1

I currently have a rancher(v2.6.8) cluster (k8s version: v.1.24.2) running on a docker container on a CentOS Linux 7 (core) machine with the ingress class defined as below -

apiVersion: networking.k8s.io/v1
kind: IngressClass
metadata:
  annotations:
  creationTimestamp: "2022-09-02T08:29:41Z"
  generation: 1
  labels:
    app.kubernetes.io/component: controller
    app.kubernetes.io/instance: ingress-nginx
    app.kubernetes.io/name: ingress-nginx
    app.kubernetes.io/version: 1.2.0
  managedFields:
  name: nginx
  resourceVersion: "340"
  uid: 34bbbd2e1-jjh-asdfas-asdfa-sdf
spec:
  controller: k8s.io/ingress-nginx

I see the readiness/liveness probes failing for the nginx-ingress controller pods when you describe the pod

Readiness probe failed: Get http://10.xxx.x.x:10254/healthz: dial tcp 10.xxx.x.x:10254: getsockopt: connection refused

Liveness probe failed: Get http://10.xxx.x.x:10254/healthz: dial tcp 10.xxx.x.x:10254: getsockopt: connection refused

I could not find the reasons for the failing readiness/liveness probes and the fix for the same and also not sure how can I restore the nginx-ingress controller daemonset which got accidentally deleted when trying to debug from the kubectl shell option on the rancher UI.

Avi
  • 1,453
  • 4
  • 18
  • 43
  • Has your issue been resolved. If not,refer to this [stackpost](https://stackoverflow.com/questions/58268579/why-kubernetes-reports-readiness-probe-failed-along-with-liveness-probe-faile) which may help you to resolve the error. – Fariya Rahmat Dec 13 '22 at 12:13

0 Answers0