kubernetes - 就绪探测失败，连接在 kubernetes 中被拒绝

Question

我已经在 k8s 中准备了一些 Keycloak 的基本示例 https://github.com/PavelDemyanenko/kubernetes-keycloak/tree/master/kubernetes ，我遇到一个问题，当我在部署结束时部署 Keycloak Helm Chart 时出现这样的错误:

  Normal   Created         3m56s                   kubelet, oauth2-proxy-control-plane  Created container keycloak
  Normal   Started         3m56s                   kubelet, oauth2-proxy-control-plane  Started container keycloak
  Warning  Unhealthy       48s (x16 over 3m18s)    kubelet, oauth2-proxy-control-plane  Readiness probe failed: Get http://10.244.0.9:8080/auth/realms/master: dial tcp 10.244.0.9:8080: connect: connection refused

$ kubectl get pod -o wide

NAME                                READY   STATUS    RESTARTS   AGE   IP            NODE                         NOMINATED NODE   READINESS GATES
oauth2-proxy-example-0              0/1     Running   3          8h    10.244.0.9    oauth2-proxy-control-plane   <none>           <none>
oauth2-proxy-example-postgresql-0   1/1     Running   1          8h    10.244.0.10   oauth2-proxy-control-plane   <none>           <none>

配置有什么问题以及为什么它不能检查就绪探测？

更新

  readinessProbe: |
    httpGet:
      path: /auth/realms/master
      port: 9080

现在我得到:

  Warning  Unhealthy  83s (x2 over 93s)    kubelet, oauth2-proxy-control-plane  Readiness probe failed: Get http://10.244.0.18:9080/auth/realms/master: net/http: request canceled (Client.Timeout exceeded while awaiting headers)

Answer 1

所以基本上我发现超时设置有帮助

 readinessProbe: |
    httpGet:
      path: /auth/realms/master
      port: 9080
    initialDelaySeconds: 30
    timeoutSeconds: 30