I’m experiencing persistent 502 “Application failed to respond” errors on my deployed Node/Express API, even though everything works locally and I’ve followed all best practices. Here’s a summary of my setup and what I’ve tried:

• Repo: My GitHub repo is just the service folder (parent-api), not a monorepo root.

• No root directory set in Railway (repo root is service root).

• No Dockerfile, .dockerignore, or .nixpacks.toml in the repo.

• package.json scripts:

• "build": "prisma generate && tsc"

• "start": "node dist/index.js"

• App listens on:

0.0.0.0:${process.env.PORT} (I see “Parent API running on http://0.0.0.0:8080” in logs)

• No custom build/start commands in Railway.

• No custom PORT env variable set in Railway.

• Build logs show tsc runs and dist/index.js is created.

• App works locally with the Railway Postgres DB (can sign up, see users in DB).

• On Railway:

• App starts, logs “running”, but all requests (including /health) return 502.

• No request logs appear in the app.

• No crash or error is shown in the logs.

• Tried deploying a minimal Express app (just /health route), still get 502.

What I need:

• Help diagnosing why Railway’s proxy cannot reach my app, despite it running and listening on the correct port/host.

• Any platform-specific issues or settings I should check.

Thank you for your help!