Skip to content

Latest commit

 

History

History
216 lines (162 loc) · 12.8 KB

README.md

File metadata and controls

216 lines (162 loc) · 12.8 KB
Raw

Google Cloud Foundation Blueprint

This repository contains Terraform configuration files based on Google Cloud best practices to deploy a foundational environment on Google Cloud. It provisions resources adhering to the recommendations outlined in the setup checklist. For a complete set of instructions, please refer to the official deployment guide

Architecture

This blueprint was automatically generated by GCP Cloud Setup. The specific resources in your Terraform module can vary based on your initial cloud configuration decisions.

  • folders:

    • Creates a hierarchical structure under your organization for logical grouping of projects (e.g., "Production," "Non-Production," "Development"). Learn more.

  • projects:

    • Provisions projects within the defined folders for isolating resources and applying specific policies. Learn more.

  • network:

    • Configures VPC networks and subnets, including firewall rules for secure communication. Learn more
    • Configures Shared VPC for centralized network management and efficient resource utilization across projects. Learn more

  • iam:

    • Defines Identity and Access Management roles and bindings at the organization, folder, and project levels to control access to resources. Learn more

  • groups:

    • Leverages Google Groups to simplify IAM management by assigning permissions to groups instead of individual users. Learn more

  • service-projects:

    • Service projects are created for both production and non-production environments to provide granular access control to specific subnets. Learn more

  • vpn:

    • Creates hybrid connectivity to your on premises networks using a optional highly available VPN. Learn more

  • log-export:

  • org-policy:

  • monitoring:

    • Centralizes metrics scope so that metrics from all of your projects can be viewed through a single scoping project. Learn more

  • groups:

    • Groups who will be able to use the service projects. Learn more

Prerequisites

Before deploying this blueprint, ensure you have the following:

  • Google Cloud Account: A Google Cloud Platform account with billing enabled.

  • GCP Cloud Shell: Click "Activate Cloud Shell" at the top of the Google Cloud console or navigate to shell.cloud.google.com.

  • Active Billing Account: If you haven't set up a working billing account at cloud-setup/billing, please refer to the Bootstrap section in the appendix.

  • State File: This blueprint utilizes a Google Cloud Storage (GCS) backend to store the Terraform state file. Cloud Setup usually creates a GCS bucket for you as part of generating this blueprint. If the bucket in your backends.tf file does not point to a valid GCS bucket, you will need to create one yourself. See the Bootstrap step in the appendix.

  • Git repository: Code should be committed to a git repository such as GCP Secure Source Manager, GitHub, AzureDevOps or similar depending on your needs. For detailed instructions, see Git Repository in the appendix.

  • Update Variables: Verify the cloud-setup.auto.tfvars file contains your desired organization ID and billing account number.

        org_id          = "<please confirm your organization id here>"
    billing_account = "<please confirm your billing account number here>"

Deployment

To run this Terraform code, you'll need to use either a service account or a user account. This account must have the appropriate IAM roles to provision the resources defined in the blueprint and needs to be authenticated with Google Cloud.

Option 1: Impersonate Service Account (recommended)

This option allows Cloud Shell to impersonate a service account with the necessary permissions to deploy your Terraform infrastructure. This approach aligns with the principle of least privilege, as recommended by Google Cloud's security best practices. Steps:

  1. Create a Service Account: If the terraform deployer service account has not been created:

    gcloud iam service-accounts create terraform-deployer --display-name "Terraform Deployer"

  2. Grant Necessary IAM Permissions:

    • Grant permission for your user to impersonate the newly created service account. Your will need the roles/iam.serviceAccountUser permission to run this command. Learn More
    gcloud iam service-accounts add-iam-policy-binding "serviceAccount:terraform-deployer@PROJECT_ID.iam.gserviceaccount.com" \
--member="user:<YOUR_GCP_EMAIL>" \
--role="roles/iam.serviceAccountTokenCreator"
    • Assign Required Roles below:
    • roles/billing.projectManager
    • roles/billing.user
    • roles/compute.xpnAdmin
    • roles/config.agent
    • roles/logging.configWriter
    • roles/orgpolicy.policyAdmin
    • roles/resourcemanager.folderIamAdmin
    • roles/resourcemanager.folderCreator
    • roles/resourcemanager.folderEditor
    • roles/resourcemanager.projectIamAdmin
    • roles/resourcemanager.projectCreator
    • roles/resourcemanager.projectDeleter
    • roles/serviceusage.serviceUsageConsumer
    • roles/secretmanager.secretAccessor
    • roles/iam.serviceAccountUser
    • roles/storage.objectUser
    gcloud organizations add-iam-policy-binding ORGANIZATION_ID \
--member="serviceAccount:terraform-deployer@PROJECT_ID.iam.gserviceaccount.com" \
--role="ROLE_NAME"  

  3. Setup Service Account Impersonation:

    gcloud auth application-default login --impersonate-service-account terraform-deployer@PROJECT_ID.iam.gserviceaccount.com

  4. Initialize and Apply Terraform:

    • Run terraform init and terraform apply as usual. The gcloud CLI will automatically use the impersonated service account's credentials for authentication with GCP.
  • Note: After you're done with the deployment, it's recommended to revoke the impersonation by closing the terminal session or running gcloud auth revoke.

Option 2: User Deploy on Cloud Shell or Local Machine

This option assumes that your user account (in Cloud Shell or locally) already has the necessary IAM permissions for the Terraform deployment. This approach is convenient for quick deployments or experimentation but can pose security risks if your user account has excessive privileges.

Steps (using Cloud Shell):

  1. Open Cloud Shell: Go to the GCP console and click the Cloud Shell icon in the top right corner.

  2. Verify Permissions:

  • Check your user account's IAM permissions using:
gcloud organizations get-iam-policy ORG_ID
  • Ensure you have the required roles for your Terraform deployment.
  1. Initialize and Apply Terraform:
  • Navigate to your Terraform project directory in Cloud Shell.
  • Run terraform init and terraform apply. Terraform will utilize your user account's credentials for authentication.

Recommendation: If you're using this option, it's highly recommended to follow the principle of least privilege:

  • Assign Permissions Temporarily: Grant your user account the required roles only during the deployment process. Revoke the roles immediately after the deployment is complete.
  • Use gcloud config set auth/impersonate_service_account: If possible, impersonate a service account with limited permissions even within your Cloud Shell session, similar to Option 1.

Important Security Considerations:

  • Principle of Least Privilege: Always grant only the minimum necessary permissions to service accounts and user accounts involved in your Terraform deployments. Regularly review and revoke unnecessary permissions.
  • Secure Secret Management: Use Google Cloud Secret Manager to store sensitive information like API keys and passwords referenced in your Terraform code.
  • Follow Google Cloud's Security Best Practices: Refer to the security checklist and other security documentation for comprehensive guidance on securing your GCP environment.
  • Terraform State Management: Store your Terraform state remotely in a secure location like Google Cloud Storage and enable encryption at rest.

Next Steps

  1. Return to the Cloud Setup experience to set up your team-Level identity and access management.
  2. Deploy your workloads on top of this foundation.
  3. Use GitOps and CICD to manage deployments Learn more.

Appendix

Bootstrap

If you did not configure a billing account admin during the initial setup, you might need to manually create the following resources:

  1. Terraform Host Project and Google Cloud Storage (GCS) Bucket: A dedicated Google Cloud Project is recommended to host your Terraform state files and run Terraform deployments. Additionally, a GCS bucket within this project will store the state files, enabling collaboration and version control.

    # Create the Terraform host project
PROJECT_ID="your-terraform-host-project"
gcloud projects create $PROJECT_ID

# Enable required APIs for the project
gcloud services enable cloudresourcemanager.googleapis.com storage-api.googleapis.com storage.googleapis.com --project=$PROJECT_ID

# Create a Cloud Storage bucket to store Terraform state
BUCKET_NAME="your-terraform-state-bucket"
gsutil mb -p $PROJECT_ID -l US gs://$BUCKET_NAME

#  Configure versioning the bucket for Terraform state
gsutil versioning set on gs://$BUCKET_NAME

    Replace with a bucket name of your configured in your backends.tf. Location can be changed to one of the options listed here

  2. Terraform Service Account Service account to access deploy terraform

    # Create service account
gcloud iam service-accounts create terraform-deployer --display-name "Terraform Deployer"

# Assign permissions to terraform bucket
gsutil iam ch role roles/storage.objectAdmin serviceAccount:terraform-deployer@<PROJECT_ID>.iam.gserviceaccount.com gs://<bucketname>

Configure a Git Repository

  1. Initialize a Git repository (if not already done):

    git init

  2. Add your Terraform files:

    git add .

  3. Commit your changes:

    git commit -m "Initial commit of Terraform foundation blueprint"

  4. Create a Git repository and push your local repository:

    git remote add origin <your-git-repository-url>
git push -u origin main

  5. Clone the Repository (optional): Verify by cloning the repository to your local machine.

    git clone <your-github-repository-url>
cd <your-github-repository>