For me, a homelab only truly comes to life once you start hosting applications on it. My previous homelab only ran a handful of applications and I basically ignored it. This incarnation already hosts several times as many and I make use of them almost every day.

Setting up a hosting environment, be it Kubernetes, Proxmox, a Cloud environment such as Azure, AWS or GCP, or any other variety of VPS, PaaS or IaaS1 offerings is a good training exercise, but in my opinion one of the main points of a homelab is to host your own copies of applications that would otherwise require you to trust and pay2 others to do so.

In this series I plan to cover what I am hosting, why I chose it over alternatives and possibly aspects of the application configuration.

What I won’t be covering

I won’t typically be going into the Kubernetes yaml or Helm Chart values, for the most part I use the same base templates and just add application specific values. I’ll go into the specifics of the base templates in my homelab setup series when I talk about K3s and Kustomize.

I also will not be going over any of what I see as the purely backend deployments (PostgreSQL, Redis/Valkey and Minio). They are deployed as single instances with the default or recommended settings and to me they are as much part of the homelab overhead as backups, Persistent Volume replication through Longhorn and traffic routing.

  1. Platform or Infrastructure as a Service. These allow you to run your applications and software without having to worry about the underlying environment. ↩︎

  2. If there is no actual cost to the service then in almost every case your data, or data about you, is the payment. This shows up as targeted advertisements or having your anonymous usage data sold to other companies for further use in ads. ↩︎