GSP345 | Automating Infrastructure on Google Cloud with Terraform: Challenge Lab

GSP345 | Automating Infrastructure on Google Cloud with Terraform: Challenge Lab.txt

######################################################################################
## Automating Infrastructure on Google Cloud with Terraform: Challenge Lab # GSP345 ##
######################################################################################

====================== Setup : Create the configuration files ======================
Make the empty files and directories in Cloud Shell or the Cloud Shell Editor.
------------------------------------------------------------------------------------

touch main.tf
touch variables.tf
mkdir modules
cd modules
mkdir instances
cd instances
touch instances.tf
touch outputs.tf
touch variables.tf
cd ..
mkdir storage
cd storage
touch storage.tf
touch outputs.tf
touch variables.tf
cd

--------------------------------------------------------------------------------
Add the following to the each variables.tf file, and fill in the GCP Project ID:
--------------------------------------------------------------------------------

variable "region" {
 default = "us-central1"
}

variable "zone" {
 default = "us-central1-a"
}

variable "project_id" {
 default = "<FILL IN PROJECT ID>"
}

------------------------------------------
Add the following to the main.tf file :
------------------------------------------

terraform {
  required_providers {
    google = {
      source = "hashicorp/google"
      version = "3.55.0"
    }
  }
}

provider "google" {
  project     = var.project_id
  region      = var.region

  zone        = var.zone
}


module "instances" {

  source     = "./modules/instances"

}

---------------------------------------------------------------------------------
Run " terraform init " in Cloud Shell in the root directory to initialize terraform.
---------------------------------------------------------------------------------

====================== TASK 1: Import infrastructure ======================

Navigate to Compute Engine > VM Instances. Click on tf-instance-1. Copy the Instance ID down somewhere to use later.

Navigate to Compute Engine > VM Instances. Click on tf-instance-2. Copy the Instance ID down somewhere to use later.

Next, navigate to modules/instances/instances.tf. Copy the following configuration into the file:

--------------------------------------------------------------

resource "google_compute_instance" "tf-instance-1" {
  name         = "tf-instance-1"
  machine_type = "n1-standard-1"
  zone         = var.zone

  boot_disk {
    initialize_params {
      image = "debian-cloud/debian-10"
    }
  }

  network_interface {
 network = "default"
  }
}

resource "google_compute_instance" "tf-instance-2" {
  name         = "tf-instance-2"
  machine_type = "n1-standard-1"
  zone         = var.zone

  boot_disk {
    initialize_params {
      image = "debian-cloud/debian-10"
    }
  }

  network_interface {
 network = "default"
  }
}

--------------------------------------------------------------------------------------------

To import the first instance, use the following command, using the Instance ID for tf-instance-1 you copied down earlier.

------------------------------------------------------------------------------------------
terraform import module.instances.google_compute_instance.tf-instance-1 <Instance ID - 1>
------------------------------------------------------------------------------------------

To import the second instance, use the following command, using the Instance ID for tf-instance-2 you copied down earlier.

------------------------------------------------------------------------------------------
terraform import module.instances.google_compute_instance.tf-instance-2 <Instance ID - 2>
------------------------------------------------------------------------------------------

The two instances have now been imported into your terraform configuration. You can now optionally run the commands to update the state of Terraform. Type yes at the dialogue after you run the apply command to accept the state changes.

----------------
terraform plan
terraform apply
----------------

====================== TASK 2: Configure a remote backend ======================

Add the following code to the modules/storage/storage.tf file:

-------------------------------------------------------------------
resource "google_storage_bucket" "storage-bucket" {
  name          = var.project_id
  location      = "US"
  force_destroy = true
  uniform_bucket_level_access = true
}
-------------------------------------------------------------------

Next, add the following to the main.tf file:

------------------------------------------------------------------
module "storage" {
  source     = "./modules/storage"
}
----------------------------------------------------------------------------

Run the following commands to initialize the module and create the storage bucket resource. Type yes at the dialogue after you run the apply command to accept the state changes.

------------------------
terraform init
terraform apply
------------------------

Next, update the main.tf file so that the terraform block looks like the following. Fill in your GCP Project ID for the bucket argument definition.

-------------------------------------------
terraform {
  backend "gcs" {
    bucket  = "<FILL IN PROJECT ID>"
 prefix  = "terraform/state"
  }
  required_providers {
    google = {
      source = "hashicorp/google"
      version = "3.55.0"
    }
  }
}
--------------------------------------------

Run the following to initialize the remote backend. Type yes at the prompt.

----------------
terraform init
----------------

====================== TASK 3: Modify and update infrastructure ======================

Navigate to modules/instances/instance.tf. Replace the entire contents of the file with the following:

--------------------------------------------------------
resource "google_compute_instance" "tf-instance-1" {
  name         = "tf-instance-1"
  machine_type = "n1-standard-2"
  zone         = var.zone
  allow_stopping_for_update = true

  boot_disk {
    initialize_params {
      image = "debian-cloud/debian-10"
    }
  }

  network_interface {
 network = "default"
  }
}

resource "google_compute_instance" "tf-instance-2" {
  name         = "tf-instance-2"
  machine_type = "n1-standard-2"
  zone         = var.zone
  allow_stopping_for_update = true

  boot_disk {
    initialize_params {
      image = "debian-cloud/debian-10"
    }
  }

  network_interface {
 network = "default"
  }
}

resource "google_compute_instance" "tf-instance-3" {
  name         = "tf-instance-3"
  machine_type = "n1-standard-2"
  zone         = var.zone
  allow_stopping_for_update = true

  boot_disk {
    initialize_params {
      image = "debian-cloud/debian-10"
    }
  }

  network_interface {
 network = "default"
  }
}
--------------------------------------------------------------------------------------------------

Run the following commands to initialize the module and create/update the instance resources. Type yes at the dialogue after you run the apply command to accept the state changes.

----------------
terraform init
terraform apply
----------------

====================== TASK 4: Taint and destroy resources ======================

Taint the tf-instance-3 resource by running the following command:

------------------------------------------------------------------------
terraform taint module.instances.google_compute_instance.tf-instance-3
------------------------------------------------------------------------

Run the following commands to apply the changes:

----------------
terraform init
terraform apply
----------------

Remove the tf-instance-3 resource from the instances.tf file. Delete the following code chunk from the file.

-----------------------------------------------------------
resource "google_compute_instance" "tf-instance-3" {
  name         = "tf-instance-3"
  machine_type = "n1-standard-2"
  zone         = var.zone
  allow_stopping_for_update = true

  boot_disk {
    initialize_params {
      image = "debian-cloud/debian-10"
    }
  }

  network_interface {
 network = "default"
  }
}
--------------------------------------------------------------------

Run the following commands to apply the changes. Type yes at the prompt.

----------------
terraform apply
----------------

====================== TASK 5: Use a module from the Registry ======================
Copy and paste the following into the main.tf file:

----------------------------------------------------------------
module "vpc" {
    source  = "terraform-google-modules/network/google"
    version = "~> 3.2.2"

    project_id   = var.project_id
    network_name = "terraform-vpc"
    routing_mode = "GLOBAL"

    subnets = [
        {
            subnet_name           = "subnet-01"
            subnet_ip             = "10.10.10.0/24"
            subnet_region         = "us-central1"
        },
        {
            subnet_name           = "subnet-02"
            subnet_ip             = "10.10.20.0/24"
            subnet_region         = "us-central1"
            subnet_private_access = "true"
            subnet_flow_logs      = "true"
            description           = "This subnet has a description"
        }
    ]
}
-------------------------------------------------------------------------------

Run the following commands to initialize the module and create the VPC. Type yes at the prompt.

---------------
terraform init
terraform apply
----------------

Navigate to modules/instances/instances.tf. Replace the entire contents of the file with the following:

-------------------------------------------------------
resource "google_compute_instance" "tf-instance-1" {
  name         = "tf-instance-1"
  machine_type = "n1-standard-2"
  zone         = var.zone
  allow_stopping_for_update = true

  boot_disk {
    initialize_params {
      image = "debian-cloud/debian-10"
    }
  }

  network_interface {
 network = "terraform-vpc"
    subnetwork = "subnet-01"
  }
}

resource "google_compute_instance" "tf-instance-2" {
  name         = "tf-instance-2"
  machine_type = "n1-standard-2"
  zone         = var.zone
  allow_stopping_for_update = true

  boot_disk {
    initialize_params {
      image = "debian-cloud/debian-10"
    }
  }

  network_interface {
 network = "terraform-vpc"
    subnetwork = "subnet-02"
  }
}
--------------------------------------------------------------------------------------------

Run the following commands to initialize the module and update the instances. Type yes at the prompt.

---------------
terraform init
terraform apply
----------------

====================== TASK 6: Configure a firewall ======================
Add the following resource to the main.tf file and fill in the GCP Project ID:

------------------------------------------------------------------
resource "google_compute_firewall" "tf-firewall" {
  name    = "tf-firewall"
 network = "projects/<PROJECT_ID>/global/networks/terraform-vpc"

  allow {
    protocol = "tcp"
    ports    = ["80"]
  }

  source_tags = ["web"]
  source_ranges = ["0.0.0.0/0"]
}
-------------------------------------------------------------------------

Run the following commands to configure the firewall. Type yes at the prompt.

---------------------
terraform init
terraform apply
----------------------

######################################################################################
## Automating Infrastructure on Google Cloud with Terraform: Challenge Lab # GSP345 ##
######################################################################################

I've just completed the lab today. deveshase is right. The start-up script makes the task stuck. We just need to remove the metadata_startup_script and keep allow_stopping_for_update.
Remember to double check subnet-01 and subnet-02 in instances.tf to pass the corresponding part.

Thanks for this, you saved me several wasted hours and credits!

please help !!!
for task 2 when i make changes in instances.tf file and try to do terraform import. it gives me error saying - cannot import non-existent remote object. can someone please guide what am i doing wrong?

If someone else happens to struggle with Task 1, the lab instructions are quite misleading.

I completed the lab successfully today (14th October 2024) and I was struggling with that Task 1 as in first attempts I edited the code in a way that it would make only an in-place upgrade, which is the way anyone would do it in the real world, but this will make the Task 1 to fail when running the check on it.

After quite a bit of Googling about this and reading the lab reviews, I found out that the VM's must be destroyed by Terraform and then to be re-created. Which is of course completely wrong way if the intention is to simulate a real world scenario where you want to add manually created VM's to be managed by Terraform.

Adding the metadata according to lab instructions and allowing the VM's to be destroyed and re-created because of this made the lab check go through successfully.

This instruction https://gist.github.com/Syed-Hassaan/e41a83345832666846ee6be0f69c1f36?permalink_comment_id=4388061#gistcomment-4388061 helped me through the rest of the lab.

So in comparison, snippet from deveshase's code without the lab metadata:

""""
resource "google_compute_instance" "tf-instance-2"{
name = "tf-instance-2"
machine_type = "n1-standard-1"
boot_disk {
initialize_params {
image = "debian-10-buster-v20221102"
}
}
network_interface {
network = "default"
access_config {
}
}
allow_stopping_for_update = true
}
""""

and same code with the metadata as it is in the lab instructions which will cause the VM's to be re-created (makes sure both VM's will have this lab metadata):

""""
resource "google_compute_instance" "tf-instance-2"{
name = "tf-instance-2"
machine_type = "n1-standard-1"
boot_disk {
initialize_params {
image = "debian-10-buster-v20221102"
}
}
network_interface {
network = "default"
access_config {
}
}
metadata_startup_script = <<-EOT
#!/bin/bash
EOT
allow_stopping_for_update = true
}
"""""

I also had some version issues with the bucket and VPC, but I just removed the "version" line from the VPC module as well and running terraform init -upgrade fixed the issue. I'm not sure if the key is to leave the versions out from both main.tf provider block and from the VPC module, but anyways that worked for me.