My Homelab Current Setup

This is the current state of my homelab after the latest build session.

The goal of this phase was not to make everything perfect. The goal was to build a real working foundation: compute, shared storage, local DNS, downloads, monitoring, dashboards, and a simple homepage to access everything from one place.

And now the base is alive.

What we have built so far

The homelab now has these core services running:

Proxmox cluster for compute
Shared NFS storage from node2
Pi-hole for local DNS and ad blocking
Homepage dashboard for quick access to services
Torrent/download LXC on node3
qBittorrent Web UI for downloads
File Browser for browsing completed downloads
Monitoring VM on node4
Prometheus for metrics collection
Grafana for dashboards
Node Exporter on Proxmox nodes and the monitoring VM
Local DNS domain using home.lan

Architecture overview

Proxmox Cluster
│
├── node1
│   ├── CT 102: Pi-hole
│   └── CT 103: Homepage
│
├── node2
│   └── Shared NVMe NFS Storage
│       └── /shared
│           ├── template/iso
│           ├── template/cache
│           ├── dump
│           └── snippets
│
├── node3
│   └── CT 233: torrent-lxc
│       ├── qBittorrent Web UI
│       └── File Browser UI
│
├── node4
│   └── VM 240: monitor-vm
│       ├── Prometheus
│       ├── Grafana
│       └── Node Exporter
│
├── node5
│   └── Future workloads
│
└── Proxmox Nodes
    └── Node Exporter on each node

Network plan

The homelab is currently using the LAN subnet:

192.168.0.0/24

The primary local domain is:

home.lan

Pi-hole is used as the local DNS server. The idea is simple: instead of remembering every IP address and port, services get readable local names like grafana.home.lan, pihole.home.lan, and downloads.home.lan.

Known service inventory

Service	IP / Host	ID	Purpose
Pi-hole	`192.168.0.200`	CT `102`	DNS and ad blocking
Homepage	`192.168.0.210`	CT `103`	Main homelab dashboard
Torrent LXC	`192.168.0.233`	CT `233`	qBittorrent and File Browser
Monitor VM	`192.168.0.240`	VM `240`	Prometheus, Grafana, Node Exporter
Node2 Storage	`node2 IP`	Host	Shared NFS storage
Proxmox main record	`192.168.0.121`	Host	`proxmox.home.lan` currently points here

Current DNS records

Domain	IP	Purpose
`home.lan`	`192.168.0.210`	Homepage dashboard
`pihole.home.lan`	`192.168.0.200`	Pi-hole admin
`torrent.home.lan`	`192.168.0.233`	Torrent server base host
`qbittorrent.home.lan`	`192.168.0.233`	qBittorrent Web UI
`downloads.home.lan`	`192.168.0.233`	File Browser downloads UI
`monitor.home.lan`	`192.168.0.240`	Monitoring VM
`prometheus.home.lan`	`192.168.0.240`	Prometheus UI
`grafana.home.lan`	`192.168.0.240`	Grafana UI
`storage.home.lan`	`node2 IP`	Shared NFS storage node
`proxmox.home.lan`	`192.168.0.121`	Main Proxmox UI record

Current service URLs

Some services still use ports directly. Later, a reverse proxy like Caddy or Nginx Proxy Manager can clean this up.

Service	URL
Homepage	`http://home.lan`
Pi-hole	`http://pihole.home.lan/admin`
qBittorrent	`http://qbittorrent.home.lan:8080`
File Browser	`http://downloads.home.lan:8081`
Prometheus	`http://prometheus.home.lan:9090`
Prometheus targets	`http://prometheus.home.lan:9090/targets`
Grafana	`http://grafana.home.lan:3000`
Proxmox main	`https://proxmox.home.lan:8006`
Proxmox node direct	`https://NODE-IP:8006`

Shared storage: node2 NFS

node2 is currently acting as the shared storage server for cluster assets.

This storage is for:

ISO images
LXC templates
VZDump backups
Snippets

It is not currently being used for live VM disks.

Storage mount

The NVMe storage on node2 is mounted at:

/shared

Expected folder structure:

/shared
├── template
│   ├── iso
│   └── cache
├── dump
└── snippets

NFS export

Current NFS export:

/shared 192.168.0.0/24(rw,sync,no_subtree_check,no_root_squash)

Check exports:

exportfs -v

Expected output should include something like:

/shared 192.168.0.0/24(...,rw,...,no_root_squash,...)

Proxmox storage entry

In the Proxmox GUI:

Datacenter → Storage → shared-nfs

Enabled content types:

ISO image
Container template
VZDump backup file
Snippets

Important rule for now:

Do not use shared-nfs for live VM disks yet.

Torrent/download server: node3 `torrent-lxc`

The torrent server is a dedicated LXC container for Linux ISOs and other legal downloads.

Container details

Setting	Value
CT ID	`233`
Hostname	`torrent-lxc`
IP	`192.168.0.233/24`
Gateway	`192.168.0.1`
DNS	`192.168.0.200`
Storage	node3 NVMe / `nvme3`
Disk	~400 GB
CPU	2 cores
RAM	2 GB

qBittorrent

qBittorrent Web UI:

http://192.168.0.233:8080
http://qbittorrent.home.lan:8080

Service check:

systemctl status qbittorrent-nox

Download folders:

/downloads/incomplete
/downloads/complete

File Browser

File Browser UI:

http://192.168.0.233:8081
http://downloads.home.lan:8081

File Browser root:

/downloads/complete

Service check:

systemctl status filebrowser

Change File Browser root:

systemctl stop filebrowser
filebrowser config set -d /etc/filebrowser/filebrowser.db -r /downloads/complete
systemctl start filebrowser

Copy completed ISO files to shared Proxmox storage

From inside torrent-lxc:

rsync -avh --progress /downloads/complete/*.iso root@NODE2-IP:/shared/template/iso/

After copying, the ISO should appear in Proxmox:

Datacenter → shared-nfs → ISO Images

Monitoring VM: node4 `monitor-vm`

The monitoring VM is the central observability box.

It currently runs:

Prometheus
Grafana
Node Exporter

VM details

Setting	Value
VM ID	`240`
Name	`monitor-vm`
Node	`node4`
IP	`192.168.0.240/24`
Gateway	`192.168.0.1`
DNS	`192.168.0.200`
Search domain	`home.lan`
CPU	2 cores
RAM	4 GB
Disk	60 GB
OS	Ubuntu Server

Firewall ports

The monitoring VM allows:

22/tcp    SSH
9090/tcp  Prometheus
9100/tcp  Node Exporter
3000/tcp  Grafana

Check UFW:

sudo ufw status

Prometheus

Prometheus is used for metrics collection.

Access

http://192.168.0.240:9090
http://prometheus.home.lan:9090

Service commands

Check service:

sudo systemctl status prometheus

Health check:

curl http://localhost:9090/-/healthy

Expected output:

Prometheus Server is Healthy.

Config path:

/etc/prometheus/prometheus.yml

Current scrape jobs

Prometheus currently scrapes:

Prometheus itself
monitor-vm Node Exporter
Proxmox node exporters

Example config shape:

global:
  scrape_interval: 15s
  evaluation_interval: 15s

scrape_configs:
  - job_name: "prometheus"
    static_configs:
      - targets: ["localhost:9090"]

  - job_name: "monitor-vm"
    static_configs:
      - targets: ["localhost:9100"]

  - job_name: "proxmox-nodes"
    static_configs:
      - targets:
          - "NODE1-IP:9100"
          - "NODE2-IP:9100"
          - "NODE3-IP:9100"
          - "NODE4-IP:9100"
          - "NODE5-IP:9100"
        labels:
          group: "proxmox"

Validate config before restarting:

sudo promtool check config /etc/prometheus/prometheus.yml

Restart Prometheus:

sudo systemctl restart prometheus

Open targets page:

http://prometheus.home.lan:9090/targets

Node Exporter

Node Exporter exposes system metrics like:

CPU usage
RAM usage
Disk usage
Network traffic
Load average
Uptime

It is installed on:

monitor-vm
Proxmox node1
Proxmox node2
Proxmox node3
Proxmox node4
Proxmox node5
Additional Proxmox endpoint 192.168.0.125, visible in Prometheus targets

Each machine exposes metrics at:

http://NODE-IP:9100/metrics

Test locally on each node:

curl http://localhost:9100/metrics | head

Check service:

systemctl status node_exporter

Grafana

Grafana is used for dashboards and visualization.

Access

http://192.168.0.240:3000
http://grafana.home.lan:3000

Service

sudo systemctl status grafana-server

Data source

Grafana uses Prometheus as the data source.

Prometheus URL inside Grafana:

http://localhost:9090

Dashboard

Imported dashboard:

Node Exporter Full
Dashboard ID: 1860

This dashboard shows:

CPU usage
RAM usage
Disk usage
Network traffic
Load average
Uptime
Per-node metrics

Homepage dashboard

Homepage runs as CT 103 on node1.

Current DNS record:

home.lan → 192.168.0.210

Access URL:

http://home.lan

Config path:

/opt/homepage/config/services.yaml

Project folder:

/opt/homepage

Restart Homepage:

cd /opt/homepage
docker compose restart

Current Homepage services

- Downloads:
    - qBittorrent:
        href: http://qbittorrent.home.lan:8080
        description: Torrent download manager
        icon: qbittorrent.png
        siteMonitor: http://qbittorrent.home.lan:8080

    - File Browser:
        href: http://downloads.home.lan:8081
        description: Browse completed downloads
        icon: filebrowser.png
        siteMonitor: http://downloads.home.lan:8081

- Monitoring:
    - Grafana:
        href: http://grafana.home.lan:3000
        description: Monitoring dashboards
        icon: grafana.png
        siteMonitor: http://grafana.home.lan:3000

    - Prometheus:
        href: http://prometheus.home.lan:9090
        description: Metrics and time-series database
        icon: prometheus.png
        siteMonitor: http://prometheus.home.lan:9090

- Storage:
    - Shared NFS Storage:
        href: http://storage.home.lan
        description: Node2 shared NVMe storage for ISOs, templates, backups
        icon: proxmox.png

Note: storage.home.lan may not show as online unless there is a web service running on that host. Right now it mostly acts as a DNS/link reference for the storage node.

Current Proxmox inventory

Datacenter: myworld
│
├── node1
│   ├── CT 102: pihole
│   ├── CT 103: homepage
│   ├── VM 100: test VM, ignored
│   ├── VM 101: test VM, ignored
│   ├── local
│   ├── local-lvm
│   └── shared-nfs
│
├── node2
│   ├── local
│   ├── local-lvm
│   └── shared-nfs
│
├── node3
│   ├── CT 233: torrent-lxc
│   ├── local
│   ├── local-lvm
│   ├── nvme3
│   └── shared-nfs
│
├── node4
│   ├── VM 240: monitor-vm
│   ├── local
│   ├── local-lvm
│   ├── nvme4
│   └── shared-nfs
│
└── node5
    ├── local
    ├── local-lvm
    └── shared-nfs

VM 100 and VM 101 are test VMs and should be ignored in the main documentation unless they become real services later.

Validation checklist

Shared storage

On node2:

df -h | grep /shared
exportfs -v

In Proxmox:

Datacenter → Storage → shared-nfs

Confirm that ISO images, container templates, backups, and snippets are enabled.

Torrent LXC

Inside torrent-lxc:

systemctl status qbittorrent-nox
systemctl status filebrowser

Browser checks:

http://qbittorrent.home.lan:8080
http://downloads.home.lan:8081

Monitoring VM

Inside monitor-vm:

sudo systemctl status prometheus
sudo systemctl status grafana-server
sudo systemctl status node_exporter

Browser checks:

http://prometheus.home.lan:9090/targets
http://grafana.home.lan:3000

Proxmox node exporters

On each Proxmox node:

systemctl status node_exporter
curl http://localhost:9100/metrics | head

What this build taught me

The biggest lesson from this build is that every homelab service follows the same pattern:

Storage → Network → User/Permissions → Service/Systemd → Firewall/DNS → Monitoring → Homepage

This pattern showed up again and again:

NFS storage needed mount points, exports, permissions, and Proxmox storage mapping.
qBittorrent needed storage folders, service setup, ports, DNS, and Homepage entry.
File Browser needed a root folder, systemd service, port, DNS, and Homepage entry.
Prometheus needed scrape configs, service validation, firewall rules, and target checks.
Grafana needed Prometheus as a data source, dashboards, and local DNS access.
Node Exporter needed service installation on every node and Prometheus scraping.

That reusable mental model is now the operating system for future homelab work.

Next planned build: Personal Vault VM

The next planned service group is:

personal-vault-vm

Planned IP:

192.168.0.250

Planned services:

Tailscale subnet router
RustDesk server
Possibly Nginx Proxy Manager or Caddy

Planned DNS records:

Domain	IP
`vault.home.lan`	`192.168.0.250`
`tailscale.home.lan`	`192.168.0.250`
`rustdesk.home.lan`	`192.168.0.250`

Future improvements

Potential next improvements:

Add Proxmox API exporter for VM/container-level metrics
Add alerting in Grafana or Prometheus Alertmanager
Add Nginx Proxy Manager or Caddy for clean no-port URLs
Add Tailscale for secure remote access
Add RustDesk server for remote desktop relay
Add backups for configs and VMs
Add a documentation repo for the homelab setup
Add Ansible scripts for repeatable installs
Consider Ceph later for true distributed storage across node NVMe drives

Current status

This homelab now has a real foundation:

Compute:      Proxmox cluster
Storage:      node2 shared NFS
DNS:          Pi-hole + home.lan
Dashboard:    Homepage
Downloads:    qBittorrent + File Browser
Monitoring:   Prometheus + Grafana + Node Exporter
Next step:    Personal Vault VM

The important part is that it is no longer just a group of machines. It is becoming a small, organized platform.

My Homelab Current Setup

My Homelab Current Setup

What we have built so far

Architecture overview

Network plan

Known service inventory

Current DNS records

Current service URLs

Shared storage: node2 NFS

Storage mount

NFS export

Proxmox storage entry

Torrent/download server: node3 torrent-lxc

Container details

qBittorrent

File Browser

Copy completed ISO files to shared Proxmox storage

Monitoring VM: node4 monitor-vm

VM details

Firewall ports

Prometheus

Access

Service commands

Current scrape jobs

Node Exporter

Grafana

Access

Service

Data source

Dashboard

Homepage dashboard

Current Homepage services

Current Proxmox inventory

Validation checklist

Shared storage

Torrent LXC

Monitoring VM

Proxmox node exporters

What this build taught me

Next planned build: Personal Vault VM

Future improvements

Current status

Torrent/download server: node3 `torrent-lxc`

Monitoring VM: node4 `monitor-vm`