< fast piano 59234> Rancher pods CrashLoopBackOff and I cann rancher-users #general

Title

faint-dentist-51157

07/27/2022, 12:14 AM

@fast-piano-59234 Rancher pods CrashLoopBackOff and I cannot Ping 10.43.0.1 using

dnsutils

pod. Can you please ping me back when you have time? Thanks for your help. I appreciate it. RKE: v1.3.8, Rancher: v2.6.6, Kubernetes: v1.22.7, Docker: v20.10.12

My issue is similar to the URL: https://github.com/rancher/rancher/issues/10322

I appreciate if anyone can help on this issue. I can nslookup any hosts like google or intranet vm's, but ping 10.43.0.1 says no route to host.

root@dnsutils:/# ping 10.43.0.1
PING 10.43.0.1 (10.43.0.1) 56(84) bytes of data.
From 10.194.252.245 icmp_seq=1 Destination Host Unreachable
From 10.194.252.245 icmp_seq=2 Destination Host Unreachable

Additionally, when I am creating Ingress manually (Helm failed): I am getting the following error:

kubectl apply -f ingress.yaml

Error from server (InternalError): error when creating "ingress.yaml": Internal error occurred: failed calling webhook "<http://validate.nginx.ingress.kubernetes.io|validate.nginx.ingress.kubernetes.io>": Post "<https://ingress-nginx-controller-admission.ingress-nginx.svc:443/networking/v1/ingresses?timeout=10s>": remote error: tls: internal error

fast-piano-59234

08/19/2022, 8:49 AM

What OS are you using? Is there a host firewall that you can disable? Sounds a lot like EL like with firewalld or custom iptables

faint-dentist-51157

08/19/2022, 1:23 PM

It is CentOS 7.9 Kernel 3.10.0-1160.36.2.el7.x86_64. I finally gave up on RKE and switched to REK2 as it avoids Docker dependency. I successfully brought up Rancher v2.6.6 on RKE2 Kubernetes v1.23.9+rke2r1, with exceptions of lot of helm-operations pods in cattle-system erroring.

10 Views

#general Join Slack