Hayden VanderWeele
← All work

Enterprise Proxmox Homelab & Network Infrastructure

Role —
Systems Administrator & Network Architect
Date —
July 1, 2026
Stack —
Proxmox, Docker, Linux, TrueNAS/ZFS, Grafana

Objective

The goal of this ongoing infrastructure project is to build and maintain a scalable, fault-tolerant home server ecosystem capable of hosting containerized services, high-bandwidth media streaming, and secure remote access. The primary engineering constraint is ensuring strict service isolation, data integrity, and high availability on bare-metal hardware.

My Role & Contribution

As the sole architect and administrator, I designed the system from the bare metal up to the application layer:

  • Provisioning a bare-metal hypervisor cluster (Dell R710 and HP ProDesk).
  • Designing network isolation and remote tunnel access via WireGuard.
  • Deploying a comprehensive Docker/Portainer container stack across isolated Debian VMs.
  • Implementing enterprise-grade telemetry and log aggregation using Prometheus, Loki, and Grafana.

Engineering Process

1. Bare-Metal Virtualization & Storage Architecture

I deployed a Proxmox VE cluster across my bare-metal hardware to act as the foundational hypervisor. To ensure robust data integrity and snapshot capabilities, I virtualized TrueNAS and configured a ZFS storage pool (“Angry-Storage”) with direct disk pass-through, creating a resilient backbone for all hosted services and continuous file synchronization via Syncthing from my Arch Linux daily-driver workstations.

2. Container Orchestration & Secure Networking

Instead of running services directly on the host, I spun up isolated Debian VMs to serve as Docker hosts managed via Portainer. I deployed a complex media and automation stack (Jellyfin, Navidrome, the *arr suite) and engineered secure network routing—specifically tunneling sensitive container traffic (like qBittorrent) exclusively through a Gluetun VPN container to prevent IP leaks. I also configured a WireGuard VPN to provide seamless, secure remote access to the internal network.

3. Telemetry & Fail-Safe Management

To maintain observability over the infrastructure, I built a production-tier monitoring stack. I deployed Node Exporter and cAdvisor to scrape hardware and container metrics, aggregating them into Prometheus and visualizing them on custom Grafana dashboards alongside Loki for centralized logging. Furthermore, to ensure I could recover the server from anywhere, I configured out-of-band management via the Dell R710’s iDRAC 6 controller to handle power losses and dynamic IP shifts.

Results & Impact

  • Production-Grade Reliability: Successfully architected an environment that mirrors enterprise cloud infrastructure, achieving high uptime and automated recovery capabilities.
  • Full-Stack Competency: The homelab serves as a continuous, practical sandbox that bridges my knowledge of physical hardware constraints (power, thermals, iDRAC) with scalable software deployment and network administration.