Monitoring vehicle metrics.
Make sure the Docker engine is running (Mac and Windows need to download the Docker app). To start, run
docker compose up --build
This starts the frontend react app, backend, and a Redis server all at the same time on Docker. The specific environments for the frontend and backend is configured in frontend/Dockerfile and backend/Dockerfile, respectively. Refer to docker-compose.yml for more details on what port and what environment variables to pass into these microservices.
Since Heroku only allows you to deploy one container onto their server (no SSH or anything either), we must consolidate everything into one single container: Dockerfile.heroku. Since each Dockerfile only has one command to run at the end, we will use supervisord.config to run various commands in the background all at once, with other features like restart, logging and such (better than a bash script).
docker build --platform linux/amd64 -f Dockerfile.heroku -t registry.heroku.com/live-timing-dash/web .
docker push registry.heroku.com/live-timing-dash/web
heroku container:release web --app live-timing-dash
Build the image with linux/amd64 to make it compatible with the actual servers running at Heroku. We need to tag each docker container with registry.heroku.com/live-timing-dash/web for it to release properly. Then we push it over to the registry what heroku uses for this specific live-timing-dash app. Then we can then use it to release to the linux.