scrutiny: Monitor Drive Health and SMART Trends

Overview

Scrutiny is a self-hosted dashboard for monitoring hard-drive health through S.M.A.R.T. data collected by smartctl. It adds historical tracking and a focused view of critical metrics to help address gaps in command-line monitoring, such as difficulty spotting gradual changes over time.

It is aimed at people managing servers or other systems with multiple drives who want a central view of drive status and configurable alerts. Scrutiny combines manufacturer-provided metrics with real-world failure thresholds; it does not replace backups or guarantee that a drive failure can be predicted.

Key Features

Web dashboard focused on critical S.M.A.R.T. metrics.
Uses smartd and smartctl for drive discovery and monitoring.
Tracks S.M.A.R.T. metrics and temperature over time.
Applies customized thresholds informed by real-world failure rates.
Sends configurable failure and collector-error notifications through webhooks and supported services.
Supports an all-in-one Docker image or separate collector, web/API, and InfluxDB components.

Use Cases

A home-server operator wants a dashboard of drive health and temperature across several disks.
A NAS or server administrator wants historical metrics to notice gradual changes that a one-time status check may miss.
An operator wants notifications when a drive reports a failure or the collector cannot read its S.M.A.R.T. data.
A multi-server administrator wants to run collectors on individual hosts and send results to a central web interface.

Getting Started

The quickest route is the Docker omnibus image. Pass through the disks you want monitored and the required device metadata access:

docker run -p 8080:8080 -p 8086:8086 --restart unless-stopped \
  -v `pwd`/scrutiny:/opt/scrutiny/config \
  -v `pwd`/influxdb2:/opt/scrutiny/influxdb \
  -v /run/udev:/run/udev:ro \
  --cap-add SYS_RAWIO \
  --device=/dev/sda \
  --name scrutiny \
  ghcr.io/analogj/scrutiny:latest-omnibus

Open http://localhost:8080 to view the dashboard. Pin a specific image version for predictable updates, and consult the README for device passthrough, configuration, and deployment details.

Considerations

Scrutiny's README describes the project as a work in progress, so some rough edges may remain.
The collector needs access to the monitored devices. Docker deployments require device passthrough, /run/udev, and SYS_RAWIO; NVMe drives may also require SYS_ADMIN.
RAID and virtual-drive visibility depends on what smartctl --scan detects and whether the controller exposes underlying S.M.A.R.T. data. Device type overrides may be needed.
S.M.A.R.T. data and thresholds are not a guarantee of advance warning. Keep independent backups and assess drive health using your broader operational practices.