Soundwav

A full production backend API built with these tech stacks:

• REST API: Django and Django REST Framework.
• Database: PostgresSQL.
• Unit Testing: Pytest.
• Packaging Management: Poetry.
• Containerization: Docker and Docker Compose.
• Cloud Provider: GCP (Google Cloud):
  • Google Cloud Compute Engine.
  • Google Cloud Storage.
  • Google Cloud SQL.
  • Google Cloud Container Registry.
• Infrastructure as Code: Terraform.
• CI/CD: Jenkins.
• Version Control: Git and GitHub.

---

Table of Contents

• Run The Backend
• Setup The Infrastructure
• Deployment
  • Deploy Manually

Backend

Set the environment variables:

• Copy backend/.env.sample/ folder and rename it to backend/.env/.

Run the base environment locally:

• Update the backend/.env/.env.base file.
• Run Docker Compose:

  docker compose -f backend/.docker-compose/base.yml up -d --build

• Run Pytest:

  docker exec -it soundwav_base_django /bin/bash -c "/opt/venv/bin/pytest"

Run the production environment locally:

• Get the environment variables from the infrastructure:

  python scripts/get_infra_output.py --compose=infrastructure/.docker-compose.yml --module=gcp

• Update the backend/.env/.env.production file.
• Run Docker Compose:

  docker compose -f backend/.docker-compose/production.yml up -d --build

---

Infrastructure

Setup Terraform Backend:

• Create a new project on Google Cloud Platform.
• Create a service account and download the service key JSON file and rename it to .gcp_creds.json.
• Create a storage on Google Cloud Storage.
• Create a file and name it to .backend.hcl under infrastructure folder.
• Copy the content of file .backend.hcl.sample inside it and fill the values.

Setup Secrets:

• Create a file with the name .secrets.auto.tfvars under infrastructure folder.
• Copy the contents of file .secrets.auto.tfvars.sample inside it and fill the values.

Setup SSH:

• Generate an SSH Key.
• Create a folder with the name .ssh under infrastructure folder.
• Copy id_rsa.pub and id_rsa file to infrastructure/.ssh.

Run Terraform Commands:

• terraform init

  docker compose -f infrastructure/.docker-compose.yml run --rm terraform init -backend-config=.backend.hcl

• terraform plan all

  docker compose -f infrastructure/.docker-compose.yml run --rm terraform plan

• terraform plan gcp

  docker compose -f infrastructure/.docker-compose.yml run --rm terraform plan -target="module.gcp"

• terraform apply all

  docker compose -f infrastructure/.docker-compose.yml run --rm terraform apply --auto-approve

• terraform apply gcp

  docker compose -f infrastructure/.docker-compose.yml run --rm terraform apply -target="module.gcp" --auto-approve

• terraform destroy all

  docker compose -f infrastructure/.docker-compose.yml run --rm terraform destroy --auto-approve

• terraform destroy gcp

  docker compose -f infrastructure/.docker-compose.yml run --rm terraform destroy -target="module.gcp" --auto-approve

• terraform output gcp

  docker compose -f infrastructure/.docker-compose.yml run --rm terraform output gcp

---

Deployment

Deploy Manually

• Create the GCP resources by following the infrastructure section.

• Export values and change them according to your infrastructure:

  export KEY_FILE=.gcp_creds.json;
  export KEY_TYPE=_json_key;
  export HOSTNAME=gcr.io;
  
  export PROJECT_ID=soundwav;
  export IMG_NAME=soundwav;
  export IMG_TAG=latest;
  
  export FINAL_IMAGE=$HOSTNAME/$PROJECT_ID/$IMG_NAME:$IMG_TAG;
  
  export ENVIRONMENT=production;
    
  export INSTANCE_USER=<YOUR_INSTANCE_USER>;
  export INSTANCE_IP=<YOUR_INSTANCE_IP>;

• Login to GCP Container Registry:

  cat $KEY_FILE | docker login -u $KEY_TYPE --password-stdin https://$HOSTNAME

• Build a Docker image:

  docker build -t $FINAL_IMAGE -f backend/Dockerfile backend --build-arg ENVIRONMENT=$ENVIRONMENT

• Push the Docker image to GCP Container Registry:

  docker push $FINAL_IMAGE

• Copy the env file and the run script to the server:

  rsync backend/.gcp_creds.json backend/.env/.env.$ENVIRONMENT scripts/run_backend.py $INSTANCE_USER@$INSTANCE_IP:/home/$INSTANCE_USER

• Login to GCP Container Registry on the server:

  ssh $INSTANCE_USER@$INSTANCE_IP "cat $KEY_FILE | docker login -u $KEY_TYPE --password-stdin https://$HOSTNAME"

• Run the script on the server:

  ssh $INSTANCE_USER@$INSTANCE_IP "python3 run_backend.py --env=.env.$ENVIRONMENT --image=$FINAL_IMAGE"

---