How do you troubleshoot Helm and Tiller?

February 21, 2019 2.7k views
Kubernetes Ubuntu

I was able to deploy several charts a few days ago but starting yesterday I cannot get helm to connect with tiller. I’ve tried updating to 2.12.3 but still having the issue.

There are no logs on the tiller pod, and debug doesn’t show any verbose output.

What works:

  • helm init with service account
  • helm init upgrade (it does install tiller v2.12.3

What doesn’t work:

  • helm version (the server portion hangs)
  • helm list and every other helm command that interacts with tiller

What I did:

  • deleted the service account, clusterrolebinding, service, deployment, secret, etc
  • re-setup everything using the directions in the post below

Just as a reference, I followed all the instructions in this post: https://www.digitalocean.com/community/tutorials/how-to-install-software-on-kubernetes-clusters-with-the-helm-package-manager

Did something change on the clusters or is there something that I’m missing?

1 Answer

Hey friend!

I ran this by Mike & Nicholas on our support team and they recognized the question from a ticket today. It sounds like this was some kind of local issue on the client system, but we never figured out what it was. Any chance it could be Windows Firewall or an anti-virus blocking traffic? It’s an off the wall theory, of course.

If you find out what it was, feel free to update here in case anyone else experiences the same thing and finds this later :)

Jarland

Have another answer? Share your knowledge.