Skip to content

Latest commit

 

History

History
155 lines (110 loc) · 5.56 KB

README-external-host.md

File metadata and controls

155 lines (110 loc) · 5.56 KB

External Host Initialization for Two-Node Toolbox

This document explains how to use Two-Node Toolbox (TNT) to install Two-Node OpenShift clusters on external RHEL hosts that are not provisioned through the AWS hypervisor automation. This workflow is designed for environments like Beaker, lab systems, or any pre-existing RHEL 9 hosts.

Overview

The init-host.yml playbook provides the same host initialization functionality as the AWS hypervisor creation scripts, preparing your external RHEL host to run OpenShift two-node cluster deployments. It replaces the AWS-specific initialization steps with Ansible automation that works on any RHEL 9 system.

Prerequisites

Host Requirements

  • Operating System: RHEL 9.x with minimal installation
  • Hardware: 64GB+ RAM, 500GB+ storage (with sufficient space in /home)
  • Network: Internet access for package downloads and registry access
  • Access: SSH access with sudo privileges

Controller Requirements

  • Ansible installed on your local machine
  • SSH key pair for authentication
  • Valid Red Hat subscription credentials (activation key recommended)

Setup Process

1. Configure Inventory

Copy the sample inventory file and configure it with your host details:

cd deploy/openshift-clusters
cp inventory.ini.sample inventory.ini

Edit inventory.ini with your external host information:

[metal_machine]
root@your-host-ip ansible_ssh_extra_args='-o ServerAliveInterval=30 -o ServerAliveCountMax=120'

[metal_machine:vars]
ansible_become_password=""

Important: Replace your-host-ip with the actual IP address or hostname of your RHEL system.

Tip: To skip the -i inventory.ini argument in all ansible commands, copy the inventory file to Ansible's default location (/etc/ansible/hosts on Linux, may vary on other operating systems).

2. Configure RHSM Credentials

You have several options for providing Red Hat subscription credentials:

Option A: Environment Variables (Recommended)

export RHSM_ACTIVATION_KEY="your-activation-key"
export RHSM_ORG="your-organization-id"

See hands-off deployment for more details on how to obtain these values

Option B: Local Variable File

cp vars/init-host.yml vars/init-host.yml.local
# Edit vars/init-host.yml.local with your credentials

Option C: Command Line

ansible-playbook init-host.yml -i inventory.ini \
  -e "rhsm_activation_key=your-key" \
  -e "rhsm_org=your-org"

3. Run Host Initialization

Execute the initialization playbook:

# Using environment variables or local config file
ansible-playbook init-host.yml -i inventory.ini

# Or with command line parameters
ansible-playbook init-host.yml -i inventory.ini \
  -e "rhsm_activation_key=your-key" \
  -e "rhsm_org=your-org"

4. What the Playbook Does

The init-host.yml playbook performs the following tasks to replicate AWS hypervisor initialization:

Host Configuration

  • Sets system hostname to match your deployment environment
  • Adds SSH host keys to prevent connection prompts
  • Creates pitadmin user with sudo access and random password

Subscription Management

  • Configures Red Hat Subscription Manager
  • Registers system using activation key or interactive credentials
  • Enables required repositories:
    • RHEL 9 BaseOS and AppStream
    • OpenShift Container Platform repositories

Package Installation

  • Installs essential development tools:
    • git - Required for dev-scripts
    • make - Essential for running dev-scripts Makefiles
    • golang - Required for Go-based tooling
    • cockpit - Web-based system management
    • lvm2 - Logical volume management
    • jq - JSON processing tool

Storage Configuration

  • If you need to configures dev-scripts to use a different path (/home/dev-scripts instead of /opt/dev-scripts, for example), add the following variable to your config_XXX.sh file export WORKING_DIR="/home/dev-scripts"
  • This might help you have sufficient disk space for OpenShift cluster deployment (80GB+ required)

Transition to OpenShift Deployment

After successful host initialization, your external RHEL system is ready for OpenShift cluster deployment. You can now proceed with the standard Two-Node Toolbox workflow:

Deploy Two-Node Cluster

Choose your preferred topology and run the setup playbook:

Arbiter Topology (Two-Node with Arbiter)

# Interactive mode
ansible-playbook setup.yml -i inventory.ini

# Non-interactive mode
ansible-playbook setup.yml -e "topology=arbiter" -e "interactive_mode=false" -i inventory.ini

Fencing Topology (Two-Node with Fencing)

# Non-interactive mode
ansible-playbook setup.yml -e "topology=fencing" -e "interactive_mode=false" -i inventory.ini

Alternative: kcli-based Deployment

For fencing topology using kcli:

ansible-playbook kcli-install.yml -i inventory.ini

Post-Deployment Features

Cluster VM Inventory Access

After successful cluster deployment (using either setup.yml or kcli-install.yml), the inventory file is automatically updated to include the cluster VMs. This allows you to run Ansible playbooks directly on the cluster nodes from your local machine.

The deployment automatically discovers running cluster VMs and adds them to the inventory with ProxyJump configuration through the hypervisor. For more details on using this feature, see: