Install Terraform

If you use a package manager on macOS, Windows, or Linux, you can use it to install Terraform.

First, install the HashiCorp tap (official Homebrew repository):

brew tap hashicorp/tap

Install Terraform:

brew install hashicorp/tap/terraform

Check your Terraform version:

terraform -version

Write configuration

Terraform configuration files are plain text files written in HashiCorp Configuration Language (HCL) and end with .tf.

We recommend using consistent formatting. Run:

terraform fmt

Initialize your workspace

Before applying, initialize the working directory so Terraform can download and install providers:

terraform init

Validate your configuration:

terraform validate

Apply infrastructure

Terraform applies changes in two steps: create an execution plan, then apply it.

Terraform creates an execution plan for the changes it will make. Review this plan to ensure that Terraform will make the changes you expect.

Once you approve the execution plan, Terraform applies those changes using your workspace's providers.

This workflow ensures that you can detect and resolve any unexpected problems with your configuration before Terraform makes changes to your infrastructure.

cd deployment/envs/gcp/prod
terraform plan -out tf.plan
terraform apply tf.plan

GCP Deployment (dev / staging / prod)

Terraform configurations for deploying Mega to GCP are under:

• deployment/envs/gcp/prod

Each environment directory contains:

• main.tf
• variables.tf
• providers.tf
• versions.tf
• terraform.tfvars.example

Prerequisites

• Install gcloud and authenticate to the target project.
• Ensure you have permissions to create: VPC, Cloud Run, Cloud SQL, Memorystore (Redis), Filestore, GCS, IAM, Cloud Logging/Monitoring.

Recommended API enablement:

gcloud services enable \
  run.googleapis.com \
  artifactregistry.googleapis.com \
  sqladmin.googleapis.com \
  servicenetworking.googleapis.com \
  redis.googleapis.com \
  file.googleapis.com \
  logging.googleapis.com \
  monitoring.googleapis.com

Configure variables

Copy the example file and edit values:

cd deployment/envs/gcp/prod
cp terraform.tfvars.example terraform.tfvars

Sensitive values should be provided via environment variables when using CI/CD:

export TF_VAR_db_username="mega_user"
export TF_VAR_db_password="your-db-password"
export TF_VAR_rails_master_key="your-rails-master-key"

Apply

terraform init
terraform plan
terraform apply

Outputs

Example outputs after deployment:

• app_cloud_run_url – Cloud Run backend (mono) URL
• ui_cloud_run_url – Cloud Run UI (Next.js SSR) URL
• cloud_sql_connection_name – Cloud SQL connection name for the application

Images (ECR Public -> Cloud Run)

• backend (mono): public.ecr.aws/m8q5m4u3/mega/mono-engine:latest
• UI (Next.js): public.ecr.aws/m8q5m4u3/mega/mega-ui:<env>-latest (e.g. staging, demo, openatom, gitmono)

Notes:

• Pulling images from ECR Public does not require additional Terraform resources; just set app_image / ui_image in terraform.tfvars.
• If you need more stable pulls and better in-region performance, you can mirror images into GCP Artifact Registry and switch app_image / ui_image to Artifact Registry URLs.

Verify logging & monitoring

Cloud Run stdout/stderr is exported to Cloud Logging by default. You can verify in Cloud Console:

• Logging: Logs Explorer (resource type cloud_run_revision)
• Monitoring: Cloud Run dashboards

Destroy / rollback

terraform destroy

Inspect state

Terraform stores infrastructure state in terraform.tfstate.

List tracked resources:

terraform state list

Print the full state:

terraform show