I setup a minimal k8s cluster with 3 f1-micro nodes on GCE via the GitLab UI. The cluster is created successfully, but when I try to install helm via the UI, I experience the error described here: https://stackoverflow.com/questions/54163478/internal-certificate-used-when-installing-helm-tiller-kubernetes
In one of the comments someone says the root cause of this error is that the machines are too small and that the smallest machine type required is “n1-standard-1”. Is that really the case, because I don’t see that documented anywhere?
Is there something I can do to make this work? A 3x “n1-standard-1” setup would be too expensive for my use case.