0

I am trying to set up linuxserver/ddclient within mikrok8s v1.23.16 on ubuntu. I found that ddclient has issues connecting to checkip.dyndns.org.

When I create a shell instance in the ddclient-pod and try to ping any IP, I get no response - so the pod seems to have no connection to the outside world.

Any ideas how to solve this? At this point, I don't even know where to start looking.

Edit: typo

petwri
  • 553
  • 1
  • 4
  • 11
  • Do you have a DNS server like `CoreDNS` up and running in the cluster? – Sibtain Feb 24 '23 at 14:55
  • Yes, CoreDNS is up and running. `kubectl get pods -n kube-system` shows the pod `coredns-64c6478b6c-jzsq4` in the state `Running`. – petwri Feb 24 '23 at 16:48
  • But I have also seen that there are 2 `calico-node`-pods, one is `Running`, the second one is in the State `CrashLoopBackOff`. – petwri Feb 24 '23 at 16:53
  • Have you tried running `microk8s enable dns`? Also go over [Debugging DNS Resolution](https://kubernetes.io/docs/tasks/administer-cluster/dns-debugging-resolution/) – Sibtain Feb 24 '23 at 17:06
  • CoreDNS is running. I have enabled it. It is not an nslookup issue, I also cannot ping raw IP-addresses. – petwri Feb 24 '23 at 17:32

1 Answers1

0

I ended up to completely uninstall the microk8s-cluster, re-apply all manifests, then it worked.

Seems like some recent hardware changes and together with that new IPs for all nodes in the cluster messed up a few things. I had several conflicting replicas for calico as well as some services with messed up endpoints.

Fresh install fixed it all.

petwri
  • 553
  • 1
  • 4
  • 11