n8n becomes an operational system
Once workflows touch CRM, payments, messaging or customer data, n8n must be operated like any other production service—with controlled access, backups and monitoring.
Infrastructure and access
Place the service behind HTTPS and a reverse proxy, restrict editor access and manage credentials independently from exported workflows.
- HTTPS and HSTS
- Restricted administration
- Encrypted credentials
- Private database network
- Regular updates
Queue mode and scaling
Queue mode with Redis and separate workers improves concurrency control and recovery. Worker counts should reflect provider limits and workflow characteristics.
Webhooks and untrusted input
Public webhooks need signatures or secrets, size limits, validation and idempotency. Never trust incoming JSON directly.
Failures, backups and runbooks
Use error workflows, alerts, controlled retries, database backups and restore tests. A backup that has never been restored is not a recovery plan.
Planning a serious digital product?
Share the current problem and expected outcome. We will suggest a practical architecture and phased delivery path.
Start a project