GKE Deploy fails in Australia as no NVIDIA GPUs available unless we modify cluster.jinja

[snowlover173]

/kind bug

What steps did you take and what happened:
Deployed Kubeflow 0.7, 1.0.0 and 1.0.1 , using the documentation for deploying Kubeflow with cli on GCP.
It failed with an error NVIDIA GPUs not available.
In gcp_config/cluster-kubeflow.yaml changed "gpu-pool-max-nodes" from 10 to 0 and it ran.

What did you expect to happen:
no error

Anything else you would like to add:
[Miscellaneous information that will assist in solving the issue.]

Environment:

• Kubeflow version: 0.7, 1.0.0, 1.0.1
• kfctl version: accordingly
• Kubernetes platform: GCP
• Kubernetes version: 1.14
• OS (e.g. from /etc/os-release):

[issue-label-bot]

Issue-Label Bot is automatically applying the labels:

Label	Probability
kind/bug	0.99

Please mark this comment with 👍 or 👎 to give our bot feedback!
Links: app homepage, dashboard and code for this bot.

[jlewi]

Which zone did you try to use?

[jlewi]

I'm closing this as obsolete because with 1.1 we are switching to blueprints which will no longer use
deployment manager.
https://github.com/kubeflow/gcp-blueprints/tree/master/kubeflow

Blueprints use Cloud Config Connector to configure your GKE clusters and you can configure settings like node autoprovisioning using tools like kustomize to generate appropriate YAML specs.