Deploy a basic Flask web server by using Terraform

In this tutorial, you learn how to get started with Terraform by using Terraform to create a basic web server on Compute Engine.

In this tutorial, you do the following:

  • Use Terraform to create a VM in Google Cloud.
  • Start a basic Python Flask server.


In this document, you use the following billable components of Google Cloud:

Compute Engine

To generate a cost estimate based on your projected usage, use the pricing calculator. New Google Cloud users might be eligible for a free trial.

When you finish the tasks that are described in this document, you can avoid continued billing by deleting the resources that you created. For more information, see Clean up.

Before you begin

Prepare to start the tutorial.

Select or create a project

  1. In the Google Cloud console, go to the project selector page.

    Go to project selector

  2. Select or create a Google Cloud project.

Set up permissions

Make sure that you have the necessary Compute Engine permissions on your user account:

  • compute.instances.*
  • compute.firewalls.*

Go to the IAM page

Learn more about roles and permissions.

Enable the API

Enable the Compute Engine API.

Enable the API

Start Cloud Shell

Cloud Shell is a Compute Engine virtual machine.

The service credentials associated with this virtual machine are automatic, so there is no need to set up or download a service account key.

Terraform is integrated with Cloud Shell, and Cloud Shell automatically authenticates Terraform, letting you get started with less setup.

Create the Compute Engine VM

First, you define the VM's settings in a Terraform configuration file. Then, you run Terraform commands to create the VM in your project.

Create the directory

Create a new directory. In your new directory, create a file for the Terraform configuration. The contents of this file describe all of the Google Cloud resources to be created in the project.

In Cloud Shell:

mkdir tf-tutorial && cd tf-tutorial

Create the Virtual Private Cloud network and subnet

In this section, you create a Virtual Private Cloud (VPC) network and subnet for the VM's network interface.

Add the following Terraform resources to the file that you created:

resource "google_compute_network" "vpc_network" {
  name                    = "my-custom-mode-network"
  auto_create_subnetworks = false
  mtu                     = 1460

resource "google_compute_subnetwork" "default" {
  name          = "my-custom-subnet"
  ip_cidr_range = ""
  region        = "us-west1"
  network       =

Create the Compute Engine VM resource

In this section, you create a single Compute Engine instance running Debian. In this tutorial, you use the smallest machine type that's available. Later, you can upgrade to a larger machine type.

Add the following google_compute_instance Terraform resource to the file that you created.

# Create a single Compute Engine instance
resource "google_compute_instance" "default" {
  name         = "flask-vm"
  machine_type = "f1-micro"
  zone         = "us-west1-a"
  tags         = ["ssh"]

  boot_disk {
    initialize_params {
      image = "debian-cloud/debian-11"

  # Install Flask
  metadata_startup_script = "sudo apt-get update; sudo apt-get install -yq build-essential python3-pip rsync; pip install flask"

  network_interface {
    subnetwork =

    access_config {
      # Include this section to give the VM an external IP address

The sample code sets the Google Cloud zone to us-west1-a. You can change this to a different zone.

Initialize Terraform

At this point, you can run terraform init to add the necessary plugins and build the .terraform directory.

terraform init


Initializing the backend...

Initializing provider plugins...

Terraform has been successfully initialized!

Validate the Terraform configuration

Optionally, you can validate the Terraform code that you've built so far. Run terraform plan, which does the following:

  • Verifies that the syntax of is correct
  • Shows a preview of the resources that will be created
terraform plan



Plan: 1 to add, 0 to change, 0 to destroy.

Note: You didn't use the -out option to save this plan, so Terraform can't
guarantee to take exactly these actions if you run "terraform apply" now.

Apply the configuration

To create the VM, run terraform apply.

terraform apply

When prompted, enter yes.

Terraform calls Google Cloud APIs to set up the new VM. Check the VM instances page to see the new VM.

Run a web server on Google Cloud

Your next steps are getting a web application created, deploying it to the VM, and creating a firewall rule to allow client requests to the web application.

Add a custom SSH firewall rule

The default-allow-ssh firewall rule in the default network lets you use SSH to connect to the VM. If you'd rather use your own custom firewall rule, you can add the following resource at the end of your file:

resource "google_compute_firewall" "ssh" {
  name = "allow-ssh"
  allow {
    ports    = ["22"]
    protocol = "tcp"
  direction     = "INGRESS"
  network       =
  priority      = 1000
  source_ranges = [""]
  target_tags   = ["ssh"]

Run terraform apply to create the firewall rule.

Connect to the VM with SSH

Validate that everything is set up correctly at this point by connecting to the VM with SSH.

  1. Go to the VM Instances page.

  2. Find the VM with the name flask-vm.

  3. In Connect column, click SSH.

    An SSH-in-browser terminal window opens for the running VM.

For more information, see Connecting to VMs.

Build the Flask app

You build a Python Flask app for this tutorial so that you can have a single file describing your web server and test endpoints.

  1. In the SSH-in-browser terminal, create a file called

  2. Add the following to the file:

    from flask import Flask
    app = Flask(__name__)
    def hello_cloud():
      return 'Hello Cloud!''')
  3. Run


    Flask serves traffic on localhost:5000 by default.

  4. Open a second SSH connection:

    1. Go to the VM Instances page.
    2. Find the VM named flask-vm and click SSH.
  5. In the second SSH connection, run curl to confirm that the greeting that you configured in is returned.


    The output from this command is Hello Cloud.

Open port 5000 on the VM

To connect to the web server from your local computer, the VM must have port 5000 open. Google Cloud lets you open ports to traffic by using firewall rules.

Add the following google_compute_firewall Terraform resource at the end of your file.

resource "google_compute_firewall" "flask" {
  name    = "flask-app-firewall"
  network =

  allow {
    protocol = "tcp"
    ports    = ["5000"]
  source_ranges = [""]

In Cloud Shell, run terraform apply to create the firewall rule.

Add an output variable for the web server URL

  1. At the end of, add a Terraform output variable to output the web server URL:

    // A variable for extracting the external IP address of the VM
    output "Web-server-URL" {
     value = join("",["http://",google_compute_instance.default.network_interface.0.access_config.0.nat_ip,":5000"])
  2. Run terraform apply.

    terraform apply

    When prompted, enter yes. Terraform prints the VM's external IP address and port 5000 to the screen, as follows:

    Web-server-URL = "http://IP_ADDRESS:5000"

    At any time, you can run terraform output to return this output:

    terraform output
  3. Click the URL from the previous step, and see the "Hello Cloud!" message.

    This means that your server is running.


  • If a required API isn't enabled, Terraform returns an error. The error message includes a link to enable the API. After enabling the API, you can rerun terraform apply.

  • If you can't connect to your VM through SSH:

    • Make sure to add the SSH firewall rule.
    • Make sure that your VM includes the tags = ["ssh"] argument.

Clean up

After completing the tutorial, you can delete everything that you created so that you don't incur any further costs.

Terraform lets you remove all the resources defined in the configuration file by running the terraform destroy command:

terraform destroy

Enter yes to allow Terraform to delete your resources.

What's next