terraform-google-examples

Once the deployment is complete, use the provided test script to verify the setup and retrieve the Load Balancer's IP address.

1. Run ./test.sh and wait for the provisioning to finish.
2. Retrieve the HTTP URL using the load-balancer-ip Terraform output.

./test.sh
echo http://$(terraform output load-balancer-ip)

If Terraform is not already installed on your system, you can use the provided installation script from the parent directory.

../terraform-install.sh

Before running Terraform, you must configure your Google Cloud project and authentication. Replace YOUR_PROJECT with your actual Google Cloud Project ID.

1. Set the project ID in your shell.
2. Configure gcloud to use that project.
3. Authenticate for Terraform using Application Default Credentials (ADC) if you are not running in Cloud Shell.

PROJECT=YOUR_PROJECT

gcloud config set project ${PROJECT}

[[ $CLOUD_SHELL ]] || gcloud auth application-default login
export GOOGLE_PROJECT=$(gcloud config get-value project)

After running terraform apply, you can verify the deployment by performing the following tests:

1. Access the Bastion Host

Use SSH agent forwarding to access the bastion host. This allows you to forward ports to services like Cerebro or Kibana:

eval $(ssh-agent)
ssh-add ~/.ssh/google_compute_engine
eval $(terraform output bastion)

2. Access the Custom Machine

SSH directly into the custom instance:

ssh tf-custom-1

3. Verify NAT Gateway Routing

To ensure the custom machine is correctly routing its external traffic through the NAT gateway, check its external IP address. The returned IP should match the IP of the NAT gateway:

curl http://ipinfo.io/ip

Before running Terraform, you must configure your Google Cloud project and authentication. Replace YOUR_PROJECT with your actual Google Cloud Project ID.

1. Set the project ID in your shell.
2. Configure gcloud to use that project.
3. Authenticate for Terraform using Application Default Credentials (ADC). If you are not in Cloud Shell, run gcloud auth application-default login.

PROJECT=YOUR_PROJECT
gcloud config set project ${PROJECT}

[[ $CLOUD_SHELL ]] || gcloud auth application-default login
export GOOGLE_PROJECT=$(gcloud config get-value project)

Once Terraform has finished applying, you can verify the deployment by running the provided test script and then querying the load balancer's IP address directly via curl.

# Wait for the load balancer to be provisioned
./test.sh

# Verify response from load balancer
curl http://$(terraform output load-balancer-ip)

To avoid ongoing costs, delete all resources created by this Terraform configuration using the destroy command.

terraform destroy

To avoid leaving orphaned resources (like forwarding rules or firewall rules), you must delete the nginx-ingress Helm release specifically before destroying the rest of the infrastructure.

Run the following command to target the release, wait for the GCE controller to clean up, and then destroy everything else:

terraform destroy -input=false -auto-approve -target helm_release.nginx-ingress && sleep 60 && \
terraform destroy

To interact with the GKE cluster directly, configure your local kubectl context using the outputs from Terraform.

1. Get Credentials:

   gcloud container clusters get-credentials $(terraform output cluster_name) --zone $(terraform output cluster_zone)

2. **Verify Connectivity**:
   ```bash
kubectl get pods
helm list

To deploy the infrastructure, navigate to the example directory, ensure Terraform is installed, and run the standard Terraform initialization and application commands.

# Change to the example directory
[[ `basename $PWD` != example-g8s-service-lb ]] && cd example-gke-k8s-service-lb

# Install Terraform (if not present)
../terraform-install.sh

# Initialize and apply
terraform init
terraform apply