【云原生-Kubernetes篇】Kubernetes入门扫盲实战

本文是在官网的教程中进行的，官方提供了在线的命令行工具和安装好的minikube工具及环境，方便新手快速入门，如下是主要包含的内容。

一、创建集群

1.1 通过minikube搭建集群

假设已经预先安装好了minikube，执行如下命令初始化一个单节点的集群：

minikube start

试试k8s集群是否初始化成功：

kubectl version

然后可以查看集群的信息和节点信息：

$ kubectl cluster-info
Kubernetes control plane is running at https://172.17.0.24:8443
KubeDNS is running at https://172.17.0.24:8443/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy

$ kubectl get nodes
NAME       STATUS   ROLES                  AGE     VERSION
minikube   Ready    control-plane,master   8m25s   v1.20.2

二、部署应用

在部署应用之前，我们查看pod的信息：

$ kubectl get pods
No resources found in default namespace.

然后我们使用如下命令部署一个应用并查看应用的信息：

$ kubectl create deployment kubernetes-bootcamp --image=gcr.io/google-samples/kubernetes-bootcamp:v1
deployment.apps/kubernetes-bootcamp created

$ kubectl get deployments
NAME                  READY   UP-TO-DATE   AVAILABLE   AGE
kubernetes-bootcamp   1/1     1            1           53s

$ kubectl get pods
NAME                                   READY   STATUS    RESTARTS   AGE
kubernetes-bootcamp-57978f5f5d-wk8kg   1/1     Running   0          62s

应用虽然部署完成了，但是默认情况下我们只能在集群内部才能访问pod（参考下面3.3节内容），命令行curl访问应用的接口则是属于集群外部访问pod了，所以要实现curl接口需要做如下的设置：

• 方案一

新启动一个该节点的命令行，执行如下命令：

$ kubectl proxy
Starting to serve on 127.0.0.1:8001

表示节点的8001端口能访问到集群中的pod了：

$ curl http://localhost:8001/version
{
  "major": "1",
  "minor": "20",
  "gitVersion": "v1.20.2",
  "gitCommit": "faecb196815e248d3ecfb03c680a4507229c2a56",
  "gitTreeState": "clean",
  "buildDate": "2021-01-13T13:20:00Z",
  "goVersion": "go1.15.5",
  "compiler": "gc",
  "platform": "linux/amd64"
}

• 方案二

主动将应用的端口暴露出去（参考如下第四章节内容）：

$ kubectl expose deployment/kubernetes-bootcamp --type="NodePort" --port 8080
service/kubernetes-bootcamp exposed

$ kubectl get services
NAME                  TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)          AGE
kubernetes            ClusterIP   10.96.0.1       <none>        443/TCP          15m
kubernetes-bootcamp   NodePort    10.106.152.94   <none>        8080:31281/TCP   12s

$ curl http://localhost:31281/version
Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-57978f5f5d-wk8kg | v=1

三、应用程序探索

应用部署完成后，我们需要通过一些手段来监控和运维我们的应用。

3.1 检查

$ kubectl get pods
NAME                                  READY   STATUS    RESTARTS   AGE
kubernetes-bootcamp-fb5c67579-q8mrb   1/1     Running   0          26s

$ kubectl describe pods
Name:         kubernetes-bootcamp-fb5c67579-q8mrb
Namespace:    default
Priority:     0
Node:         minikube/172.17.0.129
Start Time:   Tue, 08 Feb 2022 08:39:44 +0000
Labels:       app=kubernetes-bootcamp
              pod-template-hash=fb5c67579
Annotations:  <none>
Status:       Running
IP:           172.18.0.6
IPs:
  IP:           172.18.0.6
Controlled By:  ReplicaSet/kubernetes-bootcamp-fb5c67579
Containers:
  kubernetes-bootcamp:
    Container ID:   docker://e243e942198e2d1a1d37db5820e210ada432221c1ca915b7859fbc3af5543d72
    Image:          gcr.io/google-samples/kubernetes-bootcamp:v1
    Image ID:       docker-pullable://jocatalin/kubernetes-bootcamp@sha256:0d6b8ee63bb57c5f5b6156f446b3bc3b3c143d233037f3a2f00e279c8fcc64af
    Port:           8080/TCP
    Host Port:      0/TCP
    State:          Running
      Started:      Tue, 08 Feb 2022 08:39:49 +0000
    Ready:          True
    Restart Count:  0
    Environment:    <none>
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-kqhb5 (ro)
Conditions:
  Type              Status
  Initialized       True 
  Ready             True 
  ContainersReady   True 
  PodScheduled      True 
Volumes:
  default-token-kqhb5:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-kqhb5
    Optional:    false
QoS Class:       BestEffort
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
                 node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events:
  Type    Reason     Age    From               Message
  ----    ------     ----   ----               -------
  Normal  Scheduled  3m38s  default-scheduler  Successfully assigned default/kubernetes-bootcamp-fb5c67579-q8mrb to minikube
  Normal  Pulled     3m34s  kubelet            Container image "gcr.io/google-samples/kubernetes-bootcamp:v1" already present on machine
  Normal  Created    3m34s  kubelet            Created container kubernetes-bootcamp
  Normal  Started    3m33s  kubelet            Started container kubernetes-bootcamp

3.2 查看日志

使用容器名称查看容器中的运行日志。

$ kubectl logs kubernetes-bootcamp-fb5c67579-q8mrb
Kubernetes Bootcamp App Started At: 2022-02-08T08:39:49.327Z | Running On:  kubernetes-bootcamp-fb5c67579-q8mrb 

3.3 容器内执行命令

# 进入容器内部并执行bash命令
$ kubectl exec -ti kubernetes-bootcamp-fb5c67579-q8mrb -- bash

# 测试容器内部命令
root@kubernetes-bootcamp-fb5c67579-q8mrb:/# pwd 

# 容器内部curl访问pod
root@kubernetes-bootcamp-fb5c67579-q8mrb:/# curl localhost:8080
Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-fb5c67579-q8mrb | v=1

# 退出容器
root@kubernetes-bootcamp-fb5c67579-q8mrb:/# exit
exit

四、访问应用

在第二章中我们介绍了，部署完成pod之后，默认情况下只能在集群内部才能访问到该pod，想要通过外部访问该pod，就一定要使用Service。

我们如上已经创建了service，并将其暴露在外了，此处不再赘述，可以参考第二章节的方案二。

我们可以通过使用label来筛选想要查询的资源信息：

$ kubectl get pods -l app=kubernetes-bootcamp
NAME                                  READY   STATUS    RESTARTS   AGE
kubernetes-bootcamp-fb5c67579-rvbxv   1/1     Running   0          2m15s

$ kubectl get services -l app=kubernetes-bootcamp
NAME                  TYPE       CLUSTER-IP     EXTERNAL-IP   PORT(S)          AGE
kubernetes-bootcamp   NodePort   10.97.252.94   <none>        8080:31926/TCP   76s

当我们不再需要该service的时候，可以通过如下命令删除，删除后，外部将无法再访问运行的pod。

$ curl localhost:31926
Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-fb5c67579-rvbxv | v=1

$ kubectl delete service -l app=kubernetes-bootcamp
service "kubernetes-bootcamp" deleted

$ kubectl get services
NAME         TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   5m13s

$ curl localhost:31926
curl: (7) Failed to connect to localhost port 31926: Connection refused

五、扩展应用

当流量有变化的时候，我们就需要动态地对pod进行扩容：

$ kubectl get deployments
NAME                  READY   UP-TO-DATE   AVAILABLE   AGE
kubernetes-bootcamp   1/1     1            1           6m10s

$ kubectl get rs
NAME                            DESIRED   CURRENT   READY   AGE
kubernetes-bootcamp-fb5c67579   1         1         1       6m6s

$ kubectl scale deployments/kubernetes-bootcamp --replicas=4
deployment.apps/kubernetes-bootcamp scaled

$ kubectl get deployments
NAME                  READY   UP-TO-DATE   AVAILABLE   AGE
kubernetes-bootcamp   4/4     4            4           8m32s

$ kubectl get rsNAME                            DESIRED   CURRENT   READY   AGE
kubernetes-bootcamp-fb5c67579   4         4         1       7m27s

$ kubectl get pods -o wide
NAME                                  READY   STATUS    RESTARTS   AGE     IP           NODE       NOMINATED NODE   READINESS GATES
kubernetes-bootcamp-fb5c67579-82nhh   1/1     Running   0          2m13s   172.18.0.8   minikube   <none>           <none>
kubernetes-bootcamp-fb5c67579-96t49   1/1     Running   0          9m6s    172.18.0.5   minikube   <none>           <none>
kubernetes-bootcamp-fb5c67579-jz5vw   1/1     Running   0          2m13s   172.18.0.9   minikube   <none>           <none>
kubernetes-bootcamp-fb5c67579-vfpn8   1/1     Running   0          2m13s   172.18.0.7   minikube   <none>           <none>

$ kubectl describe deployments/kubernetes-bootcamp
Name:                   kubernetes-bootcamp
Namespace:              default
CreationTimestamp:      Tue, 08 Feb 2022 09:16:30 +0000
Labels:                 app=kubernetes-bootcamp
Annotations:            deployment.kubernetes.io/revision: 1
Selector:               app=kubernetes-bootcamp
Replicas:               4 desired | 4 updated | 4 total | 4 available | 0 unavailable
StrategyType:           RollingUpdate
MinReadySeconds:        0
RollingUpdateStrategy:  25% max unavailable, 25% max surge
Pod Template:
  Labels:  app=kubernetes-bootcamp
  Containers:
   kubernetes-bootcamp:
    Image:        gcr.io/google-samples/kubernetes-bootcamp:v1
    Port:         8080/TCP
    Host Port:    0/TCP
    Environment:  <none>
    Mounts:       <none>
  Volumes:        <none>
Conditions:
  Type           Status  Reason
  ----           ------  ------
  Progressing    True    NewReplicaSetAvailable
  Available      True    MinimumReplicasAvailable
OldReplicaSets:  <none>
NewReplicaSet:   kubernetes-bootcamp-fb5c67579 (4/4 replicas created)
Events:
  Type    Reason             Age    From                   Message
  ----    ------             ----   ----                   -------
  Normal  ScalingReplicaSet  9m45s  deployment-controller  Scaled up replica set kubernetes-bootcamp-fb5c67579 to 1
  Normal  ScalingReplicaSet  2m50s  deployment-controller  Scaled up replica set kubernetes-bootcamp-fb5c67579 to 4

此时我们已经将pod扩展到4个副本了，此时我们在集群外部访问Service，那么请求会被自动地以负载均衡的方式分发到各个pod上。

$ kubectl describe services/kubernetes-bootcamp
Name:                     kubernetes-bootcamp
Namespace:                default
Labels:                   app=kubernetes-bootcamp
Annotations:              <none>
Selector:                 app=kubernetes-bootcamp
Type:                     NodePort
IP Families:              <none>
IP:                       10.104.229.171
IPs:                      10.104.229.171
Port:                     <unset>  8080/TCP
TargetPort:               8080/TCP
NodePort:                 <unset>  30848/TCP
Endpoints:                172.18.0.5:8080,172.18.0.7:8080,172.18.0.8:8080 + 1 more...
Session Affinity:         None
External Traffic Policy:  Cluster
Events:                   <none>


# 如下请求将会被负载均衡地分发到各个pod上
$ curl localhost:30848
Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-fb5c67579-96t49 | v=1
$ curl localhost:30848
Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-fb5c67579-vfpn8 | v=1
$ curl localhost:30848
Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-fb5c67579-jz5vw | v=1
$ curl localhost:30848
Hello Kubernetes bootcamp! | Running on: kubernetes-bootcamp-fb5c67579-82nhh | v=1

当我们需要缩容的时候也很简单：

$ kubectl scale deployments/kubernetes-bootcamp --replicas=2
deployment.apps/kubernetes-bootcamp scaled

$ kubectl get deployments
NAME                  READY   UP-TO-DATE   AVAILABLE   AGE
kubernetes-bootcamp   2/2     2            2           17m

$ kubectl get pods -o wide
NAME                                  READY   STATUS        RESTARTS   AGE     IP           NODE       NOMINATED NODE   READINESS GATES
kubernetes-bootcamp-fb5c67579-82nhh   1/1     Terminating   0          9m38s   172.18.0.8   minikube   <none>           <none>
kubernetes-bootcamp-fb5c67579-96t49   1/1     Running       0          16m     172.18.0.5   minikube   <none>           <none>
kubernetes-bootcamp-fb5c67579-jz5vw   1/1     Running       0          9m38s   172.18.0.9   minikube   <none>           <none>
kubernetes-bootcamp-fb5c67579-vfpn8   1/1     Terminating   0          9m38s   172.18.0.7   minikube   <none>           <none>

$ kubectl get pods -o wide
NAME                                  READY   STATUS    RESTARTS   AGE   IP           NODE       NOMINATED NODE   READINESS GATES
kubernetes-bootcamp-fb5c67579-96t49   1/1     Running   0          18m   172.18.0.5   minikube   <none>           <none>
kubernetes-bootcamp-fb5c67579-jz5vw   1/1     Running   0          11m   172.18.0.9   minikube   <none>           <none>

六、更新应用

如果新版本的应用镜像已经发布，如何将旧版本的deployment升级呢？

# 查看当前pod的信息，检查使用的应用镜像版本等信息
$ kubectl describe pods
...

# 告诉deployment中的kubernetes-bootcamp，需要将镜像替换为指定的其它镜像
$ kubectl set image deployments/kubernetes-bootcamp kubernetes-bootcamp=jocatalin/kubernetes-bootcamp:v2
deployment.apps/kubernetes-bootcamp image updated


# 开始替换
$ kubectl get pods
NAME                                   READY   STATUS        RESTARTS   AGE
kubernetes-bootcamp-7d44784b7c-55vrt   1/1     Running       0          4s
kubernetes-bootcamp-7d44784b7c-cftsl   1/1     Running       0          4s
kubernetes-bootcamp-7d44784b7c-dhc42   1/1     Running       0          6s
kubernetes-bootcamp-7d44784b7c-lg2bd   1/1     Running       0          7s
kubernetes-bootcamp-fb5c67579-9bddh    1/1     Terminating   0          5m42s
kubernetes-bootcamp-fb5c67579-sfwwh    1/1     Terminating   0          5m42s
kubernetes-bootcamp-fb5c67579-sx9kt    1/1     Terminating   0          5m42s
kubernetes-bootcamp-fb5c67579-x4n6h    1/1     Terminating   0          5m42s

# 替换完成
$ kubectl get pods
NAME                                   READY   STATUS    RESTARTS   AGE
kubernetes-bootcamp-7d44784b7c-55vrt   1/1     Running   0          5m
kubernetes-bootcamp-7d44784b7c-cftsl   1/1     Running   0          5m
kubernetes-bootcamp-7d44784b7c-dhc42   1/1     Running   0          5m2s
kubernetes-bootcamp-7d44784b7c-lg2bd   1/1     Running   0          5m3s

# 查看当前pod的信息，检查使用的应用镜像版本等信息
$ kubectl describe pods
...

如果新版本镜像部署有问题，还可以及时回滚：

# 回滚deployments/kubernetes-bootcamp到上一个版本
$ kubectl rollout undo deployments/kubernetes-bootcamp
deployment.apps/kubernetes-bootcamp rolled back

# 回滚中
$ kubectl get pods
NAME                                   READY   STATUS              RESTARTS   AGE
kubernetes-bootcamp-7d44784b7c-55vrt   1/1     Terminating         0          10m
kubernetes-bootcamp-7d44784b7c-cftsl   1/1     Terminating         0          10m
kubernetes-bootcamp-7d44784b7c-dhc42   1/1     Terminating         0          10m
kubernetes-bootcamp-7d44784b7c-lg2bd   1/1     Running             0          10m
kubernetes-bootcamp-fb5c67579-6lbbh    0/1     ContainerCreating   0          1s
kubernetes-bootcamp-fb5c67579-84hrk    0/1     ContainerCreating   0          1s
kubernetes-bootcamp-fb5c67579-n8xnp    1/1     Running             0          4s
kubernetes-bootcamp-fb5c67579-qgrs7    1/1     Running             0          4s

# 回滚完成
$ kubectl get pods
NAME                                  READY   STATUS    RESTARTS   AGE
kubernetes-bootcamp-fb5c67579-6lbbh   1/1     Running   0          95s
kubernetes-bootcamp-fb5c67579-84hrk   1/1     Running   0          95s
kubernetes-bootcamp-fb5c67579-n8xnp   1/1     Running   0          98s
kubernetes-bootcamp-fb5c67579-qgrs7   1/1     Running   0          98s

# 查看当前pod的信息，检查使用的应用镜像版本等信息
$ kubectl describe pods
...

七、参考文档

学习 Kubernetes 基础知识 | Kubernetes