Add cluster node setup documentation and scripts for K3s and Talos

README.md

@@ -24,57 +24,9 @@ This project provides a complete, production-ready Kubernetes infrastructure des
 - Deploy new applications with a single command
 - Manage everything through a slick web dashboard
 
-## What Can You Run?
-
-The possibilities are endless! Here are just a few ideas:
-
-- **Personal Websites & Blogs** (WordPress, Ghost, Hugo)
-- **Photo Storage & Sharing** (PhotoPrism, Immich)
-- **Document Management** (Paperless-ngx)
-- **Media Servers** (Jellyfin, Plex)
-- **Home Automation** (Home Assistant)
-- **Password Managers** (Bitwarden, Vaultwarden)
-- **Note Taking Apps** (Joplin, Trilium)
-- **Productivity Tools** (Nextcloud, Gitea, Plausible Analytics)
-- **Database Servers** (PostgreSQL, MariaDB, MongoDB)
-- **And much more!**
-
-## Key Features
-
-- **One-Command Setup**: Get a complete Kubernetes infrastructure with a single script
-- **Secure by Default**: Automatic HTTPS certificates for all services
-- **Split-Horizon DNS**: Access services internally or externally with the same domain
-- **Custom Domains**: Use your own domain name for all services
-- **Service Templates**: Deploy new applications with a simple command
-- **Dashboard**: Web UI for monitoring and managing your infrastructure
-- **No Cloud Vendor Lock-in**: Run on your own hardware, from a Raspberry Pi to old laptops
-
 ## Getting Started
 
-For detailed instructions, check out our documentation:
-
-- [**Setup Guide**](./docs/SETUP.md) - Step-by-step instructions for setting up your infrastructure
-- [**Applications Guide**](./apps/README.md) - How to deploy and manage applications on your cloud
-- [**Maintenance Guide**](./docs/MAINTENANCE.md) - Troubleshooting, backups, updates, and security
-
-After setup, visit your dashboard at `https://dashboard.internal.yourdomain.com` to start exploring your new personal cloud infrastructure!
-
-## Project Structure
-
-```
-.
-├── bin/                      # Helper scripts
-├── apps/                     # Apps
-├── docs/                     # Documentation
-│   ├── SETUP.md              # Setup instructions
-│   ├── APPS.md               # Application deployment guide
-│   ├── MAINTENANCE.md        # Maintenance and troubleshooting
-│   ├── OPS.md                # Operations guide
-│   └── INGRESS.md            # Network configuration guide
-├── infrastructure_setup/     # Infrastructure setup scripts
-├── services/                 # Custom service templates and deployed services
-└── load-env.sh               # Environment variable loader
-```
+See [Setup](./docs/SETUP.md).
 
 ## Contributing
 

cluster-node-setup/README.md

@@ -0,0 +1,90 @@
+# Cluster Node Setup
+
+Cluster node setup is WIP. Any kubernetes setup will do. Currently, we have a working cluster using each of these methods and are moving towards Talos.
+
+## k3s cluster node setup
+
+K3s provides a fully-compliant Kubernetes distribution in a small footprint.
+
+To set up control nodes:
+
+```bash
+# Install K3s without the default load balancer (we'll use MetalLB)
+curl -sfL https://get.k3s.io | sh -s - --write-kubeconfig-mode=644 --disable servicelb --disable metallb
+
+# Set up kubectl configuration
+mkdir -p ~/.kube
+sudo cat /etc/rancher/k3s/k3s.yaml > ~/.kube/config
+chmod 600 ~/.kube/config
+```
+
+Set up the infrastructure services after these are running, then you can add more worker nodes with:
+
+```bash
+# On your master node, get the node token
+NODE_TOKEN=`sudo cat /var/lib/rancher/k3s/server/node-token`
+MASTER_IP=192.168.8.222
+# On each new node, join the cluster
+
+curl -sfL https://get.k3s.io | K3S_URL=https://$MASTER_IP:6443 K3S_TOKEN=$NODE_TOKEN sh -
+```
+
+## Talos cluster node setup
+
+This is a new experimental method for setting up cluster nodes. We're currently working through the simplest bootstrapping experience.
+
+Currently, though, all these steps are manual.
+
+Copy this entire directory to your personal cloud folder and modify it as necessary as you install. We suggest putting it in `cluster/bootstrap`.
+
+```bash
+
+# Install kubectl
+curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"
+curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl.sha256"
+echo "$(cat kubectl.sha256)  kubectl" | sha256sum --check
+sudo install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl
+
+# Install talosctl
+curl -sL https://talos.dev/install | sh
+
+# In your LAN Router (which is your DHCP server),
+
+CLUSTER_NAME=test-cluster
+VIP=192.168.8.20 # Non-DHCP
+
+# Boot your nodes with the ISO and put their IP addresses here. Pin in DHCP.
+# Nodes must all be on the same switch.
+# TODO: How to set these static on boot?
+CONTROL_NODE_1=192.168.8.21
+CONTROL_NODE_2=192.168.8.22
+CONTROL_NODE_3=192.168.8.23
+
+# Generate cluster config files (including pki and tokens)
+cd generated
+talosctl gen secrets -o secrets.yaml
+talosctl gen config --with-secrets secrets.yaml $CLUSTER_NAME https://$VIP:6443
+talosctl config merge ./talosconfig
+cd ..
+
+# If the disk you want to install Talos on isn't /dev/sda, you should
+# update to the disk you want in patch/controlplane.yml and patch/worker.yaml. If you have already attempted to install a node and received an error about not being able to find /dev/sda, you can see what disks are available on it with:
+#
+# talosctl -n $VIP get disks --insecure
+
+# See https://www.talos.dev/v1.10/talos-guides/configuration/patching/
+talosctl machineconfig patch generated/controlplane.yaml --patch @patch/controlplane.yaml -o final/controlplane.yaml
+talosctl machineconfig patch generated/worker.yaml --patch @patch/worker.yaml -o final/worker.yaml
+$
+
+# Apply control plane config
+talosctl apply-config --insecure -n $CONTROL_NODE_1,$CONTROL_NODE_2,$CONTROL_NODE_3 --file final/controlplane.yaml
+
+# Bootstrap cluster on control plan
+talosctl bootstrap -n $VIP
+
+# Merge new cluster information into kubeconfig
+talosctl kubeconfig
+
+# You are now ready to use both `talosctl` and `kubectl` against your new cluster.
+```

cluster-node-setup/patch/controlplane.yaml

@@ -0,0 +1,17 @@
+machine:
+  install:
+    disk: /dev/sdc
+  network:
+    interfaces:
+      - interface: eth0
+        vip:
+          ip: 192.168.8.20
+      - interface: eth1
+        dhcp: true
+cluster:
+  discovery:
+    enabled: true
+    registries:
+      service:
+        disabled: true
+  allowSchedulingOnControlPlanes: true

cluster-node-setup/patch/worker.yaml

@@ -0,0 +1,3 @@
+machine:
+  install:
+    disk: /dev/sdc

docs/SETUP.md

@@ -1,27 +1,25 @@
 # Setting Up Your Wild Cloud
 
-## Set up your cloud bootstrapper
+## Set up your personal cloud operations directory
 
-See [Dnsmasq Setup](./guides/dnsmasq-setup.md).
+- Create a directory somewhere. We recommend you use an Ubuntu machine.
+- Inside it, run `wild-init`. This will scaffold your cloud directory.
+- In your cloud directory, update `.wildcloud/config.yaml`. Use the same values in this dir in a `.env`
+
+## Set up your Cloud Central
+
+See [Central Setup](../central-setup/README.md).
 
 ## Set up Control Nodes
 
 ### 2. Install K3s (Lightweight Kubernetes)
 
-K3s provides a fully-compliant Kubernetes distribution in a small footprint:
-
-```bash
-# Install K3s without the default load balancer (we'll use MetalLB)
-curl -sfL https://get.k3s.io | sh -s - --write-kubeconfig-mode=644 --disable servicelb --disable metallb
-
-# Set up kubectl configuration
-mkdir -p ~/.kube
-sudo cat /etc/rancher/k3s/k3s.yaml > ~/.kube/config
-chmod 600 ~/.kube/config
-```
+See [Cluster Node Setup](../cluster-node-setup/README.md).
 
 ## Install Infrastructure Components
 
+> Currently, these are set up to run from this directory. This will be moved to (1) a `bin/wild-generate-infrastructure-setup` script to copy them all to your personal cloud dir, (2) `wild-cli` (to do the same), or (3) `wild-central`, once I get my mind made up.
+
 One command sets up your entire cloud infrastructure:
 
 ```bash
@@ -38,19 +36,6 @@ This installs and configures:
 - **ExternalDNS**: Updates DNS records automatically
 - **Kubernetes Dashboard**: Web UI for managing your cluster
 
-## Set up worker nodes
-
-For larger workloads or high availability, you can add more nodes:
-
-```bash
-# On your master node, get the node token
-NODE_TOKEN=`sudo cat /var/lib/rancher/k3s/server/node-token`
-MASTER_IP=192.168.8.222
-# On each new node, join the cluster
-
-curl -sfL https://get.k3s.io | K3S_URL=https://$MASTER_IP:6443 K3S_TOKEN=$NODE_TOKEN sh -
-```
-
 ## Next Steps
 
 Now that your infrastructure is set up, you can:

docs/guides/talos-setup.md

@@ -1,32 +0,0 @@
-# Talos
-
-This is an alternate setup to using ks that uses talos and bare kubernetes. IN PROGRESS.
-
-From https://www.talos.dev/v1.10/introduction/getting-started/
-
-```bash
-# Install kubectl
-curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"
-curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl.sha256"
-echo "$(cat kubectl.sha256)  kubectl" | sha256sum --check
-sudo install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl
-
-# Install talosctl
-curl -sL https://talos.dev/install | sh
-
-# Generate cluster config files (and pki and tokens)
-talosctl gen config test-cluster https://192.168.8.238:6443
-
-talosctl -n 192.168.8.238 get disks --insecure
-# Update disk in controlplane.yml
-
-# Apply control plane config
-talosctl apply-config --insecure --nodes 192.168.8.238 --file controlplane.yaml
-
-# Bootstrap cluster on control plan
-talosctl bootstrap --nodes 192.168.8.238 --endpoints 192.168.8.238 --talosconfig=./talosconfig
-
-# Merge into kubeconfig
-talosctl kubeconfig --nodes 192.168.8.238 --endpoints 192.168.8.238 --talosconfig=./talosconfig
-
-```