Kubernetes进阶（一）

文章目录

一、Kubernetes实现蓝绿部署

1.1 什么是蓝绿部署

蓝绿部署，英文名Blue Green Deployment，是一种可以保证系统在不间断提供服务的情况下上线的部署方式。

蓝绿部署的模型中包含两个集群，就好比海豚的左脑和右脑。

在没有上线的正常情况下，集群A和集群B的代码版本是一致的，并且同时对外提供服务。
在系统升级的时候下，我们首先把一个集群（比如集群A）从负载列表中摘除，进行新版本的部署。集群B仍然继续提供服务。

当集群A升级完毕，我们把负载均衡重新指向集群A，再把集群B从负载列表中摘除，进行新版本的部署。集群A重新提供服务。

1.2 滚动更新

// 创建模块
[root@master ~]# mkdir apache

// 编写 dockerfile文件
[root@master ~]# cat apache/Dockerfile 
FROM busybox

RUN mkdir /data && \
echo "hello amu,This is test1" > /data/index.html

ENTRYPOINT ["/bin/httpd","-f","-h","/data"]

#v1版本
// 创建两个不同版本的镜像
[root@master ~]# docker build -t gaofan1225/httpd:v1 apache/
Sending build context to Docker daemon  2.048kB
Step 1/3 : FROM busybox
latest: Pulling from library/busybox
3cb635b06aa2: Pull complete 
Digest: sha256:b5cfd4befc119a590ca1a81d6bb0fa1fb19f1fbebd0397f25fae164abe1e8a6a
Status: Downloaded newer image for busybox:latest
 ---> ffe9d497c324
Step 2/3 : RUN mkdir /data &&     echo "hello amu,This is test1" > /data/index.html
 ---> Running in 2400a9edc3b7
Removing intermediate container 2400a9edc3b7
 ---> b596ff036149
Step 3/3 : ENTRYPOINT ["/bin/httpd","-f","-h","/data"]
 ---> Running in 9bad15d04393
Removing intermediate container 9bad15d04393
 ---> 21d6bd11fd0d
Successfully built 21d6bd11fd0d
Successfully tagged gaofan1225/httpd:v1

#v2版本
// 创建之前修改dockerfile这一行：echo "hello amu,This is test2" > /data/index.html
[root@master ~]# cat apache/Dockerfile 
FROM busybox

RUN mkdir /data && \
echo "hello amu,This is test2" > /data/index.html	// 修改这里

ENTRYPOINT ["/bin/httpd","-f","-h","/data"]

[root@master ~]# docker build -t gaofan1225/httpd:v2 apache/ 
Sending build context to Docker daemon  2.048kB
Step 1/3 : FROM busybox
 ---> ffe9d497c324
Step 2/3 : RUN mkdir /data &&     echo "hello amu,This is test2" > /data/index.html
 ---> Using cache
 ---> b596ff036149
Step 3/3 : ENTRYPOINT ["/bin/httpd","-f","-h","/data"]
 ---> Using cache
 ---> 21d6bd11fd0d
Successfully built 21d6bd11fd0d
Successfully tagged gaofan1225/httpd:v2

// 登录 docker.hub仓库
[root@master ~]# docker login
Login with your Docker ID to push and pull images from Docker Hub. If you don't have a Docker ID, head over to https://hub.docker.com to create one.
Username: gaofan1225
Password: 
WARNING! Your password will be stored unencrypted in /root/.docker/config.json.
Configure a credential helper to remove this warning. See
https://docs.docker.com/engine/reference/commandline/login/#credentials-store

Login Succeeded

// 上传镜像当 docker.hub仓库 
[root@master ~]# docker push gaofan1225/httpd:v1
The push refers to repository [docker.io/gaofan1225/httpd]
88be387e4d70: Pushed 
64cac9eaf0da: Mounted from library/busybox 
v1: digest: sha256:074bdf0f11e9ee6f4efb2c12f67fdfab4eae8472fae1a75eb5004f3e5884ad3e size: 734

[root@master ~]# docker push gaofan1225/httpd:v2
The push refers to repository [docker.io/gaofan1225/httpd]
88be387e4d70: Pushed 
64cac9eaf0da: Layer already exists 
v2: digest: sha256:074bdf0f11e9ee6f4efb2c12f67fdfab4eae8472fae1a75eb5004f3e5884ad3e size: 734

// 创建一个httpd容器，使用v1版本的，并复制成三个
[root@master ~]# kubectl create deployment httpd --image gaofan1225/httpd:v1 --replicas 3
deployment.apps/httpd created

[root@master ~]# kubectl get pods
NAME                     READY   STATUS    RESTARTS   AGE
httpd-8556845dfd-gct9q   1/1     Running   0          58s
httpd-8556845dfd-hcqrk   1/1     Running   0          58s
httpd-8556845dfd-zv9qz   1/1     Running   0          58s

// 暴露端口
[root@master ~]# kubectl expose deployment httpd --port 80 --target-port 80
service/httpd exposed
[root@master ~]# kubectl get svc
NAME         TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)   AGE
httpd        ClusterIP   10.96.192.41   <none>        80/TCP    6s
kubernetes   ClusterIP   10.96.0.1      <none>        443/TCP   3m35s

// 访问测试
[root@master ~]# curl 10.96.192.41
hello amu,This is test1


// 做一个死循环
[root@master ~]# while :;do curl 10.96.192.41;done
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1				// v1版本这里会一直滚动
......以下省略

// 升级版本
[root@master ~]# kubectl set image deployment/httpd httpd=gaofan1225/httpd:v2
deployment.apps/httpd image updated

// 正在升级中
[root@master ~]# kubectl get pods
NAME                     READY   STATUS        RESTARTS   AGE
httpd-59b875c9dc-kkbdh   1/1     Running       0          4s
httpd-59b875c9dc-lhpjz   1/1     Running       0          6s
httpd-59b875c9dc-s7gmz   1/1     Running       0          8s
httpd-8556845dfd-fb887   1/1     Terminating   0          5m9s
httpd-8556845dfd-fhvlx   1/1     Terminating   0          5m6s
httpd-8556845dfd-l4rww   1/1     Terminating   0          5m7s


[root@master ~]# while :;do curl 10.96.192.41;done
hello amu,This is test1
hello amu,This is test2
hello amu,This is test1
hello amu,This is test2
hello amu,This is test1
hello amu,This is test2
hello amu,This is test1
hello amu,This is test2
......以下省略				// 升级过程中，两个版本的服务会同时滚动

// 升级结束
[root@master ~]# kubectl get pods
NAME                     READY   STATUS    RESTARTS   AGE
httpd-59b875c9dc-kkbdh   1/1     Running   0          71s
httpd-59b875c9dc-lhpjz   1/1     Running   0          73s
httpd-59b875c9dc-s7gmz   1/1     Running   0          75s


// 升级完成再去看死循环
[root@master ~]# while :;do curl 10.96.192.41;done
hello amu,This is test2
hello amu,This is test2
hello amu,This is test2
hello amu,This is test2
hello amu,This is test2
hello amu,This is test2
hello amu,This is test2
hello amu,This is test2
......以下省略			// 升级完成，v2版本一直滚动

1.2 回滚

// 回滚，返回上个版本
[root@master ~]# kubectl rollout undo deployment/httpd
deployment.apps/nginx rolled back

// 开始回滚
[root@master ~]# kubectl get pods
NAME                     READY   STATUS              RESTARTS   AGE
httpd-59b875c9dc-kkbdh   1/1     Terminating         0          3m
httpd-59b875c9dc-lhpjz   1/1     Terminating         0          3m2s
httpd-59b875c9dc-s7gmz   1/1     Running             0          3m4s
httpd-8556845dfd-nzkvg   1/1     Running             0          1s
httpd-8556845dfd-zqqz4   1/1     Running             0          3s
httpd-8556845dfd-zt4b2   0/1     ContainerCreating   0          0s

// 回滚成功
[root@master ~]# kubectl get pods
NAME                     READY   STATUS    RESTARTS   AGE
httpd-8556845dfd-nzkvg   1/1     Running   0          44s
httpd-8556845dfd-zqqz4   1/1     Running   0          46s
httpd-8556845dfd-zt4b2   1/1     Running   0          43s

// 回滚完成再去看死循环
[root@master ~]# while :;do curl 10.96.192.41;done
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1
hello amu,This is test1·		//  回滚成功，这里 v1 版本会一直滚动
......以下省略

二、使用yaml（yml）文件创建、删除容器

k8s是一个容器编排引擎，使用YAML文件编排要部署应用，因此在学习之前，应先了解YAML语法格式：

缩进表示层级关系
不支持制表符 ”tab“ 缩进，使用空格缩进
通常开头缩进 2 个空格
字符后缩进 1 个空格，如冒号、逗号等
“—” 表示YAML格式，一个文件的开始
“#” 注释

必写参数：

参数名字	意思
apiVersion	API版本
Kind	资源类型
metadata	资源元数据
spec	资源规格
replicas	副本数量
selector	标签选择器
template	Pod模板
metadata	Pod元数据
spec	Pod规格
containers	容器配置

当你不知道命令怎么编写的时候使用 explain查看注释
下面这个就是查看 metadata的注释

Pod容器的字段拼写

kubectl + explain + 类型 + 对象命令
kubectl + explain + 类型 + 对象命令 + 对象命令分类

// 查看 metadata使用方式
[root@master ~]# kubectl explain deployment.metadata

// 查看 metadata对象的namespace的使用方式
[root@master ~]# kubectl explain deployment.metadata.namespace

// 列表对象开头必须添加 "-" 
<[]Object>

可参考的一个yaml文件图：

实例中

[root@master ~]# cd /opt/

// 创建目录
[root@master opt]# mkdir manifes
[root@master opt]# cd manifest/


// 编写 dockerfile
# v1.1
[root@master manifest]# cat /root/apache/Dockerfile
FROM busybox

RUN mkdir /data && \
echo "Hi amu,This is v1" > /data/index.html  

ENTRYPOINT ["/bin/httpd","-f","-h","/data"]

[root@master manifest]# docker build -t gaofan1225/httpd:v1.1 /root/apache/


# v1.2
[root@master manifest]# cat /root/apache/Dockerfile
FROM busybox

RUN mkdir /data && \
echo "Hi amu,This is v2" > /data/index.html

ENTRYPOINT ["/bin/httpd","-f","-h","/data"]
[root@master manifest]# docker build -t gaofan1225/httpd:v1.2 /root/apache/

// 查看镜像
ENTRYPOINT ["/bin/httpd","-f","-h","/data"]
[root@master manifest]# docker images
REPOSITORY                                                        TAG       IMAGE ID       CREATED        SIZE
gaofan1225/httpd                                                  v1.2      c69e150e53dd   4 hours ago    1.24MB
gaofan1225/httpd                                                  v1.1      234a7ae0ec78   4 hours ago    1.24MB


// 编写 deployment类型的 yaml文件
[root@master manifest]# vi deploy.yml
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web
  namespace: default
spec:
  replicas: 3
  selector:			#选择
    matchLabels:	#匹配标签
      app: amu
  template:			#模板
    metadata:		#元数据
      labels:		#标签
        app: amu
    spec:
      containers:
      - image: gaofan1225/httpd:v1.1		#容器对象
        imagePullPolicy: IfNotPresent		#镜像拉取策略
        name: httpd

等同于：kubectl create deployment web --image=gaofan1225/httpd:v1.1 --replicas=3 -n default


[root@master manifest]# vi svc-deploy.yaml
apiversion: v1
kind: Service
metadata: 
  name: web
  namespace: default
spec:
  ports: 
  - port: 8001
    protocol: TCP							#协议
    targetPort: 80
  selector:
    app: amu
  type: NodePort							#类型

等同于：kubectl expose deployment web --port=8001 --target-port=80
 
// 查看pod、svc、deployment类型容器
[root@master manifest]# kubectl get pods
No resources found in default namespace. 
[root@master manifest]# kubectl get svc
NAME         TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   28m

[root@master ~]# kubectl get deployment
No resources found in default namespace.


// 创建容器
[root@master manifest]# kubectl create -f deploy.yaml 
deployment.apps/web created

[root@master manifest]# kubectl get pods
NAME                   READY   STATUS    RESTARTS   AGE
web-685c878c69-jkz62   1/1     Running   0          7s
web-685c878c69-nf968   1/1     Running   0          7s
web-685c878c69-vb7x6   1/1     Running   0          7s


[root@master ~]# kubectl get deploy
NAME   READY   UP-TO-DATE   AVAILABLE   AGE
web    3/3     3            3           79s

[root@master manifest]# kubectl get svc
NAME         TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)          AGE
kubernetes   ClusterIP   10.96.0.1       <none>        443/TCP          6h11m
web          NodePort    10.108.194.44   <none>        8001:30531/TCP   9s


// 访问测试
[root@master manifest]# curl 10.108.194.44:8001
Hi amu,This is v1

页面访问

访问：本机IP + 30531

$[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-BpHflaxb-1640080009887)(C:\Users\HP\AppData\Roaming\Typora\typora-user-images\image-20211221160119232.png)]$

// 版本升级
[root@master manifest]# vi deploy.yml
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web
  namespace: default
spec:
  replicas: 3
  selector:
    matchLabels:
      app: amu
  template:
    metadata:
      labels:
        app: amu
    spec:				
      containers:
      - image: gaofan1225/httpd:v1.2	// 修改版本，删除 imagePullPolicy
        name: httpd

[root@master manifest]# kubectl apply -f deploy.yaml 
Warning: resource deployments/web is missing the kubectlt-applied-configuration annotation which is required by ectl apply should only be used on resources created declr kubectl create --save-config or kubectl apply. The misll be patched automatically.
deployment.apps/web configured

[root@master manifest]# kubectl get svc
NAME         TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)          AGE
kubernetes   ClusterIP   10.96.0.1       <none>        443/TCP          6h46m
web          NodePort    10.108.194.44   <none>        8001:30531/TCP   35m

[root@master manifest]# curl 10.108.194.44:8001
Hi amu,This is v2

// 删除资源
[root@master manifest]# kubectl delete -f deploy.yaml 
deployment.apps "web" deleted

[root@master manifest]# kubectl delete -f svc-deploy.yaml 
service "web" deleted

[root@master manifest]# kubectl get pods
No resources found in default namespace.

// 文件还是存在的
[root@master manifest]# ls
deploy.yaml  svc-deploy.yaml

页面访问

$[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-73CK1V2b-1640080009888)(C:\Users\HP\AppData\Roaming\Typora\typora-user-images\image-20211221163557398.png)]$

注意：版本更新只需要更改文件里面的容器名字版本，升级也是一样，所以使用yaml文件部署容器，非常方便！！！

两个文件也可以合并写

[root@master manifest]# vi deploy.yml
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web
  namespace: default
spec:
  replicas: 3
  selector:			
    matchLabels:	
      app: amu
  template:			
    metadata:		
      labels:		
        app: amu
    spec:
      containers:
      - image: gaofan1225/httpd:v1.1		
        imagePullPolicy: IfNotPresent		
        name: httpd
        
---						// 中间使用 “---”隔开
apiversion: v1
kind: Service
metadata: 
  name: web
  namespace: default
spec:
  ports: 
  - port: 8001
    protocol: TCP							
    targetPort: 80
  selector:
    app: amu
  type: NodePort

三、自动创建 YAML文件

4.1 用create命令生成

kubectl create deployment nginx –image==nginx:1.16 -o yaml –dry-run=client > my-deploy.yaml

示例：

[root@master manifest]# kubectl create deployment nginx --image=nginx --dry-run=client -o yaml > nginx.yaml

[root@master manifest]# ls
deploy.yaml  nginx.yaml  svc-deploy.yaml

[root@master manifest]# cat nginx.yaml 
apiVersion: apps/v1
kind: Deployment
metadata:
  creationTimestamp: null
  labels:
    app: nginx
  name: nginx
spec:
  replicas: 1
  selector:
    matchLabels:
      app: nginx
  strategy: {}
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: nginx
    spec:
      containers:
      - image: nginx
        name: nginx
        resources: {}
status: {}

4.2 用get命令导出

kubectl get deployment nginx -o yaml > my-deploy.yaml

示例：

[root@master manifest]# kubectl create deployment nginx --image nginx --replicas 3
deployment.apps/nginx created
[root@master manifest]# kubectl get pods
NAME                     READY   STATUS              RESTARTS   AGE
nginx-85b98978db-8b9xj   0/1     ContainerCreating   0          11s
nginx-85b98978db-dds2s   0/1     ContainerCreating   0          11s
nginx-85b98978db-sg9qn   0/1     ContainerCreating   0          11s

[root@master manifest]# kubectl get deployment nginx -o yaml > nginx.yaml

# get获取的信息比create获取的信息更详细
[root@master manifest]# cat nginx.yaml 
apiVersion: apps/v1
kind: Deployment
metadata:
  annotations:
    deployment.kubernetes.io/revision: "1"
  creationTimestamp: "2021-12-21T09:16:57Z"
  generation: 1
  labels:
    app: nginx
  name: nginx
  namespace: default
  resourceVersion: "85342"
  uid: ea5653ea-9fd2-424e-9267-2b1112771835
spec:
  progressDeadlineSeconds: 600
  replicas: 3
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app: nginx
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: nginx
    spec:
      containers:
      - image: nginx
        imagePullPolicy: Always
        name: nginx
        resources: {}
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
      dnsPolicy: ClusterFirst
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      terminationGracePeriodSeconds: 30
status:
  conditions:
  - lastTransitionTime: "2021-12-21T09:16:57Z"
    lastUpdateTime: "2021-12-21T09:16:57Z"
    message: Deployment does not have minimum availability.
    reason: MinimumReplicasUnavailable
    status: "False"
    type: Available
  - lastTransitionTime: "2021-12-21T09:16:57Z"
    lastUpdateTime: "2021-12-21T09:16:57Z"
    message: ReplicaSet "nginx-85b98978db" is progressing.
    reason: ReplicaSetUpdated
    status: "True"
    type: Progressing
  observedGeneration: 1
  replicas: 3
  unavailableReplicas: 3
  updatedReplicas: 3

注意：不建议使用以上两种方法，还是建议手写，手写使用起来会更方便！

将你需要创建的资源描述到YAML文件中
部署：kubectl apply -f xxx.yaml
卸载：kubectl delete -f xxx.yaml

四、名称空间创建

// 没有pod就没有控制器
[root@master manifest]# kubectl get pods
No resources found in default namespace.

// 干跑查看格式
[root@master manifest]# kubectl create namespace runtime --dry-run=client
namespace/runtime created (dry run)
[root@master manifest]# kubectl create namespace runtime --dry-run=client -o yaml
apiVersion: v1
kind: Namespace
metadata:
  creationTimestamp: null
  name: runtime
spec: {}
status: {}

[root@master manifest]# cat test.yaml 
---
apiVersion: v1
kind: Namespace			// 名称空间
metadata:
  name: runtime

---
apiVersion: apps/v1
kind: Deployment
metadata: 
  name: web
  namespace: default
spec:
  replicas: 3
  selector:
    matchLabels:
      app: amu
  template:
    metadata: 
      labels:
        app: amu
    spec: 
      containers: 
      - image: gaofan1225/httpd:v1.1
        imagePullPolicy: IfNotPresent
        name: httpd

---
apiVersion: v1
kind: Service
metadata: 
  name: web
  namespace: default
spec:
  ports: 
  - port: 8001
    protocol: TCP
    targetPort: 80
  selector:
    app: amu
  type: NodePort

[root@master manifest]# kubectl create -f test.yaml 
namespace/runtime created
deployment.apps/web created
service/web created

[root@master manifest]# kubectl get pods
NAME                   READY   STATUS    RESTARTS   AGE
web-6f5fb9dd4d-5bmnl   1/1     Running   0          6s
web-6f5fb9dd4d-5rzg5   1/1     Running   0          6s
web-6f5fb9dd4d-vvcbf   1/1     Running   0          6s

// runtime名称空间创建成功
[root@master manifest]# kubectl get ns
NAME              STATUS   AGE
default           Active   3d5h
kube-node-lease   Active   3d5h
kube-public       Active   3d5h
kube-system       Active   3d5h
runtime           Active   14s

[root@master manifest]# kubectl get deployment
NAME   READY   UP-TO-DATE   AVAILABLE   AGE
web    3/3     3            3           5m55s

[root@master manifest]# kubectl get svc
NAME         TYPE        CLUSTER-IP    EXTERNAL-IP   PORT(S)          AGE
kubernetes   ClusterIP   10.96.0.1     <none>        443/TCP          7h53m
web          NodePort    10.111.5.34   <none>        8001:31860/TCP   9m59s

// 访问测试
[root@master manifest]# curl 10.111.5.34:8001
Hi amu,This is v1

// 删除名称空间
[root@master manifest]# kubectl delete -f test.yaml 
namespace "runtime" deleted
deployment.apps "web" deleted
service "web" deleted

[root@master manifest]# kubectl get ns
NAME              STATUS   AGE
default           Active   3d5h
kube-node-lease   Active   3d5h
kube-public       Active   3d5h
kube-system       Active   3d5h

总结：创建名称空间也可以靠yaml文件实现，不需要的时候直接删除yaml文件资源就可以删除，启用的时候直接指定yaml文件就行，非常方便！

五、健康检查

强大的自愈能力是 Kubernetes 这类容器编排引擎的一个重要特性。自愈的默认实现方式是自动重启发生故障的容器。除此之外，用户还可以利用 Liveness 和 Readiness 探测机制设置更精细的健康检查，进而实现如下需求：

零停机部署。

避免部署无效的镜像。

更加安全的滚动升级。

六、探针类型

Liveness探测：

Liveness 探测让用户可以自定义判断容器是否健康的条件。如果探测失败，Kubernetes 就会重启容器。

Readiness探测：

除了 Liveness 探测，Kubernetes Health Check 机制还包括 Readiness 探测。
用户通过 Liveness 探测可以告诉 Kubernetes 什么时候通过重启容器实现自愈；Readiness 探测则是告诉 Kubernetes 什么时候可以将容器加入到 Service 负载均衡池中，对外提供服务。

Liveness 和 Readiness的比较：

Liveness 探测和 Readiness 探测是两种 Health Check 机制，如果不特意配置，Kubernetes 将对两种探测采取相同的默认行为，即通过判断容器启动进程的返回值是否为零来判断探测是否成功。
两种探测的配置方法完全一样，支持的配置参数也一样。不同之处在于探测失败后的行为：Liveness 探测是重启容器；Readiness 探测则是将容器设置为不可用，不接收 Service 转发的请求。
Liveness 探测和 Readiness 探测是独立执行的，二者之间没有依赖，所以可以单独使用，也可以同时使用。用 Liveness 探测判断容器是否需要重启以实现自愈；用 Readiness 探测判断容器是否已经准备好对外提供服务。