Kubernetes 服务入口管理与 Nginx Ingress Controller

Kubernetes 具有强大的副本，动态扩容等特性，每一次 Pod 的变化 IP 地址都会发生变化，所以 Kubernetes 引进了 Service 的概念。Kubernetes 中使用 Service 对象抽象出来的机制来管理同一组标签的 Pod ，而不需要关心 Pod 发生了什么变化并为其分配了一个虚拟的 IP，当外界需要访问 Pod 里的容器提供的功能时，不直接使用 Pod 的 IP 地址和端口，而是访问 Service 的这个虚拟 IP 和端口，由 Service 把请求转发给它背后的 Pod。

Service 暴露服务类型

通常情况下，我们会定义一个 Service 来管理一组 Pod 暴露相关的服务，如果要对外暴露服务的话，只需要定义相应的端口即可（NodePort模式），但如果定义了很多 Service 对象并暴露服务的话就需要配置很多端口，后续维护起来就会变的很复杂，所以 Kubernetes 中还使用了 Ingress 的机制，比如使用 Nginx 绑定一个固定端口 80，后续的请求通过转发到 Service 即可。这样如果每次新增服务的话，还需要修改 Nginx 的配置,为了解决这个问题 Kubernetes 中还使用了 Ingress Controller 组件。简单理解就是原先需要修改 Nginx 配置，然后配置不同的转发规则到 Service 这个过程抽象出来变成一个 Ingress 对象，后续 Nginx 的变更再通过 Ingress Controoler 与 Kubernetes API 交互，动态的去感知集群中 Ingress 规则变化，再写到 Nginx Pod 里。

kind: Service
apiVersion: v1
metadata:
  name: test-service
spec:
  selector:
    app: test—-app
  ports:
    - protocol: TCP
      port: 80
      targetPort: 8080

由于 Service 中的服务仅可以在容器内部中通讯，如果需要外部能访问到，还需要暴露 Service , 如下 ：

    internet
        |
  ------------
  [ Services ]

kubernetes 中定义了如下的一些服务类型：

Proxy

使用 Kubernetes 代理来访问服务，一般内网查看 dashboard，调试远程查看时可能会用到。

{
  "kind": "Service",
  "apiVersion": "v1",
  "metadata": {
    "name": "kubernetes",
    "namespace": "default",
    "selfLink": "/api/v1/namespaces/default/services/kubernetes",
    "uid": "f594405b-dc19-11e8-90ea-0050569f4a19",
    "resourceVersion": "6",
    "creationTimestamp": "2018-10-30T08:01:08Z",
    "labels": {
      "component": "apiserver",
      "provider": "kubernetes"
    }
  },
  "spec": {
    "ports": [
      {
        "name": "https",
        "protocol": "TCP",
        "port": 443,
        "targetPort": 6443
      }
    ],
    "clusterIP": "10.96.0.1",
    "type": "ClusterIP",
    "sessionAffinity": "None"
  },
  "status": {
    "loadBalancer": {}
  }
}

NodePort Service

指定服务类型为 NodePort 是将外部流量直接发送给服务最原始的方式，需要在 Node 服务节点主机上开放特定的端口，通过绑定主机的某个端口,然后进行 Pod 的请求转发和负载均衡，但这种方式下缺陷是 Service 可能有很多个，如果每个都绑定一个 Node 主机端口的话，主机需要开放外围一堆的端口进行服务调用，管理混乱，并且只能使用 30000-32767 之间的端口，适用与临时暴露某一个服务的端口演示的场景。

{
  "kind": "Service",
  "apiVersion": "v1",
  "metadata": {
    "name": "kubernetes-dashboard",
    "namespace": "kube-system",
    "selfLink": "/api/v1/namespaces/kube-system/services/kubernetes-dashboard",
    "uid": "edd78318-dc1a-11e8-90ea-0050569f4a19",
    "resourceVersion": "1076",
    "creationTimestamp": "2018-10-30T08:08:05Z",
    "labels": {
      "k8s-app": "kubernetes-dashboard"
    }
  },
  "spec": {
    "ports": [
      {
        "protocol": "TCP",
        "port": 443,
        "targetPort": 8443,
        "nodePort": 32151
      }
    ],
    "selector": {
      "k8s-app": "kubernetes-dashboard"
    },
    "clusterIP": "10.103.60.159",
    "type": "NodePort",
    "sessionAffinity": "None",
    "externalTrafficPolicy": "Cluster"
  },
  "status": {
    "loadBalancer": {}
  }
}

LoadBlancer Service

一般配合公用云使用，指定的端口上的所有流量都将转发到该服务，使用 LoadBlancer Service 暴露服务时实际上是向云平台申请创建一个负载均衡器来向外暴露服务，可能需要支付额外的费用。

Nginx Ingress Controller

Nginx Ingress Controller 是由 Nginx 与 Ingress Controller 两部分组成。

Ingress

与上述的几种类型不同，Ingress 实际上不是一种服务。它是位于多前面服务的不同，可以转发不同的域名请求到集群中不同的服务上，简单的理解，Ingress 就是从 Kubernetes 集群外访问集群的入口，将用户的URL请求转发到不同的 Service 上。Ingress 相当于 Nginx、Apache 等负载均衡方向代理服务器，其中还包括规则定义，即 URL 的路由信息，路由信息得的刷新由 Ingress controller 来提供。

Ingress controller

Ingress Controller 通过不断地跟 kubernetes API 打交道，实时的感知后端 service、pod 等变化，比如新增和减少 pod，service 增加与减少等；当得到这些变化信息后，Ingress Controller 再结合下文的 Ingress 生成配置，然后更新反向代理负载均衡器，并刷新其配置，达到服务发现的作用。对于 Ingress controller 来说，创建一个 Ingress 相当于在 nginx.conf 中添加一个 server 入口，并 nginx -s reload 重新生效。。

比如想要通过负载均衡器实现不同子域名到不同服务的访问：

foo.bar.com --|                 |-> foo.bar.com s1:80
              | 178.91.123.132  |
bar.foo.com --|                 |-> bar.foo.com s2:80

定义 Ingress

apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  name: test
spec:
  rules:
  - host: foo.bar.com
    http:
      paths:
      - backend:
          serviceName: s1
          servicePort: 80
  - host: bar.foo.com
    http:
      paths:
      - backend:
          serviceName: s2
          servicePort: 80

Ingress本身并不会自动创建负载均衡器，需要运行一个 Ingress controller 来根据 Ingress 的定义来管理负载均衡器。为了 Ingress 正常工作，集群中必须运行 Ingress controller ,支持 Ingress 的类型的 kube-controller 有如下几种：

Kubernetes currently supports and maintains GCE and nginx controllers.

F5 Networks provides support and maintenance for the F5 BIG-IP Controller for Kubernetes.

Kong offers community or commercial support and maintenance for the Kong Ingress Controller for Kubernetes

Traefik is a fully featured ingress controller (Let’s Encrypt, secrets, http2, websocket…), and it also comes with commercial support by Containous

NGINX, Inc. offers support and maintenance for the NGINX Ingress Controller for Kubernetes

HAProxy based ingress controller jcmoraisjr/haproxy-ingress which is mentioned on this blog post HAProxy Ingress Controller for Kubernetes

Istio based ingress controller Control Ingress Traffic

如下是第三方 Nginx 支持的模式 NGINX Ingress Controller for Kubernetes ，非官方（虽然官方也依赖开源版本的 Nginx）。

关于 nginxinc/kubernetes-ingress 与 kubernetes/ingress-nginx 两者的差异: https://github.com/nginxinc/kubernetes-ingress/blob/master/docs/nginx-ingress-controllers.md ，如果不是使用的不是 NGINX Plus 商业版本，还是官方的支持要丰富一些，另外基于软件的 Traefik，HAProxy，Kong 也是不错的选择，不差钱的也可以考虑商业性质的硬件负载设备如 F5 等。个人比较看好 kubernetes/ingress-nginx 与 Traefik 。这里选择使用官方的 kubernetes/ingress-nginx 作为 ingress controller ，后续再尝试其他第三方的组件。

安装 nginx-ingress-controller (host network 模式根据 Node 数量设置副本数，取决于部署的方式 Deployment or DaemonSet)

$ kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/mandatory.yaml
$ kubectl -n ingress-nginx get pod -o wide
NAME                                        READY   STATUS    RESTARTS   AGE   IP          NODE                NOMINATED NODE
nginx-ingress-controller-5bdbdc5657-bcn2f   1/1     Running   0          74m   10.38.0.1   kubernetes-node-2   <none>
nginx-ingress-controller-5bdbdc5657-mmtph   1/1     Running   0          79m   10.40.0.1   kubernetes-node-1   <none>
$ kubectl get pods --all-namespaces -l app.kubernetes.io/name=ingress-nginx --watch
NAMESPACE       NAME                                        READY   STATUS    RESTARTS   AGE
ingress-nginx   nginx-ingress-controller-6457d975c8-6twqf   1/1     Running   0          78m
ingress-nginx   nginx-ingress-controller-5bdbdc5657-mmtph   1/1     Running   0          83m

$ kubectl describe pod nginx-ingress-controller-6457d975c8-6twqf -n ingress-nginx
Name:               nginx-ingress-controller-6457d975c8-6twqf
Namespace:          ingress-nginx
Priority:           0
PriorityClassName:  <none>
Node:               kubernetes-node-2/172.23.216.50
Start Time:         Wed, 31 Oct 2018 20:04:44 +0800
Labels:             app.kubernetes.io/name=ingress-nginx
                    app.kubernetes.io/part-of=ingress-nginx
                    pod-template-hash=6457d975c8
Annotations:        prometheus.io/port: 10254
                    prometheus.io/scrape: true
Status:             Running
IP:                 172.23.216.50
Controlled By:      ReplicaSet/nginx-ingress-controller-6457d975c8
Containers:
  nginx-ingress-controller:
    Container ID:  docker://f4d5b69cf579752799d6d7e92c547ed9a5a0ba9154b3683c4956079ea9e77304
    Image:         quay.io/kubernetes-ingress-controller/nginx-ingress-controller:0.20.0
    Image ID:      docker-pullable://quay.io/kubernetes-ingress-controller/nginx-ingress-controller@sha256:f6180c5397d2361c317aff1314dc192ab0f9f515346a5319422cdc264f05d2d9
    Ports:         80/TCP, 443/TCP
    Host Ports:    80/TCP, 443/TCP
    Args:
      /nginx-ingress-controller
      --configmap=$(POD_NAMESPACE)/nginx-configuration
      --publish-service=$(POD_NAMESPACE)/ingress-nginx
      --annotations-prefix=nginx.ingress.kubernetes.io
    State:          Running
      Started:      Wed, 31 Oct 2018 20:04:45 +0800
    Ready:          True
    Restart Count:  0
    Liveness:       http-get http://:10254/healthz delay=10s timeout=1s period=10s #success=1 #failure=3
    Readiness:      http-get http://:10254/healthz delay=0s timeout=1s period=10s #success=1 #failure=3
    Environment:
      POD_NAME:       nginx-ingress-controller-6457d975c8-6twqf (v1:metadata.name)
      POD_NAMESPACE:  ingress-nginx (v1:metadata.namespace)
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from nginx-ingress-serviceaccount-token-rhpsb (ro)
Conditions:
  Type              Status
  Initialized       True
  Ready             True
  ContainersReady   True
  PodScheduled      True
Volumes:
  nginx-ingress-serviceaccount-token-rhpsb:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  nginx-ingress-serviceaccount-token-rhpsb
    Optional:    false
QoS Class:       BestEffort
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/not-ready:NoExecute for 300s
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:
  Type     Reason            Age                From                        Message
  ----     ------            ----               ----                        -------
  Warning  FailedScheduling  49m (x3 over 49m)  default-scheduler           0/3 nodes are available: 1 node(s) had taints that the pod didn't tolerate, 2 node(s) didn't have free ports for the requested pod ports.
  Normal   Scheduled         49m                default-scheduler           Successfully assigned ingress-nginx/nginx-ingress-controller-6457d975c8-6twqf to kubernetes-node-2
  Normal   Pulled            48m                kubelet, kubernetes-node-2  Container image "quay.io/kubernetes-ingress-controller/nginx-ingress-controller:0.20.0" already present on machine
  Normal   Created           48m                kubelet, kubernetes-node-2  Created container
  Normal   Started           48m                kubelet, kubernetes-node-2  Started container

查看安装的版本

POD_NAMESPACE=ingress-nginx
POD_NAME=$(kubectl get pods -n $POD_NAMESPACE -l app.kubernetes.io/name=ingress-nginx -o jsonpath='{.items[0].metadata.name}')
kubectl exec -it $POD_NAME -n $POD_NAMESPACE -- /nginx-ingress-controller --version

-------------------------------------------------------------------------------
NGINX Ingress controller
  Release:    0.20.0
  Build:      git-e8d8103
  Repository: https://github.com/kubernetes/ingress-nginx.git-------------------------------------------------------------------------------

根据不同的网络环境与场景，需要选择不同的策略，具体可以参考官方的文档，如下列举常用的几种：

Cloud environments 模式

如果是使用公有云，可以使用公有云的负载到 Node 节点即可，需要支付额外费用。

NodePort Service 模式（临时测试,一般不建议使用）

apiVersion: v1
kind: Service
metadata:
  name: ingress-nginx
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
spec:
  type: NodePort
  ports:
    - name: http
      port: 80
      targetPort: 80

# nodePort: 30000
      protocol: TCP
    - name: https
      port: 443
      targetPort: 443
      protocol: TCP
  selector:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx

---

执行

$ kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/master/deploy/provider/baremetal/service-nodeport.yaml
$ kubectl -n ingress-nginx get svc
NAME            TYPE       CLUSTER-IP    EXTERNAL-IP   PORT(S)                      AGE
ingress-nginx   NodePort   10.98.42.64   <none>        80:31460/TCP,443:31200/TCP   6m50s

host network 模式

修改 hostNetwork: true

vi nginx-ingress-controller.yaml

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: nginx-ingress-controller
  namespace: ingress-nginx
  labels:
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/part-of: ingress-nginx
spec:
  replicas: 2
  selector:
    matchLabels:
      app.kubernetes.io/name: ingress-nginx
      app.kubernetes.io/part-of: ingress-nginx
  template:
    metadata:
      labels:
        app.kubernetes.io/name: ingress-nginx
        app.kubernetes.io/part-of: ingress-nginx
      annotations:
        prometheus.io/port: "10254"
        prometheus.io/scrape: "true"
    spec:
      hostNetwork: true
      serviceAccountName: nginx-ingress-serviceaccount
      containers:
        - name: nginx-ingress-controller
          image: quay.io/kubernetes-ingress-controller/nginx-ingress-controller:0.20.0
          args:
            - /nginx-ingress-controller
            - --configmap=$(POD_NAMESPACE)/nginx-configuration
            - --publish-service=$(POD_NAMESPACE)/ingress-nginx
            - --annotations-prefix=nginx.ingress.kubernetes.io
          securityContext:
            capabilities:
              drop:
                - ALL
              add:
                - NET_BIND_SERVICE
            # www-data -> 33
            runAsUser: 33
          env:
            - name: POD_NAME
              valueFrom:
                fieldRef:
                  fieldPath: metadata.name
            - name: POD_NAMESPACE
              valueFrom:
                fieldRef:
                  fieldPath: metadata.namespace
          ports:
            - name: http
              containerPort: 80
            - name: https
              containerPort: 443
          livenessProbe:
            failureThreshold: 3
            httpGet:
              path: /healthz
              port: 10254
              scheme: HTTP
            initialDelaySeconds: 10
            periodSeconds: 10
            successThreshold: 1
            timeoutSeconds: 1
          readinessProbe:
            failureThreshold: 3
            httpGet:
              path: /healthz
              port: 10254
              scheme: HTTP
            periodSeconds: 10
            successThreshold: 1
            timeoutSeconds: 1

---

$ kubectl apply -f nginx-ingress-controller.yaml

查看 ingress-nginx

$ kubectl -n ingress-nginx get pod -o wide
NAME                                        READY   STATUS    RESTARTS   AGE    IP              NODE                NOMINATED NODE
nginx-ingress-controller-6457d975c8-6twqf   1/1     Running   0          3m7s   172.23.216.50   kubernetes-node-2   <none>
nginx-ingress-controller-6457d975c8-smjsv   1/1     Running   0          3m7s   172.23.216.49   kubernetes-node-1   <none>

测试访问

$ curl -D- 172.23.216.50
HTTP/1.1 404 Not Found
Server: nginx/1.15.5
Date: Wed, 31 Oct 2018 12:08:40 GMT
Content-Type: text/html
Content-Length: 153
Connection: keep-alive

<html>
<head><title>404 Not Found</title></head>
<body>
<center><h1>404 Not Found</h1></center>
<hr><center>nginx/1.15.5</center>
</body>
</html>

创建 kubernetes-dashboard-ingress（默认 https ，annotations 头必须设置 ）

vi kubernetes-dashboard-ingress.yaml

apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  name: kubernetes-dashboard-ingress
  namespace: kube-system
  annotations:
    nginx.ingress.kubernetes.io/ssl-redirect: "true"
    nginx.ingress.kubernetes.io/rewrite-target: /
    nginx.ingress.kubernetes.io/secure-backends: "true"

spec:
#  tls:
#   - secretName: k8s-dashboard-secret
  rules:
   - http:
      paths:
      - path: /
        backend:
          serviceName: kubernetes-dashboard
          servicePort: 443
      - path: /test
        backend:
          serviceName: test-nginx
          servicePort: 80

执行

$ kubectl apply -f kubernetes-dashboard-ingress.yaml
$ kubectl get ingress -o wide --all-namespaces
NAMESPACE     NAME                           HOSTS   ADDRESS   PORTS   AGE
kube-system   kubernetes-dashboard-ingress   *                 80      3h53m

最后访问：https://172.23.216.49/ 或 https://172.23.216.50/

备注：其他功能参考官方文档：

• Ingress Path Matching

• baremetal
• monitoring

其他命令

#删除 Ingress Controller 命名空间
$ kubectl delete namespace nginx-ingress
#安装网络组件
$ yum install net-tools
#查看开放的端口
$ netstat -ntlp

REFER:
https://kubernetes.io/docs/concepts/services-networking/ingress/
https://medium.com/google-cloud/kubernetes-nodeport-vs-loadbalancer-vs-ingress-when-should-i-use-what-922f010849e0
https://www.nginx.com/products/nginx/kubernetes-ingress-controller
https://github.com/kubernetes/ingress-nginx
https://github.com/containous/traefik
https://github.com/nginxinc/kubernetes-ingress/
https://my.oschina.net/u/2306127/blog/1930169