terraform-gcp

Terraform for Google Cloud Platform

Initial Setup - IAM Service Account

Download and install Google Cloud CLI for your platform, following the instructions on the page.

On the GCP Console, create a new Service Account with the roles of:

• Editor
• Service Usage Admin

Next, access the created service_account, and create a 'New Key' with Key type: JSON, and save it somewhere safe on your workstation

Now, export the environment variables GOOGLE_APPLICATION_CREDENTIALS pointing to the full path where the .json credentials file was downloaded/saved:

export GOOGLE_APPLICATION_CREDENTIALS=/path/to/gcp-credentials.json

Up & Running with Terraform

1.: Save Terraform state on GCP:

In Google Cloud Storage, create a bucket that Terraform will use as its backend to save state:

2. Configure Terraform backend for GCS:

2.1. In backend.tf, edit the bucket to the name of the bucket you created in the step above

terraform {
    backend "gcs" {
        bucket  = "iobruno-gcp-labs-tfstate"
    }
}

2.2. Initialize Terraform modules and backend with:

terraform init

2.3. Edit the variables on terraform.tfvars:

Edit the values of the variables (project_id, lakehouse_raw_bucket, bigquery_raw_nyc_tlc) to your suit your project

2.4. Create the resources with Terraform

terraform plan

terraform apply --auto-approve

Terraform Best Practises

Following the best practises for Terraform,

Variables that might contain sensitive information were set on terraform.tfvars, which, for educational purposes only, is set not to be ignored in version control.

TODO:

• Configure Google Cloud Storage as the backend for Terraform States
• Extract sensitive data from variables.tf into *.tfvars
• Consider using Terraform modules as the project expands further