5_import_test

Example: Importing Existing Resources

This example demonstrates how to import existing CloudPilot AI-managed resources into Terraform state, allowing you to manage them with Terraform going forward.

Overview

If you have already set up CloudPilot AI for your EKS cluster (e.g. through the console or API), you can import those resources into Terraform. After importing, Terraform will track and manage the existing configuration — no need to recreate anything.

Import only requires a valid CloudPilot AI API key. No AWS credentials, profile, or kubeconfig are needed — all configuration is fetched from the CloudPilot AI API.

Prerequisites

• Terraform >= 1.5 (required for import blocks and terraform plan -generate-config-out)
• CloudPilot AI API key — See CloudPilot AI API Key Documentation
• Cluster ID — Find this in the CloudPilot AI Console under your cluster settings

Usage

Step 1: Configure Variables

cp terraform.tfvars.example terraform.tfvars

Edit terraform.tfvars with your CloudPilot AI API key.

Step 2: Set the Cluster ID

Edit main.tf and replace <CLUSTER_ID> with your actual cluster ID in both import blocks:

import {
  to = cloudpilotai_eks_cluster.imported
  id = "your-cluster-id-here"
}

import {
  to = cloudpilotai_workload_autoscaler.imported
  id = "your-cluster-id-here"
}

You can find the cluster ID in the CloudPilot AI Console under your cluster settings.

Step 3: Initialize Terraform

terraform init

Step 4: Generate Configuration

Run terraform plan with -generate-config-out to auto-generate Terraform configuration from the remote state:

terraform plan -generate-config-out=generated.tf

This creates a generated.tf file containing the full resource configuration as it currently exists in CloudPilot AI. Review this file carefully before proceeding.

Step 5: Apply (Import into State)

terraform apply

This imports the existing resources into your Terraform state without making any changes to the actual infrastructure.

Step 6: Verify

terraform state show cloudpilotai_eks_cluster.imported
terraform state show cloudpilotai_workload_autoscaler.imported
terraform plan   # Should show no changes

After Importing

Once the import is complete:

1. Merge configuration — Move the generated configuration from generated.tf into your main .tf files
2. Remove import blocks — Delete or comment out the import blocks in main.tf (they are only needed once)
3. Add operational parameters — If you plan to manage the cluster with Terraform (not just track it), add parameters like aws_profile or kubeconfig to the resource configuration for future CRUD operations
4. Verify — Run terraform plan to confirm there are no unintended changes

Alternative: CLI Import

You can also import resources one at a time using the Terraform CLI:

terraform import cloudpilotai_eks_cluster.imported <CLUSTER_ID>
terraform import cloudpilotai_workload_autoscaler.imported <CLUSTER_ID>

Note: The CLI approach requires you to write the resource configuration manually before importing. The import block approach with -generate-config-out is recommended as it auto-generates the configuration for you.