kubernetes-控制器statefulset和Job（十三）

   从前面的学习我们知道使用Deployment创建的pod是无状态的，当挂载了Volume之后，如果该pod挂了，Replication Controller会再启动一个pod来保证可用性，但是由于pod是无状态的，pod挂了就会和之前的Volume的关系断开，新创建的Pod无法找到之前的Pod。但是对于用户而言，他们对底层的Pod挂了是没有感知的，但是当Pod挂了之后就无法再使用之前挂载的存储卷。

    为了解决这一问题，就引入了StatefulSet用于保留Pod的状态信息。

    StatefulSet是为了解决有状态服务的问题（对应Deployments和ReplicaSets是为无状态服务而设计），其应用场景包括：

• 1、稳定的持久化存储，即Pod重新调度后还是能访问到相同的持久化数据，基于PVC来实现
• 2、稳定的网络标志，即Pod重新调度后其PodName和HostName不变，基于Headless Service（即没有Cluster IP的Service）来实现
• 3、有序部署，有序扩展，即Pod是有顺序的，在部署或者扩展的时候要依据定义的顺序依次依次进行（即从0到N-1，在下一个Pod运行之前所有之前的Pod必须都是Running和Ready状态），基于init containers来实现
• 4、有序收缩，有序删除（即从N-1到0）
• 5、有序的滚动更新

参考：https://kubernetes.io/docs/tutorials/stateful-application/basic-stateful-set/

 创建statefulSet示例

[root@k8s-master1 volume]# cat nginx-demo.yaml
apiVersion: v1
kind: Service
metadata:
  name: nginx
  labels:
    app: nginx
spec:
  ports:
  - port: 80
    name: web
  clusterIP: None
  selector:
    app: nginx
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: web
spec:
  selector:
    matchLabels:
      app: nginx
  serviceName: "nginx"
  replicas: 3
  template:
    metadata:
      labels:
        app: nginx
    spec:
      terminationGracePeriodSeconds: 10
      containers:
      - name: nginx
        image: nginx 
        ports:
        - containerPort: 80
          name: web
        volumeMounts:
        - name: www
          mountPath: /usr/share/nginx/html
  volumeClaimTemplates:
  - metadata:
      name: www
    spec:
      accessModes: [ "ReadWriteOnce" ]
      storageClassName: "managed-nfs-storage"
      resources:
        requests:
          storage: 1Gi

[root@k8s-master1 volume]# kubectl apply -f nginx-demo.yaml 
service/nginx created
statefulset.apps/web created
[root@k8s-master1 volume]# kubectl get svc,sts　　　　#查看无头服务，svc的CLUSTER—IP为None
NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
service/kubernetes   ClusterIP   10.0.0.1     <none>        443/TCP   9d
service/nginx        ClusterIP   None         <none>        80/TCP    13s

NAME                   READY   AGE
statefulset.apps/web   1/3     13s
[root@k8s-master1 volume]# kubectl get pv,pvc　　　　#查看pv动态供给
NAME                                                                          CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS   CLAIM               STORAGECLASS          REASON   AGE
persistentvolume/default-www-web-0-pvc-7fe5d833-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-0   managed-nfs-storage            18s
persistentvolume/default-www-web-1-pvc-85d17f15-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-1   managed-nfs-storage            14s
persistentvolume/default-www-web-2-pvc-8aa41937-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-2   managed-nfs-storage            6s

NAME                              STATUS   VOLUME                                                       CAPACITY   ACCESS MODES   STORAGECLASS          AGE
persistentvolumeclaim/www-web-0   Bound    default-www-web-0-pvc-7fe5d833-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   24s
persistentvolumeclaim/www-web-1   Bound    default-www-web-1-pvc-85d17f15-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   14s
persistentvolumeclaim/www-web-2   Bound    default-www-web-2-pvc-8aa41937-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   6s
[root@k8s-master1 volume]# kubectl get pod　　　　#查看pod
NAME                                    READY   STATUS    RESTARTS   AGE
nfs-client-provisioner-f69cd5cf-rfbdb   1/1     Running   0          2m16s
web-0                                   1/1     Running   0          33s
web-1                                   1/1     Running   0          23s
web-2                                   1/1     Running   0          15s

删除statefulSet，PVC依旧存在的，再重新创建pod时，依旧会重新去绑定原来的pvc

[root@k8s-master1 volume]# kubectl delete -f nginx-demo.yaml 
service "nginx" deleted
statefulset.apps "web" deleted
[root@k8s-master1 volume]# kubectl get pods -w
NAME                                    READY   STATUS        RESTARTS   AGE
nfs-client-provisioner-f69cd5cf-rfbdb   1/1     Running       0          7m58s
web-0                                   0/1     Terminating   0          6m15s
web-1                                   0/1     Terminating   0          6m5s
web-2                                   0/1     Terminating   0          5m57s
web-0   0/1   Terminating   0     6m17s
web-0   0/1   Terminating   0     6m17s
web-2   0/1   Terminating   0     5m59s
web-2   0/1   Terminating   0     5m59s
web-1   0/1   Terminating   0     6m11s
web-1   0/1   Terminating   0     6m11s
[root@k8s-master1 volume]# kubectl apply -f nginx-demo.yaml 
service/nginx created
statefulset.apps/web created
[root@k8s-master1 volume]# kubectl get pv,pvc
NAME                                                                          CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS   CLAIM               STORAGECLASS          REASON   AGE
persistentvolume/default-www-web-0-pvc-7fe5d833-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-0   managed-nfs-storage            9m34s
persistentvolume/default-www-web-1-pvc-85d17f15-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-1   managed-nfs-storage            9m30s
persistentvolume/default-www-web-2-pvc-8aa41937-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-2   managed-nfs-storage            9m22s

NAME                              STATUS   VOLUME                                                       CAPACITY   ACCESS MODES   STORAGECLASS          AGE
persistentvolumeclaim/www-web-0   Bound    default-www-web-0-pvc-7fe5d833-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   9m40s
persistentvolumeclaim/www-web-1   Bound    default-www-web-1-pvc-85d17f15-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   9m30s
persistentvolumeclaim/www-web-2   Bound    default-www-web-2-pvc-8aa41937-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   9m22s

滚动更新

RollingUpdate 更新策略在 StatefulSet 中实现 Pod 的自动滚动更新。 当StatefulSet的 .spec.updateStrategy.type 设置为 RollingUpdate 时，默认为：RollingUpdate。StatefulSet 控制器将在 StatefulSet 中删除并重新创建每个 Pod。 它将以与 Pod 终止相同的顺序进行（从最大的序数到最小的序数），每次更新一个 Pod。 在更新其前身之前，它将等待正在更新的 Pod 状态变成正在运行并就绪。如下操作的滚动更新是有2-0的顺序更新。

[root@k8s-master1 volume]# vim nginx-demo.yaml 
[root@k8s-master1 volume]# kubectl apply -f nginx-demo.yaml 
service/nginx unchanged
statefulset.apps/web configured
[root@k8s-master1 volume]# kubectl get pods -w
NAME                                    READY   STATUS        RESTARTS   AGE
nfs-client-provisioner-f69cd5cf-rfbdb   1/1     Running       0          14m
web-0                                   1/1     Running       0          3m42s
web-1                                   1/1     Running       0          3m34s
web-2                                   0/1     Terminating   0          3m24s
web-2   0/1   Terminating   0     3m35s
web-2   0/1   Terminating   0     3m35s
web-2   0/1   Pending   0     0s
web-2   0/1   Pending   0     0s
web-2   0/1   ContainerCreating   0     0s
web-2   1/1   Running   0     7s
web-1   1/1   Terminating   0     3m52s
web-1   0/1   Terminating   0     3m53s
web-1   0/1   Terminating   0     3m54s
web-1   0/1   Terminating   0     3m54s
web-1   0/1   Pending   0     0s
web-1   0/1   Pending   0     0s
web-1   0/1   ContainerCreating   0     0s
web-1   1/1   Running   0     23s
web-0   1/1   Terminating   0     4m25s
web-0   0/1   Terminating   0     4m26s
web-0   0/1   Terminating   0     4m33s
web-0   0/1   Terminating   0     4m33s
web-0   0/1   Pending   0     0s
web-0   0/1   Pending   0     0s
web-0   0/1   ContainerCreating   0     0s
web-0   1/1   Running   0     4s

扩展伸缩

[root@k8s-master1 volume]# kubectl scale sts web --replicas=5  #扩展到5个副本
statefulset.apps/web scaled
[root@k8s-master1 volume]# kubectl get pod -w　　　　#动态查看扩容
NAME                                    READY   STATUS              RESTARTS   AGE
nfs-client-provisioner-f69cd5cf-rfbdb   1/1     Running             0          25m
web-0                                   1/1     Running             0          9m40s
web-1                                   1/1     Running             0          10m
web-2                                   1/1     Running             0          10m
web-3                                   0/1     ContainerCreating   0          12s
web-3   1/1   Running   0     21s
web-4   0/1   Pending   0     0s
web-4   0/1   Pending   0     0s
web-4   0/1   Pending   0     1s
web-4   0/1   ContainerCreating   0     1s
web-4   1/1   Running   0     9s
[root@k8s-master1 volume]# kubectl get pv,pvc　　　　#查看pv绑定
NAME                                                                          CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS   CLAIM               STORAGECLASS          REASON   AGE
persistentvolume/default-www-web-0-pvc-7fe5d833-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-0   managed-nfs-storage            25m
persistentvolume/default-www-web-1-pvc-85d17f15-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-1   managed-nfs-storage            25m
persistentvolume/default-www-web-2-pvc-8aa41937-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-2   managed-nfs-storage            25m
persistentvolume/default-www-web-3-pvc-c7efc39f-0a47-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-3   managed-nfs-storage            2m10s
persistentvolume/default-www-web-4-pvc-d498cf02-0a47-11e9-b58a-000c298a2b5f   1Gi        RWO            Delete           Bound    default/www-web-4   managed-nfs-storage            112s

NAME                              STATUS   VOLUME                                                       CAPACITY   ACCESS MODES   STORAGECLASS          AGE
persistentvolumeclaim/www-web-0   Bound    default-www-web-0-pvc-7fe5d833-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   25m
persistentvolumeclaim/www-web-1   Bound    default-www-web-1-pvc-85d17f15-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   25m
persistentvolumeclaim/www-web-2   Bound    default-www-web-2-pvc-8aa41937-0a44-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   25m
persistentvolumeclaim/www-web-3   Bound    default-www-web-3-pvc-c7efc39f-0a47-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   2m13s
persistentvolumeclaim/www-web-4   Bound    default-www-web-4-pvc-d498cf02-0a47-11e9-b58a-000c298a2b5f   1Gi        RWO            managed-nfs-storage   112s

[root@k8s-master1 volume]# kubectl patch sts web -p '{"spec":{"replicas":2}}'　　#缩容到两个副本
statefulset.apps/web patched
[root@k8s-master1 volume]# kubectl get pod -w　　　　#动态查看缩容
NAME                                    READY   STATUS        RESTARTS   AGE
nfs-client-provisioner-f69cd5cf-rfbdb   1/1     Running       0          28m
web-0                                   1/1     Running       0          12m
web-1                                   1/1     Running       0          13m
web-2                                   1/1     Running       0          13m
web-3                                   1/1     Running       0          3m16s
web-4                                   0/1     Terminating   0          2m55s
web-4   0/1   Terminating   0     2m57s
web-4   0/1   Terminating   0     2m57s
web-3   1/1   Terminating   0     3m18s
web-3   0/1   Terminating   0     3m19s
web-3   0/1   Terminating   0     3m22s
web-3   0/1   Terminating   0     3m22s
web-2   1/1   Terminating   0     13m
web-2   0/1   Terminating   0     13m
web-2   0/1   Terminating   0     13m
web-2   0/1   Terminating   0     13m

更新策略

[root@k8s-master1 volume]# kubectl patch sts web -p '{"spec":{"updateStrategy":{"rollingUpdate":{"partition":2}}}}'
statefulset.apps/web patched
[root@k8s-master1 volume]# kubectl describe sts web
Name:               web
Namespace:          default
CreationTimestamp:  Fri, 28 Dec 2018 10:11:01 +0800
Selector:           app=nginx
Labels:             <none>
Annotations:        kubectl.kubernetes.io/last-applied-configuration:
                      {"apiVersion":"apps/v1","kind":"StatefulSet","metadata":{"annotations":{},"name":"web","namespace":"default"},"spec":{"replicas":3,"select...
Replicas:           824638349112 desired | 2 total
Update Strategy:    RollingUpdate
  Partition:        824638947612
Pods Status:        2 Running / 0 Waiting / 0 Succeeded / 0 Failed
。。。

版本升级

[root@k8s-master1 volume]# kubectl get sts -o wide
NAME   READY   AGE   CONTAINERS   IMAGES
web    2/2     29m   nginx        nginx:1.14
[root@k8s-master1 volume]# kubectl set image sts/web nginx=nginx:1.15
statefulset.apps/web image updated
[root@k8s-master1 volume]# kubectl get sts -o wide
NAME   READY   AGE   CONTAINERS   IMAGES
web    2/2     32m   nginx        nginx:1.15

headless service做dns解析是直接解析到pod的

[root@k8s-master1 volume]# kubectl get pod,svc -o wide
NAME                                        READY   STATUS    RESTARTS   AGE   IP            NODE            NOMINATED NODE   READINESS GATES
pod/nfs-client-provisioner-f69cd5cf-rfbdb   1/1     Running   0          74m   172.17.73.3   192.168.0.126   <none>           <none>
pod/web-0                                   1/1     Running   0          59m   172.17.32.3   192.168.0.125   <none>           <none>
pod/web-1                                   1/1     Running   0          59m   172.17.73.5   192.168.0.126   <none>           <none>

NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE   SELECTOR
service/kubernetes   ClusterIP   10.0.0.1     <none>        443/TCP   9d    <none>
service/nginx        ClusterIP   None         <none>        80/TCP    63m   app=nginx
[root@k8s-master1 volume]# kubectl run -it --image=busybox:1.28.4 --rm --restart=Never sh
If you don't see a command prompt, try pressing enter.
/ # nslookup web-0.nginx
Server:    10.0.0.2
Address 1: 10.0.0.2 kube-dns.kube-system.svc.cluster.local

Name:      web-0.nginx
Address 1: 172.17.32.3 web-0.nginx.default.svc.cluster.local
/ # nslookup web-1.nginx
Server:    10.0.0.2
Address 1: 10.0.0.2 kube-dns.kube-system.svc.cluster.local

Name:      web-1.nginx
Address 1: 172.17.73.5 web-1.nginx.default.svc.cluster.local

JOB

https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/

Job分为普通任务（Job）和定时任务（CronJob）

一次性执行

应用场景：离线数据处理，视频解码等业务

job示例

[root@k8s-master1 ~]# vim job.yaml 
apiVersion: batch/v1
kind: Job
metadata:
  name: pi
spec:
  template:
    spec:
      containers:
      - name: pi
        image: perl
        command: ["perl",  "-Mbignum=bpi", "-wle", "print bpi(2000)"]
      restartPolicy: Never
  backoffLimit: 4

[root@k8s-master1 ~]# kubectl apply -f job.yaml
job.batch/pi created
[root@k8s-master1 ~]# kubectl get pod
NAME                                    READY   STATUS      RESTARTS   AGE
nfs-client-provisioner-f69cd5cf-rfbdb   1/1     Running     0          87m
pi-nslj8                                0/1     Completed   0          119s
web-0                                   1/1     Running     0          71m
web-1                                   1/1     Running     0          72m
[root@k8s-master1 ~]# kubectl logs pi-nslj8
3.1415926535897932384626433832795028841971693993751058209749445923078164062862089986280348253421170679821480865132823066470938446095505822317253594081284811174502841027019385211055596446229489549303819644288109756659334461284756482337867831652712019091456485669234603486104543266482133936072602491412737245870066063155881748815209209628292540917153643678925903600113305305488204665213841469519415116094330572703657595919530921861173819326117931051185480744623799627495673518857527248912279381830119491298336733624406566430860213949463952247371907021798609437027705392171762931767523846748184676694051320005681271452635608277857713427577896091736371787214684409012249534301465495853710507922796892589235420199561121290219608640344181598136297747713099605187072113499999983729780499510597317328160963185950244594553469083026425223082533446850352619311881710100031378387528865875332083814206171776691473035982534904287554687311595628638823537875937519577818577805321712268066130019278766111959092164201989380952572010654858632788659361533818279682303019520353018529689957736225994138912497217752834791315155748572424541506959508295331168617278558890750983817546374649393192550604009277016711390098488240128583616035637076601047101819429555961989467678374494482553797747268471040475346462080466842590694912933136770289891521047521620569660240580381501935112533824300355876402474964732639141992726042699227967823547816360093417216412199245863150302861829745557067498385054945885869269956909272107975093029553211653449872027559602364806654991198818347977535663698074265425278625518184175746728909777727938000816470600161452491921732172147723501414419735685481613611573525521334757418494684385233239073941433345477624168625189835694855620992192221842725502542568876717904946016534668049886272327917860857843838279679766814541009538837863609506800642251252051173929848960841284886269456042419652850222106611863067442786220391949450471237137869609563643719172874677646575739624138908658326459958133904780275898

cronjob

https://kubernetes.io/docs/tasks/job/automated-tasks-with-cron-jobs/

定时任务，像Linux的Crontab一样。

应用场景：通知，备份

cronjob示例

[root@k8s-master1 ~]# vim cronjob.yaml 
apiVersion: batch/v1beta1
kind: CronJob
metadata:
  name: hello
spec:
  schedule: "*/1 * * * *"
  jobTemplate:
    spec:
      template:
        spec:
          containers:
          - name: hello
            image: busybox
            args:
            - /bin/sh
            - -c
            - date; echo Hello from the Kubernetes cluster
          restartPolicy: OnFailure

[root@k8s-master1 ~]# kubectl apply -f cronjob.yaml
cronjob.batch/hello created
[root@k8s-master1 ~]# kubectl get cronjob
NAME    SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
hello   */1 * * * *   False     1        5s              8s
[root@k8s-master1 ~]# kubectl get pod
NAME                                    READY   STATUS              RESTARTS   AGE
hello-1545968100-zhf2w                  0/1     Completed           0          67s
hello-1545968160-zftxj                  0/1     ContainerCreating   0          7s
nfs-client-provisioner-f69cd5cf-rfbdb   1/1     Running             0          96m
pi-nslj8                                0/1     Completed           0          10m
web-0                                   1/1     Running             0          80m
web-1                                   1/1     Running             0          81m
[root@k8s-master1 ~]# kubectl logs hello-1545968100-zhf2w
Fri Dec 28 03:35:14 UTC 2018
Hello from the Kubernetes cluster