Holocron Logs

The original plan was Ghost Pro to Azure VM. A real AZ-104 deployment, production operations, unified monitoring. The actual decision was Ghost Pro to Astro Content Collections on Netlify. Here is why the plan changed, what was built instead, and what that decision says about how infrastructure choices should actually be made.

A three-layer backup strategy for a three-node Proxmox cluster. Proxmox snapshots with retention tiers, nightly vzdump to a local external drive on Node-B, and an azcopy pipeline pushing everything to Azure Blob Storage at 3AM. The bugs that nearly killed the first run are worth documenting.

RustDesk deployed as Bothawui (LXC CT 114) on Node-B at 192.168.20.81. The Azure relay was rejected because Tailscale already handles NAT traversal. Self-hosted means no subscription, no data leaving the fleet, and a relay server that actually understands the network it is running on.

Home Assistant (Kashyyyk, VM 116) on VLAN 30 Mustafar. The deployment took an hour. The debugging took three more. A missing VLAN tag on Proxmox net0, Isolate Network silently blocking all inter-VLAN traffic, and a firewall rule source type that only matched the gateway. All three resolved before InfluxDB data started flowing.

Built a Windows Server 2022 domain controller from scratch, stood up a full OU structure with cascading GPOs, domain-joined a Windows 11 Pro workstation, and connected the on-premises forest to Azure via Entra Connect Password Hash Sync. The Server Core mistake cost four hours. The rebuild was worth documenting.

Wazuh already detects vulnerabilities. It scans every enrolled agent, inventories installed packages, and cross-references them against the National Vulnerability Database. This workflow bridges the gap!

A unified observability bot that replaces the sprawl with a single, heavily isolated entry point. It reads everything, queries anything, and can't modify a thing.

A custom Discord bot powered by Anthropic's Claude API. It lives in the Alliance Fleet Discord server and serves as an interactive knowledge assistant.

Most homelab automation falls into two categories: scheduled scripts in cron that hope for the best, or one-off webhooks that fire alerts nobody reads. Neither reflects how production environments actually work.

Uptime Kuma runs lightweight HTTP/TCP/ping checks against every service endpoint and fires a Discord webhook the moment something goes down. It's the canary - fast, dumb, and reliable.

The goal is to pass the entire NVIDIA RTX 4000 Ada GPU through to a single VM. So it has exclusive, bare-metal-equivalent access to the hardware.

Twelve services, twelve passwords. No centralized authentication. No audit trail showing who logged into what and when.

Why Build an Observability Stack Running 25+ services across 3 nodes without observability is running blind. You find out something's wrong when a user (you) notices a service is down. By then, the problem has been...

The Problem My portfolio at tima.dev , hosted on Netlify, a fully public site started showing a Chrome prompt: 'Access other devices on your local network.' When I clicked Block, every Ghost blog feature image broke....

Job Radar is an n8n workflow that runs every 6 hours and posts matching jobs to a Discord channel.

Nginx Proxy Manager stopped accepting logins. I spent an hour trying to fix it before realizing the whole container was compromised from a bad initial deployment.

Access every homelab service from anywhere (office, phone, travel) over HTTPS with valid SSL certificates, without exposing a single port on my home router.

Partition Expansion on a Live Proxmox VM - From 3GB to 128GB Without Data Loss

This post documents every failure I hit while enrolling 10 agents across a 3-node Proxmox cluster with 4-VLAN segmentation, and the reusable deployment script that came out of it.

Grafana OAuth with Authentik: The `root_url` Gotcha

This post covers how I deployed Authentik as a centralized identity provider for the fleet: the architecture decisions, the integration work, and the debugging that came with it.

How I built a complete security monitoring pipeline across a 3-node Proxmox homelab, from Wazuh agent enrollment to real-time Discord alerts, and the silent HTTP 400 bug that almost killed the whole thing.

Running LLM inference and image generation on a homelab GPU node, and everything that went wrong along the way.

Wazuh an open-source security platform that combines endpoint monitoring, log analysis, vulnerability detection, and regulatory compliance into a single manager-agent architecture.

NPM + UniFi Firewall Rule Ordering - The Silent Traffic Drop

Wazuh: When to Stop Fighting and Use the Script The Goal Deploy a full Wazuh SIEM stack - manager, indexer, and dashboard - on a dedicated LXC container in the homelab. Centralized log collection, threat detection, file...

Diagnosing a Silent Crash with No Logs Category: Node-A went down on a Sunday morning and left me nothing. No kernel panic. No crash dump. No syslog entries from the window that mattered. The machine was just... off....