HW Monitor

Small web dashboard for watching how hard your Linux box is working: temps, voltages, disk SMART, GPU load. Built for home servers and Unraid; runs in Docker or straight on the host.

Metrics are collected in a background thread so the UI stays snappy. Short history lives in memory, longer ranges in SQLite.

What it shows

Live and historical graphs for CPU and GPU utilization, motherboard-style voltages from lm-sensors, disk temperatures via smartctl, plus SMART details when you need them.

Auto-detection

• Disks: SMART_DEVICES=auto walks block devices (/dev/sda, /dev/nvme0n1, …).
• NVIDIA: ENABLE_NVIDIA=auto turns on when nvidia-smi exists.
• AMD / Intel GPUs: sysfs + hwmon when the stack exposes them (ENABLE_AMD, ENABLE_INTEL).

Pinned drives look like:

SMART_DEVICES=/dev/sda,/dev/nvme0n1

Docker Compose

cp .env.example .env

Set at least AUTH_PASS and SECRET_KEY, tweak SMART_DEVICES if you want. Then:

docker compose up -d --build

Open http://<host>:<PORT> (default port 8181). History goes under ./data.

Running without Docker

You need Linux, Python 3.11+, lm-sensors and smartmontools. NVIDIA metrics need nvidia-smi on the PATH.

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
cp .env.example .env
python server.py

Then browse to http://127.0.0.1:8181.

Environment

See .env.example for everything. Most people touch AUTH_PASS, SECRET_KEY, disk list, and the ENABLE_* toggles (auto, true, or false).

Image

Prebuilt containers live on GHCR (ghcr.io/antond645-cpu/hw-monitor). Each git tag publishes matching semver tags; latest follows the newest release.

Unraid Community Applications

Included templates:

• unraid/hw-monitor.xml
• ca_profile.xml

Own Apps feed: Apps → Settings → add a custom repository with this XML:

https://raw.githubusercontent.com/antond645-cpu/hw_monitor/main/unraid/hw-monitor.xml

Refresh, search “HW Monitor”, install.

Global CA listing: submit the repo at ca.unraid.net/submit so it appears for everyone after review.

Caveats

Apple and Windows hosts won’t expose the same sysfs/sensor surface—this targets Linux.

In Docker, SMART needs /dev (and usually SYS_ADMIN + SYS_RAWIO caps) mounted so smartctl can talk to drives. AMD and Intel GPU numbers depend heavily on kernel and Mesa/i915 versions.