Managed Kubernetes on Hetzner with Syself

Kubernetes is the gold standard — but you probably think that it will be too hard to manage. This is a valid concern that we can take away from you. With Syself, everyone can manage an Enterprise-grade Kubernetes infrastructure. If you want support for the migration, we and our partner companies will be there to help.

Hetzner's strategy is to make their products as cost-efficient as possible. They do that by cutting down on services that are highly relevant for many companies, such as close and personal support, managed databases and identity management. However, Syself bridges the gap and builds a professional and enterprise-grade infrastructure on top of Hetzner, delivering these additional services. We know how to bring the best out of Hetzner's products, so that you get a cost-efficient AND reliable infrastructure.

Syself doesn't charge anything for network traffic. On Hetzner, each virtual machine comes with 20 TB of free network traffic (ingress and egress). Hetzner charges an additional € 0.50 per month for public IPv4 addresses. Virtual machines come with public IPv6 address free or charge.

Hetzner will launch a Managed Kubernetes product on their own at some point. Our platform on Hetzner has a completely different approach. On Syself Autopilot, the entire lifecycle of clusters is automated. We optimize Kubernetes and operating system configuration, do security hardening and audits, and heavily test upgrades. You can also manage clusters declaratively, in a Kubernetes-native way. Hetzner will not deliver a similar enterprise-grade product, but rather focus on giving quick access to provisioning clusters, without covering day-2 operations, self-healing, and <a href="/features/dedicated-solutions">support</a>. If anything goes wrong, no one will be there to help you. Furthermore, there is no forecast for when Hetzner's Kubernetes solution will be launched. With Syself, you can start now.

Rancher and k3s are Kubernetes distributions, with a focus on cluster provisioning. They simplify day-1 operations by making the experience of deploying Kubernetes clusters more straightforward, and include patches to make the initial setup easier. You still need to provide the machines to install Rancher or k3s on. Syself Autopilot, on the other hand, covers not only the Kubernetes layer, but also virtual machine and bare metal servers provisioning. Furthermore, operations such as scaling, updates, encryption, security hardening and storage setup are also managed by Syself Autopilot. If you want production-ready clusters, with sane default configurations and self-healing capabilities, while retaining full ownership of your infrastructure, Syself is a better option than DYI or "homebrew" Kubernetes solutions for the long term.

Terraform is a infrastructure as code tool that interfaces with multiple providers. When talking about Kubernetes on Hetzner, Terraform allows you to automate the creation of resources such as Load Balancers and Virtual Machines on Hetzner Cloud, as well as the installation of Kubernetes in these VMs. While it is possible to provision dedicated servers (Hetzner Robot) with Terraform, there is no straightforward way to install Kubernetes on them. Different from Terraform, which only handles the initial setup, Syself Autopilot continuously monitors the state of your infrastructure, and ensures it's always matching the desired state. Our platform also handles updates, scaling and self-healing, in addition to providing production-ready configurations by default. This makes Syself Autopilot a better long-term solution for companies looking for building reliable, scalable Kubernetes infrastructure on Hetzner.

Yes, you can. Most of our customers use them to save money and get an optimal cost-benefit ratio.

Setting up a Kubernetes cluster involves a lot of different steps. Managed Kubernetes is a solution offered by some companies to simplify the deployment and operation of Kubernetes clusters. Instead of handling the complex setup and maintenance of control planes, load balancing, and node management yourself, the provider manages these tasks. Syself Autopilot also automates upgrades, bare metal integration, database, encryption at rest, and networking aspects such as subnet, hostname, firewall and network interface configuration.

With declarative management, you specify all the configuration of your clusters in a configuration file, written in YAML, which is a human-readable configuration format. Instead of manually configuring resources, you declare how your clusters should look like, and our platform automatically ensures that you get exactly what you want. This configuration file can then be used for infrastructure as code, where tools like Terraform can manage resources in a repeatable and predictable manner. Another option is to export the configuration file to a Git repository, and use tools like Argo CD to sync them with our platform automatically. This also enables you to audit, review and rollback every piece of your infrastructure.

On Syself Autopilot, dedicated servers are first-class citizens. You can mix them with virtual machines in the same cluster, and use the same declarative management approach to manage dedicated servers — our platform will ensure that they are always matching the desired state. On a technical level, our software uses the Hetzner Robot API key for booting the machine and installing the operating system, after which they are set up with the same configuration as VMs, and join the cluster as a node. To ensure we only manage the servers you want, you add them to an inventory, which contains metadata to identify the server.