Hostinger n8n Hosting: Self-Host n8n on a VPS

What is the Hostinger n8n VPS template?

The n8n template is a preconfigured OS image for Hostinger’s VPS plans. Instead of a blank Linux server, you get Ubuntu with Docker already installed and n8n running inside a container, plus an Nginx reverse proxy and automatic Let’s Encrypt SSL once you point a domain at the server. What normally takes the better part of an hour, provisioning, Docker, Compose files, reverse proxy, certificates, collapses into choosing a template during checkout or switching to it later from hPanel.

There are three n8n templates to choose from: the standard n8n install, an n8n queue-mode variant for higher-concurrency workloads, and an n8n template bundled with more than a hundred prebuilt workflow templates. All three run the same software, n8n’s Community Edition, which is free to use for your own workflows with no execution meter. That last point matters more than it sounds: on n8n Cloud you pay per workflow run, while self-hosted runs are limited only by what your server can handle.

Which VPS size fits your n8n workload?

n8n itself is not demanding, its documented minimum is roughly 1 vCPU, 1 GB of RAM, and 10 GB of storage. But minimum and comfortable are different things, and RAM is the resource that decides how your instance feels. Here’s how the VPS tiers map to real workloads, using each plan’s specs from the live cards on this page.

Light use: testing and personal automations

The smallest VPS tier (1 vCPU, 4 GB RAM, 50 GB NVMe) is plenty for learning n8n, running a handful of personal workflows, or automating a side project. Scheduled syncs, form-to-spreadsheet flows, and notification bots all fit comfortably here. This is where most people should start, you can resize upward later without rebuilding anything.

The sweet spot: daily business workflows and AI agents

The mid-size tier (2 vCPU, 8 GB RAM, 100 GB NVMe) is where we point most buyers. It handles dozens of active workflows, webhook-heavy setups, and always-on AI agent loops without breaking a sweat. If your automations run a business, lead routing, order processing, content pipelines, the headroom buys you reliability during traffic spikes, which is exactly when you can’t afford a stalled queue.

Heavy use: high volume and queue mode

When workflows fire hundreds of times an hour or process large datasets, step up to the larger tier and switch to the queue-mode template. Queue mode moves execution off n8n’s main process (with Redis as the message broker), so the editor stays responsive while workers chew through runs in the background. Pair it with PostgreSQL instead of the default SQLite once you’re past a few thousand executions a month, database write speed becomes the bottleneck, and NVMe storage earns its keep.

How do you install n8n on Hostinger?

Below is the setup outline from Hostinger’s official tutorial on self-hosting n8n, the exact screens and menu names, condensed. The whole process takes about 10-15 minutes of waiting plus a few minutes of clicking.

1. Buy a VPS plan. Shared hosting cannot run n8n, you need a VPS. During checkout you can already pick the n8n template; if you skip it, you can switch later.
2. Open hPanel and go to VPS. Use the left-hand menu, find your server, and click Manage.
3. Go to OS & Panel → Operating System. This is where Hostinger lists its OS and application templates.
4. Use Change OS and search for “n8n”. Pick the template you want (standard, queue mode, or the +100 workflows bundle) and click Change OS.
5. Confirm the wipe. Check the box acknowledging that all files will be deleted and cannot be restored, then click Next and set your new VPS password.
6. Wait, then open Manage App. Once installation finishes, scroll up and hit Manage App to reach your n8n dashboard, your first visit creates the owner account.

Two things to do right after install: point a domain at the server so the automatic SSL certificate provisions (running automation over plain HTTP is asking for credential leaks), and set up the firewall in hPanel under VPS → Security → Firewall so only the ports you need are open. Neither step is optional for anything beyond tinkering.

n8n Cloud vs self-hosting: the honest cost comparison

This is the comparison most n8n hosting articles fudge, so let’s do it with real numbers. n8n lists its Cloud prices in euros, below are the figures from n8n.io’s pricing page as reported in October 2026 breakdowns. Every Cloud plan includes unlimited users and unlimited active workflows; only the execution allowance changes.

Note: n8n bills Cloud plans in EUR, there is no USD price list on n8n.io (checked October 2026).

The math that matters: a single month of Cloud Starter costs more than the monthly server rate shown in the live VPS card below, and the Community software on that server has no execution meter at all. At roughly 80 workflow runs a day you’d already be brushing Cloud Starter’s ceiling; on your own server those same runs cost nothing extra. n8n Cloud makes sense when you never want to think about a server; self-hosting wins the moment your automations run daily and you’d rather own the infrastructure than rent the meter.

What the one-click template won’t do for you

One-click means the install, not the ownership. Updates, backups, firewall rules, and SSL renewal monitoring are yours. That’s the honest trade of self-hosting, and it’s worth naming the failure modes n8n’s own community forum sees most on Hostinger servers so you recognize them fast.

• 502 Bad Gateway on long workflows. The classic: a workflow runs longer than the reverse proxy’s default timeout and Nginx drops the connection. Raising the proxy timeouts in the Nginx config fixes it, this is configuration, not a broken server.
• The server hangs mid-execution. Usually memory pressure, the Docker container gets killed when RAM runs out. Check with your container stats, trim heavy workflows into batches, or move up a tier and switch to queue mode.
• SQLite starts to strain. The default database is fine for hundreds of executions a month. Past a few thousand, migrate to PostgreSQL, the template’s NVMe storage makes this a genuine upgrade, not a lateral move.
• Webhook URLs break after changes. n8n needs to know its own public URL (the environment variables for host and webhook URL must match the domain you actually use). Set them once, correctly, and most “my webhooks stopped working” mysteries never happen.

None of this requires a DevOps background, it requires a search engine and twenty unhurried minutes per issue, the first time each. After that, a self-hosted n8n instance is famously boring in the best way: it just runs.

Who should self-host n8n on Hostinger, and who shouldn’t

Self-host if your workflows run daily, you care where your data lives, or you’re building AI agent loops that would burn through a Cloud execution allowance. It’s also the right call if you want the server for other things too, the same VPS can host side projects alongside n8n.

Don’t self-host if the thought of SSH makes you break out in hives and your usage is tiny, n8n Cloud’s free trial and Starter exist for exactly that case. Also skip it if you need n8n’s Enterprise features (SSO, Git environments, audit logs): those are licensed products, not something the Community template can unlock.

Pros and cons

FAQ

Related