Self-host n8n on your own VPS
n8n is the open-source workflow automation tool that everyone is using to wire up AI agents, scrapers, CRMs and APIs without vendor lock-in. Self-hosting it on a VPS skips the n8n Cloud subscription, keeps your workflows private, and gives you unlimited executions. Servury is built for it: anonymous signup, crypto payments, full root, 7 locations.
n8n runs comfortably on a 2 GB plan. Starting at $15.59/mo.
Why self-host n8n
Quick start: n8n on Servury via Docker
Tested on Ubuntu 24. Pick a 2 GB+ plan, deploy, SSH in.
# 1. Install Docker
curl -fsSL https://get.docker.com | sh
# 2. Run n8n with persistent storage
docker volume create n8n_data
docker run -d --name n8n --restart unless-stopped \
-p 5678:5678 \
-v n8n_data:/home/node/.n8n \
-e N8N_HOST=$(curl -s ifconfig.me) \
-e WEBHOOK_URL=http://$(curl -s ifconfig.me):5678/ \
n8nio/n8n
# 3. Open http://YOUR_SERVER_IP:5678 and finish onboarding
# 4. Recommended: put nginx + Let's Encrypt or Caddy in front for HTTPS
What people run on self-hosted n8n
Chain OpenAI/Anthropic calls with vector DBs, scraping, and downstream actions. Most popular n8n use case in 2026.
Webhook in → enrich via APIs → LLM classification → CRM out. Replaces $200/mo SaaS with a $5 VPS.
Reply triggers, scheduled summaries, AI-powered replies. n8n's HTTP nodes make it trivial.
Cron-style scrapers feeding databases, search indexes, and AI training data sets.
Connect Notion, Linear, GitHub, Stripe, anything with a REST API. The "Zapier alternative" everyone wanted.
Email parsing, calendar sync, RSS to Telegram, document workflows. Anything you'd normally pay for monthly.
Frequently asked questions
Why self-host n8n instead of using n8n Cloud?
n8n Cloud bills per active workflow and execution. A self-hosted instance on a $5 VPS gives you unlimited workflows and unlimited executions. You also keep all credentials and workflow data on a host you control.
Will a 2 GB VPS handle n8n?
Yes for personal use and light production workloads (a few thousand executions per day). For heavier loads with PostgreSQL backend and Redis queue mode, jump to a 4 GB plan.
Do I need to use Docker?
Not strictly. n8n can also be installed via npm. Docker is the recommended path because the official image is maintained and updates are one-command. Both work fine on Servury.
How do I expose n8n over HTTPS?
Put Caddy or nginx in front. Caddy is the simplest: one line of Caddyfile and Lets Encrypt certs are automatic. We have full root, you can install whatever reverse proxy you like.
Can I run n8n with a real database instead of SQLite?
Yes. n8n supports PostgreSQL and MySQL via env vars. For production we recommend Postgres (run it locally on the same VPS or as a separate Docker container with a shared network).
Can I run AI workflows that hit OpenAI/Anthropic from n8n?
Yes. n8n has built-in nodes for OpenAI, Anthropic, OpenRouter, and any HTTP-based LLM API. Your API keys stay encrypted in your n8n instance, never sent anywhere else.
Does Servury log my n8n traffic?
No. Servury runs no application-level logging on customer servers. With anonymous signup and crypto payment, the entire stack stays private.
Where should I deploy for best webhook latency?
Pick the location closest to the services that will hit your webhooks. New York and Frankfurt are the most common picks for global webhook traffic.
