Design Proxmox High Availability cluster Hetzner

Proxmox Cluster High Reliability Hetzner

1 February 2024 / Commenti

Find out how we created a high-reliability Proxmox Cluster on Hetzner, optimized for parallel computation, LLM development, and workload continuity.

Introduction

In this article, we will explore the process of creating a High-Reliability Proxmox Cluster on Hetzner. This project was designed to support parallel computation for clients, provide pre-configured images with development environments for AI and LLM, ensure continuity of workloads, and minimize costs.

Objectives and Challenges

The goal of this project was to create a high-reliability hosting environment to support the following business needs:

  • Parallel Computation for Clients: The need to run computationally intensive workloads in parallel mode to speed up processing times.
  • Preconfigured images for LLM: The provision of VM images preconfigured with development environments for large language models (LLMs).
  • Workload Continuity: Ensuring business continuity of critical VMs in the event of hardware or software failures.
  • Cost Minimization: Optimizing IT infrastructure costs using open source solutions and efficient hardware components.

The main challenge was to design and implement a reliable, scalable, and manageable cluster configuration while keeping costs down.

Business Requirements

The business requirements for this project were as follows:

  • High Availability:Critical VMs must be available at all times, with little or no downtime.
  • Fault Tolerance: The system must be able to tolerate hardware or software failures on a single node without service interruption.
  • Scalability:The infrastructure must be scalable to support future workload growth.
  • Cost-efficiency:The solution must be built with open source hardware and software components and with low cost of ownership.
  • Ease of Management: The infrastructure must be easy to manage and monitor.

Project Description

To meet the business requirements, a high-reliability Proxmox cluster on Hetzner servers was designed and implemented. The cluster configuration includes the following elements:

  • Number of Proxmox nodes: 3
  • Network architecture: Using vSwitch for networking
  • Delocalized backup on QNap NAS

Architecture

The cluster consists of three Proxmox nodes, each hosting a dedicated Hetzner server. The nodes are connected to each other via a high-speed network.

Network

A virtual vSwitch is used to create a private network within the provider. VMs and hypervisors are interconnected through this private network and can communicate with each other without having to traverse the public network.

Delocalized backup

Backups of VMs are performed regularly on a QNap NAS outside the cluster and on object storage buckets. This ensures data protection in case of disaster recovery.

Replication VM

Replication of virtual machines is done by synchronization across the network. In case of failure of the primary node, the VM is automatically triggered on the backup node via the Proxmox HA stack.

High reliability

High reliability is achieved using a combination of VM replication and network failover. In the event of a primary node failure, the VM is automatically activated on the backup node and the network is configured to direct traffic to the new primary node.

Advantages and benefits

  • Ease of management:The infrastructure is easy to manage and monitor using the Proxmox web interface and standard monitoring tools.
  • Cost-effective:The Hetzner offering has excellent pricing and good reliability of bare metal hardware.

Technical details

Configuration of Proxmox VE:

  • Version of Proxmox VE: 8
  • Storage: local SSD and network storage
  • Network: private vSwitch with VLAN
  • Security: Proxmox VE built-in firewall

Network settings:

  • Static IP addresses for cluster nodes
  • VLANs for the cluster's private network

External resources:

Frequently Asked Questions (FAQs)

D: What are the advantages of a high-reliability Proxmox cluster?

R: The advantages of a high-reliability Proxmox cluster include:

  • High availability: Critical VMs are always available, with little or no downtime.
  • Fault Tolerance: The system can tolerate hardware or software failures on a single node without service interruption.
  • Scalability: The infrastructure is scalable to support future workload growth.
  • Cost efficiency: The solution is built with open source hardware and software components and low cost of ownership.
  • Ease of Management: The infrastructure is easy to manage and monitor.

D: What are the components needed to create a high-reliability Proxmox cluster?

R: The components needed to create a high-reliability Proxmox cluster include:

  • Three or more dedicated Hetzner servers
  • Proxmox VE software
  • Network storage

D: How can I monitor my high-reliability Proxmox cluster?

R: There are several ways to monitor a Proxmox high-reliability cluster, including:

  • The Proxmox VE web interface
  • Third-party monitoring tools
  • CustomScripts

Call to action

Are you looking for a reliable and scalable hosting solution for your critical VMs? Contact us today to find out how we can help you set up a high-reliability Proxmox cluster on Hetzner servers.

Benefits of working with us:

  • Many years of experience in configuring and managing Proxmox clusters
  • In-depth knowledge of Proxmox VE, networking and scripting
  • Customer-oriented approach and dedicated support service
  • Custom solutions to meet your specific needs

Visit our Proxmox Consulting and Support for more details, and contact us for a free personalized consultation!

Proxmox