n8n Hosting on VPS

Self-hosted workflow automation. Run n8n on a VPS to wire up 400+ services and AI models with no per-execution fees and full data ownership.

Use cases

Replace Zapier and Make for high-volume flows

Run thousands of executions a day with zero metered cost. Same drag-drop interface, same triggers, none of the per-task billing.

Wire AI models into existing business processes

Native nodes for OpenAI, Anthropic, Ollama. Build agent loops, RAG pipelines, content workflows without writing glue code.

Sync data across SaaS without ETL bills

CRM ↔ database ↔ spreadsheet ↔ Slack flows that run on cron or webhook. Same job as Airbyte/Fivetran, your own box, your own cost.

Build internal API gateways

Expose webhook URLs that transform, enrich and route incoming payloads. n8n becomes the glue between legacy systems and modern APIs.

Schedule reports, monitoring, alerts

Cron-triggered flows that pull metrics, format them, and push to Slack/Discord/email. Same patterns as Grafana alerting but for arbitrary data sources.

Power customer-facing automations

Trigger flows from your product's webhooks. Onboarding sequences, billing reconciliation, support ticket routing – all without dev time per flow.

Recommended plans for n8n

Medium plan handles n8n + Postgres + Redis for moderate workflow volume.

  • 40 Countries
    Linux
    Intel Xeon
    AMD Ryzen

    From$10.19 / 1 month
    $9.68 with is*smart
    • Xeon 2x2.20 GHz CPU
    • 2 Gb RAM
    • 30GB SSD Drive
    • 3 Tb Bandwidth
    • IPV4 optional
    • 1Gbps Port
    • Weekly Backup
    • KVM Virtualization
    • Control panels
  • 40 Countries
    Linux
    Intel Xeon
    AMD Ryzen

    From$21.24 / 1 month
    $20.18 with is*smart
    • Xeon 3x2.20 GHz CPU
    • 4 Gb RAM
    • 40GB SSD Drive
    • Unmetered Bandwidth
    • IPV4 optional
    • 1Gbps Port
    • Weekly Backup
    • KVM Virtualization
    • Control panels
  • 40 Countries
    Linux
    Intel Xeon
    AMD Ryzen

    From$33.99 / 1 month
    $32.29 with is*smart
    • Xeon 4x2.20 GHz CPU
    • 8 Gb RAM
    • 50GB SSD Drive
    • Unmetered Bandwidth
    • IPV4 optional
    • 1Gbps Port
    • Weekly Backup
    • KVM Virtualization
    • Control panels

n8n hosting options that scale without painful migrations

Hosting requirements changed? Skip unnecessary migration. Add RAM, extend the drive, increase backup frequency, or attach new IPs — all on the same n8n VPS. Upgrade via ticket or pre-configure before checkout.

Setup guide

Five steps. Public webhook URL working at the end.

Step 01 — Order your VPS

Medium plan handles n8n + Postgres + Redis for moderate workflow volume. Pick a region close to the services your workflows talk to most.

ssh root@your-server-ip

Step 02 — Install Docker and Docker Compose

n8n runs cleanest as a Docker Compose stack – Postgres for state, Redis for queue, n8n itself, and Caddy for HTTPS termination.

curl -fsSL https://get.docker.com | shsystemctl enable --now docker

Step 03 — Configure the n8n stack

Create a docker-compose.yml with n8n, Postgres, and Redis. Set WEBHOOK_URL to your public HTTPS endpoint so external services can hit your flows.

services:  n8n:    image: n8nio/n8n    environment:      - WEBHOOK_URL=https://n8n.yourdomain.com      - DB_TYPE=postgresdb      - DB_POSTGRESDB_HOST=postgres    depends_on: [postgres, redis]

Step 04 — Set up HTTPS

Point a subdomain at your VPS IP. Caddy or Coolify reverse-proxy can handle Let's Encrypt automatically.

A   n8n.yourdomain.com   →   your-server-ip

Step 05 — Verify your first workflow

Open the n8n UI, create a Webhook → Set → Respond workflow, hit the webhook URL with curl. If you get the expected JSON back, the public endpoint, queue and DB are all wired correctly.

curl -X POST https://n8n.yourdomain.com/webhook-test/hello

Expected output:

{"status":"ok"}

Common issues

  • Webhooks return 404 – n8n needs the public URL set via WEBHOOK_URL env var, not just listening on localhost.
  • Workflows lose state on restart – make sure Postgres is the configured DB, not the default SQLite, before you put it in production.

FAQ

How much RAM does n8n need?

n8n itself starts at ~500 MB. Add Postgres (~200 MB), Redis (~100 MB), and your workflow execution overhead. Medium plan (4 GB) is the sane minimum for production; Lite is fine for dev/playground.

Do I need a separate worker node?

Not at low volume. Once you're past ~10 executions per second, n8n's queue mode with a separate worker container helps. Same VPS, different container. Scale up to a multi-VPS setup only if a single box can't keep up.

Can I import workflows from n8n Cloud?

Yes. Export from your n8n Cloud workspace as JSON, import into the self-hosted instance. Credentials don't transfer (encryption keys differ) – you'll re-enter API keys once.

Ready to deploy n8n?

Webhook-ready in under 5 minutes. Public HTTPS, persistent Postgres, all included.

Launch n8n on VPS

More from is*hosting

VPS

Virtual servers with dedicated KVM cores — scale workers and queues on demand.
Explore
From $5.94 /mo

Dedicated Servers

Bare-metal throughput for high-volume workflow execution.
Explore
From $75.00 /mo

Recent Posts

  • Mailcow on VPS

    Mailcow on a VPS: Your Own Mail Server Without the Per-Seat Bill

    A full guide to self-hosting email with Mailcow on a VPS: Docker setup, DNS and DKIM records, spam filtering, and webmail, with no per-seat fees to pay.

    UTC 2026-10-01 9 min reading
  • How to Install Ghost on a VPS: Self-Hosting Guide

    How to Install Ghost on a VPS: Self-Hosting Guide

    Install Ghost 6 on a VPS and run your own publication. Native newsletters, Stripe paid tiers, and free SSL, replacing the $29/mo Ghost(Pro) Publisher plan.

    UTC 2026-09-29 8 min reading
  • Nextcloud on VPS: Self-Hosted File Sync Without the Per-Seat Bill

    Nextcloud on VPS: Self-Hosted File Sync Without the Per-Seat Bill

    Installing Nextcloud on a VPS using the official All-in-One method via Docker. From a blank VPS to a working instance in less than 10 minutes.

    UTC 2026-09-22 8 min reading