Skip to content

Latest commit

 

History

History
117 lines (71 loc) · 3.89 KB

deployment.md

File metadata and controls

117 lines (71 loc) · 3.89 KB
Raw

Deployment

This is the end-to-end process for deploying out the clinch-home cloud.

Packer template

  • Local network access to the server
  • Running the packer template when you can connect to the Proxmox server directly
  • Setting up the template

Creating the virtual infrastructure

Run the "create" workflow in this repository. To do this, go to the create actions page of the clinch-home repository. Press the "Run workflow" button and then the same named green button that appears.

clinch-home-create-workflow.png

The run will take around 15 minutes and at the end there will be 4 servers created on the Proxmox host in Geddes.

Setup network access

Go to the Unifi portal and select the Geddes network.

clinch-home-unifi-dashboard.png

On the left click "Settings" > "Firewall & Security" and scroll down to the "Port Forwarding" section. Using the "Create New Forwarding Rule" or by editing an existing rule make sure these rules exist:

Name: GitHub Runner SSH / Kubernetes Worker (1,2,3) SSH
Enable Forward Rule: Enable
From: Any
Port: 1600(1,2,3,4)
Forward IP: 192.168.2.16(1,2,3,4)
Forward Port: 22
Protocol: Both
Logging: Enable

Configuring the hosts

Configuring virtual GitHub runner

Once that's completed then the first step is to sort out the GitHub runner that runs on the virtual machine on the Proxmox host. Go to the runners page for the clinch-home repository and press "New self-hosted runner". Select "Linux" as the OS and keep the instructions on the page while we connect to the host.

Open a terminal and SSH into the server. Navigate to the /usr/local/bin directory so that SELinux doesn't block the service script from being run. Run the commands from the GitHub help page. After running the first command in the "configure" section dependencies will likely need to be installed. Run this command:

sudo ./bin/installdependencies.sh

Finally, instead of running the run script, start a service instead by running the following:

sudo ./svc.sh install

sudo ./svc.sh start

Install the required packages

Hashicorp (Packer/Terraform)

sudo yum install -y yum-utils

sudo yum-config-manager --add-repo https://rpm.releases.hashicorp.com/RHEL/hashicorp.repo

sudo yum -y install terraform packer

Ansible

sudo yum -y install python39

sudo update-alternatives --config python3

sudo python3 -m pip install ansible-core requests

NodeJS

dnf groupinstall "Development Tools"

dnf module install nodejs

az cli

sudo rpm --import https://packages.microsoft.com/keys/microsoft.asc

sudo dnf install -y https://packages.microsoft.com/config/rhel/8/packages-microsoft-prod.rpm

sudo dnf install azure-cli

Configuring Kubernetes

SSH into kubeworker-1 and run the following command to initialise the cluster:

sudo kubeadm init

Copy the last line of the output and run it on the other two nodes. It should look something like this:

sudo kubeadm join 192.168.2.161:6443 --token <SOME_TOKEN> \
        --discovery-token-ca-cert-hash <SOME_HASH>>

Once that's done go back to the master node and run these commands

mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
kubectl get nodes

You should see something similar to this:

NAME               STATUS   ROLES           AGE     VERSION
edi-kubeworker-1   Ready    control-plane   3m10s   v1.24.3
edi-kubeworker-2   Ready    <none>          79s     v1.24.3
edi-kubeworker-3   Ready    <none>          74s     v1.24.3