Developer
Deployment
Staging Docker stack and production considerations.
Staging compose (all services)
File: ragnarock-backend/docker/staging/docker-compose.yml
cd ragnarock-backend
make staging-up-d| Service | Port | Image |
|---|---|---|
main | 8000 | Dockerfile.main |
admin | 3001 | Dockerfile.admin |
agent | 8100 | Dockerfile.agent |
nginx | 80 | Reverse proxy |
| Postgres, Redis, Typesense, Mailpit | (internal) | Shared infra |
The agent image builds from agent/ inside the backend repository (same source as ragnarock-agent).
Production checklist
Nginx / local hostnames
Development nginx configs map main.localhost, app.localhost, mailpit.localhost. Production replaces this with real DNS and TLS certificates.
Health checks
| Endpoint | Expect |
|---|---|
GET / (API) | 200 |
GET /health (agent) | 200 |
| Postgres / Redis | Docker healthchecks in compose |
Observability
- Winston logging in NestJS (
libs/logger) - Linear sync runs and activity feed for user-visible audit
- Mailpit only for development — configure real SMTP (Brevo, etc.) in production via
libs/mailer