ReplicationController初体验

本节，我们通过一些例子来演示ReplicationController的创建、查看、更新、删除等操作。

创建

首先我们准备一个名为replication_controller_simple.yaml的配置文件，其内容如下所示：

[root@ecs-d8b6 manifests]# cat replication_controller_simple.yaml 
apiVersion: v1
kind: ReplicationController
metadata:
  name: replication-controller-runs-pod
spec:
  replicas: 3
  selector:
    app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:1.19.0

该ReplicationController配置保证同时有3个Pod副本在运行。

使用kubectl create命令创建资源，如下所示：

[root@ecs-d8b6 manifests]# kubectl create -f replication_controller_simple.yaml 
replicationcontroller/replication-controller-runs-pod created

查看

我们先查看刚刚创建的ReplicationController资源：

[root@ecs-d8b6 manifests]# kubectl get replicationcontrollers 
NAME                              DESIRED   CURRENT   READY   AGE
replication-controller-runs-pod   3         3         3       3m4s

可以看到期望的Pod数量（DESIRED）为3，当前已创建的Pod数量（CURRENT）也为3，并且处于运行状态的Pod数量（READY）同样是3，正是我们所期望的状态。

接着，我们查看实际运行的Pod：

[root@ecs-d8b6 manifests]# kubectl get pods
NAME                                    READY   STATUS    RESTARTS   AGE
replication-controller-runs-pod-gv4l6   1/1     Running   0          6m11s
replication-controller-runs-pod-pcpmb   1/1     Running   0          6m11s
replication-controller-runs-pod-qxrhz   1/1     Running   0          6m11s

可以看到Pod运行也是正常的。

最后，我们查看ReplicationController的详细信息：

[root@ecs-d8b6 manifests]# kubectl describe replicationcontrollers replication-controller-runs-pod 
Name:         replication-controller-runs-pod
Namespace:    default
Selector:     app=nginx
Labels:       app=nginx
Annotations:  <none>
Replicas:     3 current / 3 desired
Pods Status:  3 Running / 0 Waiting / 0 Succeeded / 0 Failed
Pod Template:
  Labels:  app=nginx
  Containers:
   nginx:
    Image:        nginx:1.19.0
    Port:         <none>
    Host Port:    <none>
    Environment:  <none>
    Mounts:       <none>
  Volumes:        <none>
Events:
  Type    Reason            Age   From                    Message
  ----    ------            ----  ----                    -------
  Normal  SuccessfulCreate  10m   replication-controller  Created pod: replication-controller-runs-pod-gv4l6
  Normal  SuccessfulCreate  10m   replication-controller  Created pod: replication-controller-runs-pod-qxrhz
  Normal  SuccessfulCreate  10m   replication-controller  Created pod: replication-controller-runs-pod-pcpmb

通过最后部分的Events信息可知，上面显示的Pod确实是名为replication-controller-runs-pod的ReplicationController创建的。

更新

改变副本个数

我们把上面replication_controller_simple.yaml配置文件中的副本数由3改为5，然后使用kubectl apply提交更改：

[root@ecs-d8b6 manifests]# kubectl apply -f replication_controller_simple.yaml 
replicationcontroller/replication-controller-runs-pod configured

然后再次查看Pod，可以发现Pod数量也会相应地调整为5。

[root@ecs-d8b6 manifests]# kubectl get pods
NAME                                    READY   STATUS    RESTARTS   AGE
replication-controller-runs-pod-gv4l6   1/1     Running   0          25m
replication-controller-runs-pod-pcpmb   1/1     Running   0          25m
replication-controller-runs-pod-qtd4b   1/1     Running   0          6s
replication-controller-runs-pod-qxrhz   1/1     Running   0          25m
replication-controller-runs-pod-zz6fj   1/1     Running   0          6s

同样地，当副本数变小时，多出的Pod会被删除。

模拟Pod异常

假定当前名为replication-controller-runs-pod的ReplicationController维护的Pod副本数为3，我们通过删除Pod的方式模拟Pod异常，如下所示：

[root@ecs-d8b6 manifests]# kubectl get pods
NAME                                    READY   STATUS    RESTARTS   AGE
replication-controller-runs-pod-gv4l6   1/1     Running   0          30m
replication-controller-runs-pod-pcpmb   1/1     Running   0          30m
replication-controller-runs-pod-qxrhz   1/1     Running   0          30m
[root@ecs-d8b6 manifests]# 
[root@ecs-d8b6 manifests]# 
[root@ecs-d8b6 manifests]# 
[root@ecs-d8b6 manifests]# kubectl delete pods replication-controller-runs-pod-gv4l6 
pod "replication-controller-runs-pod-gv4l6" deleted
[root@ecs-d8b6 manifests]# kubectl get pods
NAME                                    READY   STATUS    RESTARTS   AGE
replication-controller-runs-pod-4598b   1/1     Running   0          10s
replication-controller-runs-pod-pcpmb   1/1     Running   0          31m
replication-controller-runs-pod-qxrhz   1/1     Running   0          31m

可以看到，当我们删除了名为replication-controller-runs-pod-gv4l6的Pod后，新的Pod对象replication-controller-runs-pod-4598b马上被创建了出来，从而保证副本数维持不变。

删除

当删除ReplicationController对象时，由该对象创建的Pod默认也会被删除，如下所示：

[root@ecs-d8b6 manifests]# kubectl delete -f replication_controller_simple.yaml 
replicationcontroller "replication-controller-runs-pod" deleted
[root@ecs-d8b6 manifests]# kubectl get pods
NAME                                    READY   STATUS        RESTARTS   AGE
replication-controller-runs-pod-ktrxr   0/1     Terminating   0          18s
replication-controller-runs-pod-rcdhk   0/1     Terminating   0          18s
replication-controller-runs-pod-sc9mj   0/1     Terminating   0          18s
[root@ecs-d8b6 manifests]# kubectl get pods
No resources found in default namespace.