Kubernetes练习<2>---Pod管理及资源请单

命令运行容器
一、重新开启

[wjjk8s@server1 ~]$ kubectl get node
error: error loading config file "/etc/kubernetes/admin.conf": open /etc/kubernetes/admin.conf: permission denied
[wjjk8s@server1 ~]$ su - root 
[wjjk8s@server1 ~]$ su - root 
[root@server1 ~]# chmod 666 /etc/kubernetes/admin.conf		#权限问题添加即可

[wjjk8s@server1 ~]# kubectl run nginx --image=nginx		#命令运行容器
[wjjk8s@server1 ~]# kubectl get pod				#创建一个pod
[wjjk8s@server1 ~]# kubectl get pod -n default 
[wjjk8s@server1 ~]# kubectl get all -n default 
[wjjk8s@server1 ~]$ kubectl get pod -n kube-system 		#全部running即可
[wjjk8s@server1 ~]$ kubectl describe pod nginx

二、命令运行容器

[wjjk8s@server1 ~]$ kubectl run test --image=busybox
pod/test created
[wjjk8s@server1 ~]$ kubectl delete pod test			#删除

集群内部任意节点可以访问pod，但集群外部无法访问。

[wjjk8s@server1 ~]$ kubectl run test -it --image=busyboxplus --restart=Never
curl 10.244.1.4							#集群内部可以访问nginx，

Pod管理
Pod是可以创建和管理Kubernetes计算的最小部署单元，一个Pod代表着2集群运行一个进程，每个Pod都有唯一的ip。
一个Pod类似于一个豌豆荚，包含一个或多个容器（docker），同一个Pod多个容器间共享IPC，Network和UTC namespace。

kubectl命令集官方文档：
https://kubernetes.io/docs/reference/generated/kubectl/kubectl-commands

1、自主式访问：开启两个容器

[wjjk8s@server1 ~]$ kubectl get pod
[wjjk8s@server1 ~]$ kubectl run myapp --image=myapp:v1
[wjjk8s@server1 ~]$ kubectl get pod -o wide			#两台node分摊调度
NAME    READY   STATUS              RESTARTS   AGE     IP           NODE      NOMINATED NODE   READINESS GATES
myapp   0/1     ContainerCreating   0          14s     <none>       server4   <none>           <none>
nginx   1/1     Running             1          5h32m   10.244.1.4   server2   <none>           <none>
[wjjk8s@server1 ~]$ kubectl delete pod myapp
pod "myapp" deleted

2、deployment的方式运行

[wjjk8s@server1 ~]$ kubectl create deployment myapp --image=myapp:v1
deployment.apps/myapp created
[wjjk8s@server1 ~]$ kubectl get all

[wjjk8s@server1 ~]$ kubectl delete pod nginx
pod "nginx" deleted
[wjjk8s@server1 ~]$ kubectl delete pod myapp-7449b5b68f-xslkh 
pod "myapp-7449b5b68f-xslkh" deleted
[wjjk8s@server1 ~]$ kubectl get pod				#删除不干净，会重新自动创建
NAME                     READY   STATUS              RESTARTS   AGE
myapp-7449b5b68f-d8g9r   0/1     ContainerCreating   0          7s

deployment的方式运行还可以实现拉伸为多个：

[wjjk8s@server1 ~]$ kubectl scale --replicas=2 deployment myapp	#创建副本

3、pod间的互相访问：创建service来实现
service是一个抽象概念，定义了一个服务的多个pod逻辑集合和访问pod策略，一般把service称为微服务。

（1）创建service

错误

[wjjk8s@server1 ~]$ kubectl expose deployment nginx --port=80 --target-port=80
Error from server (NotFound): deployments.apps "nginx" not found

#必须找一个已存在的容器

[wjjk8s@server1 ~]$ kubectl expose deployment myapp --port=80 --target-port=80
service/myapp exposed

[wjjk8s@server1 ~]$ kubectl get deployments.apps 
NAME    READY   UP-TO-DATE   AVAILABLE   AGE
myapp   0/2     2            0           16m

[wjjk8s@server1 ~]$ kubectl describe svc myapp
[wjjk8s@server1 ~]$ kubectl get svc myapp		#集群分给我们myapp的集群内部使用的ip地址
NAME    TYPE        CLUSTER-IP    EXTERNAL-IP   PORT(S)   AGE
myapp   ClusterIP   10.102.86.1   <none>        80/TCP    9m56s

此时的pod客户端可以通过service的名称访问后端的两个pod

(2)使用nodeport类型暴露端口，让外部客户端访问pod

[wjjk8s@server1 ~]$ kubectl edit svc myapp		#修改
service/myapp edited

kubectl expose deployment myapp --port=80 --target-port=80  --type=NodePort	#创建时指定

在ClusterIP基础上为service每台机器绑定一个端口，这样可以通过–type=NodePort访问


4、pod的滚动更新以及回滚

（1）更新pod

[wjjk8s@server1 ~]$ kubectl set image deployments myapp myapp=myapp:v2 --record
deployment.apps/myapp image updated
[wjjk8s@server1 ~]$ kubectl rollout history deployment myapp	#查看

（2）回滚

[wjjk8s@server1 ~]$ kubectl rollout undo deployment myapp --to-revision=1
deployment.apps/myapp rolled back

资源请单

[wjjk8s@server1 ~]$ kubectl delete deployments.apps myapp		#删除deployments
deployment.apps "myapp" deleted
[wjjk8s@server1 ~]$ kubectl get pod	
NAME   READY   STATUS    RESTARTS   AGE
test   1/1     Running   0          169m

（1）创建清单

[wjjk8s@server1 ~]$ vim .vimrc
autocmd filetype yaml setlocal ai ts=2 sw=2 et
[wjjk8s@server1 manifest]$ kubectl api-versions 			#查看版本
[wjjk8s@server1 ~]$ mkdir manifest
[wjjk8s@server1 manifest]$ kubectl explain pod				#编写pod资源设计内容
[wjjk8s@server1 manifest]$kubectl explain pod.spec			#期望spec状态
[wjjk8s@server1 manifest]$ kubectl explain pod.spec.containers
[wjjk8s@server1 manifest]$ vim pod.yaml					#编写清单
apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v1
      imagePullPolicy: IfNotPresent

[wjjk8s@server1 manifest]$ kubectl create -f pod.yaml 			#创建成功
pod/myapp created
[wjjk8s@server1 manifest]$ kubectl delete -f pod.yaml 			#删除
pod "myapp" deleted

（2）一个pod创建2个容器

[wjjk8s@server1 manifest]$ vim pod.yaml
apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v1
      imagePullPolicy: IfNotPresent
    - name: busybox
      image: busyboxplus
      imagePullPolicy: IfNotPresent
      tty: true
[wjjk8s@server1 manifest]$ kubectl create -f pod.yaml
pod/myapp created
[wjjk8s@server1 manifest]$ kubectl get pod
NAME    READY   STATUS    RESTARTS   AGE
myapp   2/2     Running   0          7s
test    1/1     Running   0          3h43m
 
[wjjk8s@server1 ~]$ kubectl exec -it myapp -c busybox -- sh		#连接进入容器

Pod管理*