Group 13

Follow the steps below to set up the entire project using Docker Compose.

1. Turn on Docker Desktop
2. Copy the .env.sample file into a .env and fill in the environment variables with the required secrets (Only the .env at root directory is needed, .env in subdirectories are deprecated)
3. Change directory to the scripts folder and run chmod +rwx <script> for both scripts
4. Ensure that all the directories have their environment variables setup as well by following the respective README.md's in the directories
5. For TA reviewing this assignment or production environment, simply run docker compose -f docker-compose.prod.yaml up --build. For development, run docker compose up --build
6. Go to http://localhost:3000 and interact with Peerprep!

For development

For development, we are using Dockerfile.dev for all backend microservices to allow for hot-reload using air. Since we do not employ multi-stage builds for development environments, the image size is usually much larger than it should be.

For production

For production, we are using Dockerfile.prod for all services. These files employ a multistage build process which ensures the final image is lean and only contains the minimum amount of files to containerise the service.

Google Cloud CLI helpful commands

To use kubectl with a specific project

gcloud container clusters get-credentials ${CLUSTER_NAME} \
--project=${PROJECT} \
--region=${REGION}

For the purpose of this project, an example of the above command would be

gcloud container clusters get-credentials peerpreps --project=smooth-research-397708 --region=asia-southeast1

Now, you can run commands like:

• kubectl get nodes: Gets the nodes available in your cluster
• kubectl get deployments: Track status of deployments
• kubectl get pod: See pods created by deployment
• kubectl get services: Get all services

Gets the project ID

gcloud config get-value project

FAQ / Troubleshooting

1. Running docker compose up --build returns a "database service unhealthy" error and terminates.

   Simply rerun the command and this issue should go away. The reason for this is because the script is designed to terminate on error, and database existing is considered an error.