Key Points
- Research suggests Harvester Hypervisor setup involves checking hardware, downloading the ISO, booting from it, and following an installation wizard.
- It seems likely that the process is straightforward for users with basic server knowledge, but may require attention to network and disk configurations.
- The evidence leans toward needing at least 8 cores, 32 GB RAM, and 250 GB disk for testing, with higher specs for production.
Prerequisites and Setup
Before starting, ensure your server meets Harvester’s hardware requirements:
- CPU: x86_64 with virtualization support, at least 8 cores for testing, 16 for production.
- Memory: At least 32 GB for testing, 64 GB for production.
- Disk: At least 250 GB (or 180 GB with multiple disks) for testing, 500 GB for production, using SSD/NVMe with 5,000+ IOPS.
- Network: 1 Gbps Ethernet minimum for testing, 10 Gbps for production, with at least 2 NICs recommended for bonding.
Download the latest Harvester ISO from the official releases page: Harvester Releases.
Installation Steps
Follow these steps to set up Harvester:
- Boot from ISO: Insert the bootable USB/CD with the Harvester ISO and boot your server, selecting Harvester Installer.
- Choose Installation Mode: Select Create a new Harvester cluster for a fresh setup, or join an existing cluster if applicable.
- Configure Disks and Network: Choose installation and data disks, set persistent size (minimum 150 GiB), and configure network (DHCP or static IP).
- Set Host and Admin Details: Enter a hostname, cluster token, and admin password.
- Complete Installation: Review settings, confirm, and wait for the node to restart.
- Access Web Interface: Log in at the provided URL (e.g., https://your-virtual-ip) with the admin account, setting the password on first login.
Survey Note: Detailed Guide to Setting Up Harvester Hypervisor
This detailed guide explores the process of setting up Harvester Hypervisor, an open-source hyperconverged infrastructure (HCI) solution built on Kubernetes, designed for running both virtual machines (VMs) and containerized workloads. The information is drawn from the official Harvester documentation, GitHub releases, and related resources, ensuring a comprehensive understanding as of April 2025.
Background and Context
Harvester is a modern, open-source alternative to proprietary HCI solutions like vSphere and Nutanix, built on cutting-edge technologies such as Linux, KVM, Kubernetes, KubeVirt, and Longhorn. It runs on bare-metal servers, providing integrated virtualization and distributed storage capabilities. This guide focuses on setting up Harvester using the ISO installation method, based on version v1.4, which is assumed to be the latest stable version given the available documentation.
Detailed Setup Process
Step 1: Check Hardware Requirements
Before proceeding, ensure your server meets the minimum hardware requirements for Harvester, as outlined in the official documentation:
| Component | Development/Testing | Production |
| CPU | x86_64 with hardware-assisted virtualization, 8 cores minimum | x86_64 with hardware-assisted virtualization, 16 cores minimum |
| Memory | 32 GB minimum | 64 GB minimum |
| Disk | 250 GB minimum (180 GB with multiple disks), SSD/NVMe, 5,000+ random IOPS per disk | 500 GB minimum, SSD/NVMe, 5,000+ random IOPS per disk |
| Network | 1 Gbps Ethernet minimum, 1 NIC required, 2 NICs recommended for bonding | 10 Gbps Ethernet minimum, 1 NIC required, at least 2 NICs recommended for bonding |
- Important Notes:
- Use YES-certified hardware for SUSE Linux Enterprise Server (SLES) compatibility, as recommended for SLES 15 SP3 or SP4.
- Laptops and nested virtualization are not supported.
- Each node must have a unique product_uuid (fetched from /sys/class/dmi/id/product_uuid).
- For high availability and best performance, use at least two NICs in each node for a bonded NIC for the management network.
Step 2: Download the Harvester ISO
- Visit the official Harvester GitHub releases page to download the latest stable ISO image:
- Harvester Releases
- Select the most recent stable version (e.g., v1.4.2) and download the ISO file. Ensure you choose the correct architecture (x86_64).
Step 3: Prepare Bootable Media
- Create a bootable USB drive or CD/DVD using the downloaded Harvester ISO. Tools like Rufus (for Windows) or dd (for Linux) can be used for this purpose.
- Insert the bootable media into your server and ensure it is detected.
Step 4: Boot from the ISO
- Power on your server and enter the BIOS/UEFI settings (if necessary) to set the boot order to prioritize the USB drive or CD/DVD.
- Save the changes and restart the server.
- When the boot menu appears, select Harvester Installer. The installer will perform initial hardware checks, and if requirements are not met, it will display warnings.
Step 5: Follow the Installation Wizard
The Harvester installer provides a guided wizard for configuration. Below are the detailed steps, as extracted from the documentation:
| Step | Description | Notes/Options |
| 0 | Mount Harvester ISO and boot server, select Harvester Installer. | Hardware checks displayed if minimum requirements not met. |
| 1 | Choose installation mode using arrow keys (default is first node as management node). | Options: Create a new Harvester cluster, Join an existing Harvester cluster, Install Harvester binaries only. |
| 2 | Choose node role (required if joining existing cluster). | Options: Default Role, Management Role, Witness Role, Worker Role. |
| 3 | Select installation disk and data disk, configure persistent size (default/minimum 150 GiB). | Uses GPT partitioning schema; can select MBR for BIOS boot. |
| 4 | Configure HostName of the node. | - |
| 5 | Configure network interface(s) for management network, IP via DHCP or static. | Bonded NIC mgmt-bo created; DHCP requires static MAC-to-IP mapping, must include option routers. |
| 6 | (Optional) Configure DNS Servers, use commas for multiple, blank for default. | - |
| 7 | Configure virtual IP (VIP) by selecting VIP Mode, ensure unique if using DHCP. | - |
| 8 | Configure Cluster token for adding other nodes. | - |
| 9 | Configure and confirm a Password for access, default SSH user is rancher. | - |
| 10 | Configure NTP servers, defaults to 0.suse.pool.ntp.org, use commas for multiple. | - |
| 11 | (Optional) Enter Proxy address for HTTP proxy, leave blank if not needed. | - |
| 12 | (Optional) Import SSH keys via HTTP URL, e.g., https://github.com/<username>.keys. | - |
| 13 | (Optional) Customize host with Harvester configuration file via HTTP URL. | Refer to /v1.4/install/harvester-configuration for config details. |
| 14 | Review and confirm installation options, installation takes a few minutes. | - |
| 15 | Node restarts after installation, access web interface at https://your-virtual-ip. | Use F12 to switch to Shell, type exit to return to console. |
| 16 | Set password for default admin user on first login. | Additional setup needed if Install Harvester binaries only chosen. |
- Network Configuration Notes:
- If using DHCP, ensure your DHCP server is configured with static MAC-to-IP mapping and includes the option routers directive for proper routing.
- For static IP, manually enter the IP address, subnet mask, gateway, and DNS servers.
Step 6: Access the Harvester Web Interface
- After the node restarts, the Harvester console will display the management URL, typically in the format https://your-virtual-ip.
- Open a web browser and navigate to this URL.
- Log in with:
- Username: admin
- Password: The password you set during installation (you will be prompted to set it on first login if not already done).
- Use F12 to switch from the Harvester console to the Shell if needed, and type exit to return to the console.
Step 7: Complete Initial Setup and Expand Cluster (Optional)
- Once logged into the web interface, you can manage VMs, storage, and other resources.
- For production environments, it is recommended to add additional nodes to the cluster for high availability. Use the cluster token set during installation to join new nodes, following the same installation process but selecting Join an existing Harvester cluster.
Additional Considerations and Troubleshooting
- CPU Compatibility: All nodes in the cluster must have CPUs with the same specifications for live migration to function correctly, especially for operations like automatic VM migration in Maintenance Mode.
- Known Issues:
- Installer crashes with older graphics cards: Workaround by appending vga=792 to the GRUB entry and forcing a resolution of 1024x768.
- Node join failures with FQDN and custom SSL: Fix by setting agent-tls-mode to system-store and adding the CA to the trust list.
- Community and Support: Harvester is community-supported via GitHub and forums, with optional enterprise support from SUSE for production use.
Suitability and Recommendations
This setup process is suitable for users with basic server administration knowledge, particularly those familiar with Kubernetes or looking to bridge VMs and containers. It is ideal for home labs, small to medium enterprises, and testing environments, with the option to scale for production by adding nodes. Users should ensure they meet hardware requirements and are prepared for potential learning curves, especially with network and disk configurations.
Conclusion
This guide provides a comprehensive step-by-step process for setting up Harvester Hypervisor, covering hardware checks, ISO installation, and initial configuration. By following these steps, users can establish a modern HCI solution for both virtualized and containerized workloads, with flexibility for future expansion and community support for troubleshooting.