civo/terraform-provider-civo

civo_kubernetes_cluster broken in 1.0.2

dirsigler opened this issue · 2 comments

There seems to be a big fuckup with the civo_kubernetes_cluster in the Civo Terraform Provider Version 1.0.2

If you use the Civo Provider Version 1.0.2 and want to create a Kubernetes Cluster you receive following error:

civo_network.cluster_network: Refreshing state... [id=1dff45ff-2a28-4b66-aa07-9c83c887b7bb]
civo_firewall.cluster_firewall: Refreshing state... [id=fdc9d187-435f-44c9-90cd-13dbfaf8e8fa]
civo_kubernetes_cluster.k3s_cluster: Refreshing state... [id=ff87caa7-2313-463f-9f17-3056d678a671]
╷
│ Error: [ERR] error retrieving the pool for kubernetes cluster error: &errors.errorString{s:"Invalid address to set: []string{\"pools\", \"0\", \"id\"}"}
│
│   with civo_kubernetes_cluster.k3s_cluster,
│   on main.tf line 18, in resource "civo_kubernetes_cluster" "k3s_cluster":
│   18: resource "civo_kubernetes_cluster" "k3s_cluster" {
│

You can test that easily out with:

terraform {
  required_version = "~> 1.0"
  required_providers {
    civo = {
      source  = "civo/civo"
      version = "1.0.1"
    }
  }
}
provider "civo" {
  region = "FRA1"
  token  = var.CIVO_TOKEN
}

variable "CIVO_TOKEN" {
  type        = string
  description = "API Token used for Civo Cloud."
}

resource "civo_network" "cluster_network" {
  label  = "test-3-vpc"
  region = "FRA1"
}

resource "civo_firewall" "cluster_firewall" {
  name       = "test-3-fw"
  region     = "FRA1"
  network_id = civo_network.cluster_network.id
  depends_on = [civo_network.cluster_network]
}

resource "civo_kubernetes_cluster" "k3s_cluster" {
  name             = "test-3-k3s"
  num_target_nodes = 3
  target_nodes_size = "g3.k3s.xsmall"
  applications      = "-Traefik"
  firewall_id       = civo_firewall.cluster_firewall.id
  network_id        = civo_network.cluster_network.id
  region            = "FRA1"
}

If you use above configuration you can create the setup without any problems.
Then change the Provider Version to 1.0.2 and terraform init -upgrade.
During the apply you receive above error.

You can also right away use Version 1.0.2 and instantly receive the error.

Checking now

I received the same error and gladly using version 1.0.1 cluster creation works fine.