https://github.com/atmatto/weatherapp
https://github.com/atmatto/weatherapp
Last synced: 4 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/atmatto/weatherapp
- Owner: atmatto
- Created: 2024-06-05T11:33:15.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2024-08-27T11:58:38.000Z (10 months ago)
- Last Synced: 2024-08-27T13:15:01.143Z (10 months ago)
- Language: JavaScript
- Size: 186 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Weather App
## Development
Run the `dev.sh` script to start up the development environment. This hosts the frontend on port 8000 and exposes the API on port 9000. Hot reload is enabled.
Use the `./dev.sh build` command to force Docker to build the images. Remember to insert the OpenWeatherMap API key into the file `owm-appid.txt` for everything to be functional.
## Production
Static files are served using Nginx in a Docker container. API server is run in another Docker container. They are coordinated using Docker Compose and exposed to the Internet with a [Caddy](https://caddyserver.com/) reverse proxy. The UFW firewall is used for protection. Everything is managed using a Systemd service (`ansible/weatherapp.service`; it uses the `prod.sh` script to turn the app on and off) and will run automatically after a reboot.
## Deployment
The deployment has been tested on Ubuntu 24.04 Minimal. At least 5 GB of disk space is recommended.
Insert the OpenWeatherMap API key into the file `owm-appid.txt`.
Run `terraform init` in the `terraform/` directory.
Insert AWS API keys into the files named `terraform/aws_*_key` and run the `setup-server.sh` script. This will set up the required AWS infrastructure using terraform (`terraform/`). The file `ansible/inventory.yaml` will be automatically generated and the old one backed up.
To ensure proper configuration of HTTPS, add an `A` DNS record for your domain pointing to the newly created EC2 instance's public IP address. Then, update the domain name in `Caddyfile`. Caddy will automatically configure the required certificates.
The server is automatically configured using a [cloud-init](https://cloud-init.io/) file. The only thing that is left is running the script `deploy.sh` which runs the Ansible playbook. It uploads the latest commit on the `main` branch to the server, prepares everything and runs the application.
To manually inspect the server, connect to it using SSH to the account `maintainer`, using an appropriate private key.
### Test virtual machine
The `vm/` directory contains everything needed to simulate a real cloud server using QEMU/KVM. The directory contains an SSH key pair (`id_ed25519`) to be used only for testing purposes. The file `user-data` is identical to the file `terraform/cloud-init.yaml`, but contains the private key.
Run `./qemu.sh setup` to download the OS image and generate a disk image containing cloud-init files. Then, use `./qemu.sh run` command to turn on the VM. Use the `deploy.sh` script (from the `vm/` directory) to deploy the application to the VM using the real Ansible playbook. The application will be exposed on `localhost:2280` (forwarded from port 80 on the VM). Use the `connect.sh` script to inspect the VM with SSH.
## Backend
### Usage
`npm install` – install dependencies
`npm start` – run server
### Environment variables
- `APPID_FILE` – Path to a file containing the OpenWeatherMap API key (optional)
- `APPID` – OpenWeatherMap API key (this overrides the value from APPID_FILE)
- `MAP_ENDPOINT` – `http://api.openweathermap.org/data/2.5`
- `TARGET_CITY` – `Helsinki,fi`
- `PORT` – 9000
### API
`GET /api/weather`
## Frontend
### Usage
`NODE_OPTIONS=--openssl-legacy-provider npm start` – run Webpack in development mode
`NODE_OPTIONS=--openssl-legacy-provider npm run build` – build in production mode into `dist/`
### Environment variables
- `HOST` (devServer) – `0.0.0.0`
- `PORT` (devServer) – `8000`
- `ENDPOINT` – base URL, should end with `/api`, default: `http://0.0.0.0:9000/api`
## Possible improvements:
- Make a better Systemd service which would check the status of the application and restart it automatically in case of any issues.
- Enhance the monitoring and logging of the application.