当前位置：首页 > article >正文

K8S中Pod控制器之Deployment(Deploy)控制器

article 2025/1/20 2:40:35

Deployment(Deploy)控制器

从Kubernetes 1.2版本开始，引入了Deployment控制器，它提供了一种更高级别的抽象来管理应用的生命周期，包括应用的声明式更新、回滚、暂停和恢复等。Deployment的设计目标是简化应用的部署和扩展，同时提供强大的应用更新和版本管理功能。
这种控制器并不直接管理pod，而是通过管理ReplicaSet来间接管理Pod，即：Deployment管理ReplicaSet，ReplicaSet管理Pod。所以Deployment比ReplicaSet功能更加强大。

Deployment主要功能

ReplicaSet管理：Deployment控制器会创建一个或多个ReplicaSet对象来管理Pod的副本，确保指定数量的Pod副本始终运行。
声明式更新：用户可以声明他们希望Pod如何更新，Deployment会自动将当前状态变更为用户所期望的状态。
滚动更新：Deployment支持滚动更新（Rolling Update），这意味着在更新过程中，旧版本的Pod会逐个被新版本的Pod替换，直到所有Pod都更新到新版本。
版本控制：Deployment记录了Pod模板的历史版本，允许用户回滚到之前的版本，如果新版本存在问题。
暂停和恢复：用户可以暂停Deployment的更新，进行维护或修改，之后可以恢复更新。
扩展和缩减：用户可以根据需要对Deployment控制的Pod数量进行扩展或缩减。
健康检查：Deployment可以使用liveness和readiness探针来确定Pod的健康状态，并在必要时进行适当的处理。
策略配置：用户可以为Deployment配置更新策略，包括更新的速度和模式。
历史记录：Deployment控制器会保留Pod模板的变更历史，通常保留最近的几次更新，以便于回滚操作。
状态检查：提供了观察Deployment状态的能力，包括它的更新进度和条件。

创建Deployment

# vim pc-deployment.yaml
[root@k8s-master ~]# cat  pc-deployment.yaml
---
apiVersion: apps/v1 # 版本号
kind: Deployment # 类型
metadata: # 元数据
  name: pc-deployment # rs名称
  namespace: test # 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
[root@k8s-master ~]# kubectl apply -f pc-deployment.yaml 
deployment.apps/pc-deployment created

查看deploy控制器

[root@k8s-master ~]# kubectl get deploy -n test
NAME            READY   UP-TO-DATE   AVAILABLE   AGE
pc-deployment   0/3     3            0           8s
[root@k8s-master ~]# kubectl get deploy -n test -w
NAME            READY   UP-TO-DATE   AVAILABLE   AGE
pc-deployment   0/3     3            0           11s
pc-deployment   1/3     3            1           34s
pc-deployment   2/3     3            2           34s
pc-deployment   3/3     3            3           34s

查看rs控制器

[root@k8s-master ~]# kubectl get rs -n test
NAME                       DESIRED   CURRENT   READY   AGE
pc-deployment-5ffc5bf56c   3         3         3       88s

查看pod

[root@k8s-master ~]# kubectl get pods -n test
NAME                             READY   STATUS    RESTARTS   AGE
pc-deployment-5ffc5bf56c-hj285   1/1     Running   0          80s
pc-deployment-5ffc5bf56c-l5ccs   1/1     Running   0          80s
pc-deployment-5ffc5bf56c-pmg8t   1/1     Running   0          80s

Deployment扩缩容

命令方式

[root@k8s-master ~]# kubectl scale deploy pc-deployment --replicas=5 -n test
deployment.apps/pc-deployment scaled
[root@k8s-master ~]# kubectl get pods -n test
NAME                             READY   STATUS    RESTARTS   AGE
pc-deployment-5ffc5bf56c-hj285   1/1     Running   0          8m7s
pc-deployment-5ffc5bf56c-l5ccs   1/1     Running   0          8m7s
pc-deployment-5ffc5bf56c-pmg8t   1/1     Running   0          8m7s
pc-deployment-5ffc5bf56c-tmmhz   1/1     Running   0          5s
pc-deployment-5ffc5bf56c-tqt7j   1/1     Running   0          5s

编辑副本

#将副本数改为3
[root@k8s-master ~]# kubectl edit deployment.apps pc-deployment -n test
deployment.apps/pc-deployment edited

[root@k8s-master ~]# kubectl get pods -n test  -w
NAME                             READY   STATUS    RESTARTS   AGE
pc-deployment-5ffc5bf56c-hj285   1/1     Running   0          10m
pc-deployment-5ffc5bf56c-pmg8t   1/1     Running   0          10m
pc-deployment-5ffc5bf56c-tmmhz   1/1     Running   0          2m13s

Deployment镜像更新

Deployment控制器确实支持两种主要的更新策略：重建更新（Recreate）和滚动更新（RollingUpdate）。这两种策略定义了如何替换Deployment中的Pods以进行更新。可以通过strategy字段指定更新策略类型，并设置相关的属性。以下是两种策略的简要说明：

重建更新（Recreate）：
- 在这种策略下，Deployment控制器会先终止所有正在运行的Pods，然后再创建新的Pods。
- 这种方法会导致短暂的停机时间，因为旧的Pods全部停止后才进行新的Pods的创建。
- 这种策略适用于小型应用或对停机时间不敏感的场景。
- 属性设置示例：
```
spec:
  strategy:   # 策略
    type: Recreate  # 重建更新
```
滚动更新（RollingUpdate）：
- 这是默认的更新策略，它允许你指定两个属性：maxUnavailable和maxSurge。
- maxUnavailable定义了在更新过程中可以处于不可用状态的最大Pod数量。这可以设置为绝对数量或是Pod总数的百分比。
- maxSurge定义了可以超出期望数量创建的新Pod的最大数量。这同样可以设置为绝对数量或是Pod总数的百分比。
- 滚动更新策略可以确保在更新过程中始终有一定数量的Pod在线，从而减少或避免停机时间。
- 属性设置示例：
```
strategy:
  type: RollingUpdate
  rollingUpdate:
    maxUnavailable: 1   #在更新过程中，可以有一个Pod处于不可用状态
    maxSurge: 1     #可以有一个额外的Pod在更新完成前被创建
```

重建更新（Recreate）

[root@k8s-master ~]# vim pc-deployment.yaml 
[root@k8s-master ~]# kubectl delete -f pc-deployment.yaml 
deployment.apps "pc-deployment" deleted
[root@k8s-master ~]# kubectl apply -f pc-deployment.yaml 
deployment.apps/pc-deployment created
[root@k8s-master ~]# kubectl get pods -n test  -w
NAME                             READY   STATUS    RESTARTS   AGE
pc-deployment-5ffc5bf56c-cgk2b   1/1     Running   0          5s
pc-deployment-5ffc5bf56c-gvqgr   1/1     Running   0          5s
pc-deployment-5ffc5bf56c-hp7mc   1/1     Running   0          4s
^C[root@k8s-master ~]# kubectl set image deploy pc-deployment nginx=nginx:1.17.4 -n ttdeployment.apps/pc-deployment image updated
[root@k8s-master ~]# kubectl get pods -n test  -w
NAME                             READY   STATUS              RESTARTS   AGE
pc-deployment-5ffc5bf56c-cgk2b   1/1     Running             0          60s
pc-deployment-5ffc5bf56c-gvqgr   1/1     Running             0          60s
pc-deployment-5ffc5bf56c-hp7mc   1/1     Running             0          59s
pc-deployment-d79885fd8-pfxzf    0/1     ContainerCreating   0          2s
pc-deployment-d79885fd8-pfxzf    0/1     ContainerCreating   0          2s
pc-deployment-d79885fd8-pfxzf    1/1     Running             0          37s
pc-deployment-5ffc5bf56c-hp7mc   1/1     Terminating         0          94s
pc-deployment-d79885fd8-2n875    0/1     Pending             0          0s
pc-deployment-d79885fd8-2n875    0/1     Pending             0          0s
pc-deployment-d79885fd8-2n875    0/1     ContainerCreating   0          0s
pc-deployment-5ffc5bf56c-hp7mc   1/1     Terminating         0          95s
pc-deployment-5ffc5bf56c-hp7mc   0/1     Terminating         0          95s
pc-deployment-d79885fd8-2n875    0/1     ContainerCreating   0          2s
pc-deployment-5ffc5bf56c-hp7mc   0/1     Terminating         0          101s
pc-deployment-5ffc5bf56c-hp7mc   0/1     Terminating         0          101s
pc-deployment-d79885fd8-2n875    1/1     Running             0          30s
pc-deployment-5ffc5bf56c-cgk2b   1/1     Terminating         0          2m5s
pc-deployment-d79885fd8-lsjff    0/1     Pending             0          0s
pc-deployment-d79885fd8-lsjff    0/1     Pending             0          0s
pc-deployment-d79885fd8-lsjff    0/1     ContainerCreating   0          0s
pc-deployment-5ffc5bf56c-cgk2b   1/1     Terminating         0          2m6s
pc-deployment-5ffc5bf56c-cgk2b   0/1     Terminating         0          2m6s
pc-deployment-d79885fd8-lsjff    0/1     ContainerCreating   0          2s
pc-deployment-d79885fd8-lsjff    1/1     Running             0          2s
pc-deployment-5ffc5bf56c-gvqgr   1/1     Terminating         0          2m7s
pc-deployment-5ffc5bf56c-gvqgr   1/1     Terminating         0          2m8s
pc-deployment-5ffc5bf56c-gvqgr   0/1     Terminating         0          2m10s
pc-deployment-5ffc5bf56c-gvqgr   0/1     Terminating         0          2m11s
pc-deployment-5ffc5bf56c-gvqgr   0/1     Terminating         0          2m11s
pc-deployment-5ffc5bf56c-cgk2b   0/1     Terminating         0          2m19s
pc-deployment-5ffc5bf56c-cgk2b   0/1     Terminating         0          2m19s
^C[root@k8s-master ~]# kubectl get pods -n test 
NAME                            READY   STATUS    RESTARTS   AGE
pc-deployment-d79885fd8-2n875   1/1     Running   0          92s
pc-deployment-d79885fd8-lsjff   1/1     Running   0          62s
pc-deployment-d79885fd8-pfxzf   1/1     Running   0          2m9s
[root@k8s-master ~]# kubectl get deployment.apps pc-deployment -n test -o wide
NAME            READY   UP-TO-DATE   AVAILABLE   AGE     CONTAINERS   IMAGES         SELECTOR
pc-deployment   3/3     3            3           3m38s   nginx        nginx:1.17.4   app=nginx-pod,app in (nginx-pod)

滚动更新（RollingUpdate）

先启动一个新pod，然后停一个旧pod，再起一个新pod......

[root@k8s-master ~]# vim pc-deployment.yaml 
[root@k8s-master ~]# kubectl apply -f pc-deployment.yaml 
deployment.apps/pc-deployment created
[root@k8s-master ~]# kubectl get pods -n test -w
NAME                             READY   STATUS    RESTARTS   AGE
pc-deployment-5ffc5bf56c-6d9nk   1/1     Running   0          5s
pc-deployment-5ffc5bf56c-m2mgv   1/1     Running   0          5s
pc-deployment-5ffc5bf56c-tfzx2   1/1     Running   0          5s
^C[root@k8s-master ~]kubectl set image deploy pc-deployment nginx=nginx:1.17.2 -n test 
deployment.apps/pc-deployment image updated
[root@k8s-master ~]# kubectl get pods -n test -w
NAME                             READY   STATUS              RESTARTS   AGE
pc-deployment-574bcdf54b-9m25z   0/1     ContainerCreating   0          2s
pc-deployment-5ffc5bf56c-6d9nk   1/1     Running             0          19s
pc-deployment-5ffc5bf56c-m2mgv   1/1     Running             0          19s
pc-deployment-5ffc5bf56c-tfzx2   1/1     Running             0          19s
pc-deployment-574bcdf54b-9m25z   0/1     ContainerCreating   0          2s
pc-deployment-574bcdf54b-9m25z   1/1     Running             0          36s
pc-deployment-5ffc5bf56c-tfzx2   1/1     Terminating         0          53s
pc-deployment-574bcdf54b-wlcpz   0/1     Pending             0          0s
pc-deployment-574bcdf54b-wlcpz   0/1     Pending             0          0s
pc-deployment-574bcdf54b-wlcpz   0/1     ContainerCreating   0          0s
pc-deployment-5ffc5bf56c-tfzx2   1/1     Terminating         0          53s
pc-deployment-574bcdf54b-wlcpz   0/1     ContainerCreating   0          2s
pc-deployment-5ffc5bf56c-tfzx2   0/1     Terminating         0          55s
pc-deployment-5ffc5bf56c-tfzx2   0/1     Terminating         0          57s
pc-deployment-5ffc5bf56c-tfzx2   0/1     Terminating         0          58s
pc-deployment-574bcdf54b-wlcpz   1/1     Running             0          26s
pc-deployment-5ffc5bf56c-m2mgv   1/1     Terminating         0          79s
pc-deployment-574bcdf54b-b9lbf   0/1     Pending             0          0s
pc-deployment-574bcdf54b-b9lbf   0/1     Pending             0          0s
pc-deployment-574bcdf54b-b9lbf   0/1     ContainerCreating   0          0s
pc-deployment-5ffc5bf56c-m2mgv   1/1     Terminating         0          79s
pc-deployment-5ffc5bf56c-m2mgv   0/1     Terminating         0          80s
pc-deployment-574bcdf54b-b9lbf   0/1     ContainerCreating   0          1s
pc-deployment-574bcdf54b-b9lbf   1/1     Running             0          2s
pc-deployment-5ffc5bf56c-6d9nk   1/1     Terminating         0          81s
pc-deployment-5ffc5bf56c-6d9nk   1/1     Terminating         0          81s
pc-deployment-5ffc5bf56c-6d9nk   0/1     Terminating         0          83s
pc-deployment-5ffc5bf56c-m2mgv   0/1     Terminating         0          87s
pc-deployment-5ffc5bf56c-m2mgv   0/1     Terminating         0          87s
pc-deployment-5ffc5bf56c-6d9nk   0/1     Terminating         0          94s
pc-deployment-5ffc5bf56c-6d9nk   0/1     Terminating         0          94s
^C[root@k8s-master ~]# kubectl get pods -n test 
NAME                             READY   STATUS    RESTARTS   AGE
pc-deployment-574bcdf54b-9m25z   1/1     Running   0          2m4s
pc-deployment-574bcdf54b-b9lbf   1/1     Running   0          62s
pc-deployment-574bcdf54b-wlcpz   1/1     Running   0          88s
[root@k8s-master ~]# kubectl get pods -n test  -o wide
NAME                             READY   STATUS    RESTARTS   AGE     IP               NODE        NOMINATED NODE   READINESS GATES
pc-deployment-574bcdf54b-9m25z   1/1     Running   0          2m19s   10.244.36.80     k8s-node1   <none>           <none>
pc-deployment-574bcdf54b-b9lbf   1/1     Running   0          77s     10.244.36.81     k8s-node1   <none>           <none>
pc-deployment-574bcdf54b-wlcpz   1/1     Running   0          103s    10.244.169.134   k8s-node2   <none>           <none>

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
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）
查看当前升级版本的状态

[root@k8s-master ~]# kubectl rollout status deployment pc-deployment -n test
deployment "pc-deployment" successfully rolled out
[root@k8s-master ~]# kubectl set image deploy pc-deployment nginx=nginx:1.17.5 -n testdeployment.apps/pc-deployment image updated
[root@k8s-master ~]# kubectl set image deploy pc-deployment nginx=nginx:1.17.6 -n test 
deployment.apps/pc-deployment image updated

[root@k8s-master ~]# kubectl get pod -n test -o wide
NAME                             READY   STATUS              RESTARTS   AGE     IP               NODE        NOMINATED NODE   READINESS GATES
pc-deployment-574bcdf54b-9m25z   1/1     Running             0          10m     10.244.36.80     k8s-node1   <none>           <none>
pc-deployment-574bcdf54b-b9lbf   1/1     Running             0          9m44s   10.244.36.81     k8s-node1   <none>           <none>
pc-deployment-574bcdf54b-wlcpz   1/1     Running             0          10m     10.244.169.134   k8s-node2   <none>           <none>
pc-deployment-768755cc6b-dbzh2   0/1     ContainerCreating   0          18s     <none>           k8s-node2   <none>           <none>
 
^C[root@k8s-master ~]# kubectl get pods -n test  -o wide  
NAME                             READY   STATUS    RESTARTS   AGE   IP               NO
pc-deployment-768755cc6b-9xrmz   1/1     Running   0          36s   10.244.36.83     k8
pc-deployment-768755cc6b-ctlw9   1/1     Running   0          10s   10.244.36.84     k8
pc-deployment-768755cc6b-dbzh2   1/1     Running   0          63s   10.244.169.135   k8
[root@k8s-master ~]# 
[root@k8s-master ~]# kubectl get deployment.apps pc-deployment -n test -o wide

NAME            READY   UP-TO-DATE   AVAILABLE   AGE   CONTAINERS   IMAGES         SELE
pc-deployment   3/3     3            3           12m   nginx        nginx:1.17.6   app=
[root@k8s-master ~]# 
[root@k8s-master ~]# kubectl rollout status deployment pc-deployment -n test
deployment "pc-deployment" successfully rolled out

#查看历史版本升级
[root@k8s-master ~]# kubectl rollout history deploy pc-deployment -n test
deployment.apps/pc-deployment 
REVISION  CHANGE-CAUSE
1         <none>
2         <none>
3         <none>
4         <none>

版本回退

[root@k8s-master ~]# kubectl rollout undo deployment pc-deployment --to-revision=1 -n test
deployment.apps/pc-deployment rolled back
[root@k8s-master ~]# kubectl get deployment.apps pc-deployment -n test -o wide
NAME            READY   UP-TO-DATE   AVAILABLE   AGE   CONTAINERS   IMAGES         SELECTOR
pc-deployment   3/3     3            3           15m   nginx        nginx:1.17.1   app=nginx-pod,app in (nginx-pod)
[root@k8s-master ~]# kubectl rollout history deploy pc-deployment -n test
deployment.apps/pc-deployment 
REVISION  CHANGE-CAUSE
2         <none>
3         <none>
4         <none>
5         <none>

[root@k8s-master ~]# kubectl get rs -n test
NAME                       DESIRED   CURRENT   READY   AGE
pc-deployment-574bcdf54b   0         0         0       15m
pc-deployment-5ffc5bf56c   3         3         3       15m
pc-deployment-768755cc6b   0         0         0       4m49s
pc-deployment-84756ff8f5   0         0         0       4m59s
# 其实deployment之所以可是实现版本的回滚，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

金丝雀发布（Canary Release）是一种渐进式的部署策略，用于逐步将新版本的应用程序或服务部署到生产环境中。这种方法允许开发者在全面推广之前，先在一小部分用户或服务器上测试新版本，以确保新版本在实际使用中的表现符合预期。
金丝雀发布的关键步骤包括：
- 准备新版本：开发并测试新版本的应用程序。
- 部署新版本：将新版本的应用程序部署到生产环境，但只对一小部分用户或服务器可见。
- 监控和测试：密切监控新版本的性能和用户反馈，确保它按预期工作。
- 逐步扩展：如果新版本表现良好，逐步增加访问新版本的用户比例，直到完全替换旧版本。
- 回滚：如果在测试过程中发现问题，可以立即回滚到旧版本，以避免影响所有用户。
在Kubernetes环境中，金丝雀发布可以通过Deployment控制器来实现。Deployment控制器允许你定义应用的期望状态，包括应用的版本和运行的副本数量。通过使用Deployment，你可以控制更新过程，包括暂停和恢复更新，以及设置滚动更新的策略。
例如，你可以创建一个新的Deployment来部署新版本的Pod，然后使用Kubernetes的服务（如Service或Ingress）将一部分流量路由到新版本的Pod上。如果一切顺利，你可以继续更新，逐步增加新版本Pod的数量，直到完全替换旧版本。如果出现问题，可以暂停更新，甚至回滚到旧版本。
金丝雀发布是一种有效的风险管理策略，可以帮助减少新版本部署可能导致的负面影响。


[root@k8s-master ~]# kubectl set image deploy pc-deployment nginx=nginx:1.17.4 -n test && kubectl rollout pause deployment pc-deployment -n test 
deployment.apps/pc-deployment paused
[root@k8s-master ~]# kubectl set image deploy pc-deployment nginx=nginx:1.17.5 -n test && kubectl rollout pause deployment pc-deployment -n test 
deployment.apps/pc-deployment image updated
error: deployments.apps "pc-deployment" is already paused
[root@k8s-master ~]# kubectl get rs -n test -o wide
NAME                       DESIRED   CURRENT   READY   AGE     CONTAINERS   IMAGES         SELECTOR
pc-deployment-5ffc5bf56c   0         0         0       20m     nginx        nginx:1.17.1   app=nginx-pod,app in (nginx-pod),pod-template-hash=5ffc5bf56c
pc-deployment-768755cc6b   0         0         0       9m39s   nginx        nginx:1.17.6   app=nginx-pod,app in (nginx-pod),pod-template-hash=768755cc6b
pc-deployment-84756ff8f5   0         0         0       9m49s   nginx        nginx:1.17.5   app=nginx-pod,app in (nginx-pod),pod-template-hash=84756ff8f5
pc-deployment-d79885fd8    3         3         3       68s     nginx        nginx:1.17.4   app=nginx-pod,app in (nginx-pod),pod-template-hash=d79885fd8
[root@k8s-master ~]# kubectl get rs -n test -o wide
NAME                       DESIRED   CURRENT   READY   AGE     CONTAINERS   IMAGES         SELECTOR
pc-deployment-5ffc5bf56c   0         0         0       20m     nginx        nginx:1.17.1   app=nginx-pod,app in (nginx-pod),pod-template-hash=5ffc5bf56c
pc-deployment-768755cc6b   0         0         0       9m42s   nginx        nginx:1.17.6   app=nginx-pod,app in (nginx-pod),pod-template-hash=768755cc6b
pc-deployment-84756ff8f5   0         0         0       9m52s   nginx        nginx:1.17.5   app=nginx-pod,app in (nginx-pod),pod-template-hash=84756ff8f5
pc-deployment-d79885fd8    3         3         3       71s     nginx        nginx:1.17.4   app=nginx-pod,app in (nginx-pod),pod-template-hash=d79885fd8
[root@k8s-master ~]# kubectl get rs -n test -o wide
NAME                       DESIRED   CURRENT   READY   AGE    CONTAINERS   IMAGES         SELECTOR
pc-deployment-5ffc5bf56c   0         0         0       20m    nginx        nginx:1.17.1   app=nginx-pod,app in (nginx-pod),pod-template-hash=5ffc5bf56c
pc-deployment-768755cc6b   0         0         0       10m    nginx        nginx:1.17.6   app=nginx-pod,app in (nginx-pod),pod-template-hash=768755cc6b
pc-deployment-84756ff8f5   0         0         0       10m    nginx        nginx:1.17.5   app=nginx-pod,app in (nginx-pod),pod-template-hash=84756ff8f5
pc-deployment-d79885fd8    3         3         3       100s   nginx        nginx:1.17.4   app=nginx-pod,app in (nginx-pod),pod-template-hash=d79885fd8
[root@k8s-master ~]# kubectl get pod -n test -o wide
NAME                            READY   STATUS    RESTARTS   AGE     IP               NODE        NOMINATED NODE   READINESS GATES
pc-deployment-d79885fd8-2dps9   1/1     Running   0          2m6s    10.244.36.88     k8s-node1   <none>           <none>
pc-deployment-d79885fd8-6zddq   1/1     Running   0          2m8s    10.244.169.137   k8s-node2   <none>           <none>
pc-deployment-d79885fd8-ccz9b   1/1     Running   0          2m12s   10.244.36.87     k8s-node1   <none>           <none>

#当前pod暂停更新，UP-TO-DATE为0，已更新到最新版本的副本数为0
[root@k8s-master ~]# kubectl get deployment.apps pc-deployment -n test -o wide
NAME            READY   UP-TO-DATE   AVAILABLE   AGE   CONTAINERS   IMAGES         SELECTOR
pc-deployment   3/3     0            3           21m   nginx        nginx:1.17.5   app=nginx-pod,app in (nginx-pod)

#继续更新操作
[root@k8s-master ~]# kubectl rollout resume deploy pc-deployment -n test
deployment.apps/pc-deployment resumed
[root@k8s-master ~]# kubectl get pod -n test -o wide
NAME                             READY   STATUS              RESTARTS   AGE     IP               NODE        NOMINATED NODE   READINESS GATES
pc-deployment-84756ff8f5-b5gzk   0/1     ContainerCreating   0          0s      <none>           k8s-node1   <none>           <none>
pc-deployment-84756ff8f5-gp97h   1/1     Running             0          17s     10.244.169.138   k8s-node2   <none>           <none>
pc-deployment-84756ff8f5-zw85b   1/1     Running             0          19s     10.244.36.89     k8s-node1   <none>           <none>
pc-deployment-d79885fd8-6zddq    1/1     Terminating         0          3m14s   10.244.169.137   k8s-node2   <none>           <none>
pc-deployment-d79885fd8-ccz9b    1/1     Running             0          3m18s   10.244.36.87     k8s-node1   <none>           <none>

#已更新最新版本的副本为3
[root@k8s-master ~]# kubectl get deployment.apps pc-deployment -n test -o wide
NAME            READY   UP-TO-DATE   AVAILABLE   AGE   CONTAINERS   IMAGES         SELECTOR
pc-deployment   3/3     3            3           22m   nginx        nginx:1.17.5   app=nginx-pod,app in (nginx-pod)

查看全文

http://www.kler.cn/a/510094.html