Fails with "gcloud command not found"

[sdevang]

Hi,

I am running terraform apply from inside terraform cloud (inside the container) and it fails with following error.

Error: Error running command '.terraform/modules/gke/terraform-google-kubernetes-engine-9.0.0/modules/private-cluster/scripts/wait-for-cluster.sh example-prod-infra-core-175c example-prod-infra-core-gke01': exit status 127. Output: Waiting for cluster example-prod-infra-core-gke01 in project example-prod-infra-core-175c to reconcile...
.terraform/modules/gke/terraform-google-kubernetes-engine-9.0.0/modules/private-cluster/scripts/wait-for-cluster.sh: line 28: gcloud: command not found

My GKE config is,

module "gke" {
  source                     = "terraform-google-modules/kubernetes-engine/google//modules/private-cluster"
  project_id                 = module.core_services_project.project_id
  name                       = "example-prod-infra-core-gke01"
  region                     = "europe-west2"
  zones                      = ["europe-west2-a", "europe-west2-b", "europe-west2-c"]
  network_project_id         = var.host_project
  network                    = "shared-vpc"
  subnetwork                 = "example-prod-infra-core-subnet01"
  ip_range_pods              = "example-prod-infra-core-subnet01-secondary"
  ip_range_services          = "example-prod-infra-core-subnet01-secondary"
  http_load_balancing        = true
  horizontal_pod_autoscaling = true
  network_policy             = true
  enable_private_endpoint    = true
  enable_private_nodes       = true
  create_service_account     = false
  add_cluster_firewall_rules = true
  remove_default_node_pool   = true
  service_account            = module.kubernetes_service_account.email
  default_max_pods_per_node  = 32
  master_authorized_networks = [
    {
      cidr_block   = "10.10.20.0/24"
      display_name = "VPC"
    }
  ]
  node_pools = [
    {
      name                      = "default-node-pool"
      machine_type              = "n1-standard-1"
      min_count                 = 1
      max_count                 = 1
      disk_size_gb              = 100
      disk_type                 = "pd-standard"
      image_type                = "UBUNTU"
      auto_repair               = true
      auto_upgrade              = true
      preemptible               = false
      #   initial_node_count = 80
    },
  ]

  node_pools_labels = {
    all = {}

    default-node-pool = {
      default-node-pool = "true"
    }
  }

  node_pools_metadata = {
    all = {}

    default-node-pool = {
      node-pool-metadata-custom-value = "default-node-pool"
    }
  }

  node_pools_tags = {
    all = []

    default-node-pool = [
      "default-node-pool",
    ]
  }
}

[bharathkkb]

Hi @sdevang
You can set skip_provisioners=true to skip those local exec provisioners.

@morgante thoughts on using the gcloud module to install gcloud on terraform cloud? Not sure if that is the right long term approach.
Edit: Looks like we were planning to in #401

[sdevang]

@bharathkkb Thanks for the reply.

I did set skip_provisioner=true and it did work after that.

I also tried installing gcloud in terraform cloud using "terraform-google-modules/terraform-google-gcloud" but it did not help.

[morgante]

Yeah we'd need to fix this in the module via #401. Closing for now.