.如果無法訪問google , 無法獲取pause-amd則做如下處理

docker pull kubeguide/pause-amd64:3.0
在每臺node上為kubelet的啟動引數新增: 
--pod-infra-container-image=kubeguide/pause-amd64:3.0
修改skydns-rc.yaml,修改image映象為如下
outrider/etcd-amd64
outrider/kube2sky
outrider/skydns
outrider/exechealthz

2.建立rc成功，卻沒有pod，kubectl describe rc mysql顯示No API token found for service account “default”
vim /etc/kubernetes/apiserver
將“–admission-control=NamespaceLifecycle,NamespaceExists, LimitRanger, SecurityContextDeny, ServiceAccount, ResourceQuota”
中的SecurityContextDeny, ServiceAccount刪掉並重啟kube-apiserver

3.no nodes，排程失敗，沒有可用的節點
node和master節點通訊故障，可以通過journalctl -u kubelet | tail -300分析日誌排錯

4.certificate is valid for kubernetes.default.svc, kubernetes.default, kubernetes, localhost, not kube-master 或者Get https://192.168.56.102:6443/api/v1/services?resourceVersion=0: x509: certificate is valid for 192.168.124.17, 169.169.0.1, not 192.168.56.102 ， certificate signed by unknown authority。
建議不使用證書實驗，或者重新做證書