Can not deploy helm chart in Terraform (Kubernetes cluster unreachable)

Question

I am started GKE cluster using Terraform (link), Now I am trying to release the helm charts on the cluster, and tried for "Nginx Ingress" helm chart which is as follow:

resource "helm_release" "ingress" {
  name       = "ingress"
  repository = "https://kubernetes.github.io/ingress-nginx"
  chart      = "ingress-nginx"
}

Terraform Plan:

Terraform will perform the following actions:

  # helm_release.ingress will be created
  + resource "helm_release" "ingress" {
      + chart            = "ingress-nginx"
      + disable_webhooks = false
      + force_update     = false
      + id               = (known after apply)
      + metadata         = (known after apply)
      + name             = "ingress"
      + namespace        = "default"
      + recreate_pods    = false
      + repository       = "https://kubernetes.github.io/ingress-nginx"
      + reuse            = false
      + reuse_values     = false
      + status           = "DEPLOYED"
      + timeout          = 300
      + verify           = false
      + version          = "2.3.0"
      + wait             = true
    }

But I am getting an error

Error: Kubernetes cluster unreachable: Get https://35.232.164.12/version?timeout=32s: dial tcp 35.232.164.12:443: i/o timeout

  on helm.tf line 36, in resource "helm_release" "ingress":
  36: resource "helm_release" "ingress" {

How is this error related to deploying multiple Helm charts? The error is saying you don't have network connectivity to the API server. — ydaetskcoR
@ydaetskcoR, I am new to this terraform and I don't know much about it — Ravindra Gupta

Harsh Manvar Harsh Manvar · Accepted Answer · 2020-05-27T12:38:17

Here terraform not able to create a connection with Kubernetes cluster. cluster is unreachable.

Also using terraform also have to first setup tiller then and then helm chart will work.

in terraform, you can use helm provider, if are using service account you add existing one or create one if required

provider "helm" {
  service_account = "${kubernetes_service_account.helm_account.metadata.0.name}"
  tiller_image = "gcr.io/kubernetes-helm/tiller:${var.helm_version}"
  #install_tiller = false

  kubernetes {
    host                   = "${google_container_cluster.<name>.endpoint}"
    token                  = "${data.google_client_config.current.access_token}"

    client_certificate     = "${base64decode(google_container_cluster.data-dome-cluster.master_auth.0.client_certificate)}"
    client_key             = "${base64decode(google_container_cluster.data-dome-cluster.master_auth.0.client_key)}"
    cluster_ca_certificate = "${base64decode(google_container_cluster.data-dome-cluster.master_auth.0.cluster_ca_certificate)}"
  }

one helm started running you can start setting up other releases using helm package manager.

you can also check this answer for more details : Deploying Helm workloads with Terraform on GKE cluster

Can not deploy helm chart in Terraform (Kubernetes cluster unreachable)

1 Answers