cockpit-smart — SMART Diskstatus for Cockpit

Cockpit plugin that displays S.M.A.R.T. health and attributes for all connected disks, with history tracking and trend detection.

This is a personal hobby project I build for my own use and publish in case it's useful to someone else. I work on it in my spare time, so issues and PRs are welcome but replies may be slow. Use at your own risk.

Features

• Health status (PASSED/FAILED) per disk
• Automatic disk type identification (HDD/SSD/NVMe)
• Key attributes highlighted (reallocated sectors, pending sectors, etc.)
• NVMe-specific attributes (available spare, percentage used, media errors)
• Warning markers for critical values
• Expandable details for all SMART attributes
• Manual refresh button

History and Trend Detection

• Daily logging: smart-history.sh collects SMART data and appends it as JSONL to /var/log/smart-history.json
• Change detection: Automatically flags attributes that changed since the previous measurement
  • SATA: raw value changes, value drops, threshold proximity (< 2x threshold)
  • NVMe: numeric changes in all attributes
  • Generalized logic — works for any disk type without hardcoded attribute lists
• Sparklines: Inline SVG mini-graphs showing trends over the last 30 data points
• Severity levels: Danger (red), warning (yellow), info (blue) — sorted by severity
• Log rotation: 365 days retention with compression

Installation

sudo bash install.sh

This will:

• Install smartmontools if not present
• Copy plugin files to /usr/share/cockpit/cockpit-smart/
• Set up a daily cron job (08:00) for history collection
• Configure logrotate (365 days retention)

Requirements: smartmontools, python3, cockpit

Access

Open Cockpit in your browser and click SMART Status in the menu.

URL: https://<your-server>:9090/cockpit-smart/

File Structure

cockpit-smart/
├── manifest.json              # Cockpit plugin manifest
├── index.html                 # UI (HTML + CSS + JS)
├── scripts/
│   ├── smart-collect.sh       # Data collection (smartctl → JSON)
│   └── smart-history.sh       # History logging (JSONL with timestamp)
├── install.sh                 # Installation script (plugin + cron + logrotate)
└── README.md

How It Works

1. index.html uses cockpit.spawn() to run smart-collect.sh with superuser privileges
2. smart-collect.sh runs smartctl against all disks and collects info/health/attributes
3. Output is parsed as JSON and rendered as cards per disk in the web UI
4. History is loaded from /var/log/smart-history.json and compared to current data
5. Changes are displayed in a "Forandringar" (Changes) section per disk card with sparklines

History Format (JSONL)

Each line in /var/log/smart-history.json is a complete JSON object:

{"timestamp":"2025-03-15T08:00:00Z","disks":[...]}

The disks array has the same structure as smart-collect.sh output.

Testing Without Cockpit

Open index.html directly in a browser — it falls back to mock data with simulated history for testing the trend detection and sparkline features.

Cron Schedule

The history collection runs daily at 08:00 via /etc/cron.d/smart-history. Adjust the schedule by editing that file if your server has different wake/sleep times.

License

MIT

Part of a Cockpit plugin suite

I build a small family of Cockpit plugins for home servers, all dependency-light and made to be readable at a glance:

• cockpit-temps — hardware temperature history with thresholds
• cockpit-smart — S.M.A.R.T. disk health with trend tracking (this plugin)
• cockpit-pcloud — pCloud storage quota and backup folder status
• cockpit-tailscale — plain-language Tailscale network overview

Browse them all via the cockpit-plugin topic.