K8S/Kubernetes

namespace资源名称空间

    删除namespace资源会级联删除其所包含的所有其它资源对象
    名称空间仅仅只是用来限制资源名称的作用域      并不能实现Pod的通信隔离

在名称空间下操作service 实现service资源隔离
 [root@k8s-master ~]# kubectl create service nodeport my-ns --tcp=5678:8080 -n qa
service/my-ns created
[root@k8s-master ~]# kubectl get svc
NAME         TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.96.0.1        <none>        443/TCP        231d
myapp        NodePort    10.100.165.177   <none>        80:32185/TCP   21h
[root@k8s-master ~]# kubectl get svc -n default
NAME         TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.96.0.1        <none>        443/TCP        231d
myapp        NodePort    10.100.165.177   <none>        80:32185/TCP   21h
[root@k8s-master ~]# kubectl get svc -n qa
NAME      TYPE       CLUSTER-IP     EXTERNAL-IP   PORT(S)          AGE
my-ns     NodePort   10.102.116.9   <none>        5678:30170/TCP   24s

[root@k8s-master ~]# kubectl delete svc my-ns
Error from server (NotFound): services "my-ns" not found
[root@k8s-master ~]# kubectl delete svc my-ns -n qa
service "my-ns" deleted
[root@k8s-master ~]# kubectl delete all --all -n qa
service "my-ns" deleted

[root@k8s-master ~]# kubectl get svc
NAME         TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.96.0.1        <none>        443/TCP        230d
myapp        NodePort    10.100.165.177   <none>        80:32185/TCP   5m
[root@k8s-master ~]# kubectl run client --image=busybox --restart=Never -it /bin/sh
If you don't see a command prompt, try pressing enter.
/ # wget -O -q http://myapp.default:80
Connecting to myapp.default:80 (10.100.165.177:80)
-q                   100% |*****************************************|    65  0:00:00 ETA
/ # wget -O  - -q http://myapp.default:80
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>
/ # wget -O  - -q http://myapp.default:80
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>

[root@k8s-master ~]# wget -O - -q http://myappd.default:80/hostname.html
[root@k8s-master ~]# kubectl get pods
NAME                     READY     STATUS    RESTARTS   AGE
client                   1/1       Running   0          46m
myapp-6865459dff-c59qp   1/1       Running   0          1h
myapp-6865459dff-zd6wg   1/1       Running   0          10m
[root@k8s-master ~]# kubectl exec -it client /bin/sh
/ # wget -O - -q http://myapp.default:80/hostname.html
myapp-6865459dff-zd6wg

pod和pods deployment和deployments都可以
[root@k8s-master ~]# kubectl get deployment
NAME      DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
myapp     2         2         2            2           1h
[root@k8s-master ~]# kubectl get deployments
NAME      DESIRED   CURRENT   UP-TO-DATE   AVAILABLE   AGE
myapp     2         2         2 


kubectl打补丁包
[root@k8s-master ~]# kubectl patch deployment myapp-deploy -p '{"spec":{"strategy":{"rollingUpdate":{"maxSurge":1,"maxUnavailable":0}}}}'
deployment.extensions/myapp-deploy patched

[root@k8s-master ~]# kubectl set image deployment myapp-deploy myapp=ikubernetes:v3 && kubectl rollout pause deployment myapp-deploy
deployment.extensions/myapp-deploy image updated
deployment.extensions/myapp-deploy paused

 

pod介绍

pod的种类

   1.自主式pod
      自主式pod在非正常结束的情况下无法自动重新启动一个新的Pod

   2.受控Pod
      由控制器管理启动的Pod在异常退出的时候 控制器会自动创建新Pod

   1.pod存活性探测
         存活性探测一旦检测到异常就会不断的重启容器 直到服务恢复正常
    2.pod就绪性探测
         探测失败后不会杀死或者重启容器 而是通知其尚未就绪   并触发依赖于其就绪状态的操作如从Service中移除此pod对象
         就绪性探测和service把用户请求调度到pod上有着重要的关系

 3.httpGet
    livenessProbe:
      httpGet:
        path: /healthz
        port: 80
        scheme: HTTP
 4.execAction
    livenessProbe:
      exec:
        command: ["test","-e","/tmp/healthy"]
 5.TcpSocket
    livenessProbe:
      tcpSocket:
        port: 80
 6.pod生命周期钩子函数
    lifecycle:
      postStart:
        exec:
          command: ["/bin/sh","-c","echo 'lifecycle hooks handler ' > /usr/share/nginx/html/test.html"]

 

pod控制器

       自主式pod由对应节点上的kubelet负责监控其容器的存活性   容器主进程崩溃后 kubelet能够自动重启相应的容器

       kubelet对非主进程崩溃类的容器错误却是无法感知的  比如index.html不存在 这种异常检测依赖于pod自定义的存活性检测

       pod对象遭意外删除或者pod所在的节点发生故障    节点故障后kubelet服务 也会异常导致节点上的pod运行将无法得到保证

       这种情况需要用到Pod控制器来保证所有pod来正常运行  pod的存活性和就绪性探测只能保证pod所在节点上pod正常运行

       pod控制器保证可以是跨集群节点上pod正常运行

 

资源注解  Annotations

       资源注解和资源标签类似 不过注解不能用于标签和挑选K8S资源对象      仅可以用于资源提供元数据信息.元数据不受字符数量的限制