Terraform error for kubernetes_manifest resource on GKE cluster

Question

I have a pretty standard custom GKE cluster deployed by Terraform. The cluster has following resource was deployed:

resource "kubernetes_manifest" "backendconfig_health_check" {
  manifest = {
    "apiVersion" = "cloud.google.com/v1"
    "kind"       = "BackendConfig"
    "metadata"   = {
      "name"      = "svc-health-check"
      "namespace" = "svc-primary"
    }
    "spec" = {
      "healthCheck" = {
        "healthyThreshold" = 1
        "port"             = 8090
        "requestPath"      = "/api/v1/health"
        "type"             = "HTTP"
      }
    }
  }

  depends_on = [ google_container_cluster.primary]
}

All went well but after some time upon terraform plan I started to get the following error:

╷
│ Warning: Attribute not found in schema
│ 
│   with module.gke.kubernetes_manifest.backendconfig_health_check,
│   on gke/kubernetes_manifest_backendconfig_health_check.tf line 1, in resource "kubernetes_manifest" "backendconfig_health_check":
│    1: resource "kubernetes_manifest" "backendconfig_health_check" {
│ 
│ Unable to find schema type for attribute:
│ metadata.clusterName
│ 
│ (and 6 more similar warnings elsewhere)
╵
Failed generating plan JSON
Exit code: 1

Failed to marshal plan to json: error in marshalResourceDrift: failed to encode refreshed data for module.gke.kubernetes_manifest.backendconfig_health_check as JSON: attribute "object": attribute "metadata": attribute "clusterName" is required

And looks like that clusterName is not a required metadata attribute. It does not exist.

Terraform has the following providers config:

required_providers {
  google = {
    source  = "hashicorp/google"
    version = "4.76.0"
  }

  google-beta = {
    source  = "hashicorp/google-beta"
    version = "4.76.0"
  }

  #  https://registry.terraform.io/providers/hashicorp/kubernetes/latest/docs
  kubernetes = {
    source  = "hashicorp/kubernetes"
    version = "2.19.0"
  }

Also terraaform state show module.gke.kubernetes_manifest.backendconfig_health_check command returns in object attribute like below:

...
  object   = {
    apiVersion = "networking.gke.io/v1beta1"
    kind       = "FrontendConfig"
    metadata   = {
      ... ... ...
      clusterName                = null
      ... ... ...
 
    } 
...

I haven't changed anything in Terraform scripts So I am searching a solution for this.

Any help will be appreciated! Thank you!

Answer 1

After all the solution was to remove invalid Terraform states with terraform state rm __resource__ and then import them with terraform import.