Node.js
Deploy Node apps and front ends — Express, Next, Vite and more — with workers and build steps handled.
Step by step: Deploy Node.js to DigitalOcean · Deploy Node.js to Hetzner · Deploy Node.js to Vultr · Deploy Node.js to OVHcloud · Deploy Node.js to Scaleway · Deploy Node.js to UpCloud
- Start:
npm startwhenpackage.jsonhas a start script; the app must listen on$PORTand0.0.0.0(HOST,BIND_ADDRESSand similar documented variables are set for you). - Packager: npm, pnpm, yarn or bun, chosen from your lockfile.
- Workers: an
npm run workerscript is offered as a worker. - Static front ends: use a static site instead — no server process needed.
- Large front-end builds are fitted to the build server’s memory automatically (build methods).
Build step
Your package.json build script runs before start with no configuration from you — for Next.js, Nuxt, SvelteKit and similar server-rendered frameworks, that’s next build/nuxt build/vite build detected straight from your dependencies, and the server starts with the matching command for each (next start, node .output/server/index.mjs…). An app that only produces static output (Gatsby, Create React App, a plain Vite SPA) is served as a static build with no server process at all — see static sites to skip the server entirely instead of running one just to serve files.
Workers are a second process, not a second app
web and worker are the same image, started with a different script — one codebase and one build, several process types, each scaled on its own. An npm run worker script is offered automatically; declare any others yourself:
web: npm start
worker: npm run worker
or add one from the app’s Workers tab without a repo change, and scale either from the CLI:
railyard scale my-app worker=2
Node has no single queue convention, so a BullMQ or similar Redis-backed job runner just needs its own script — REDIS_URL is injected automatically once ioredis, bullmq or bull is a dependency. See Workers and scheduled jobs for cron.
Config and secrets
Everything process.env reads is a plain var: set it under Variables, or railyard env:set my-app KEY=VALUE from the CLI, and the same build runs unchanged from staging through production — dev/prod parity, not a different bundle per environment. Point at a database or Redis running as its own service on the canvas with ${{redis.RAILYARD_PRIVATE_URL}} — resolved per environment. See Variables and secrets.
Deploy it
- From a repo: paste the Git URL, or pick it from the picker once GitHub is connected — Railyard reads
package.json, fills in the start command, packager and any database your dependencies declare, and deploys on every push. - From the CLI:
railyard create my-app https://github.com/me/my-app --addons postgres,redis, orrailyard git:remote && git push railyard mainfrom inside the repo. - One click: add the deploy button to your README, or publish to the template catalog so anyone can launch it with its addons and variables already filled in.