Simple application to train some programming and DevOps skills. It contains:
- Backend API (Python, FastAPI, SQLAlchemy, Pydantic)
- Frontend (NodeJS, React)
- Database (default - local SQLite file)
Base API CRUD Python application, working with fastapi, sqlalchemy, pydantic.
By default works with SQLite file-based database stored in the container — all data will be deleted after restart.
You can set your own database connection in backend/.env file or with a DB_CONN variable (See docker-compose.yml).
You can find swagger at /api/docs. Main API enpoint is /api/servers
Fill in the .env file:
mv .env.example .env
nano .env
cd backend
python3 -m pip install -r requirements.py
uvicorn app.main:app
To load some predifined data from dataset.csv use:
python3 load.py
To set backend version which will appear at the very bottom of the frontend page, you should set BACKEND_VERSION environment variable, e.g. BACKEND_VERSION=1.1.2. This value will be in the response on /api/version API endpoint.
Fill in the .env file in backend directory:
cd backend
mv .env.example .env
nano .env
Works as non-root user worker. You can build your own image:
cd backend
docker build -t eto-backend:latest .
docker run --env-file .env -p 8000:8000 eto-backend:latest
Or run my image from DockerHub:
docker run --env-file .env -p 8000:8000 etoosamoe/eto-backend:latest
curl --request POST \
--url http://127.0.0.1:8000/api/servers \
--header 'Content-Type: application/json' \
--header 'accept: application/json' \
--data '{
"hostname": "ci00",
"type": "hw",
"cores": 32,
"ram": 64,
"disk": 1100
}'
Frontend works with NodeJS and uses React. It makes requests to backend API to get some data about servers.
By default REACT_APP_BACKEND_URL is empty, which means frontend will use relative URLs.
If you want to run frontend without container (which includes Nginx and will redirect /api requests to backend), you must set REACT_APP_BACKEND_URL env variable, e.g.:
cd frontend
npm install
REACT_APP_BACKEND_URL=http://127.0.0.1:8000 npm start
Then go to http://127.0.0.1:3000.
To build static files:
cd frontend
npm ci
npm run build
Files will be in the build directory.
To set frontend version which will appear at the very bottom of the page, you should set REACT_APP_FRONT_VERSION environment variable, e.g. REACT_APP_FRONT_VERSION=1.1.1.
This is a multi-stage build, and final image contains only Nginx web-server, static files from build image, and Nginx configuration file (frontend/nginx.conf). Nginx also proxy-passes all requests to /api/servers, /api/version, /api/docs locations to backend instance. (See frontend/Dockerfile)
Fill in the .env file in frontend directory:
cd frontend
mv .env.example .env
nano .env
Then run build container image:
cd frontend
docker build -t eto-frontend:latest .
docker run --env-file .env -p 3000:3000 eto-frontend:latest
Or run my image from DockerHub:
docker run --env-file .env -p 3000:3000 etoosamoe/eto-frontend:latest
Fill in the .env file in frontend and backend directories:
mv .env.example .env
nano .env
Tip
If you use ARM arch (e.g. mac m1 cpu), you should uncomment arch directive in docker-compose.yml file.
The run these commands from the root directory:
docker compose pull
docker compose up -d
or
docker compose up -d --no-build
Manifests in k8s directory includes:
db.ymlas Postgres deployment, with configmapbackend.ymlandfrontend.ymlto deploy application- namespace creation
If you already have a database just fill in db and host to configmap (1 step), and fill in the user and password to secrets.yml. Then go straight to 4 step.
- Fill in the configmap parameters in
k8s/db.ymlwith database name and database host. - Add base64-encoded values of
postgres-userandpostgres-passwordtosecrets.yml. - Deploy the database:
cd k8s
kubectl apply -f namespace.yml
kubectl apply -f db.yml
- Deploy the application:
# If you need a database uncomment the line below
# kubectl apply -f db.yml
kubectl apply -f namespace.yml
kubectl apply -f backend.yml
kubectl apply -f frontend.yml
WIP
Backend has /healthcheck endpoint which returns status-code 200.
You can add Sentry DSN as SENTRY_DSN env. variable, e.g. SENTRY_DSN="https://<dsn>.ingest.sentry.io/<id>".
Prometheus-like metrics endpoint: WIP
/flaky — this API enpoint will raise an 500 error sometimes. Default probability is 0.1 which means 10% of all requests. It can be modified by FLAKY_ERROR_PROBABILITY env. variable. e.g. if probability=1 every request to /flaky endpoint will raise an error.