https://rancher.com/ logo
Join Slack
Powered by
This message was deleted.
# k3s
a
This message was deleted.
q
I have a similar issue, I have a service that isn't getting any endpoints. Where did you find the "Skipping service" message?
c
that indicates that there are no ready pods that match the selector. Check that you have pods deployed, and that they are working properly.
q
In my case, I have ready pods, the selector is correct and they work fine, but still no endpoints. I can't find any errors or anything that could explain it in the logs from the api server or in any events. I'm at a loss to how to debug further.
My issue has evolved a little, only to become weirder. When I do 
kubectl get ep
, I get the endpoint, and it has an IP because it has found a target pod. Everything looks fine. But my ingress controller is still logging an error, because it can't find the endpoints for the given service ... It's like some sort of split brain, except I only have 1 server using 1 external datastore.
Everything is working fine with services that has existed for a while, but two new ones have this problem. The new ones were created after I upgraded to v1.29.3+k3s1, not entirely sure exactly which version I was running before, but it was v1.28.something.
c
are you sure that it actually can’t find the endpoints? or are the endpoints flapping when the pod goes unready or is restarted, and it logs an error about it every time that happens.
q
The pods are stable, running with no restarts or any trouble for the last 35 hours / 11 hours. One is nginx and the other is cert-manager solver.
3 Views
Open in Slack
PreviousNext
Powered by