The error suggests that the cluster might be in an upgrading state.
For that, you can use the Google SDK and determine if it's being upgraded:
gcloud container clusters describe CLUSTER_NAME --zone ZONE/--region REGION
If not, check that you're actually pointing your kubectl
tool to correct cluster:
kubectl config current-context
gcloud
has the advantage that it can automatically configure your kubectl
client:
gcloud container clusters get-credentials CLUSTER_NAME --zone ZONE/--region REGION
If the cluster is being in that state for quite a long time (usually several hours), there might be an error in the cluster backend (since GKE is a managed implementation of Kubernetes). If that's the case, you might want to reach out to support to review your case.