chore: update readme

Signed-off-by: Devin Buhl <devin@buhl.casa>
This commit is contained in:
Devin Buhl committed 2023-01-05 08:59:16 -05:00
1 parent ef4b12e66a
commit 71bbebda9e
1 file changed
+5 -6
+5 -6
View File
@@ -33,26 +33,25 @@ _Additional applications include [hajimari](https://github.com/toboshii/hajimari
For provisioning the following tools will be used: For provisioning the following tools will be used:
- [Fedora 36 Server](https://getfedora.org/en/server/download/) - Universal operating system that supports running all kinds of home related workloads in Kubernetes and has a faster release cycle - [Ansible](https://www.ansible.com) - Sets up the operating system and installs k3s
- [Ubuntu 22.04 Server](https://ubuntu.com/download/server) - Alternative operating system, limited community support - [Terraform](https://www.terraform.io) - Provisions an existing Cloudflare domain and certain DNS records to be used with your Kubernetes cluster
- [Ansible](https://www.ansible.com) - Provision Fedora Server and install k3s
- [Terraform](https://www.terraform.io) - Provision an already existing Cloudflare domain and certain DNS records to be used with your k3s cluster
## 📝 Prerequisites ## 📝 Prerequisites
**Note:** _This template has not been tested on cloud providers like AWS EC2, Hetzner, Scaleway etc... Those cloud offerings probably have a better way of provsioning a Kubernetes cluster and it's advisable to use those instead of the Ansible playbooks included here. This repository can still be tweaked for the GitOps/Flux portion if there's a cluster working in one those environments._ **Note:** _This template has not been tested on cloud providers like AWS EC2, Hetzner, Scaleway etc... Those cloud offerings probably have a better way of provsioning a Kubernetes cluster and it's advisable to use those instead of the Ansible playbooks included here. This repository can still be tweaked for the GitOps/Flux portion if there's a cluster working in one those environments._
First and foremost some experience in debugging/troubleshooting problems **and a positive attitude is required** ;)
### 📚 Reading material ### 📚 Reading material
- [Organizing Cluster Access Using kubeconfig Files](https://kubernetes.io/docs/concepts/configuration/organize-cluster-access-kubeconfig/) - [Organizing Cluster Access Using kubeconfig Files](https://kubernetes.io/docs/concepts/configuration/organize-cluster-access-kubeconfig/)
### 💻 Systems ### 💻 Systems
- One or more nodes with a fresh install of [Fedora Server 36](https://getfedora.org/en/server/download/). - One or more nodes with a fresh install of [Fedora Server 36](https://getfedora.org/en/server/download/) or [Ubuntu 22.04 Server](https://ubuntu.com/download/server).
- These nodes can be ARM64/AMD64 bare metal or VMs. - These nodes can be ARM64/AMD64 bare metal or VMs.
- An odd number of control plane nodes, greater than or equal to 3 is required if deploying more than one control plane node. - An odd number of control plane nodes, greater than or equal to 3 is required if deploying more than one control plane node.
- A [Cloudflare](https://www.cloudflare.com/) account with a domain, this will be managed by Terraform and external-dns. You can [register new domains](https://www.cloudflare.com/products/registrar/) directly thru Cloudflare. - A [Cloudflare](https://www.cloudflare.com/) account with a domain, this will be managed by Terraform and external-dns. You can [register new domains](https://www.cloudflare.com/products/registrar/) directly thru Cloudflare.
- Some experience in debugging problems and a positive attitude ;)
📍 It is recommended to have 3 master nodes for a highly available control plane. 📍 It is recommended to have 3 master nodes for a highly available control plane.