pod控制器-deploy

deployment跟别的控制器不一样的是，deployment通过管理replicaset，来管理pod，deloyment比replicaset的功能更加强大，理所应当的给划归重点

deploy的功能有以下几个

支持replicaset的所有功能
支持发布的停止和继续
支持滚动更新和回滚更新

deploy的资源清单文件大概看一下以备不时之需

apiVersion: apps/v1 # 版本号
kind: Deployment # 类型       
metadata: # 元数据
  name: # rs名称 
  namespace: # 所属命名空间 
  labels: #标签
    controller: deploy
spec: # 详情描述
  replicas: 3 # 副本数量
  revisionHistoryLimit: 3 # 保留历史版本
  paused: false # 暂停部署，默认是false
  progressDeadlineSeconds: 600 # 部署超时时间（s），默认是600
  strategy: # 策略
    type: RollingUpdate # 滚动更新策略
    rollingUpdate: # 滚动更新
      maxSurge: 30% # 最大额外可以存在的副本数，可以为百分比，也可以为整数
      maxUnavailable: 30% # 最大不可用状态的 Pod 的最大值，可以为百分比，也可以为整数
  selector: # 选择器，通过它指定该控制器管理哪些pod
    matchLabels:      # Labels匹配规则
      app: nginx-pod
    matchExpressions: # Expressions匹配规则
      - {key: app, operator: In, values: [nginx-pod]}
  template: # 模板，当副本数量不足时，会根据下面的模板创建pod副本
    metadata:
      labels:
        app: nginx-pod
    spec:
      containers:
      - name: nginx
        image: nginx:1.17.1
        ports:
        - containerPort: 80

我们创建yaml看看

vi pc-deployment.yaml

apiVersion: apps/v1
kind: Deployment      
metadata:
  name: pc-deployment
  namespace: dev
spec: 
  replicas: 3
  selector:
    matchLabels:
      run: nginx-pod
  template:
    metadata:
      labels:
        run: nginx-pod
    spec:
      containers:
      - name: nginx
        image: nginx:1.17.1

上节的pod创建以后是挂起状态我刚刚解决了，重启治百病，我重启了一下系统，又启动了一下minikube现在终于是可以正常创建pod了

[qq@k8snode1 ~]$ kubectl get pod,deploy -n dev
NAME                                 READY   STATUS    RESTARTS   AGE
pod/nginx-64b7c758f4-c4mfv           1/1     Running   0          4m14s
pod/nginx-64b7c758f4-khpp5           1/1     Running   0          4m14s
pod/nginx-64b7c758f4-rdk7w           1/1     Running   0          4m14s
pod/pc-deployment-774fccff58-lf5q2   1/1     Running   0          2m12s
pod/pc-deployment-774fccff58-sm28t   1/1     Running   0          2m12s
pod/pc-deployment-774fccff58-xxh4v   1/1     Running   0          2m12s

NAME                            READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/nginx           3/3     3            3           4m14s
deployment.apps/pc-deployment   3/3     3            3           2m12s

我们就继续了乌拉乌拉

扩缩容

我之前说过了，deploy通过管理rs来管理pod
下面我们就可以看我创建的deploy创建的rs和pod了
[qq@k8snode1 ~]$ kubectl get pod,deploy,rs -n dev
NAME                                 READY   STATUS    RESTARTS   AGE
pod/pc-deployment-774fccff58-lf5q2   1/1     Running   0          10m
pod/pc-deployment-774fccff58-sm28t   1/1     Running   0          10m
pod/pc-deployment-774fccff58-xxh4v   1/1     Running   0          10m

NAME                            READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/pc-deployment   3/3     3            3           10m

NAME                                       DESIRED   CURRENT   READY   AGE
replicaset.apps/pc-deployment-774fccff58   3         3         3       10m

教材上说的用命令手动添加pod
[root@k8s-master01 ~]# kubectl scale deploy pc-deployment --replicas=5  -n dev
deployment.apps/pc-deployment scaled


为啥啊，我直接改文件不香吗，说实话我可能写出来有点困难，看懂还是比较简单的吧
[qq@k8snode1 ~]$ kubectl edit deploy pc-deployment -n dev
deployment.apps/pc-deployment edited
------  replicas: 6  副本数量改了个6，有啥难的
[qq@k8snode1 ~]$ kubectl get pod,deploy,rs -n dev
# 可以看到加了3个pod，其他的基本没变
NAME                                 READY   STATUS    RESTARTS   AGE
pod/pc-deployment-774fccff58-bb7zx   1/1     Running   0          4m34s
pod/pc-deployment-774fccff58-j2m6w   1/1     Running   0          4m34s
pod/pc-deployment-774fccff58-lf5q2   1/1     Running   0          22m
pod/pc-deployment-774fccff58-sm28t   1/1     Running   0          22m
pod/pc-deployment-774fccff58-xxh4v   1/1     Running   0          22m
pod/pc-deployment-774fccff58-zfkkm   1/1     Running   0          4m34s

NAME                            READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/pc-deployment   6/6     6            6           22m

NAME                                       DESIRED   CURRENT   READY   AGE
replicaset.apps/pc-deployment-774fccff58   6         6         6       22m

镜像更新

[qq@k8snode1 ~]$ kubectl get pod -n dev -w
# 首先认识一下持续更新的命令，实乃神器啊，认识一下
NAME                             READY   STATUS    RESTARTS   AGE
pc-deployment-774fccff58-bb7zx   1/1     Running   0          9m48s
pc-deployment-774fccff58-j2m6w   1/1     Running   0          9m48s
pc-deployment-774fccff58-lf5q2   1/1     Running   0          27m
pc-deployment-774fccff58-sm28t   1/1     Running   0          27m
pc-deployment-774fccff58-xxh4v   1/1     Running   0          27m
pc-deployment-774fccff58-zfkkm   1/1     Running   0          9m48s

deployment支持两种更新策略:重建更新和滚动更新,可以通过strategy指定策略类型,支持两个属性:

strategy：指定新的Pod替换旧的Pod的策略，支持两个属性：
type：指定策略类型，支持两种策略
Recreate：在创建出新的Pod之前会先杀掉所有已存在的Pod
RollingUpdate：滚动更新，就是杀死一部分，就启动一部分，在更新过程中，存在两个版本Pod
rollingUpdate：当type为RollingUpdate时生效，用于为RollingUpdate设置参数，支持两个属性：
maxUnavailable：用来指定在升级过程中不可用Pod的最大数量，默认为25%。
maxSurge：用来指定在升级过程中可以超过期望的Pod的最大数量，默认为25%。

先看看重建更新

编辑pc-deployment.yaml,在spec节点下添加更新策略

spec:
  strategy: # 策略
    type: Recreate # 重建更新

删除掉deploy刚好，我们看一下删除的时候pod的状态，注意，此处开了两个终端

一个终端执行命令，一个终端查看状态
 kubectl get pod -n dev -w

[qq@k8snode1 ~]$ kubectl delete deploy pc-deployment -n dev
deployment.apps "pc-deployment" deleted
# 这边一个删除命令
另一边就有反应了
[qq@k8snode1 ~]$ kubectl get pod -n dev -w
NAME                             READY   STATUS    RESTARTS   AGE
pc-deployment-774fccff58-bb7zx   1/1     Running   0          9m48s
pc-deployment-774fccff58-j2m6w   1/1     Running   0          9m48s
pc-deployment-774fccff58-lf5q2   1/1     Running   0          27m
pc-deployment-774fccff58-sm28t   1/1     Running   0          27m
pc-deployment-774fccff58-xxh4v   1/1     Running   0          27m
pc-deployment-774fccff58-zfkkm   1/1     Running   0          9m48s
pc-deployment-774fccff58-sm28t   1/1     Terminating   0          32m
pc-deployment-774fccff58-xxh4v   1/1     Terminating   0          32m
pc-deployment-774fccff58-j2m6w   1/1     Terminating   0          14m
pc-deployment-774fccff58-zfkkm   1/1     Terminating   0          14m
pc-deployment-774fccff58-bb7zx   1/1     Terminating   0          14m
pc-deployment-774fccff58-lf5q2   1/1     Terminating   0          32m
pc-deployment-774fccff58-lf5q2   0/1     Terminating   0          32m
pc-deployment-774fccff58-lf5q2   0/1     Terminating   0          32m
pc-deployment-774fccff58-lf5q2   0/1     Terminating   0          32m
pc-deployment-774fccff58-j2m6w   0/1     Terminating   0          14m
pc-deployment-774fccff58-j2m6w   0/1     Terminating   0          14m
pc-deployment-774fccff58-j2m6w   0/1     Terminating   0          14m
pc-deployment-774fccff58-bb7zx   0/1     Terminating   0          14m
pc-deployment-774fccff58-bb7zx   0/1     Terminating   0          14m
pc-deployment-774fccff58-bb7zx   0/1     Terminating   0          14m
pc-deployment-774fccff58-xxh4v   0/1     Terminating   0          32m
pc-deployment-774fccff58-xxh4v   0/1     Terminating   0          32m
pc-deployment-774fccff58-xxh4v   0/1     Terminating   0          32m
pc-deployment-774fccff58-zfkkm   0/1     Terminating   0          14m
pc-deployment-774fccff58-zfkkm   0/1     Terminating   0          14m
pc-deployment-774fccff58-zfkkm   0/1     Terminating   0          14m
pc-deployment-774fccff58-sm28t   0/1     Terminating   0          32m
pc-deployment-774fccff58-sm28t   0/1     Terminating   0          32m
pc-deployment-774fccff58-sm28t   0/1     Terminating   0          32m
# 这个自己看一下就好

我们继续改文件

把这一段粘贴到这里来

  namespace: dev
spec:
  strategy: # 策略
    type: Recreate # 重建更新
  replicas: 3
  selector:

[qq@k8snode1 ~]$ kubectl apply -f  pc-deployment.yaml
deployment.apps/pc-deployment created
看看状态pod从挂起，到容器创建，到运行
pc-deployment-774fccff58-npx7s   0/1     Pending             0          0s
pc-deployment-774fccff58-npx7s   0/1     Pending             0          0s
pc-deployment-774fccff58-wgdpd   0/1     Pending             0          1s
pc-deployment-774fccff58-lnfr6   0/1     Pending             0          1s
pc-deployment-774fccff58-wgdpd   0/1     Pending             0          1s
pc-deployment-774fccff58-lnfr6   0/1     Pending             0          1s
pc-deployment-774fccff58-npx7s   0/1     ContainerCreating   0          1s
pc-deployment-774fccff58-wgdpd   0/1     ContainerCreating   0          1s
pc-deployment-774fccff58-lnfr6   0/1     ContainerCreating   0          1s
pc-deployment-774fccff58-npx7s   1/1     Running             0          4s
pc-deployment-774fccff58-wgdpd   1/1     Running             0          4s
pc-deployment-774fccff58-lnfr6   1/1     Running             0          4s

[qq@k8snode1 ~]$ kubectl edit deploy pc-deployment  -n dev
deployment.apps/pc-deployment edited
我把版本号改成了1.17.1

看状态
pc-deployment-774fccff58-lnfr6   1/1     Running             0          4s

pc-deployment-774fccff58-lnfr6   1/1     Terminating         0          4m44s
pc-deployment-774fccff58-npx7s   1/1     Terminating         0          4m44s
pc-deployment-774fccff58-wgdpd   1/1     Terminating         0          4m44s

pc-deployment-774fccff58-wgdpd   0/1     Terminating         0          4m45s
pc-deployment-774fccff58-wgdpd   0/1     Terminating         0          4m45s
pc-deployment-774fccff58-wgdpd   0/1     Terminating         0          4m45s

pc-deployment-774fccff58-lnfr6   0/1     Terminating         0          4m45s
pc-deployment-774fccff58-lnfr6   0/1     Terminating         0          4m46s
pc-deployment-774fccff58-lnfr6   0/1     Terminating         0          4m46s

pc-deployment-774fccff58-npx7s   0/1     Terminating         0          4m46s
pc-deployment-774fccff58-npx7s   0/1     Terminating         0          4m46s
pc-deployment-774fccff58-npx7s   0/1     Terminating         0          4m46s

pc-deployment-5d776b56ff-j4cq4   0/1     Pending             0          0s
pc-deployment-5d776b56ff-j4cq4   0/1     Pending             0          0s

pc-deployment-5d776b56ff-47qrc   0/1     Pending             0          0s

pc-deployment-5d776b56ff-mgqdk   0/1     Pending             0          0s
pc-deployment-5d776b56ff-mgqdk   0/1     Pending             0          0s

pc-deployment-5d776b56ff-47qrc   0/1     Pending             0          0s

pc-deployment-5d776b56ff-j4cq4   0/1     ContainerCreating   0          1s
pc-deployment-5d776b56ff-47qrc   0/1     ContainerCreating   0          1s
pc-deployment-5d776b56ff-mgqdk   0/1     ContainerCreating   0          2s

pc-deployment-5d776b56ff-j4cq4   1/1     Running             0          40s
pc-deployment-5d776b56ff-47qrc   1/1     Running             0          42s
pc-deployment-5d776b56ff-mgqdk   1/1     Running             0          45s


再查一下你会发现，之前的rs还在，这是为回滚准备的
[qq@k8snode1 ~]$ kubectl get pod,deploy,rs -n dev
NAME                                 READY   STATUS    RESTARTS   AGE
pod/pc-deployment-5d776b56ff-47qrc   1/1     Running   0          13m
pod/pc-deployment-5d776b56ff-j4cq4   1/1     Running   0          13m
pod/pc-deployment-5d776b56ff-mgqdk   1/1     Running   0          13m

NAME                            READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/pc-deployment   3/3     3            3           18m

NAME                                       DESIRED   CURRENT   READY   AGE
replicaset.apps/pc-deployment-5d776b56ff   3         3         3       13m
replicaset.apps/pc-deployment-774fccff58   0         0         0       18m

滚动更新

改文件

spec:
  strategy: # 策略
    type: RollingUpdate # 滚动更新策略
    rollingUpdate:
      maxSurge: 25% 
      maxUnavailable: 25%

[qq@k8snode1 ~]$ kubectl edit deploy pc-deployment -n dev
deployment.apps/pc-deployment edited



pc-deployment-5d776b56ff-mgqdk   1/1     Running             0          45s
pc-deployment-785c7c59f6-fv769   0/1     Pending             0          0s
pc-deployment-785c7c59f6-fv769   0/1     Pending             0          0s
pc-deployment-785c7c59f6-fv769   0/1     ContainerCreating   0          0s
pc-deployment-785c7c59f6-fv769   1/1     Running             0          32s
pc-deployment-785c7c59f6-j69pm   0/1     Pending             0          0s
pc-deployment-5d776b56ff-47qrc   1/1     Terminating         0          19m
pc-deployment-785c7c59f6-j69pm   0/1     Pending             0          0s
pc-deployment-785c7c59f6-j69pm   0/1     ContainerCreating   0          0s
pc-deployment-5d776b56ff-47qrc   0/1     Terminating         0          19m
pc-deployment-5d776b56ff-47qrc   0/1     Terminating         0          19m
pc-deployment-5d776b56ff-47qrc   0/1     Terminating         0          19m
pc-deployment-785c7c59f6-j69pm   1/1     Running             0          2s
pc-deployment-5d776b56ff-mgqdk   1/1     Terminating         0          19m
pc-deployment-785c7c59f6-85n4f   0/1     Pending             0          1s
pc-deployment-785c7c59f6-85n4f   0/1     Pending             0          1s
pc-deployment-785c7c59f6-85n4f   0/1     ContainerCreating   0          1s
pc-deployment-785c7c59f6-85n4f   1/1     Running             0          1s
pc-deployment-5d776b56ff-mgqdk   0/1     Terminating         0          19m
pc-deployment-5d776b56ff-j4cq4   1/1     Terminating         0          19m
pc-deployment-5d776b56ff-mgqdk   0/1     Terminating         0          19m
pc-deployment-5d776b56ff-mgqdk   0/1     Terminating         0          19m
pc-deployment-5d776b56ff-j4cq4   0/1     Terminating         0          19m
pc-deployment-5d776b56ff-j4cq4   0/1     Terminating         0          19m
pc-deployment-5d776b56ff-j4cq4   0/1     Terminating         0          19m

状态这边的意思看着有点多，也得分方法看
例
pc-deployment-5d776b56ff-mgqdk   1/1     Running             0          45s
|      deploy名  |     rs编号     |    pod随机名
虽乱，但是还能分析出来的，刚开始pod在5d的rs里
deploy创建了78rs，在78rs里创建了fv769pod
然后干掉了一台5drs里的pod，以此模式滚动更新，慢慢所以pod都被更新到了78rs里

版本回退

deployment支持版本升级过程中的暂停、继续功能以及版本回退等诸多功能，下面具体来看.

kubectl rollout：版本升级相关功能，支持下面的选项：

status 显示当前升级状态
history 显示升级历史记录
pause 暂停版本升级过程
resume 继续已经暂停的版本升级过程
restart 重启版本升级过程
undo 回滚到上一级版本（可以使用–to-revision回滚到指定版本）

[qq@k8snode1 ~]$ kubectl rollout status deploy pc-deployment -n dev
deployment "pc-deployment" successfully rolled out
查看当前的版本更新状态
[qq@k8snode1 ~]$ kubectl rollout history deploy pc-deployment -n dev
deployment.apps/pc-deployment
REVISION  CHANGE-CAUSE
1         <none>
2         <none>
3         <none>
有三次版本更新记录，说明完成过两次升级

版本回退
我们先回滚版本1看看
[qq@k8snode1 ~]$ kubectl rollout undo deploy  pc-deployment --to-revision=1 -n dev
deployment.apps/pc-deployment rolled back

[qq@k8snode1 ~]$ kubectl get deploy -n dev -o wide
NAME            READY   UP-TO-DATE   AVAILABLE   AGE     CONTAINERS   IMAGES         SELECTOR
pc-deployment   3/3     3            3           3h20m   nginx        nginx:1.17.1   run=nginx-pod

[qq@k8snode1 ~]$ kubectl get rs -n dev -o wide
NAME                       DESIRED   CURRENT   READY   AGE     CONTAINERS   IMAGES         SELECTOR
pc-deployment-5d776b56ff   0         0         0       3h19m   nginx        nginx:1.17.2   pod-template-hash=5d776b56ff,run=nginx-pod
pc-deployment-774fccff58   3         3         3       3h23m   nginx        nginx:1.17.1   pod-template-hash=774fccff58,run=nginx-pod
pc-deployment-785c7c59f6   0         0         0       3h      nginx        nginx:1.17.3   pod-template-hash=785c7c59f6,run=nginx-pod

# 查看rs，发现第一个rs中有3个pod运行，后面两个版本的rs中pod未运行
# 其实deployment之所以可以实现版本的回滚，就是通过记录下历史rs来实现的，
# 一旦想回滚到哪个版本，只需要将当前版本pod数量降为0，然后将回滚版本的pod提升为目标数量就可以了

金丝雀发布

Deployment控制器支持控制更新过程中的控制，如“暂停(pause)”或“继续(resume)”更新操作。

比如有一批新的Pod资源创建完成后立即暂停更新过程，此时，仅存在一部分新版本的应用，主体部分还是旧的版本。然后，再筛选一小部分的用户请求路由到新版本的Pod应用，继续观察能否稳定地按期望的方式运行。确定没问题之后再继续完成余下的Pod资源滚动更新，否则立即回滚更新操作。这就是所谓的金丝雀发布。

更新deployment的版本，并配置暂停deploymen
[qq@k8snode1 ~]$ kubectl set image deploy pc-deployment nginx=nginx:1.17.4 -n dev && kubectl rollout pause deployment pc-deployment  -n dev
deployment.apps/pc-deployment image updated
deployment.apps/pc-deployment paused

稍等一下，看一下状态，发现又创建了一个rs里面运行着一个容器，同时原版本没有减少
[qq@k8snode1 ~]$ kubectl get rs -n dev
NAME                       DESIRED   CURRENT   READY   AGE
pc-deployment-5d776b56ff   0         0         0       3h27m
pc-deployment-6d6bc4bd5    1         1         1       2m7s
pc-deployment-774fccff58   3         3         3       3h32m
pc-deployment-785c7c59f6   0         0         0       3h9m
在看一下更新状态，可以看到更新了1个，更新未完成
[qq@k8snode1 ~]$ kubectl rollout status deploy pc-deployment -n dev
Waiting for deployment "pc-deployment" rollout to finish: 1 out of 3 new replicas have been updated...

这种情况下在生产环境可以分配一小部分用户去访问新的版本，如果新版本没问题了，就可以完全更新了
[qq@k8snode1 ~]$ kubectl rollout resume  deploy pc-deployment -n dev
deployment.apps/pc-deployment resumed
原版本的3台机没了，全部更换为了新版本
[qq@k8snode1 ~]$ kubectl get rs -n dev
NAME                       DESIRED   CURRENT   READY   AGE
pc-deployment-5d776b56ff   0         0         0       3h40m
pc-deployment-6d6bc4bd5    3         3         3       14m
pc-deployment-774fccff58   0         0         0       3h45m
pc-deployment-785c7c59f6   0         0         0       3h22m
看一下更新状态，也结束了
[qq@k8snode1 ~]$ kubectl rollout status deploy pc-deployment -n dev
deployment "pc-deployment" successfully rolled out

结束实验删除环境

rs，deploy，pod都会被删除
[qq@k8snode1 ~]$ kubectl delete -f pc-deployment.yaml
deployment.apps "pc-deployment" deleted

相关推荐

在.net core中使用nginx做负载均衡

React 18 超全升级指南

LTUI v1.7 发布，一个基于 Lua 的跨平台字符终端 UI 界面库

生成对抗网络(GAN)的半监督学习

《若依ruoyi》第二十五章:Spring boot 上传下载封装详解二

QT进阶之路 : 布局详解

如何在 Element UI 中使用栅格布局实现响应式设计?

IDEA 中 Jetty 的配置操作手册 idea jfinal

如何在Dify平台上创建智能Agent:一步步教你实现超级智能体搭建

k8s自动化运维三