bug: When using Github Actions to deploy terraform, it returns an error due to helm and a missing environment variable.

[Vermyndax]

Summary

I'm trying to use this module to navigate the bizarre reality known as deploying the AWS load balancer. I finally got this to work locally, but it will not work in Github Actions. The error is thrown:

Error: Kubernetes cluster unreachable: invalid configuration: no configuration has been provided, try setting KUBERNETES_MASTER environment variable

I suspect this is because I have a Kubernetes alias for the provider for this cluster. Does this module support using an alias?

Issue Type

Bug Report

Terraform Version

Terraform v1.4.1
on darwin_arm64
+ provider registry.terraform.io/cloudposse/utils v1.7.1
+ provider registry.terraform.io/hashicorp/aws v4.59.0
+ provider registry.terraform.io/hashicorp/cloudinit v2.3.2
+ provider registry.terraform.io/hashicorp/helm v2.9.0
+ provider registry.terraform.io/hashicorp/kubernetes v2.19.0
+ provider registry.terraform.io/hashicorp/time v0.9.1
+ provider registry.terraform.io/hashicorp/tls v4.0.4

Steps to Reproduce

provider "kubernetes" {
  alias                  = "trogdor_kubernetes"
  host                   = module.trogdor_eks.cluster_endpoint
  cluster_ca_certificate = base64decode(module.trogdor_eks.cluster_certificate_authority_data)

  exec {
    api_version = "client.authentication.k8s.io/v1beta1"
    command     = "aws"
    # This requires the awscli to be installed locally where Terraform is executed
    args = ["eks", "get-token", "--cluster-name", module.trogdor_eks.cluster_name]
  }
}

provider "helm" {
  alias = "trogdor_helm"
  kubernetes {
    host                   = module.trogdor_eks.cluster_endpoint
    cluster_ca_certificate = base64decode(module.trogdor_eks.cluster_certificate_authority_data)

    exec {
      api_version = "client.authentication.k8s.io/v1beta1"
      command     = "aws"
      # This requires the awscli to be installed locally where Terraform is executed
      args = ["eks", "get-token", "--cluster-name", module.trogdor_eks.cluster_name]
    }
  }
}

module "trogdor_eks_load_balancer_controller" {
  source  = "lablabs/eks-load-balancer-controller/aws"
  version = "1.2.0"

  cluster_identity_oidc_issuer     = module.trogdor_eks.cluster_oidc_issuer_url
  cluster_identity_oidc_issuer_arn = module.trogdor_eks.oidc_provider_arn
  cluster_name                     = module.trogdor_eks.cluster_name
  irsa_role_name_prefix            = "${var.environment}-${var.app_name}-trogdor-eks-lb"
}

Expected Results

I wish my Github Action would work :(

Actual Results

Planning failed. Terraform encountered an error while generating this plan.

Error: Kubernetes cluster unreachable: invalid configuration: no configuration has been provided, try setting KUBERNETES_MASTER environment variable

  with module.trogdor_eks_load_balancer_controller.helm_release.this[0],
  on .terraform/modules/trogdor_eks_load_balancer_controller/helm.tf line 1, in resource "helm_release" "this":
   1: resource "helm_release" "this" {

[thetoolsmith]

@Vermyndax I adde this to the kubernetes block within the provider block and it fixed the issue for me.

provider "helm" {
  kubernetes {
    config_path = "~/.kube/config"
  }
}

[jaygridley]

Hi @Vermyndax, in your configuration you have alias defined for both Helm a Kubernetes providers. You might want to pass them explicitly to the module trogdor_eks_load_balancer_controller, see https://developer.hashicorp.com/terraform/language/modules/develop/providers#passing-providers-explicitly for the reference.