Error 400: Autopilot clusters must be regional clusters

[kzos]

1. I am trying to instantiate the ray cluster on GKE following this tutorial: https://github.com/GoogleCloudPlatform/ai-on-gke/blob/main/ray-on-gke/README.md

with below settings in workloads.tfvars:-

project_id = "deployment_work"

cluster_name = "vm-deploy-k8-fsz"

cluster_location = "me-central1-a"

create_cluster = true ================================>

2. But getting this below error:

╷ │ Error: googleapi: Error 400: Autopilot clusters must be regional clusters. │ Details: │ [ │ { │ "@type": "type.googleapis.com/google.rpc.RequestInfo", │ "requestId": "0x4861d29cd332588e" │ } │ ] │ , badRequest │ │ with module.infra[0].module.public-gke-autopilot-cluster[0].module.gke.google_container_cluster.primary, │ on .terraform/modules/infra.public-gke-autopilot-cluster.gke/modules/beta-autopilot-public-cluster/cluster.tf line 22, in resource "google_container_cluster" "primary": │ 22: resource "google_container_cluster" "primary" { │ ╵

Could you please guide me to get create_cluster working.

==================================================> NOTE: Initially i tried to create the kubernetes cluster in the GCP and gave above setting parameters with "create_cluster = false", cluster_name="vm-deploy-k8-fz" to verify my project id and cluster name, terraform apply ran without any error.

but I am not sure how do i attach to the existing cluster on the GCP(there is no guide on this in readme for newbies). $ kubectl get raycluster --> does not return anything :(

provide guide for this as well. ==================================================>

[roberthbailey]

Please set cluster_location to a GCP region, such as me-central1 rather than a zone.

[kzos]

sure, thanks I guess I did the same and went past through and getting new error,

ll mark this as closed and create as new issue.