确认[Kubernetes]的ingress功能是否正常
首先
这次我想要确认一下Ingress的运作。Ingress可以简单说是L7层的负载均衡器。
虽然与在L4层进行负载均衡的LoadBalancer服务相似，但严格来说，Ingress并不是一个服务，而是一个独立的”Ingress”资源。
值得注意的是，在当前最新的V1.18版本中，它仍然是一个Beta版。
kubernetes.io改名为Ingress。
准备之前的事项
创建Pod
为了验证Ingress的操作，创建以下Pod（Deployment）。
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx0
spec:
replicas: 2
selector:
matchLabels:
app: nginx-dep0
template:
metadata:
labels:
app: nginx-dep0
spec:
containers:
- name: nginx
image: nginx:latest
ports:
- containerPort: 80
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx1
spec:
replicas: 2
selector:
matchLabels:
app: nginx-dep1
template:
metadata:
labels:
app: nginx-dep1
spec:
containers:
- name: nginx
image: nginx:latest
ports:
- containerPort: 80
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx2
spec:
replicas: 2
selector:
matchLabels:
app: nginx-dep2
template:
metadata:
labels:
app: nginx-dep2
spec:
containers:
- name: nginx
image: nginx:latest
ports:
- containerPort: 80
---
我会申请这个宣言。
$ kubectl apply -f deployment.yaml
deployment.apps/nginx0 created
deployment.apps/nginx1 created
deployment.apps/nginx2 created
$ kubectl get pod -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nginx0-8466d57dbc-cndw8 1/1 Running 0 43s 192.168.79.124 k8s-worker01 <none> <none>
nginx0-8466d57dbc-pdpxn 1/1 Running 0 43s 192.168.69.215 k8s-worker02 <none> <none>
nginx1-68d8c5cc6d-fsmt7 1/1 Running 0 43s 192.168.69.210 k8s-worker02 <none> <none>
nginx1-68d8c5cc6d-sbcwp 1/1 Running 0 43s 192.168.79.120 k8s-worker01 <none> <none>
nginx2-55f9dc64d5-lqq8w 1/1 Running 0 43s 192.168.69.214 k8s-worker02 <none> <none>
nginx2-55f9dc64d5-rr25f 1/1 Running 0 43s 192.168.79.122 k8s-worker01 <none> <none>
做一个Pod的标记
为了进行动作确认，我们创建了一个index.html文件，并分别在其中写入了「nginx0」、「nginx1」和「nginx2」。
「nginx1」和「nginx2」将分别在更低级的目录下创建。
## nginx0
$ kubectl exec -it nginx0-8466d57dbc-cndw8 /bin/bash
# echo nginx0 > /usr/share/nginx/html/index.html
# exit
## nginx1
$ kubectl exec -it nginx1-68d8c5cc6d-fsmt7 /bin/bash
# mkdir /usr/share/nginx/html/path1
# echo nginx1 >/usr/share/nginx/html/path1/index.html
# exit
## nginx2
$ kubectl exec -it nginx2-55f9dc64d5-lqq8w /bin/bash
# mkdir /usr/share/nginx/html/path2
# echo nginx2 >/usr/share/nginx/html/path2/index.html
# exit
创建NodePort
创建用于Ingress后端的NodePort。对每个Pod（Deployment）进行创建。
---
apiVersion: v1
kind: Service
metadata:
name: node-port0
spec:
type: NodePort
ports:
- name: node-port
protocol: TCP
port: 8080
targetPort: 80
nodePort: 30010
selector:
app: nginx-dep0
---
apiVersion: v1
kind: Service
metadata:
name: node-port1
spec:
type: NodePort
ports:
- name: node-port
protocol: TCP
port: 8080
targetPort: 80
nodePort: 30011
selector:
app: nginx-dep1
---
apiVersion: v1
kind: Service
metadata:
name: node-port2
spec:
type: NodePort
ports:
- name: node-port
protocol: TCP
port: 8080
targetPort: 80
nodePort: 30012
selector:
app: nginx-dep2
我将应用此宣言。
$ kubectl apply -f nodePort.yaml
service/node-port0 created
service/node-port1 created
service/node-port2 created
$ kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 48d
node-port0 NodePort 10.103.174.12 <none> 8080:30010/TCP 6s
node-port1 NodePort 10.104.244.161 <none> 8080:30011/TCP 6s
node-port2 NodePort 10.102.124.27 <none> 8080:30012/TCP 6s
创建Ingress资源
既经过这些准备工作，现在我们开始创建Ingress。
要使用Ingress，需要准备”Ingress资源”和”Ingress控制器”。
在清单上，需要一个请求路径列表。与路径关联的后端是通过serviceName和servicePort进行定义的。还可以选择性地设置主机名。另外，最好也设置一个默认的后端。
Ingress资源 -> Ingress规则、默认的后端
apiVersion: networking.k8s.io/v1beta1
kind: Ingress
metadata:
name: sample-ingress
spec:
rules:
- host: foo.bar.com
http:
paths:
- path: /path1
backend:
serviceName: node-port1
servicePort: 8080
- path: /path2
backend:
serviceName: node-port2
servicePort: 8080
backend:
serviceName: node-port0
servicePort: 8080
我会提交这个宣言书。
$ kubectl apply -f sampleIngress.yaml
ingress.networking.k8s.io/simple-ingress created
$ kubectl get ingress
NAME HOSTS ADDRESS PORTS AGE
sample-ingress foo.bar.com 80 10s
在这个时候，“ADDRESS”是空白的。
安装 Ingress 控制器
据说在GKE上，Ingress控制器默认部署，所以以后不再需要（未确认）。
但是，这次是在本地部署（而且是在个人电脑上，而且是虚拟机），所以我们将使用Nginx Ingress。
NGINX入口控制器 -> 安装指南
请按照「Docker for Mac」的安装步骤进行安装。
※前提是需要已经安装了MetalLB。
请参考MetalLB的安装步骤。
$ kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/nginx-0.30.0/deploy/static/mandatory.yaml
namespace/ingress-nginx configured
configmap/nginx-configuration created
configmap/tcp-services created
configmap/udp-services created
serviceaccount/nginx-ingress-serviceaccount created
clusterrole.rbac.authorization.k8s.io/nginx-ingress-clusterrole created
role.rbac.authorization.k8s.io/nginx-ingress-role created
rolebinding.rbac.authorization.k8s.io/nginx-ingress-role-nisa-binding created
clusterrolebinding.rbac.authorization.k8s.io/nginx-ingress-clusterrole-nisa-binding created
deployment.apps/nginx-ingress-controller created
limitrange/ingress-nginx created
$ kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/nginx-0.30.0/deploy/static/provider/cloud-generic.yaml
service/ingress-nginx created
我会确认。 (Wǒ huì .)
$ kubectl -n ingress-nginx get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
ingress-nginx LoadBalancer 10.102.172.38 10.20.30.150 80:31001/TCP,443:32326/TCP 39s
$ kubectl get pod -n ingress-nginx
NAME READY STATUS RESTARTS AGE
nginx-ingress-controller-7f74f657bd-8l8c6 1/1 Running 0 21m
安装Ingress控制器后，Ingress资源的ADDRESS字段将显示IP地址。
$ kubectl get ingresses.
NAME HOSTS ADDRESS PORTS AGE
simple-ingress foo.bar.com 10.20.30.150 80 84m
确认动作
我們從gateway發送請求並確認回應。
$ curl http://10.20.30.150/path1/index.html -H 'Host: foo.bar.com'
nginx1
$ curl http://10.20.30.150/path2/index.html -H 'Host: foo.bar.com'
nginx2
$ curl http://10.20.30.150/ -H 'Host: foo.bar.com'
nginx0
你回答得很好嘛。 (Nǐ dé .)
总结
我已经确认了设置和操作，但由于多年来只做过低层次的基础结构，对L7层的运作还不太理解。
在这里纠结也没有意义，所以我打算先继续前进，然后再重新确认一次。
