0

When I try to create a GKE cluster it consistently fails at the last stage, when the cluster is being health checked. Both the gcloud cli and the GCP console fail with:

All cluster resources were brought up, but: only 0 nodes out of 3 have registered; this is likely due to Nodes failing to start correctly; try re-creating the cluster or contact support if that doesn't work.

The cli command used was: gcloud container clusters create gitlab-spring --num-nodes=3

Looking under Compute Engine I can see the VM instances and Instance Group, but I get the red exclamation indicator in Kubernetes Engine. I'm not sure how to resolve this, as the cli command above is very basic. I have tried recreating the cluster multiple times with the same result.

2
  • Just a follow-up in case others run into this - although this issue was reliably repeatable, I was able to address the situation by creating another GCP project and creating the cluster there. Unsure what was broken with the old project, I had created clusters in it before. Commented Jun 25, 2020 at 21:37
  • I've tried to run yout command and it works for me. There are addional error logg from the command output? From the CLI, what is the error? Commented Jun 26, 2020 at 9:23

1 Answer 1

1

There is a public documentation on this issue mentioning potential causes and possible solutions.

As per my understanding, and also considering the error message where it's failing after 83%. I suspect possibly automatically created firewall rules for ingress may have been either deleted or modified. If you check out this documentation, understand the purpose as this permits health checks of a NodePort Service or network endpoint group (NEG). Also, reconfigured DNS for *.gcr.io. would also result in the same time of issue. Therefore, I suggest carefully check each and every point.

1

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.