Skip to content

Latest commit

 

History

History
116 lines (106 loc) · 4.59 KB

README.md

File metadata and controls

116 lines (106 loc) · 4.59 KB
Raw

ILT 4 Golang Backend with Cloud Run and Cloud SQL

We'll create a simple backend service that stores user data in an SQL database. We'll use CloudSQL-managed service solution from the Google Cloud Platform

Stacks

Development Step

  1. Create .env file and export it as environment variables 
    cd ilt-4
cat << EOT > .env
DB_USER="<your_db_username>"
DB_PASSWORD="<your_db_password>"
DB_HOST="localhost" #For development
DB_HOST_COMPOSE="db" #For development on docker compose
DB_PORT="5432"
DB_NAME="<your_db_name>"
POSTGRESQL_URL="postgres://<your_db_username>:<your_db_password>@<your_db_host>:5432/<your_db_name>?sslmode=disable"
EOT
export $(xargs < .env)
  2. Build a Docker image with this command 
    cd ilt-4
docker build -t <name_ilt e.g ilt4>:<ilt_version e.g 0.1.0> -f ./Dockerfile.prod .
  3. Run docker compose in ./build/docker-compose/ folder 
    cd ilt-4/build/docker-compose
docker compose up -d
  4. Migrate database schema 
    cd ilt-4
migrate -database $POSTGRESQL_URL -path migrations/migrate up
  5. Test the API with Postman
  6. Teardown docker compose and delete image 
    cd ilt-4/build/docker-compose
docker compose down

docker image rm <name_ilt e.g ilt4>:<ilt_version e.g 0.1.0>

Production Step

  1. Push docker image from Development Step # 2 to Google Container Registry 
    docker tag <name_ilt e.g ilt4>:<ilt_version e.g 0.1.0> gcr.io/<your_project_id>/<name_ilt e.g ilt4>:<ilt_version e.g 0.1.0>
docker push gcr.io/<your_project_id>/<name_ilt e.g ilt4>:<ilt_version e.g 0.1.0>
  2. Upload the migration schema that is located on ilt-4/migrations/migrate/*.up.sql to any Google Cloud Storage bucket
  3. Update your .env file from Development Step # 1 with new envars 
    cd ilt-4
cat << EOT >> .env
PROJECT_ID="<your_gcp_project_id>"
REGION="<your_prefered_region>"
ILT_NAME="ilt-4"
CLOUDSQL_NAME_PREFIX="<you_prefered_prefix>"
SUBNET_IP_RANGE="11.1.0.0/16"
PRIVATE_IP_FIRST_ADDRESS="11.255.0.0"
SVPC_IP_RANGE="11.254.0.0/28"
EOT
export $(xargs < .env)
  4. Prepare the terraform envars 
    export GCR_VERSION="$(git describe --tags --abbrev=0 | cut -d '/' -f2)"
export TF_VAR_project="$PROJECT_ID"
export TF_VAR_region="$REGION"
export TF_VAR_ilt_name="$ILT_NAME"
export TF_VAR_image="gcr.io/$PROJECT_ID/$ILT_NAME:$GCR_VERSION"
export TF_VAR_db_user="$DB_USER"
export TF_VAR_db_password="$DB_PASSWORD"
export TF_VAR_db_port="$DB_PORT"
export TF_VAR_db_name="$DB_NAME"
export TF_VAR_name_prefix="$CLOUDSQL_NAME_PREFIX"
export TF_VAR_master_user_name="$DB_USER"
export TF_VAR_master_user_password="$DB_PASSWORD"
export TF_VAR_subnet_ip_range="$SUBNET_IP_RANGE"
export TF_VAR_subnet_region="$REGION"
export TF_VAR_private_ip_first_address="$PRIVATE_IP_FIRST_ADDRESS"
export TF_VAR_svpc_connector_ip_range="$SVPC_IP_RANGE"
  5. Got to ilt-4/infra/gcp-vpc to create a new VPC network with a subnet, private service connection, and serverless VPC access 
    terraform apply # Then type yes
  6. Got to ilt-4/infra/gcp-cloud-sql to create a new CloudSQL instance 
    terraform apply # Then type yes
  7. After CloudSQL was created. Go to GCP Console > Cloud SQL > your_instance
  8. Click Import in the top ribbon. Select the migration schema that you uploaded to Google Cloud Storage. Select the users database. Then start the migration by clicking Import
  9. Got to ilt-4/infra/gcp-cloud-run to create a new Cloud Run container 
    terraform apply # Then type yes
  10. In the terraform output there will be a variable service_url that contains your API URL. You can test your API with it

References