5、kubernetes之 Pod控制器

第五部分 Pod控制器
1、pod回顾

apiVersion, kind, metadata, spec, status（只读）
spec：
  containers
  nodeSelector
  nodeName
  restartPolicy:
    Always, Never, OnFailure
  containers：
    name
    image
    imagePullPolicy：Always、Never、IfNotPresent
    ports：
      name
      containerPort
    livenessProbe
    readinessProbe
    liftcycle
  ExecAction: exec
  TCPSocketAction：tcpSocket
  HTTPGetAction: httpGet

2、Pod控制器
几种控制器说明

Pod控制器：
ReplicationController：
ReplicaSet：
Deployment：
DaemonSet：
Job：
Cronjob：
StatefulSet
TPR: Third Party Resources, 1.2+, 1.7
CDR: Custom Defined Resources, 1.8+
Operator：

Helm：更新慢，暂时用的不多。

Pod资源以外删除，不会被重建。
由控制器代管。
Rc：过于庞大
Job，一次性作业，保证周期性任务正常退出
Cronjob：周期性运行。
Deployment：只适用于无状态应用。关注群体行为。一群鸡，吃掉后买一只鸡苗。
StatefulSet：关注个体，哈士奇，投入感情。如redis-cluter，mysql。只是封装，需要手动封装脚本实现。在k8s上要求极其高，每种应用单独对待，对运维要求高。

3、控制器操作实践
ReplicaSet控制演示

[root@k8s-master ~]# kubectl get deploy,rs,pod
No resources found.
[root@k8s-master ~]# cat yas/rs-demo.yaml
apiVersion: apps/v1
kind: ReplicaSet
metadata:
  name: myapp
  namespace: default
spec:
  replicas: 2
  selector:
    matchLabels:
      app: myapp
      release: canary
  template:
    metadata:
      name: myapp-pod
      labels:
        app: myapp
        release: canary
        envirment: qa
    spec:
      containers:
      - name: myapp-container
        image: ikubernetes/myapp:v1
        ports:
        - name: http
          containerPort: 80
[root@k8s-master ~]# kubectl create -f yas/rs-demo.yaml
replicaset.apps/myapp created
[root@k8s-master ~]# kubectl get deploy,rs,pod         
NAME                          DESIRED   CURRENT   READY   AGE
replicaset.extensions/myapp   2         2         2       3m10s

NAME              READY   STATUS    RESTARTS   AGE
pod/myapp-gwqpd   1/1     Running   0          3m9s
pod/myapp-l9lvz   1/1     Running   0          3m9s
[root@k8s-master ~]# kubectl describe pods myapp-gwqpd # 查看容器详细信息。
[root@k8s-master ~]# kubectl delete pod myapp-l9lvz  # 此时直接删除pod，控制器作用会重建。
pod "myapp-l9lvz" deleted
[root@k8s-master ~]# kubectl get pod -owide  
更改pods数量，把pods由三个变为2个。
[root@k8s-master ~]# kubectl edit rs myapp  # 编辑副本数replicas: 3，保存退出后即生效。
[root@k8s-master ~]# kubectl get rs       
NAME    DESIRED   CURRENT   READY   AGE
myapp   3         3         2       18h
[root@k8s-master ~]# kubectl get pods     
NAME          READY   STATUS    RESTARTS   AGE
myapp-6zr5r   1/1     Running   1          18h
myapp-gwqpd   1/1     Running   1          18h
myapp-h76x5   1/1     Running   0          4s
[root@k8s-master ~]# kubectl get rs -owide
NAME    DESIRED   CURRENT   READY   AGE   CONTAINERS        IMAGES                 SELECTOR
myapp   3         3         3       18h   myapp-container   ikubernetes/myapp:v1   app=myapp,release=canary

Pod升级：（ikubernetes/myapp该镜像有时候只有V1、V3、V4版本可用）
编辑文件之后，只有重建资源，才会升级更新。
$ vim yas/rs-demo.yaml 升级到v2版本-->调整image镜像：image: ikubernetes/myapp:v2
$ kubectl apply -f yas/rs-demo.yaml

[root@k8s-master ~]# kubectl get pods -owide
NAME          READY   STATUS    RESTARTS   AGE     IP             NODE        NOMINATED NODE   READINESS GATES
myapp-cklcv   1/1     Running   0          4m58s   10.244.1.223   k8s-node1   <none>           <none>
myapp-z2lsn   1/1     Running   0          4m58s   10.244.2.19    k8s-node2   <none>           <none>
[root@k8s-master ~]# curl 10.244.1.223 
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>
[root@k8s-master ~]# curl  10.244.2.19 
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>
[root@k8s-master ~]# kubectl delete pod myapp-cklcv    #配置文件先做好，然后依次删除pods，对应rs来说是无感知状态。
pod "myapp-cklcv" deleted
[root@k8s-master ~]# kubectl get pods -owide        
NAME          READY   STATUS    RESTARTS   AGE     IP             NODE        NOMINATED NODE   READINESS GATES
myapp-j7bjb   1/1     Running   0          20s     10.244.1.224   k8s-node1   <none>           <none>
myapp-z2lsn   1/1     Running   0          6m18s   10.244.2.19    k8s-node2   <none>           <none>
[root@k8s-master ~]# curl 10.244.1.224 
Hello MyApp | Version: v2 | <a href="hostname.html">Pod Name</a>
[root@k8s-master ~]# curl  10.244.2.19              
注意：不要一次性删除，会影响用户访问效果（需要手动删除pod）。

Deploymen控制器
Deployment实现自动升级，明显优于ReplicaSet，建立在RS之上。如下，

 

[root@k8s-master yas]# cat deploy-demo.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: myapp-deploy
  namespace: default
spec:
  replicas: 2
  selector:
    matchLabels:
      app: myapp
      release: canary
  template:
    metadata:
      name: myapp
      labels:
        app: myapp
        release: canary
    spec:
      containers:
      - name: myapp
        image: ikubernetes/myapp:v1
        ports:
        - name: http
          containerPort: 80	  
[root@k8s-master yas]# kubectl create -f deploy-demo.yaml 
deployment.apps/myapp-deploy created
[root@k8s-master ~]# kubectl get deploy
NAME           READY   UP-TO-DATE   AVAILABLE   AGE
myapp-deploy   2/2     2            2           11s
[root@k8s-master ~]# kubectl get rs
NAME                     DESIRED   CURRENT   READY   AGE
myapp-deploy-9699554f5   2         2         2       14s
[root@k8s-master ~]# kubectl get pods  
NAME                           READY   STATUS    RESTARTS   AGE
myapp-deploy-9699554f5-2phbb   1/1     Running   0          45s
myapp-deploy-9699554f5-xtnds   1/1     Running   0          45s

同样修改镜像版本信息，image: ikubernetes/myapp:v2
Apply创建或者更新，Create只能创建
[root@k8s-master yas]# vim deploy-demo.yaml
root@k8s-master yas]# kubectl apply -f deploy-demo.yaml
此时可以看到，只要更新完成后，所有pod自动更新镜像。
-w查看更新过程：kubectl get pods -w
history版本状态查询：kubectl rollout history deploy/myapp-deploy

通过打补丁方式修改rs数量等操作
kubectl patch deployment myapp-deploy -p '{"spec":{"replicas":5}} # 修改副本
kubectl patch deployment myapp-deploy -p '{"spec":{"strategy":{"rollingUpdate":{"maxSurge":0 ,"maxUnavailable":1}}}}'
[root@k8s-master yas]# kubectl describe deploy myapp-deploy
...
MinReadySeconds: 0
RollingUpdateStrategy: 1 max unavailable, 0 max surge
Pod Template:
  Labels: app=myapp
               release=can
...
金丝雀发布，一个一个的删除替换，通过pause逐步替换实现。

[root@k8s-master ~]#  kubectl rollout
  history     显示 rollout 历史
  pause       标记提供的 resource 为中止状态
  resume      继续一个停止的 resource
  status      显示 rollout 的状态
  undo        撤销上一次的 rollout
[root@k8s-master yas]# kubectl set image deployment/myapp-deploy myapp=ikubernetes/myapp:v3 && kubectl rollout pause deployment myapp-deploy          
deployment.extensions/myapp-deploy image updated
deployment.extensions/myapp-deploy paused
[root@k8s-master ~]# kubectl rollout status deploy/myapp-deploy  监视器1
[root@k8s-master ~]# kubectl get pods -l app -w    监视器2
[root@k8s-master ~]# kubectl rollout resume deployment/myapp-deploy && kubectl rollout pause deployment myapp-deploy  # 解除，然后继续暂停
deployment.extensions/myapp-deploy resumed
deployment.extensions/myapp-deploy paused
会继续更新第二个pod，然后暂停了。

回滚操作：
kubectrollout history deploy/myapp-deploy # 查看历史版本
$ kubectl rollout undo deploy/myapp-deploy --to-revision=1
error: you cannot rollback a paused deployment; resume it first with 'kubectl rollout resume deployment/myapp-deploy' and try again #提示上面暂停了，不能回滚，需要解除暂停后安排回滚：kubectl rollout resume deployment/myapp-deploy

$ kubectl rollout resume deployment/myapp-deploy
deployment.extensions/myapp-deploy resumed
$ kubectl rollout undo deploy/myapp-deploy  --to-revision=1
deployment.extensions/myapp-deploy rolled back
$ kubectl get pods -owide
$ kubectl get rs -owide
$ kubectl get deploy -owide
$ curl 10.244.1.232  # 验证po服务版本信息

Demonset介绍
关注gitthub仓库：https://hub.docker.com/r/ikubernetes/filebeat/tags

$ docker pull ikubernetes/filebeat:5.6.5-alpine
$ docker images
$ docker image inspect ikubernetes/filebeat:5.6.5-alpine

$ cat ds-demo.yaml 
apiVersion: apps/v1
kind: Deployment
metadata:
  name: redis
  namespace: default
spec:
  replicas: 1
  selector:
    matchLabels:
      app: redis
      release: logstor
  template:
    metadata:
      labels:
        app: redis
        release: logstor
    spec:
      containers:
      - name: reids
        image: redis:4.0-alpine
        ports:
        - name: redis
          containerPort: 6379
---
apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: filebeat-ds
  namespace: default
spec:
  selector:
    matchLabels:
      app: filebeat
      release: stable
  template:
    metadata:
      labels:
        app: filebeat
        release: stable
    spec:
      containers:
      - name: filebeat
        image: ikubernetes/filebeat:5.6.5-alpine
        env:
        - name: REDIS_HOST
          value: redis.default.svc.cluster.local
        - name: REDIS_LOG_LEVEL
          value: info

$ kubectl create -f ds-demo.yaml
$ kubectl get pods
$ kubectl expose deploy redis --port=6379
$ kubectl get deploy redis
$ kubectl get svc redis
$ kubectl exec -it redis-9fb54c59-dhbz7 -- /bin/sh
$ kubectl exec -it redis-9fb54c59-dhbz7 -- netstat -lnpt
$ kubectl exec -it redis-9fb54c59-dhbz7 -- cat /etc/resolv.conf
$ kubectl exec -it redis-9fb54c59-dhbz7 -- nslookup redis.default.svc.cluster.local # 解析ip跟kubectl get svc redis获取匹配一致。
$ kubectl exec -it redis-9fb54c59-dhbz7 -- redis-cli -h redis.default.svc.cluster.local

$ kubectl exec -it filebeat-ds-6gtjv -- ps aux
$ kubectl exec -it filebeat-ds-6gtjv -- tail -3 /etc/filebeat/filebeat.yml
$ kubectl exec -it filebeat-ds-6gtjv -- printenv |grep -i REDIS_HOST

如上，发现fillebeat有配环境变量，但是没有加载成功。应该是启动加载顺序的问题。
综上，，先是redis服务启动，并暴露服务，最后filebeat添加redis变量，启动服务。
升级，也是一个个pod依次执行...
[root@k8s-master yas]# kubectl set image ds filebeat-ds filebeat=ikubernetes/filebeat:5.6.6-alpine
daemonset.extensions/filebeat-ds image updated
[root@k8s-master yas]# kubectl get pods -w