4、kubernetes之Pod应用进阶

第四部分 Pod应用进阶
1、资源配置清单

复制代码
自主式Pod资源
资源的清单格式：
  一级字段：apiVersion（group/version），kind，netadata（name,namespace,labels,annotations，...），spec，status（只读）
Pod资源：
  spec.container <[]object>
  
  - name <string>
    image <string>
	imagePullPolicy <string>
	  Always（默认方式，每次启动都要从仓库去拉取镜像，比较费时）, Never, IfNotPresent(本地不存在就去仓库下载镜像，缺点是latest镜像更新后，他不去主动更新)
  修改镜像中的默认应用：command、args
      参考：https://kubernetes.io/docs/tasks/inject-data-application/define-command-argument-container/
  标签：
    key=value
      key:字母，数字，——、-、.
      value:可以为空，只能字母或数字开头及结尾，中间可以使用	
  标签选择器：
    等值关系：=,==,!=
    集合关系：KEY in (value1,value2,...)
	          KEY notin (value1,value2,...)
			  KEY
			  ! KEY
  许多资源支持内嵌字段定义其使用的标签选择器：
    matchLabels：直接使用给定键值
	matchExpressions：基于给定的表达式来定义使用标签选择器，{key:'',operator:'OPRRAOR',values:[VAL1,VAL2，...]}
	  操作符：
	    In，NotIn： values字段的值必须为非空列表
		Exists，NotExists：values字段的值必须为空列表；
  nodeSelector，节点标签选择器，针对打标签的node节点进行选择
  nodeName，节点名称，不常用。
  annotations，与labels不同的地方，它不能用于挑选资源对象，仅用于对象提供“元数据”
复制代码

$ kubectl explain pods.spec.containers

2、标签使用
a、多个标签查询，多个标签过滤
$ kubectl get pods --show-labels
$ kubectl get pods -l app --show-labels
$ kubectl get pods -L app
$ kubectl get pods -L app,run
$ kubectl get pods -l "relesae in (beta,stable,alpha)" --show-labels
$ kubectl get pods -l "relesae notin (beta,stable,alpha)" --show-labels
$ kubectl get pods -l "relesae" --show-labels
$ kubectl get pods -l "! relesae" --show-labels

复制代码
[root@k8s-master ~]# kubectl get pods --show-labels       
NAME                           READY   STATUS    RESTARTS   AGE     LABELS
myapp-86984b4c7c-mjlsj         1/1     Running   0          5h56m   pod-template-hash=86984b4c7c,run=myapp
myapp-86984b4c7c-n4mcj         1/1     Running   0          5h56m   pod-template-hash=86984b4c7c,run=myapp
myapp-86984b4c7c-t4c56         1/1     Running   0          5h56m   pod-template-hash=86984b4c7c,run=myapp
myappx                         2/2     Running   3          3h35m   app=pod-demo,tier=frontend
nginx-deploy-55d8d67cf-kcrhs   1/1     Running   0          7h16m   pod-template-hash=55d8d67cf,run=nginx-deploy
test-64585fcd47-2pwbk          1/1     Running   0          7h10m   pod-template-hash=64585fcd47,run=test
[root@k8s-master ~]# kubectl get pods -l app --show-labels
NAME     READY   STATUS    RESTARTS   AGE     LABELS
myappx   2/2     Running   3          3h35m   app=pod-demo,tier=frontend
[root@k8s-master ~]# kubectl get pods -L app              
NAME                           READY   STATUS    RESTARTS   AGE     APP
myapp-86984b4c7c-mjlsj         1/1     Running   0          5h56m   
myapp-86984b4c7c-n4mcj         1/1     Running   0          5h56m   
myapp-86984b4c7c-t4c56         1/1     Running   0          5h56m   
myappx                         2/2     Running   3          3h35m   pod-demo
nginx-deploy-55d8d67cf-kcrhs   1/1     Running   0          7h16m   
test-64585fcd47-2pwbk          1/1     Running   0          7h10m   
[root@k8s-master ~]# kubectl get pods -L app,run
NAME                           READY   STATUS    RESTARTS   AGE     APP        RUN
myapp-86984b4c7c-mjlsj         1/1     Running   0          5h57m              myapp
myapp-86984b4c7c-n4mcj         1/1     Running   0          5h57m              myapp
myapp-86984b4c7c-t4c56         1/1     Running   0          5h57m              myapp
myappx                         2/2     Running   3          3h36m   pod-demo   
nginx-deploy-55d8d67cf-kcrhs   1/1     Running   0          7h17m              nginx-deploy
test-64585fcd47-2pwbk          1/1     Running   0          7h11m              test
复制代码

b、标签新增或覆盖使用
$ kubectl label pods myappx relesae=canary
$ kubectl label pods myappx relesae=stable --overwrite
$ kubectl get pods -l relesae=stable --show-labels
$ kubectl get pods -l relesae=stable,app=pod-demo --show-labels
$ kubectl get pods -l relesae=stable,app!=pod-dem --show-labels

复制代码
[root@k8s-master ~]# kubectl get pods -l app --show-labels
NAME     READY   STATUS    RESTARTS   AGE     LABELS
myappx   2/2     Running   3          3h42m   app=pod-demo,tier=frontend
[root@k8s-master ~]# kubectl label pods myappx relesae=canary
pod/myappx labeled
[root@k8s-master ~]# kubectl get pods -l app --show-labels   
NAME     READY   STATUS    RESTARTS   AGE     LABELS
myappx   2/2     Running   3          3h43m   app=pod-demo,relesae=canary,tier=frontend
[root@k8s-master ~]# kubectl label pods myappx relesae=stable
error: 'relesae' already has a value (canary), and --overwrite is false
[root@k8s-master ~]# kubectl label pods myappx relesae=stable --overwrite
pod/myappx labeled
[root@k8s-master ~]# kubectl get pods -l app --show-labels   
NAME     READY   STATUS    RESTARTS   AGE     LABELS
myappx   2/2     Running   3          3h43m   app=pod-demo,relesae=stable,tier=frontend
[root@k8s-master ~]# kubectl get pods -l relesae=stable --show-labels
NAME     READY   STATUS    RESTARTS   AGE     LABELS
myappx   2/2     Running   3          3h46m   app=pod-demo,relesae=stable,tier=frontend
复制代码

c、节点标记
添加节点资源的时候，可以针对标签给定适当资源。nodeSelector节点标签选择器

复制代码
[root@k8s-master ~]# kubectl get no --show-labels
NAME         STATUS   ROLES    AGE   VERSION   LABELS
k8s-master   Ready    master   15h   v1.14.3   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=k8s-master,kubernetes.io/os=linux,node-role.kubernetes.io/master=
k8s-node1    Ready    <none>   15h   v1.14.3   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=k8s-node1,kubernetes.io/os=linux
k8s-node2    Ready    <none>   9h    v1.14.3   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=k8s-node2,kubernetes.io/os=linux
[root@k8s-master ~]# kubectl label nodes k8s-node1 disktype=ssd
node/k8s-node1 labeled
[root@k8s-master ~]# kubectl get no --show-labels              
NAME         STATUS   ROLES    AGE   VERSION   LABELS
k8s-master   Ready    master   15h   v1.14.3   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=k8s-master,kubernetes.io/os=linux,node-role.kubernetes.io/master=
k8s-node1    Ready    <none>   15h   v1.14.3   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,disktype=ssd,kubernetes.io/arch=amd64,kubernetes.io/hostname=k8s-node1,kubernetes.io/os=linux
k8s-node2    Ready    <none>   9h    v1.14.3   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=k8s-node2,kubernetes.io/os=linux
复制代码

pod创建过程直接nodeSelector选择器，挑选合适的node节点。

复制代码
[root@k8s-master ~]# cat /root/yas/pod-demo.yaml
apiVersion: v1
kind: Pod
metadata:
  name: pod-demo
  namespace: default
  labels:
    app: myapp
    tier: frontend
  annotations:
    k8s.mastet/create-by: "sunny wei"
spec:
  containers:
  - image: ikubernetes/myapp:v1
    name: myapp
  - image: busybox:latest
    name: busybox
    command:
    - "/bin/sh"
    - "-c"
    - "sleep 3600"
  nodeSelector:
    disktype: ssd
[root@k8s-master ~]# kubectl delete -f /root/yas/pod-demo.yaml
[root@k8s-master ~]# kubectl apply -f /root/yas/pod-demo.yaml      
pod/pod-demo created
[root@k8s-master ~]# kubectl get pods pod-demo
NAME       READY   STATUS    RESTARTS   AGE
pod-demo   2/2     Running   0          43s
[root@k8s-master ~]# kubectl get pods pod-demo -owide            
NAME       READY   STATUS    RESTARTS   AGE     IP             NODE        NOMINATED NODE   READINESS GATES
pod-demo   2/2     Running   0          4m23s   10.244.1.213   k8s-node1   <none>           <none>
[root@k8s-master ~]# kubectl get no -l disktype=ssd 
NAME        STATUS   ROLES    AGE   VERSION
k8s-node1   Ready    <none>   16h   v1.14.3
复制代码

3、pod 状态

Pod运行启动过程。如下截图

存活(Liveness) 探针 - 探测应用是否处于健康状态，如果不健康则删除并重新创建容器. 即在什么情况下重启pod是合适的?
就绪(Readiness) 探针 - 探测应用是否启动完成并且处于正常服务状态，如果不正常则不会接收来自 Kubernetes Service 的流量. 即在什么情况下, 我们应该从服务端点列表删除pod, 使其不再响应请求?

Pod生命周期

复制代码
 状态：
 创建pod：
 Pod生命周期重要行为
   初始化容器
   容器探测
     liveness
	 readliness
  restartPolicy：
    Always, OnFailure, Never. Default to Always.
  探针类型有三种：
    ExecAction、TCPSocketAction、HTTPGetAction
复制代码

4、存活探测案例

[root@k8s-master ~]# kubectl explain pod.spec.containers.livenessProbe
[root@k8s-master ~]# kubectl explain pod.spec.containers.livenessProbe.exec
自定义探针，只需要定义其中一个就行。
使用自定义命令探测，开启2个窗口。
[root@k8s-master ~]# kubectl get pod liveness-exec-pod -w

复制代码
[root@k8s-master ~]# cat /root/yas/liveness-exec.yaml
apiVersion: v1
kind: Pod
metadata:
  name: liveness-exec-pod
  namespace: default
spec:
  containers:
  - image: ikubernetes/myapp:v1
    name: liveness-exec-container
    imagePullPolicy: IfNotPresent
    command: ["/bin/sh","-c","touch /tmp/healthy;sleep 20;rm -f /tmp/healthy;sleep 3600"]
    livenessProbe:
      exec:
        command: ["test","-c","/tmp/healthy"]
      initialDelaySeconds: 1
      periodSeconds: 3
复制代码

[root@k8s-master ~]# kubectl create -f /root/yas/liveness-exec.yaml
pod/liveness-exec-pod created
[root@k8s-master ~]# kubectl describe pod liveness-exec-pod 查看报错信息
[root@k8s-master ~]# kubectl get pod liveness-exec-pod -w
NAME READY STATUS RESTARTS AGE
liveness-exec-pod 1/1 Running 1 47s
liveness-exec-pod 1/1 Running 2 88s
liveness-exec-pod 1/1 Running 3 2m11s
liveness-exec-pod 1/1 Running 4 2m53s
liveness-exec-pod 0/1 CrashLoopBackOff 4 3m35s
liveness-exec-pod 1/1 Running 5 4m21s
liveness-exec-pod 1/1 Running 6 5m3s
liveness-exec-pod 0/1 CrashLoopBackOff 6 5m45s

httpGet探针示例：

复制代码
[root@k8s-master ~]# cat /root/yas/liveness-httpget.yaml
apiVersion: v1
kind: Pod
metadata:
  name: liveness-httpget-pod
  namespace: default
spec:
  containers:
  - image: ikubernetes/myapp:v1
    name: liveness-httpget-container
    imagePullPolicy: IfNotPresent
    ports:
    - name: http
      containerPort: 80
    livenessProbe:
      httpGet:
        port: http
        path: /index.html
      initialDelaySeconds: 1
      periodSeconds: 3
[root@k8s-master ~]# kubectl create -f /root/yas/liveness-httpget.yaml
复制代码

此时验证容器时正常运行的，下面故意将路径修改为不存在路径，如下
调整path: /index/html，该路径不存在，容器无法启动。

====》以上是liveness探针，下面引入Readiness探针，Pod对象启动后，容器应用通常需要一段时间才能完成其初始化过程，例如加载配置或数据，甚至有些程序需要运行某类的预热过程，若在此阶段完成之前接入客户端的请求，势必会因为等待太久而影响用户体验，这时就需要就绪探针。 如果没有将就绪探针添加到pod中，它们几乎会立即成为服务端点，导致服务不可用。
就绪性探测，跟service关联。
新增的对应标签pod启动成功，但是pod内的war代码没有完成启动就绪，项目访问访问出现404.（必须要做）

就行探测，删除文件后容器未退出，只显示未就绪状态。

复制代码
[root@k8s-master ~]# cat yas/readiness-httpget.yaml
apiVersion: v1
kind: Pod
metadata:
  name: readiness-httpget-pod
  namespace: default
spec:
  containers:
  - image: ikubernetes/myapp:v1
    name: readiness-httpget-container
    imagePullPolicy: IfNotPresent
    ports:
    - name: http
      containerPort: 80
    readinessProbe:
      httpGet:
        port: http
        path: /index.html
      initialDelaySeconds: 1
      periodSeconds: 3
复制代码

[root@k8s-master ~]# kubectl create -f yas/readiness-httpget.yaml
[root@k8s-master ~]# kubectl exec -it readiness-httpget-pod -- /bin/sh
/ # ps -ef
PID USER TIME COMMAND
1 root 0:00 nginx: master process nginx -g daemon off;
5 nginx 0:00 nginx: worker process
6 root 0:00 /bin/sh
10 root 0:00 ps -ef
/ # rm -f /usr/share/nginx/html/index.html
/ # echo "hi">/usr/share/nginx/html/index.html
/ # exit
[root@k8s-master ~]# kubectl get pod readiness-httpget-pod -w
NAME READY STATUS RESTARTS AGE
readiness-httpget-pod 0/1 Running 0 7m16s

生命周期
$ kubectl explain pods.spec.containers.lifecycle.postStart
$ kubectl explain pod.spec.containers.lifecycle.preStop
命令优先级，先是执行pod command，在执行容器生命周期command。

复制代码
[root@k8s-master ~]# cat yas/pod-start.yaml
apiVersion: v1
kind: Pod
metadata:
  name: pod-start
  namespace: default
spec:
  containers:
  - image: ikubernetes/myapp:v1
    name: pod-start-container
    imagePullPolicy: IfNotPresent
    lifecycle:
      postStart:
        exec:
          command: ['mkdir','-p','/data/web/html']
    command: ['/bin/sh','-c','sleep 360']
[root@k8s-master ~]# kubectl create -f yas/pod-start.yaml
pod/pod-start created
[root@k8s-master ~]# kubectl get pod pod-start 
NAME        READY   STATUS    RESTARTS   AGE
pod-start   1/1     Running   0          14s
复制代码

以上问探测生命周期脚本，下面是查看。
[root@k8s-master ~]# kubectl exec -it pod-start -- /bin/sh
/ # ls /data/web/html/
/ # ps -ef
PID USER TIME COMMAND
1 root 0:00 /bin/sh -c sleep 360
5 root 0:00 sleep 360
9 root 0:00 /bin/sh
16 root 0:00 ps -ef
/ # exit

posted @   wang_wei123  阅读(196)  评论(0)  编辑  收藏  举报
编辑推荐：
· AI与.NET技术实操系列：基于图像分类模型对图像进行分类
· go语言实现终端里的倒计时
· 如何编写易于单元测试的代码
· 10年+ .NET Coder 心语，封装的思维：从隐藏、稳定开始理解其本质意义
· .NET Core 中如何实现缓存的预热？
阅读排行：
· 25岁的心里话
· 闲置电脑爆改个人服务器（超详细） #公网映射 #Vmware虚拟网络编辑器
· 零经验选手，Compose 一天开发一款小游戏！
· 通过 API 将Deepseek响应流式内容输出到前端
· AI Agent开发，如何调用三方的API Function，是通过提示词来发起调用的吗
点击右上角即可分享
微信分享提示