Evidence / homelab operations

Addresses, hostnames, account details, and private configuration are intentionally omitted.

Dated records from changes that did not always go cleanly.

Environment scope

This is a self-run lab, not business production experience.

A place to practice operational habits before the first IT role.

Compute: 3 Proxmox nodes
Network edge: OPNsense firewall
Switching: Managed VLAN trunks
Recovery: Backups + change notes

Simplified service path

Internet
   |
OPNsense firewall
   |
Managed switching
   |
Proxmox cluster
   |
Identity / monitoring / storage / services

Separated network roles

Management and infrastructure
Internal applications
Public ingress and reverse proxy
Storage and backup traffic
Guest, IoT, testing, and game services

Sanitized record

Three changes with checks and outcomes.

May 10, 2026 / cluster quorum

A switch migration interrupted two Proxmox node paths. I traced the affected ports, restored VLAN reachability, confirmed all three nodes were quorate, saved the switch configuration, and added the recovery order to the change record.

May 26, 2026 / guest access

The captive portal was bound to the wrong logical interface. I corrected the binding, cleared stale bypass sessions, checked the persisted configuration and service state, then validated the portal with a fresh mobile connection.

May 31, 2026 / DNS migration

I moved client-facing DNS to Technitium, updated DHCP resolver assignments across VLANs, disabled the old resolver path, and documented the new filtering and split-horizon behavior.

What this proves

Hands-on practice, with the limits stated clearly.

This evidence shows repeated practice with dependencies, validation, rollback thinking, and documentation. It does not claim enterprise scale, customer impact, uptime guarantees, or production ownership.

Return to the case study →