Setting up your n8n

Every step, including the ones that are easy to skip and expensive to skip. It takes about measurement pending .

Before you start

• A Google account to sign in with. You do not need GitHub and you do not fork anything.
• A payment method. The free allowance is smaller than this app needs.
• Nothing else. No terminal, no domain, no certificate.

1. Open the deploy form

The button below opens the form with the starting point and the name already filled in. Sign in when it asks.

2. Turn on the managed database required

Tick "Add a managed database" on the form. This is the single most important thing on this page. Without it, n8n keeps your workflows in temporary storage that is cleared on every restart, and turning the database on afterwards gives you an empty one rather than your work back.

Nothing warns you at the moment of the mistake, because at that moment nothing is wrong. What that failure looks like when it arrives.

3. Choose 1 GB of app power required

Open the size options and pick 1 GB. If your account has no capacity yet, this is where you buy some, and you come back to the form afterwards.

Leave the instance count at one. Two copies of n8n both fire every schedule and both answer every trigger, so your workflows would run twice.

4. Deploy, and wait

Your instance is being built rather than handed to you from a pool, so the first one takes a few minutes. Restarts after that are quick.

5. Set the encryption key required

n8n encrypts every credential you save with one key. If that key is not set explicitly and stored, it gets regenerated and every credential you have saved becomes unreadable. Not by us, not by anyone. This is the most expensive mistake available in self-hosted n8n and it is the reason for this step.

Generate a value here, store it in your password manager, then add it in the dashboard as N8N_ENCRYPTION_KEY and restart the app.

This whole step disappears the day the deploy form can generate and store it for you, which is what we are building next. Until then, it is here and it is honest about being here.

What happens when this key goes missing, and how to recover

6. Open your n8n and create your account

Your instance has its own address with a certificate already in place. The first screen asks you to set an owner email and password. Do that before you do anything else: until you do, anyone who knows the address can reach the setup screen.

Where the controls are

We do not build a second control panel for this. Everything lives in the Dockhold dashboard, so here is what each part of it does in plain words.

Is it running?

The app's page shows its current state. If something has gone wrong it is restarted for you, so a brief blip is normal and a persistent one is worth reading the logs about.

Reading the logs

Logs are on the app's page. This is where a failed start explains itself, and it is the first thing to look at before asking anyone anything.

Restarting

There is a restart control on the app. Your workflows and credentials live in the managed database, so a restart does not touch them.

Making it bigger

The size controls change the memory an app gets without rebuilding it. If n8n is running out of room, this is the lever, and this page covers what to try in the workflow first.

Your own domain

Add the domain in the app settings and point one DNS record at us. The certificate is handled. Do it early if you are going to do it, because anything you have already told about your old address keeps using it.

Updating n8n

Your instance runs a pinned version and it does not change underneath you. Redeploying moves you to whatever version the shared starting point currently pins. If you want to control that yourself, or bundle extra community nodes into the image, fork the starting point and deploy from your fork.

Stopping

Delete the app when you are done. Export your workflows first, from inside n8n. That export is yours and it imports into any n8n anywhere.

You now have room for other things

This is worth knowing, because most people do not: you bought capacity, not an n8n plan. Whatever n8n is not using is still yours, and putting a second app next to it costs nothing extra until you outgrow the pool.

• A status or uptime checker

  Watches the things your workflows depend on, and tells you before your automation does.

• A small site or landing page

  The forms on it can post straight into the n8n you already have running.

• A shared dashboard

  Reads the same database your workflows write to, so the numbers are already there.

What else deploys in one go