Skip to content
All the guides

Dokploy and Hetzner

Price

Render.com charges $7 for a starter server per month. For this, you get 0.5 CPU and 512mb RAM. For hosting something like N8N, there’s also the need for a Postgres database which adds another $7 per month. 

There are two places that things start to add a lot of cost on to this though if you start getting into wanting to do more:

  • Hosting of multiple different sites or services. Each one requires a separate Starter server which each add another $7 on to the monthly cost. 
  • For any services that are more intensive and need more than the pretty limited amount of compute that is provided, the cost increases sharply - the next server offered after the

Starter is $25 per month for 2GB of RAM and 1 CPU, so it soons adds up where you have multiple things running.

Enter Dokploy

Dokploy offers a very deliberate open source solution to the problem, mirroring many of the ways that Render’s UI works, but breaking apart the software/hardware link that exists with Render.

Hetzner for hosting

Compared to Render, Hetzner’s hosting is significantly cheaper. For €8 per month, there’s a 4 vCPU, 8GB of RAM and 80GB of storage server available. Add another 20% for daily backups and I’d argue that you have something much better than Render from a capacity standpoint for about the same price.

The Dokploy learning curve

There’s definitely a learning curve for getting started with Dokploy as compared to Render (although most of this is down to amnesia from the pain of getting N8N set up on Render at least two years ago and also now pushing far beyond what I was using Render for).

Some of this comes from being able to easily host multiple services on the same server, and Dokploy comes with a number of templates that make it easier to get started. The templates are not perfect though, and more often than not there’s some additional tweaking to the configuration that’s needed to get things working (I’m documenting these as I come across them).

I’ve had to go deeper into Docker to get some things to work or to fix things that I’ve accidentally broken.