Service详解

Pod是程序的载体，通过Pod的IP可以访问程序，但Pod的IP地址并不固定，不方便直接对Pod IP直接进行访问。

k8s提供Service资源，能对提供同一个服务的多个Pod进行聚合，并提供一个统一的入口地址。通过访问Service的入口地址就能访问Pod服务。

Service很多情况是概念，kube-proxy服务进程真正起作用，每个Node节点上运行kube-proxy。创建Service通过api-server向etcd写入创建的service信息，kube-proxy基于监听发现Service变动，将最新的Service信息转换成对应访问规则。

一般常用ipvs模式

#开启ipvs 
#把mode变成ipvs
[root@master ~]# kubectl edit cm kube-proxy -n kube-system
configmap/kube-proxy edited
[root@master ~]# kubectl delete pod -l k8s-app=kube-proxy -n kube-system
pod "kube-proxy-lqdks" deleted
pod "kube-proxy-wcn2x" deleted
pod "kube-proxy-x78bv" deleted
#查看规则
[root@master ~]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
  -> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  10.96.0.1:443 rr
  -> 192.168.126.132:6443         Masq    1      0          0
TCP  10.96.0.10:53 rr
  -> 10.244.0.4:53                Masq    1      0          0
  -> 10.244.0.5:53                Masq    1      0          0
TCP  10.96.0.10:9153 rr
  -> 10.244.0.4:9153              Masq    1      0          0
  -> 10.244.0.5:9153              Masq    1      0          0
TCP  10.100.107.200:443 rr
  -> 192.168.126.134:443          Masq    1      0          10
UDP  10.96.0.10:53 rr
  -> 10.244.0.4:53                Masq    1      0          0
  -> 10.244.0.5:53                Masq    1      0          0

资源清单

kind: Service  # 资源类型
apiVersion: v1  # 资源版本
metadata: # 元数据
  name: service # 资源名称
  namespace: dev # 命名空间
spec: # 描述
  selector: # 标签选择器，用于确定当前service代理哪些pod
    app: nginx
  type: # Service类型，指定service的访问方式
  clusterIP:  # 虚拟服务的ip地址
  sessionAffinity: # session亲和性，支持ClientIP、None两个选项
  ports: # 端口信息
    - protocol: TCP 
      port: 3017  # service端口
      targetPort: 5003 # pod端口
      nodePort: 31122 # 主机端口

• clusterIP:默认值，k8s系统自动分配，只能在集群内部使用
• nodePort：将Service通过指定Node上端口暴露给外部
• LoadBalancer：使用外接负载均衡器完成到服务的负载分发，注意此模式需要外部云环境支持
• ExternalName： 把集群外部的服务引入集群内部，直接使用

使用

环境准备

首先利用Deployment创建出三个Pod，为Pod设置标签

deployment.yaml

apiVersion: apps/v1
kind: Deployment      
metadata:
  name: pc-deployment
  namespace: dev
spec: 
  replicas: 3
  selector:
    matchLabels:
      app: nginx-pod
  template:
    metadata:
      labels:
        app: nginx-pod
    spec:
      containers:
      - name: nginx
        image: nginx:1.17.1
        ports:
        - containerPort: 80

#这里我创建了一个临时文件夹把yaml文件放入这里执行 防止与其他文件混淆
[root@master ~]# mkdir tmp
[root@master ~]# cd tmp/

#创建名称空间dev
[root@master tmp]# kubectl create ns dev
namespace/dev created
#创建deployment
[root@master tmp]# kubectl create -f deployment.yaml
deployment.apps/pc-deployment created
#查看Pod详情
[root@master ~]# kubectl get pod -n dev -o  wide
NAME                             READY   STATUS    RESTARTS   AGE   IP            NODE    NOMINATED NODE   READINESS GATES
pc-deployment-6696798b78-7l646   1/1     Running   0          33m   10.244.2.75   node2   <none>           <none>
pc-deployment-6696798b78-dp9t4   1/1     Running   0          33m   10.244.2.77   node2   <none>           <none>
pc-deployment-6696798b78-llzpc   1/1     Running   0          33m   10.244.2.76   node2   <none>           <none>


#方便测试，修改三台nginx的index.html页面
# kubectl exec -it pc-deployment-自动生成的后缀 -n dev /bin/sh
# echo [IP地址] > /usr/share/nginx/html/index.html

[root@master ~]# kubectl exec -it pc-deployment-6696798b78-7l646 -n dev /bin/sh
# echo "10.244.2.75" > /usr/share/nginx/html/index.html
# exit
[root@master ~]# kubectl exec -it pc-deployment-6696798b78-dp9t4  -n dev /bin/sh
# echo "10.244.2.77" > /usr/share/nginx/html/index.html
# exit
[root@master ~]# kubectl exec -it pc-deployment-6696798b78-llzpc  -n dev /bin/sh
# echo "10.244.2.76" > /usr/share/nginx/html/index.html
# exit

[root@master ~]# curl 10.244.2.75:80
10.244.2.75
[root@master ~]# curl 10.244.2.77:80
10.244.2.77
[root@master ~]# curl 10.244.2.76:80
10.244.2.76

如果无法ping通node节点或Pod节点，很有可能是虚拟机IP地址改变请设置静态IP地址

#BOOTPROTO=static 设置静态模式
IPADDR= #IP地址 你自己虚拟机的IP地址
NETMASK=255.255.255.0 #掩码
GATEWAY= #网关
DNS1=8.8.8.8
DNS2=114.114.114.114

ClusterIP

ClusterIP

通过集群的内部 IP 暴露服务，选择该值时服务只能够在集群内部访问。

创建service-clusterip.yaml

apiVersion: v1
kind: Service
metadata:
  name: service-clusterip
  namespace: dev
spec:
  selector:
    app: nginx-pod
  clusterIP: 10.97.97.97 # service的ip地址，如果不写，默认会生成一个
  type: ClusterIP
  ports:
  - port: 80  # Service端口       
    targetPort: 80 # pod端口

#创建service查看信息
[root@master tmp]# kubectl create -f service-clusterip.yaml
service/service-clusterip created
[root@master tmp]# kubectl get svc -n dev -o wide
NAME                TYPE        CLUSTER-IP    EXTERNAL-IP   PORT(S)   AGE   SELECTOR
service-clusterip   ClusterIP   10.97.97.97   <none>        80/TCP    25s   app=nginx-pod
[root@master tmp]# kubectl describe svc service-clusterip -n dev
Name:              service-clusterip
Namespace:         dev
Labels:            <none>
Annotations:       <none>
Selector:          app=nginx-pod
Type:              ClusterIP
IP:                10.97.97.97
Port:              <unset>  80/TCP
TargetPort:        80/TCP
Endpoints:         10.244.2.75:80,10.244.2.76:80,10.244.2.77:80
Session Affinity:  None
Events:            <none>

上述详细信息里的Endpoint是k8s一个资源对象，存储在etcd中，用来记录Service对应的所以Pod访问地址，根据Service的selector描述产生。

一个Service由一组Pod组成，这些Pod通过Endpoints暴露，Endpoints是实现实际服务的端点集合。

Endpoints: 10.244.2.75:80,10.244.2.76:80,10.244.2.77:80

#查看endpoints
[root@master tmp]# kubectl get endpoints -n dev -o wide
NAME                ENDPOINTS                                      AGE
service-clusterip   10.244.2.75:80,10.244.2.76:80,10.244.2.77:80   23m

#查看ipvs规则
[root@master tmp]# ipvsadm -Ln
...
TCP  10.97.97.97:80 rr#rr 轮询策略
  -> 10.244.2.75:80               Masq    1      0          0
  -> 10.244.2.76:80               Masq    1      0          0
  -> 10.244.2.77:80               Masq    1      0          0
...
#循环访问10.97.97.97:80查看效果
[root@master tmp]# while true;do curl 10.97.97.97:80; sleep 2; done;
10.244.2.76
10.244.2.75
10.244.2.77
10.244.2.76
...

负载分发策略

对Service的访问被分发到了后端的Pod上去，目前kubernetes提供了两种负载分发策略：

• 如果不定义，默认使用kube-proxy的策略，比如随机、轮询

• 基于客户端地址的会话保持模式，即来自同一个客户端发起的所有请求都会转发到固定的一个Pod上

  此模式可以使在spec中添加sessionAffinity:ClientIP选项

#首先停止service
[root@master tmp]# kubectl delete -f service-clusterip.yaml
service "service-clusterip" deleted
#修改
[root@master tmp]# vim service-clusterip.yaml
[root@master tmp]# cat service-clusterip.yaml
apiVersion: v1
kind: Service
metadata:
  name: service-clusterip
  namespace: dev
spec:
  sessionAffinity: ClientIP
  selector:
    app: nginx-pod
  clusterIP: 10.97.97.97 # service的ip地址，如果不写，默认会生成一个
  type: ClusterIP
  ports:
  - port: 80  # Service端口
    targetPort: 80 # pod端口 
    
    
#查看ipvs策略
[root@master tmp]# ipvsadm -Ln
...
TCP  10.97.97.97:80 rr persistent 10800
  -> 10.244.2.75:80               Masq    1      0          0
  -> 10.244.2.76:80               Masq    1      0          0
  -> 10.244.2.77:80               Masq    1      0          0
...

#循环访问测试
[root@master tmp]# while true;do curl 10.97.97.97; sleep 2; done;
10.244.2.77
10.244.2.77
10.244.2.77
10.244.2.77
...
#删除service
[root@master tmp]# kubectl delete -f service-clusterip.yaml
service "service-clusterip" deleted

HeadLiness

在某些场景中，开发人员可能不想使用Service提供的负载均衡功能，而希望自己来控制负载均衡策略，针对这种情况，kubernetes提供了HeadLiness Service，这类Service不会分配Cluster IP，如果想要访问service，只能通过service的域名进行查询。

创建service-headliness.yaml

apiVersion: v1
kind: Service
metadata:
  name: service-headliness
  namespace: dev
spec:
  selector:
    app: nginx-pod
  clusterIP: None # 将clusterIP设置为None，即可创建headliness Service
  type: ClusterIP
  ports:
  - port: 80    
    targetPort: 80

#创建svc
[root@master tmp]# kubectl create  -f service-headliness.yaml
service/service-headliness created
[root@master tmp]# kubectl get svc -n dev
NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
service-headliness   ClusterIP   None         <none>        80/TCP    0s
[root@master tmp]# kubectl describe svc service-headliness -n dev
Name:              service-headliness
Namespace:         dev
Labels:            <none>
Annotations:       <none>
Selector:          app=nginx-pod
Type:              ClusterIP
IP:                None
Port:              <unset>  80/TCP
TargetPort:        80/TCP
Endpoints:         10.244.2.75:80,10.244.2.76:80,10.244.2.77:80
Session Affinity:  None
Events:            <none>

#查看域名解析
[root@master tmp]# kubectl exec -it pc-deployment-6696798b78-7l646 -n dev /bin/sh
#
# cat /etc/resolv.conf
nameserver 10.96.0.10
search dev.svc.cluster.local svc.cluster.local cluster.local localdomain
options ndots:5
# exit


#访问域名
[root@master tmp]# dig @10.96.0.10 service-headliness.dev.svc.cluster.local

; <<>> DiG 9.11.4-P2-RedHat-9.11.4-26.P2.el7_9.13 <<>> @10.96.0.10 service-headliness.dev.svc.cluster.local
; (1 server found)
;; global options: +cmd
;; Got answer:
;; WARNING: .local is reserved for Multicast DNS
;; You are currently testing what happens when an mDNS query is leaked to DNS
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 22352
;; flags: qr aa rd; QUERY: 1, ANSWER: 3, AUTHORITY: 0, ADDITIONAL: 1
;; WARNING: recursion requested but not available

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
;; QUESTION SECTION:
;service-headliness.dev.svc.cluster.local. IN A

;; ANSWER SECTION:
service-headliness.dev.svc.cluster.local. 30 IN A 10.244.2.77
service-headliness.dev.svc.cluster.local. 30 IN A 10.244.2.75
service-headliness.dev.svc.cluster.local. 30 IN A 10.244.2.76

;; Query time: 39 msec
;; SERVER: 10.96.0.10#53(10.96.0.10)
;; WHEN: Sun Jun 04 05:58:29 EDT 2023
;; MSG SIZE  rcvd: 237

#删除服务
[root@master tmp]# kubectl delete -f service-headliness.yaml
service "service-headliness" deleted

NodePort

如果希望将Service暴露给集群外部使用，那么就要使用到另外一种类型的Service，称为NodePort类型。NodePort的工作原理其实就是将service的端口映射到Node的一个端口上，然后就可以通过NodeIp:NodePort来访问service了。

如果你将 type 字段设置为 NodePort，则 Kubernetes 控制平面将在 --service-node-port-range 标志指定的范围内分配端口（默认值：30000-32767）。 每个节点将那个端口（每个节点上的相同端口号）代理到你的服务中。 你的服务在其 .spec.ports[*].nodePort 字段中报告已分配的端口。

创建service-nodeport.yaml

apiVersion: v1
kind: Service
metadata:
  name: service-nodeport
  namespace: dev
spec:
  selector:
    app: nginx-pod
  type: NodePort # service类型
  ports:
  - port: 80
    nodePort: 30002 # 指定绑定的node的端口(默认的取值范围是：30000-32767), 如果不指定，会默认分配
    targetPort: 80

#创建service
[root@master tmp]# kubectl create -f service-nodeport.yaml
service/service-nodeport created
#查看
[root@master tmp]# kubectl get svc -n dev -o wide
NAME               TYPE       CLUSTER-IP      EXTERNAL-IP   PORT(S)        AGE   SELECTOR
service-nodeport   NodePort   10.101.18.114   <none>        80:30002/TCP   21s   app=nginx-pod
#通过电脑主机浏览器访问集群中任意一个nodeip：端口，即可访问pod
[root@master tmp]# kubectl get nodes -o wide
NAME     STATUS   ROLES    AGE    VERSION   INTERNAL-IP       EXTERNAL-IP   OS-IMAGE                KERNEL-VERSION          CONTAINER-RUNTIME
master   Ready    master   5d4h   v1.17.4   192.168.126.132   <none>        CentOS Linux 7 (Core)   3.10.0-957.el7.x86_64   docker://24.0.2
node1    Ready    <none>   5d4h   v1.17.4   192.168.126.133   <none>        CentOS Linux 7 (Core)   3.10.0-957.el7.x86_64   docker://18.6.3
node2    Ready    <none>   5d4h   v1.17.4   192.168.126.134   <none>        CentOS Linux 7 (Core)   3.10.0-957.el7.x86_64   docker://18.6.3

LoadBalancer

LoadBalancer和NodePort很相似，目的都是向外部暴露一个端口，区别在于LoadBalancer会在集群的外部再来做一个负载均衡设备，而这个设备需要外部环境支持的，外部服务发送到这个设备上的请求，会被设备负载之后转发到集群中。

ExternalName

ExternalName类型的Service用于引入集群外部的服务，它通过externalName属性指定外部一个服务的地址，然后在集群内部访问此service就可以访问到外部的服务了。

创建service-externalname.yaml

apiVersion: v1
kind: Service
metadata:
  name: service-externalname
  namespace: dev
spec:
  type: ExternalName # service类型
  externalName: www.baidu.com  #改成ip地址也可以

#创建
[root@master tmp]# kubectl create -f service-externalname.yaml
service/service-externalname created
#解析域名
[root@master tmp]# dig @10.96.0.10 service-externalname.dev.svc.cluster.local

; <<>> DiG 9.11.4-P2-RedHat-9.11.4-26.P2.el7_9.13 <<>> @10.96.0.10 service-externalname.dev.svc.cluster.local
; (1 server found)
;; global options: +cmd
;; Got answer:
;; WARNING: .local is reserved for Multicast DNS
;; You are currently testing what happens when an mDNS query is leaked to DNS
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 59479
;; flags: qr aa rd; QUERY: 1, ANSWER: 4, AUTHORITY: 0, ADDITIONAL: 1
;; WARNING: recursion requested but not available

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
;; QUESTION SECTION:
;service-externalname.dev.svc.cluster.local. IN A

;; ANSWER SECTION:
service-externalname.dev.svc.cluster.local. 5 IN CNAME www.baidu.com.
www.baidu.com.          5       IN      CNAME   www.a.shifen.com.
www.a.shifen.com.       5       IN      A       110.242.68.3
www.a.shifen.com.       5       IN      A       110.242.68.4

;; Query time: 51 msec
;; SERVER: 10.96.0.10#53(10.96.0.10)
;; WHEN: Sun Jun 04 07:08:19 EDT 2023
;; MSG SIZE  rcvd: 247