Kubernetes進階 -- 編寫資源清單（二）

容器列表

[root@server2 manifest]# vim pod.yml
apiVersion: v1
kind: Pod
metadata:
  name: myapp
#  namespace: default
#  labels:
spec:
  containers:
    - name: myappv1
      image: myapp:v1
    - name: myappv2
      image: myapp:v2		#在這裏定義兩個容器

[root@server2 manifest]# kubectl create -f pod.yml 
pod/myapp created

[root@server2 manifest]# kubectl get pod
NAME    READY   STATUS   RESTARTS   AGE
myapp   1/2     Error    4          2m5s      

[root@server2 manifest]# kubectl logs myapp myappv2
2020/06/26 02:57:40 [emerg] 1#1: bind() to 0.0.0.0:80 failed (98: Address in use)		# 這裏
nginx: [emerg] bind() to 0.0.0.0:80 failed (98: Address in use)

由於myapp這個容器開起的是一個服務，所以在pod當中開啓多個同樣容器時，端口就會出現搶奪，其中一個容器就會一直重啓，所以不能同時開啓端口一樣的兩個容器。

# 修改：
apiVersion: v1
kind: Pod
metadata:
  name: myapp
#  namespace: default
#  labels:
spec:
  containers:
    - name: myappv1
      image: myapp:v1
    - name: demo
      image: busyboxplus		# 這是一個交互式的終端,需要下面三個參數的支持
      tty: true			# 打開一個終端，需要 stdin 和stdinOnce的支持。
      stdin: true		
      stdinOnce: true
      
[root@server2 manifest]# kubectl delete -f pod.yml 
pod "myapp" deleted
[root@server2 manifest]# kubectl apply -f pod.yml 
pod/myapp created
[root@server2 manifest]# kubectl get pod
NAME    READY   STATUS    RESTARTS   AGE
myapp   2/2     Running   0          10s

[root@server2 manifest]# kubectl attach myapp -c demo -it		#-c指定容器，-it 交互
If you don't see a command prompt, try pressing enter.
/ # 
/ # curl localhost
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>

再這個容器卻可以本地訪問myapp：v1容器開啓的端口，這時因爲同一個pod中的容器共用網絡。

那末對於myapp打開一個服務並沒有打開一個終端的容器怎麼連接哪？

[root@server2 manifest]# kubectl exec -it myapp -c myappv1 sh	#exec讓他打開一個shell
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl kubectl exec [POD] -- [COMMAND] instead.
/ # netstat -tnlp
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State       PID/Program name    
tcp        0      0 0.0.0.0:80              0.0.0.0:*               LISTEN      1/nginx: master pro
/ # cat /etc/hostname 
myapp
/ # cd /usr/share/nginx/html/
/usr/share/nginx/html # cat index.html 
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>

關閉這些服務後，k8s會自動進行重啓。

鏡像拉取策略

[root@server2 manifest]# kubectl explain pod.spec.containers
   imagePullPolicy	<string>
     Image pull policy. One of Always, Never, IfNotPresent. Defaults to Always
     if :latest tag is specified, or IfNotPresent otherwise. Cannot be updated.

默認總會去倉庫拉取最新的，這就要求我們再更新鏡像的時候打上一個最新的標籤。

使用：

spec:
  containers:
    - name: myappv1
      image: myapp:v1
    - name: demo
      image: busyboxplus
      tty: true
      stdin: true
      stdinOnce: true
      imagePullPolicy: IfNotPresent
在這裏加上參數就行了。IfNotPresent 當我們本地沒有的時候纔去拉取

端口映射

[root@server2 manifest]# vim pod.yml
apiVersion: v1
kind: Pod
metadata:
  name: myapp
#  namespace: default
#  labels:
spec:
  containers:
    - name: myappv1
      image: myapp:v1
      ports:
      - name: http
        containerPort: 80		#端口映射，就不用再命令行中加參數了
        hostPort: 80

[root@server2 manifest]# kubectl apply -f pod.yml 
pod/myapp created
[root@server2 manifest]# kubectl get pod -o wide
NAME    READY   STATUS    RESTARTS   AGE   IP            NODE      NOMINATED NODE   READINESS GATES
myapp   1/1     Running   0          6s    10.244.2.22   server4   <none>           <none>

# 再server4上查看端口映射
[root@server4 ~]# iptables -t nat -nL| grep :80
CNI-HOSTPORT-SETMARK  tcp  --  10.244.2.22          0.0.0.0/0            tcp dpt:80
CNI-HOSTPORT-SETMARK  tcp  --  127.0.0.1            0.0.0.0/0            tcp dpt:80
DNAT       tcp  --  0.0.0.0/0            0.0.0.0/0            tcp dpt:80 to:10.244.2.22:80

使用主機網絡模式

[root@server2 manifest]# vim pod.yml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
#  namespace: default
#  labels:
spec:
  containers:
    - name: myappv1
      image: myapp:v1
  hostNetwork: true		#默認是false
#    - name: demo
#      image: busyboxplus
#      tty: true
#      stdin: true
#      stdinOnce: true
#      imagePullPolicy: IfNotPresent
[root@server2 manifest]# kubectl apply -f pod.yml 
kubepod/myapp created
[root@server2 manifest]# kubectl get pod -o wide
NAME    READY   STATUS    RESTARTS   AGE   IP             NODE      NOMINATED NODE   READINESS GATES
myapp   1/1     Running   0          5s    172.25.254.4   server4   <none>           <none>
[root@server2 manifest]# curl 172.25.254.4
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>

172.25.254.4 正是我們server4的ip地址，共用了網絡，但是這種容器只能開啓一個，不能有副本，因爲ip只有一個，會衝突。

pod重啓策略

restartPolicy，定義pod的重啓策略，有always，onfailure，never，默認爲always，就是無論容器如何終止，時中去重啓它，onfailure是隻有pod以非0退出碼終止時，纔會重啓它，就是當正常退出時（0）不重啓，never，將退出碼報告給master，永不重啓。

[root@server2 manifest]# kubectl exec -it myapp -c myappv1 sh	# 交互式訪問，
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl kubectl exec [POD] -- [COMMAND] instead.
/usr/share/nginx # nginx -s stop		# 關閉nginx
2020/06/26 03:58:21 [notice] 18#18: signal process started
/usr/share/nginx # command terminated with exit code 137
[root@server2 manifest]# kubectl get pod -o wide
NAME    READY   STATUS    RESTARTS   AGE   IP            NODE      NOMINATED NODE   READINESS GATES
myapp   1/1     Running   1          68s   10.244.2.23   server4   <none>           <none>
						## 進行了重啓

現在我們加上never參數：

[root@server2 manifest]# kubectl delete -f pod.yml 		# 刪除之前的
vim podpod "myapp" deleted

[root@server2 manifest]# vim pod.yml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
#  namespace: default
#  labels:
spec:
  containers:
    - name: myappv1
      image: myapp:v1
  restartPolicy: Never		# 永不重啓

[root@server2 manifest]# kubectl apply -f pod.yml 
pod/myapp created
[root@server2 manifest]# kubectl get pod -o wide
NAME    READY   STATUS    RESTARTS   AGE   IP            NODE      NOMINATED NODE   READINESS GATES
myapp   1/1     Running   0          4s    10.244.2.24   server4   <none>           <none>
[root@server2 manifest]# kubectl exec -it myapp -c myappv1 sh
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl kubectl exec [POD] -- [COMMAND] instead.
/ # nginx -s stop
2020/06/26 04:01:52 [notice] 12#12: signal process started
/ # command terminated with exit code 137
[root@server2 manifest]# kubectl get pod -o wide
NAME    READY   STATUS      RESTARTS   AGE   IP            NODE      NOMINATED NODE   READINESS GATES
myapp   0/1     Completed   0          20s   10.244.2.24   server4   <none>           <none>
		#容器就顯示完成了，就不再重啓

[root@server2 manifest]# kubectl get pod -o yaml |less
我們可以導出到yaml文件格式來查看具體的格式，這種方法還可以用與忘記格式時使用

resource資源限制

[root@server2 manifest]# vim pod.yml
apiVersion: v1
kind: Pod
metadata:
  name: myapp
#  namespace: default
#  labels:
spec:
  containers:
    - name: myappv1
      image: myapp:v1
      resources:
        requests:		# 下限
          memory: 100Mi
          cpu: 0.1
        limits:			# 上限
          memory: 200Mi
          cpu: 0.2

[root@server2 manifest]# kubectl apply -f pod.yml 
kubec	getpod/myapp created
[root@server2 manifest]# kubectl describe pod myapp 
Name:         myapp
Namespace:    default
...
    Limits:
      cpu:     200m
      memory:  200Mi
    Requests:
      cpu:        100m
      memory:     100Mi
...

結點選擇器

nodeSelector，定義node的label過濾標籤，以key：value格式指定。

[root@server2 manifest]# kubectl get node --show-labels 
NAME      STATUS   ROLES    AGE     VERSION   LABELS
...
server4   Ready    <none>   7d15h   v1.18.3   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,
kubernetes.io/hostname=server4,kubernetes.io/os=linux

[root@server2 manifest]# vim pod.yml
kind: Pod
metadata:
  name: myapp
#  namespace: default
#  labels:
spec:
  containers:
    - name: myappv1
      image: myapp:v1
      resources:
        requests:
          memory: 100Mi
          cpu: 0.1
        limits:
          memory: 200Mi
          cpu: 0.2

 - [ ] nodeSelector:
           kubernetes.io/hostname: server4		選擇server4的標籤

[root@server2 manifest]# kubectl get pod -o wide
NAME    READY   STATUS    RESTARTS   AGE   IP            NODE      NOMINATED NODE   READINESS GATES
myapp   1/1     Running   0          4s    10.244.2.25   server4   <none>           <none>
是在server4上.

[root@server2 manifest]# vim pod.yml 
 - [ ] nodeSelector:
           kubernetes.io/hostname: server3		改爲server3的標籤
[root@server2 manifest]# kubectl apply -f pod.yml 
pod/myapp created
[root@server2 manifest]# kubectl get pod -o wide
NAME    READY   STATUS    RESTARTS   AGE   IP            NODE      NOMINATED NODE   READINESS GATES
myapp   1/1     Running   0          3s    10.244.1.48   server3   <none>           <none>
就運行在server3上了

運行命令

指定容器的啓動命令。

apiVersion: v1
kind: Pod
metadata:
  name: pod-example
spec:
  containers:
  - name: ubuntu
    image: ubuntu:trusty
    command: ["echo"]
    args: ["Hello World"]		# 加上這兩個參數就行了，也可以都寫到command中，用，隔開。

打標籤

[root@server2 manifest]# vim pod.yml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
  labels:
    wo: myapp
spec:
  containers:
    - name: myappv1
      image: myapp:v1

[root@server2 manifest]# kubectl apply -f pod.yml 
pod/myapp created
[root@server2 manifest]# kubectl get pod --show-labels 
NAME    READY   STATUS    RESTARTS   AGE   LABELS
myapp   1/1     Running   0          14s   wo=myapp
[root@server2 manifest]# kubectl get pod -l wo		# 過濾包含wo的標籤
NAME    READY   STATUS    RESTARTS   AGE
myapp   1/1     Running   0          90s
[root@server2 manifest]# kubectl get pod -L wo
NAME    READY   STATUS    RESTARTS   AGE   WO
myapp   1/1     Running   0          2m    myapp
[root@server2 manifest]# kubectl label pod myapp version=v1		#打新的標籤
pod/myapp labeled
[root@server2 manifest]# kubectl get pod --show-labels 
NAME    READY   STATUS    RESTARTS   AGE     LABELS
myapp   1/1     Running   0          4m39s   version=v1,wo=myapp

[root@server2 manifest]# kubectl label pod myapp version=v2 --overwrite 		#更改標籤
pod/myapp labeled
[root@server2 manifest]# kubectl get pod --show-labels 
NAME    READY   STATUS    RESTARTS   AGE     LABELS
myapp   1/1     Running   0          5m56s   version=v2,wo=myapp
[root@server2 manifest]# kubectl label pod myapp wo=ni --overwrite 
pod/myapp labeled
[root@server2 manifest]# kubectl get pod --show-labels 
NAME    READY   STATUS    RESTARTS   AGE    LABELS
myapp   1/1     Running   0          6m8s   version=v2,wo=ni