502 对于每个其他请求入口 nginx controller

Question

我在入口遇到问题，每个替代请求都返回 502 Bad gateway。

apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  name: ingress-service
  annotations:
    kubernetes.io/ingress.class: nginx
    nginx.ingress.kubernetes.io/ssl-redirect: 'false'
spec:
  rules:
  - http:
      paths:
      - path: /test1/
        backend:
          serviceName: test1-service
          servicePort: 8089
      - path: /test2/
        backend:
          serviceName: test2-service
          servicePort: 4444 

--------------用户服务.yaml------------

apiVersion: v1
kind: Service
metadata:
  name: user-service
spec:
  type: NodePort
  ports:
    - port: 8089
      targetPort: 8089
      nodePort: 31520
  selector:
    app: user

----------日志----------------

192.168.99.1 - - [23/Feb/2020:17:02:16 +0000] "GET /user/ HTTP/1.1" 200 13 "-" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.116 Safari/537.36" 427 0.002 [default-user-service-8089] [] 172.17.0.13:8089 13 0.002 200 5a8eb879aca5ebf682712c30d9f4c826
2020/02/23 17:02:22 [error] 2671#2671: *60232 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.99.1, server: _, request: "GET /user/ HTTP/1.1", upstream: "http://172.17.0.7:8089/user/", host: "192.168.99.101"
2020/02/23 17:02:22 [error] 2671#2671: *60232 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.99.1, server: _, request: "GET /user/ HTTP/1.1", upstream: "http://172.17.0.4:8089/user/", host: "192.168.99.101"
2020/02/23 17:02:22 [error] 2671#2671: *60232 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.99.1, server: _, request: "GET /user/ HTTP/1.1", upstream: "http://172.17.0.3:8089/user/", host: "192.168.99.101"
192.168.99.1 - - [23/Feb/2020:17:02:22 +0000] "GET /user/ HTTP/1.1" 502 565 "-" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.116 Safari/537.36" 453 0.005 [default-user-service-8089] [] 172.17.0.7:8089, 172.17.0.4:8089, 172.17.0.3:8089 0, 0, 0 0.001, 0.004, 0.000 502, 502, 502 4e37afefb17069ea66a46e06cb74fbd7
192.168.99.1 - - [23/Feb/2020:17:02:24 +0000] "GET /user/ HTTP/1.1" 200 13 "-" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.116 Safari/537.36" 453 0.001 [default-user-service-8089] [] 172.17.0.13:8089 13 0.000 200 3595217828be889af6f8f1508cefae59

使用 NodePort 访问单个服务时没有问题。

kubectl describe svc test1-service

Name:                     user-service
Namespace:                default
Labels:                   <none>
Annotations:              <none>
Selector:                 app=web
Type:                     NodePort
IP:                       xxxxxxxxxxx
Port:                     <unset>  8089/TCP
TargetPort:               8089/TCP
NodePort:                 <unset>  31520/TCP
Endpoints:                172.17.0.13:8089,172.17.0.3:8089,172.17.0.4:8089 + 1 more...
Session Affinity:         None
External Traffic Policy:  Cluster
Events:                   <none>

Answer 1

我想通了，我不知道这种行为的确切原因，但部署之间存在 label 次冲突。

  labels:
    app: test1

我没有注意到在其他部署中也使用了相同的 label。 更改它们后问题得到解决，但不确定它是否与行为直接相关，因为对我来说每个备用请求都失败了。

Answer 2

尝试删除反斜杠并确保服务正在运行并列出到配置的端口

apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  name: ingress-service
  annotations:
    kubernetes.io/ingress.class: nginx
    nginx.ingress.kubernetes.io/ssl-redirect: 'false'
spec:
  rules:
  - http:
      paths:
      - path: /test1
        backend:
          serviceName: test1-service
          servicePort: 8089
      - path: /test2
        backend:
          serviceName: test2-service
          servicePort: 4444