local.location not set in wait-for-cluster.sh script

[pedromctech]

Hi guys. I'm having this issue from today:

module.gke-cluster.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0] (local-exec): Executing: ["/bin/sh" "-c" "PATH=/google-cloud-sdk/bin:$PATH\n.terraform/modules/gke-cluster/modules/beta-private-cluster/scripts/wait-for-cluster.sh s4nhub-production bio4n-k8s-cluster-prod\n"]
module.gke-cluster.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0] (local-exec): Waiting for cluster s4nhub-production//bio4n-k8s-cluster-prod to reconcile...
module.gke-cluster.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0] (local-exec): ERROR: (gcloud.container.clusters.list) Term operand expected [name=bio4n-k8s-cluster-prod AND location= *HERE*].
module.gke-cluster.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0] (local-exec): Unable to get status for s4nhub-production//bio4n-k8s-cluster-prod
Error: Error running command 'PATH=/google-cloud-sdk/bin:$PATH
.terraform/modules/gke-cluster/modules/beta-private-cluster/scripts/wait-for-cluster.sh s4nhub-production bio4n-k8s-cluster-prod
': exit status 1. Output: Waiting for cluster s4nhub-production//bio4n-k8s-cluster-prod to reconcile...
ERROR: (gcloud.container.clusters.list) Term operand expected [name=bio4n-k8s-cluster-prod AND location= *HERE*].

This is my cluster:

module "gke-cluster" {
  source     = "github.com/terraform-google-modules/terraform-google-kubernetes-engine//modules/beta-private-cluster"
  project_id = "s4nhub-production"
  name       = join("-", ["bio4n-k8s-cluster", terraform.workspace])
  regional   = true
  region     = "us-east1"
  ...

It seems that location parameters are not being set in wait-for-cluster.sh script. Thanks for your support!

[bharathkkb]

I think this maybe the destroy hook complaining as it recreates to accept location due to #734
Could you try tainting it
terraform taint module.gke-cluster.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0]

[morgante]

@bharathkkb I think we'll need to make the script more robust for destroy hooks (if location isn't passed, skip that part in the filter).

[bharathkkb]

@morgante yes PR incoming

[bharathkkb]

@pedrodotmc could you try with the fix branch

module "gke" {
  source                 = "github.com/terraform-google-modules/terraform-google-kubernetes-engine?ref=fallback-wait-for-cluster"
....
}

I tried an upgrade from 12.0 to this branch and it works

module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0]: Destroying... [id=6413876306447008812]
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0]: Provisioning with 'local-exec'...
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_command[0]: Destroying... [id=7056368053423075778]
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0] (local-exec): Executing: ["/bin/sh" "-c" "PATH=/google-cloud-sdk/bin:$PATH\n.terraform/modules/example.gke/scripts/wait-for-cluster.sh project-id gke-on-vpc-cluster\n"]
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_command[0]: Destruction complete after 0s
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0] (local-exec): Waiting for cluster project-id//gke-on-vpc-cluster to reconcile...
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_command[0]: Creating...
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_command[0]: Provisioning with 'local-exec'...
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_command[0] (local-exec): Executing: ["/bin/sh" "-c" "PATH=/google-cloud-sdk/bin:$PATH\n.terraform/modules/example.gke/scripts/wait-for-cluster.sh project-id gke-on-vpc-cluster us-east4 \n"]
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_command[0] (local-exec): Waiting for cluster project-id/us-east4/gke-on-vpc-cluster to reconcile...
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0] (local-exec): Cluster is ready!
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0]: Destruction complete after 1s
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0]: Creating...
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_destroy_command[0]: Creation complete after 0s [id=7768141382647091032]
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_command[0] (local-exec): Cluster is ready!
module.example.module.gke.module.gcloud_wait_for_cluster.null_resource.run_command[0]: Creation complete after 1s [id=1626491905156619176]

Apply complete! Resources: 2 added, 0 changed, 2 destroyed.

[pedromctech]

Tainting resource run_destroy_command works like a charm. Thanks @bharathkkb !