ingress-nginx无法启动

Question

老师，你好
无法启动ingress-nginx,主节点状态：

[root@server01 ingress-nginx]# kubectl get all -n ingress-nginx
NAME                                            READY   STATUS             RESTARTS   AGE
pod/default-http-backend-5c9bb94849-cbtb2       1/1     Running            0          10m
pod/nginx-ingress-controller-5d5dc7d9dd-lgt6d   0/1     CrashLoopBackOff   6          10m

NAME                           TYPE        CLUSTER-IP    EXTERNAL-IP   PORT(S)   AGE
service/default-http-backend   ClusterIP   10.99.17.59   <none>        80/TCP    10m

NAME                                       READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/default-http-backend       1/1     1            1           10m
deployment.apps/nginx-ingress-controller   0/1     1            0           10m

去从节点上面查看系统日志：

Error syncing pod cffc32c6-a0a8-11ea-bcc6-080027a4aa8b (“nginx-ingress-controller-5d5dc7d9dd-lgt6d_ingress-nginx(cffc32c6-a0a8-11ea-bcc6-080027a4aa8b)”), skipping: failed to “StartContainer” for “nginx-ingress-controller” with CrashLoopBackOff: “Back-off 5m0s restarting failed container=nginx-ingress-controller pod=nginx-ingress-controller-5d5dc7d9dd-lgt6d_ingress-nginx(cffc32c6-a0a8-11ea-bcc6-080027a4aa8b)”

查看pods日志：
kubectl logs nginx-ingress-controller-5d5dc7d9dd-lgt6d -n ingress-nginx
I0528 06:13:26.011841 6 main.go:191] Creating API client for https://10.96.0.1:443
I0528 06:13:26.020737 6 main.go:235] Running in Kubernetes cluster version v1.14 (v1.14.0) - git (clean) commit 641856db18352033a0d96dbc99153fa3b27298e5 - platform linux/amd64
F0528 06:13:26.020841 6 main.go:98] No service with name /default-http-backend found: an empty namespace may not be set when a resource name is provided

再之后不知道从哪个方向查了，从节点的系统日志不是很明确，请老师给指条明路。

Answer 1

刘果国 2020-05-29 09:29:20

你好，先用kubectl describe一下pod，最下面有一些信息。另外kubectl logs时，加一个-f，等待看一会，注意error级别的日志

• 提问者 Allen_Liu_1106 #1

  感谢老师，解决了，我居然把PID看成端口号了。。。

Answer 2

提问者 Allen_Liu_1106 2020-05-29 09:56:56

老师早上好，

我网上查了下，网上有人给出回答，增加API SERVER配置到mandatory.yaml，于是：

1. 在mandatory.yaml中增加了api server的配置- --apiserver-host=http://192.168.0.108:2653

2. kubectl delete -f  mandatory.yaml

3. kubectl create -f  mandatory.yaml

4. kubectl get all -n ingress-nginx

   pod running,一直不ready，output：

   pod/nginx-ingress-controller-f9d7cdf74-6k8gf   0/1     Running   0          28s

5.  kubectl logs -f pod/nginx-ingress-controller-f9d7cdf74-6k8gf -n ingress-nginx

   -------------------------------------------------------------------------------

   NGINX Ingress controller

    Release:    0.19.0

    Build:      git-05025d6

   Repository: https://github.com/kubernetes/ingress-nginx.git

   -------------------------------------------------------------------------------

   nginx version: nginx/1.15.3

   I0529 01:49:29.860351       6 main.go:191] Creating API client for http://192.168.0.108:2653

   查看日志，到这里就卡住不动了。

6. 去node上面，查看系统日志

   May 28 21:52:49 server04 kubelet[1574]: E0528 21:52:49.180550    1574 pod_workers.go:190] Error syncing pod 89dcea82-a14e-11ea-9b5d-080027a4aa8b ("nginx-ingress-controller-f9d7cdf74-6k8gf_ingress-nginx(89dcea82-a14e-11ea-9b5d-080027a4aa8b)"), skipping: failed to "StartContainer" for "nginx-ingress-controller" with CrashLoopBackOff: "Back-off 1m20s restarting failed container=nginx-ingress-controller pod=nginx-ingress-controller-f9d7cdf74-6k8gf_ingress-nginx(89dcea82-a14e-11ea-9b5d-080027a4aa8b)"
   

• 刘果国 #1

  192.168.0.108:2653是个啥地址了？apiserver应该是6443呀，还应该是是https的而不是http哦