Service暴露端口方式

Service存在的意义

为解决Pod的动态变化，提供统一访问入口：

• 防止Pod失联，准备找到提供同一个服务的Pod
• 定义一组Pod的访问策略
• Service通过标签关联一组Pod
• Service使用iptables或者ipvs为一组Pod提供负载均衡能力

service

将运行在一组 Pods上的应用程序公开为网络服务的抽象方法。

使用 Kubernetes，你无需修改应用程序即可使用不熟悉的服务发现机制。 Kubernetes 为 Pods 提供自己的 IP 地址，并为一组 Pod 提供相同的 DNS 名， 并且可以在它们之间进行负载均衡。

Service 资源

Kubernetes Service 定义了这样一种抽象：逻辑上的一组 Pod，一种可以访问它们的策略 —— 通常称为微服务。 Service 所针对的 Pods 集合通常是通过选择算符来确定的。 要了解定义服务端点的其他方法，请参阅不带选择算符的服务。

举个例子，考虑一个图片处理后端，它运行了 3 个副本。这些副本是可互换的 —— 前端不需要关心它们调用了哪个后端副本。 然而组成这一组后端程序的 Pod 实际上可能会发生变化， 前端客户端不应该也没必要知道，而且也不需要跟踪这一组后端的状态。

Service 定义的抽象能够解耦这种关联。

云原生服务发现

如果你想要在应用程序中使用 Kubernetes API 进行服务发现，则可以查询 API 服务器 的 Endpoints 资源，只要服务中的 Pod 集合发生更改，Endpoints 就会被更新。

对于非本机应用程序，Kubernetes 提供了在应用程序和后端 Pod 之间放置网络端口或负载均衡器的方法

实例

ClusterIP

ClusterIP：默认类型，自动分配一个仅Cluster内部能够访问的虚拟IP（只能在集群内部访问）

[root@master manifest]# cat network.yaml 
apiVersion: apps/v1 
kind: Deployment 
metadata: 
  name: myapp-deploy 
  namespace: default 
spec: 
  replicas: 2 
  selector: 
    matchLabels: 
      app: myapp 
      release: v1
  template:
    metadata: 
      labels: 
        app: myapp 
        release: v1 
    spec: 
      containers: 
      - name: myapp
        image: hyhxy0206/apache:v1
        imagePullPolicy: IfNotPresent
---
apiVersion: v1 
kind: Service 
metadata: 
  name: myapp-nodeport
  namespace: default
spec: 
  type: ClusterIP
  selector: 
    app: myapp
    release: v1
  ports: 
  - name: httpd
    port: 80
    targetPort: 80
[root@master manifest]# kubectl apply -f network.yaml 
deployment.apps/myapp-deploy created
service/myapp-nodeport created
[root@master manifest]# kubectl get pods,svc
NAME                                READY   STATUS    RESTARTS   AGE
pod/myapp-deploy-65b8f7b7fd-6df4n   1/1     Running   0          2m59s
pod/myapp-deploy-65b8f7b7fd-8v2gv   1/1     Running   0          2m59s

NAME                     TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)   AGE
service/kubernetes       ClusterIP   10.96.0.1        <none>        443/TCP   8d
service/myapp-nodeport   ClusterIP   10.100.149.133   <none>        80/TCP    2m59s

[root@master manifest]# curl 10.100.149.133
test page with v1

NodePort

NodePort：经过每一个 Node 上的 IP和静态端口（NodePort）(范围30000-32767)暴露服务。以ClusterIP为基础NodePort服务会路由到 ClusterIP服务。经过请求:能够从集群的外部访问一个集群内部的 NodePort服务ClusterIP和路由规则会自动创建

[root@master manifest]# cat network.yaml 
apiVersion: apps/v1 
kind: Deployment 
metadata: 
  name: myapp-deploy 
  namespace: default 
spec: 
  replicas: 2 
  selector: 
    matchLabels: 
      app: myapp 
      release: v1
  template:
    metadata: 
      labels: 
        app: myapp 
        release: v1 
    spec: 
      containers: 
      - name: myapp
        image: hyhxy0206/apache:v1
        imagePullPolicy: IfNotPresent
---
apiVersion: v1 
kind: Service 
metadata: 
  name: myapp-nodeport
  namespace: default
spec: 
  type: NodePort  //类型
  selector: 
    app: myapp
    release: v1
  ports: 
  - name: httpd
    port: 80
    targetPort: 80
    nodePort: 30001   //映射端口
[root@master manifest]# kubectl apply -f network.yaml 
deployment.apps/myapp-deploy created
service/myapp-nodeport created
[root@master manifest]# kubectl get pod,svc
NAME                                READY   STATUS    RESTARTS   AGE
pod/myapp-deploy-65b8f7b7fd-mmwhk   1/1     Running   0          23s
pod/myapp-deploy-65b8f7b7fd-wttt4   1/1     Running   0          23s

NAME                     TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)        AGE
service/kubernetes       ClusterIP   10.96.0.1       <none>        443/TCP        8d
service/myapp-nodeport   NodePort    10.105.76.230   <none>        80:30001/TCP   23s

[root@master manifest]# curl 192.168.143.140:30001
test page with v1

Loadbalancer

LoadBalancer：使用云提供商的负载均衡器，能够向外部暴露服务。外部的负载均衡器能够路由到 NodePort服务和 ClusterIP服务

ExternalName

ExternalName创建一个dns别名指到service name上，主要是防止service name发生变化，要配合dns插件使用

[root@master manifest]# vim network.yaml 
apiVersion: apps/v1 
kind: Deployment 
metadata: 
  name: myapp-deploy 
  namespace: default 
spec: 
  replicas: 2 
  selector: 
    matchLabels: 
      app: myapp 
      release: v1
  template:
    metadata: 
      labels: 
        app: myapp 
        release: v1 
    spec: 
      containers: 
      - name: httpd
        image: hyhxy0206/httpd:v0.1 //基于centos的源码apache
        imagePullPolicy: IfNotPresent
---
apiVersion: v1
kind: Service
metadata: 
  name: myapp-externalname
  namespace: default
spec: 
  type: ExternalName  //类型
  externalName: my.k8s.example.com
[root@master manifest]# kubectl apply -f network.yaml 
deployment.apps/myapp-deploy created
service/myapp-externalname created
root@master manifest]# kubectl get pod,svc
NAME                                READY   STATUS    RESTARTS   AGE
pod/myapp-deploy-5c8c56f8fc-4cmp9   1/1     Running   0          49s
pod/myapp-deploy-5c8c56f8fc-z4lqx   1/1     Running   0          49s

NAME                         TYPE           CLUSTER-IP   EXTERNAL-IP          PORT(S)   AGE
service/kubernetes           ClusterIP      10.96.0.1    <none>               443/TCP   8d
service/myapp-externalname   ExternalName   <none>       my.k8s.example.com   <none>    49s
//进入容器
root@master manifest]# kubectl exec -it myapp-deploy-5c8c56f8fc-4cmp9 /bin/sh
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl exec [POD] -- [COMMAND] instead.
sh-4.4# cd /usr/local/
sh-4.4# ls
apache	apr-util  etc	 include  lib64    sbin   src
apr	bin	  games  lib	  libexec  share
//安装软件
sh-4.4# #yum install -y bind-utils
sh-4.4#  nslookup myapp-externalname.default.svc.cluster.local
Server:		10.96.0.10
Address:	10.96.0.10#53

myapp-externalname.default.svc.cluster.local	canonical name = my.k8s.example.com.
** server can't find my.k8s.example.com: NXDOMAIN

// 经过 nslookup 访问
sh-4.4#  nslookup myapp-externalname.default.svc.cluster.local
Server:		10.96.0.10
Address:	10.96.0.10#53

myapp-externalname.default.svc.cluster.local	canonical name = my.k8s.example.com.
** server can't find my.k8s.example.com: NXDOMAIN
// 经过 dig 访问
sh-4.4# dig -t A myapp-externalname.default.svc.cluster.local.@10.96.0.10

; <<>> DiG 9.11.26-RedHat-9.11.26-6.el8 <<>> -t A myapp-externalname.default.svc.cluster.local.@10.96.0.10
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 50466
;; flags: qr rd; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 1
;; WARNING: recursion requested but not available

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
; COOKIE: c5b3afaa1d57af33 (echoed)
;; QUESTION SECTION:
;myapp-externalname.default.svc.cluster.local.\@10.96.0.10. IN A

;; Query time: 1014 msec
;; SERVER: 10.96.0.10#53(10.96.0.10)
;; WHEN: Sun Dec 26 15:50:02 UTC 2021
;; MSG SIZE  rcvd: 97

ExternalIP

[root@master manifest]# vim network.yaml 
apiVersion: apps/v1 
kind: Deployment 
metadata: 
  name: myapp-deploy 
  namespace: default 
spec: 
  replicas: 2 
  selector: 
    matchLabels: 
      app: myapp 
      release: v1
  template:
    metadata: 
      labels: 
        app: myapp 
        release: v1 
    spec: 
      containers: 
      - name: httpd
        image: hyhxy0206/httpd:v0.1
        imagePullPolicy: IfNotPresent
---
apiVersion: v1
kind: Service
metadata: 
  name: myapp-externalip
  namespace: default
spec: 
  selector: 
    app: myapp
    release: v1
  ports: 
  - name: httpd
    port: 80
    targetPort: 80
  externalIPs: 
    - 10.0.0.240
[root@master manifest]# kubectl apply -f network.yaml 
deployment.apps/myapp-deploy created
service/myapp-externalip created
[root@master manifest]# kubectl get pods,svc
NAME                                READY   STATUS    RESTARTS   AGE
pod/myapp-deploy-5c8c56f8fc-ppgtc   1/1     Running   0          53s
pod/myapp-deploy-5c8c56f8fc-vz9nb   1/1     Running   0          53s

NAME                       TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)   AGE
service/kubernetes         ClusterIP   10.96.0.1        <none>        443/TCP   8d
service/myapp-externalip   ClusterIP   10.111.117.175   10.0.0.240    80/TCP    53s
[root@master manifest]# curl 10.111.117.175
<html><body><h1>It works!</h1></body></html>
[root@master manifest]# curl 10.0.0.240
<html><body><h1>It works!</h1></body></html>

Service代理方式

userspace 代理模式

这种模式，kube-proxy 会监视 Kubernetes 控制平面对 Service 对象和 Endpoints 对象的添加和移除操作。 对每个 Service，它会在本地 Node 上打开一个端口（随机选择）。 任何连接到“代理端口”的请求，都会被代理到 Service 的后端 Pods 中的某个上面（如 Endpoints 所报告的一样）。 使用哪个后端 Pod，是 kube-proxy 基于 SessionAffinity 来确定的。

最后，它配置 iptables 规则，捕获到达该 Service 的 clusterIP（是虚拟 IP） 和 Port 的请求，并重定向到代理端口，代理端口再代理请求到后端Pod。

默认情况下，用户空间模式下的 kube-proxy 通过轮转算法选择后端。

iptables 代理模式

这种模式，kube-proxy 会监视 Kubernetes 控制节点对 Service 对象和 Endpoints 对象的添加和移除。 对每个 Service，它会配置 iptables 规则，从而捕获到达该 Service 的 clusterIP 和端口的请求，进而将请求重定向到 Service 的一组后端中的某个 Pod 上面。 对于每个 Endpoints 对象，它也会配置 iptables 规则，这个规则会选择一个后端组合。

默认的策略是，kube-proxy 在 iptables 模式下随机选择一个后端。

使用 iptables 处理流量具有较低的系统开销，因为流量由 Linux netfilter 处理， 而无需在用户空间和内核空间之间切换。 这种方法也可能更可靠。

如果 kube-proxy 在 iptables 模式下运行，并且所选的第一个 Pod 没有响应， 则连接失败。 这与用户空间模式不同：在这种情况下，kube-proxy 将检测到与第一个 Pod 的连接已失败， 并会自动使用其他后端 Pod 重试。

你可以使用 Pod 就绪探测器 验证后端 Pod 可以正常工作，以便 iptables 模式下的 kube-proxy 仅看到测试正常的后端。 这样做意味着你避免将流量通过 kube-proxy 发送到已知已失败的 Pod。

IPVS 代理模式

FEATURE STATE: Kubernetes v1.11 [stable]

在 ipvs 模式下，kube-proxy 监视 Kubernetes 服务和端点，调用 netlink 接口相应地创建 IPVS 规则， 并定期将 IPVS 规则与 Kubernetes 服务和端点同步。 该控制循环可确保IPVS 状态与所需状态匹配。访问服务时，IPVS 将流量定向到后端Pod之一。

IPVS代理模式基于类似于 iptables 模式的 netfilter 挂钩函数， 但是使用哈希表作为基础数据结构，并且在内核空间中工作。 这意味着，与 iptables 模式下的 kube-proxy 相比，IPVS 模式下的 kube-proxy 重定向通信的延迟要短，并且在同步代理规则时具有更好的性能。 与其他代理模式相比，IPVS 模式还支持更高的网络流量吞吐量。

IPVS 提供了更多选项来平衡后端 Pod 的流量。 这些是：

• rr：轮替（Round-Robin）
• lc：最少链接（Least Connection），即打开链接数量最少者优先
• dh：目标地址哈希（Destination Hashing）
• sh：源地址哈希（Source Hashing）
• sed：最短预期延迟（Shortest Expected Delay）
• nq：从不排队（Never Queue）

说明：

要在 IPVS 模式下运行 kube-proxy，必须在启动 kube-proxy 之前使 IPVS 在节点上可用。

当 kube-proxy 以 IPVS 代理模式启动时，它将验证 IPVS 内核模块是否可用。 如果未检测到 IPVS 内核模块，则 kube-proxy 将退回到以 iptables 代理模式运行。

在这些代理模型中，绑定到服务 IP 的流量： 在客户端不了解 Kubernetes 或服务或 Pod 的任何信息的情况下，将 Port 代理到适当的后端。

如果要确保每次都将来自特定客户端的连接传递到同一 Pod， 则可以通过将 service.spec.sessionAffinity 设置为 “ClientIP” （默认值是 “None”），来基于客户端的 IP 地址选择会话关联。 你还可以通过适当设置 service.spec.sessionAffinityConfig.clientIP.timeoutSeconds 来设置最大会话停留时间。 （默认值为 10800 秒，即 3 小时）。

k8s练习实例

#滚动升级与回滚
//制作apache镜像
[root@master apache]# cat Dockerfile 
FROM busybox

RUN mkdir /data && \
    echo "test page with v1" > /data/index.html

ENTRYPOINT ["/bin/httpd","-f","-h","/data"]
[root@master apache]# docker build -t hyhxy0206/apache:v1 .
Sending build context to Docker daemon  2.048kB
Step 1/3 : FROM busybox
 ---> ffe9d497c324
Step 2/3 : RUN mkdir /data &&     echo "test page with v1" > /data/index.html
 ---> Running in 6ef5a9999027
Removing intermediate container 6ef5a9999027
 ---> 1f55d1956ec6
Step 3/3 : ENTRYPOINT ["/bin/httpd","-f","-h","/data"]
 ---> Running in 9e953c76320f
Removing intermediate container 9e953c76320f
 ---> 8171aabbee2d
Successfully built 8171aabbee2d
Successfully tagged hyhxy0206/apache:v1
[root@master apache]# vim Dockerfile
[root@master apache]# docker build -t hyhxy0206/apache:v2 .
Sending build context to Docker daemon  2.048kB
Step 1/3 : FROM busybox
 ---> ffe9d497c324
Step 2/3 : RUN mkdir /data &&     echo "test page with v2" > /data/index.html
 ---> Running in 591e79d258a9
Removing intermediate container 591e79d258a9
 ---> 47cda9c6f56f
Step 3/3 : ENTRYPOINT ["/bin/httpd","-f","-h","/data"]
 ---> Running in 03aab99e17ce
Removing intermediate container 03aab99e17ce
 ---> 9731bddd2d85
Successfully built 9731bddd2d85
Successfully tagged hyhxy0206/apache:v2
[root@master apache]# docker images
REPOSITORY                                                        TAG        IMAGE ID       CREATED              SIZE
hyhxy0206/apache                                                  v2         9731bddd2d85   About a minute ago   1.24MB
hyhxy0206/apache                                                  v1         8171aabbee2d   About a minute ago   1.24MB

[root@master apache]# docker login
[root@master apache]# docker push hyhxy0206/apache:v1
[root@master apache]# docker push hyhxy0206/apache:v2、

//用v1版本运行3个pod
[root@master ~]# kubectl create deployment web --image hyhxy0206/apache:v1 --replicas 3
deployment.apps/web created
[root@master ~]# kubectl get pods
NAME                   READY   STATUS              RESTARTS   AGE
web-65bbbfdf58-7gfnr   0/1     ContainerCreating   0          6s
web-65bbbfdf58-k52hl   0/1     ContainerCreating   0          6s
web-65bbbfdf58-nf6tg   0/1     ContainerCreating   0          6s
[root@master ~]# kubectl expose deploy web --port 80 --target-port 80
service/web exposed
[root@master ~]# kubectl get svc,pods
NAME                 TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)   AGE
service/kubernetes   ClusterIP   10.96.0.1        <none>        443/TCP   2d23h
service/web          ClusterIP   10.106.145.169   <none>        80/TCP    13s

NAME                       READY   STATUS    RESTARTS   AGE
pod/web-65bbbfdf58-7gfnr   1/1     Running   0          60s
pod/web-65bbbfdf58-k52hl   1/1     Running   0          60s
pod/web-65bbbfdf58-nf6tg   1/1     Running   0          60s

[root@master ~]# while :;do curl 10.106.145.169;done
//设置镜像使其滚动升级
[root@master ~]# kubectl set image deploy/web apache=hyhxy0206/apache:v2
deployment.apps/web image updated

[root@master ~]# while :;do curl 10.101.67.172;done
test page with v2
test page with v2
test page with v2
test page with v2
test page with v2
test page with v2

//运行命令回滚，上一版本
[root@master ~]# kubectl rollout undo deploy/web
deployment.apps/web rolled back
[root@master ~]# while :;do curl 10.101.67.172;done
test page with v1
test page with v1
test page with v1
test page with v1
test page with v1
test page with v1

[root@master ~]# kubectl rollout undo deploy/web
deployment.apps/web rolled back
[root@master ~]# while :;do curl 10.101.67.172;done
test page with v2
test page with v2
test page with v2
test page with v2
test page with v2
test page with v2

//给一个副本扩容
[root@master manifest]# kubectl get pods
NAME                   READY   STATUS    RESTARTS   AGE
test                   1/1     Running   0          8m7s
web-64bbf59bf8-dm7c5   1/1     Running   0          9m23s
web-64bbf59bf8-gmp78   1/1     Running   0          9m23s
web-64bbf59bf8-kzx4p   1/1     Running   0          9m23s
web-64bbf59bf8-r4dlb   1/1     Running   0          2m36s

#一个pod跑不同镜像的容器
[root@master manifest]# cat redis.yaml 
apiVersion: v1
kind: Pod  
metadata:
  name: test  
  labels:
    app: good   
spec:
  containers:            
  - image: nginx
    name: nginx
  - image: redis
    name: redis
  - image: memcached
    name: memcached
[root@master manifest]# kubectl apply -f redis.yaml 
pod/test created
[root@master manifest]# kubectl get pods
NAME   READY   STATUS              RESTARTS   AGE
test   0/3     ContainerCreating   0          14s
[root@master manifest]# kubectl get pods
NAME   READY   STATUS    RESTARTS   AGE
test   3/3     Running   0          72s
[root@master manifest]#  kubectl describe pods test
Name:         test
Namespace:    default
Priority:     0
Node:         node2.example.com/192.168.143.142
Start Time:   Mon, 27 Dec 2021 00:56:38 +0800
Labels:       app=good
Annotations:  <none>
Status:       Running
IP:           10.244.2.116
IPs:
  IP:  10.244.2.116
Containers:
  nginx:
    Container ID:   docker://57fd71cacf1d8fecc6241d9c86ada46a4e4cea93f0a62db16b928bdb9a3a32f1
    Image:          nginx
    Image ID:       docker-pullable://nginx@sha256:366e9f1ddebdb844044c2fafd13b75271a9f620819370f8971220c2b330a9254
    Port:           <none>
    Host Port:      <none>
    State:          Running
      Started:      Mon, 27 Dec 2021 00:56:42 +0800
    Ready:          True
    Restart Count:  0
    Environment:    <none>
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-rbqzh (ro)
  redis:
    Container ID:   docker://99d40d9c03f3097e9096224d3cea1a588e895fcd26355db455da7667632ae2f4
    Image:          redis
    Image ID:       docker-pullable://redis@sha256:db485f2e245b5b3329fdc7eff4eb00f913e09d8feb9ca720788059fdc2ed8339
    Port:           <none>
    Host Port:      <none>
    State:          Running
      Started:      Mon, 27 Dec 2021 00:57:08 +0800
    Ready:          True
    Restart Count:  0
    
# 给一个pod创建service，并可以通过ClusterIP/NodePort
[root@master manifest]# vim nodeport.yaml 
apiVersion: apps/v1 
kind: Deployment 
metadata: 
  name: myapp-deploy 
  namespace: default 
spec: 
  replicas: 1
  selector: 
    matchLabels: 
      app: myapp 
      release: v1
  template:
    metadata: 
      labels: 
        app: myapp 
        release: v1 
    spec: 
      containers: 
      - name: myapp
        image: hyhxy0206/httpd:v0.1
        imagePullPolicy: IfNotPresent
---
apiVersion: v1 
kind: Service 
metadata: 
  name: myapp-nodeport
  namespace: default
spec: 
  type: NodePort  
  selector: 
    app: myapp
    release: v1
  ports: 
  - name: httpd
    port: 80
    targetPort: 80
    nodePort: 31960
[root@master manifest]# kubectl apply -f nodeport.yaml 
deployment.apps/myapp-deploy created
service/myapp-nodeport created
[root@master manifest]#  kubectl get pod,svc
NAME                               READY   STATUS    RESTARTS   AGE
pod/myapp-deploy-ccf9fc64b-9pjnb   1/1     Running   0          85s

NAME                     TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
service/kubernetes       ClusterIP   10.96.0.1      <none>        443/TCP        8d
service/myapp-nodeport   NodePort    10.99.165.44   <none>        80:31960/TCP   85s

[root@master manifest]# curl 10.99.165.44
<html><body><h1>It works!</h1></body></html>
[root@master manifest]# curl 192.168.143.140:31960
<html><body><h1>It works!</h1></body></html>


# 创建deployment和service，使用centos容器Nslookup解析service

[root@master manifest]# vim nslookup.yaml 
apiVersion: apps/v1 
kind: Deployment 
metadata: 
  name: myapp-deploy 
  namespace: default 
spec: 
  replicas: 1
  selector: 
    matchLabels: 
      app: myapp 
      release: v1
  template:
    metadata: 
      labels: 
        app: myapp 
        release: v1 
    spec: 
      containers: 
      - name: myapp
        image: hyhxy0206/httpd:v0.1
        imagePullPolicy: IfNotPresent
---
apiVersion: v1
kind: Service
metadata: 
  name: myapp-externalname
  namespace: default
spec: 
  type: ExternalName 
  externalName: web.test.example.com
[root@master manifest]# kubectl apply -f nslookup.yaml 
deployment.apps/myapp-deploy created
service/myapp-externalname created
[root@master manifest]# kubectl get pod,svc
NAME                               READY   STATUS    RESTARTS   AGE
pod/myapp-deploy-ccf9fc64b-lhntk   1/1     Running   0          23s

NAME                         TYPE           CLUSTER-IP   EXTERNAL-IP            PORT(S)   AGE
service/kubernetes           ClusterIP      10.96.0.1    <none>                 443/TCP   8d
service/myapp-externalname   ExternalName   <none>       web.test.example.com   <none>    23s

//辨析一个busybox资源定义文件
[root@master manifest]# cat busybox.yaml 
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: xx
  namespace: default
spec:
  replicas: 3
  selector:
    matchLabels:
      app: busybox
  template:
    metadata:
      labels:
        app: busybox
    spec:
      containers:
      - name: b4
        image: busybox
        command: ["/bin/sh","-c","sleep 9000"]
[root@master manifest]# kubectl apply -f busybox.yaml 
deployment.apps/xx created
[root@master manifest]# kubectl get pod,svc
NAME                               READY   STATUS              RESTARTS   AGE
pod/myapp-deploy-ccf9fc64b-lhntk   1/1     Running             0          2m36s
pod/xx-758696cd47-6nt24            0/1     ContainerCreating   0          7s
pod/xx-758696cd47-jsq6p            0/1     ContainerCreating   0          7s
pod/xx-758696cd47-qp6mq            1/1     Running             0          7s

NAME                         TYPE           CLUSTER-IP   EXTERNAL-IP            PORT(S)   AGE
service/kubernetes           ClusterIP      10.96.0.1    <none>                 443/TCP   8d
service/myapp-externalname   ExternalName   <none>       web.test.example.com   <none>    2m36s
[root@master manifest]# kubectl get pod,svc
NAME                               READY   STATUS    RESTARTS   AGE
pod/myapp-deploy-ccf9fc64b-lhntk   1/1     Running   0          3m14s
pod/xx-758696cd47-6nt24            1/1     Running   0          45s
pod/xx-758696cd47-jsq6p            1/1     Running   0          45s
pod/xx-758696cd47-qp6mq            1/1     Running   0          45s

NAME                         TYPE           CLUSTER-IP   EXTERNAL-IP            PORT(S)   AGE
service/kubernetes           ClusterIP      10.96.0.1    <none>                 443/TCP   8d
service/myapp-externalname   ExternalName   <none>       web.test.example.com   <none>    3m14s

//exec 交互验证
[root@master manifest]# kubectl exec -it xx /bin/sh
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl exec [POD] -- [COMMAND] instead.
Error from server (NotFound): pods "xx" not found
[root@master manifest]# kubectl exec -it xx-758696cd47-6nt24 /bin/sh
kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl exec [POD] -- [COMMAND] instead.
/ #  nslookup myapp-externalname.default.svc.cluster.local 
Server:		10.96.0.10
Address:	10.96.0.10:53

myapp-externalname.default.svc.cluster.local	canonical name = web.test.example.com

*** Can't find myapp-externalname.default.svc.cluster.local: No answer