Scrutiny 0.10.0-1

Scrutiny watches the health of the NAS's disks: their SMART data, kept over time (temperatures, reallocated and pending sectors, power-on hours...), judged both by each disk's own verdict and by Scrutiny's, from real-world failure rates of each attribute's values - so a disk that's likely to fail shows before it does. It alerts the ntfy app when one fails. The app runs it natively on the NAS (no Docker).

For the Intel models only: Scrutiny keeps its history in InfluxDB 2, which has no build for the ARM models' 32-bit processors.

• It's reached through the Nginx app (which it needs), at https://scrutiny.<domain>: Scrutiny has no login of its own, so it only listens on the NAS itself, and Nginx asks for the app's login first - user name admin and a generated password, which the management page shows, copies or replaces.
• The disks are read hourly (and when it starts, and on request from the page). A disk that's asleep is left asleep - not read at all that time - so the disks still spin down; its last reading stays until it next wakes. Each disk is read as a SATA disk, with the program's own smartctl (7.5 - newer than the NAS's).
• Alerts: set an ntfy topic on the page (e.g. nas, the one your phone subscribes to) and Scrutiny sends it an alert when a disk fails its SMART check or Scrutiny's own - with the ntfy app's token for the apps, so nothing else needs setting up. The page can send a test alert. Scrutiny's own notification settings (in its interface) add to these.
• Scrutiny and InfluxDB run as their own scrutiny user (created by the app, ID 65), on the NAS itself only (ports 8091 and 8093); reading the disks runs as root, as SMART needs. InfluxDB uses about 100-200MB of memory.
• Its data is kept next to the app, e.g. /mnt/HD/HD_a2/Nas_Prog/_scrutiny, and survives reinstalls and removal: the disks' history (db/ - Scrutiny's database - and influxdb/), the page's settings, the password (root only) and the logs (each kept under 1MB).
• Its programs are Scrutiny's official ones (its web interface, its front end and its collector), InfluxDB's official ones and smartctl built from source, all fully static (prepared by static/scrutiny, which checks them on the NAS's kind of processor, without AVX). Scrutiny doesn't update itself: updates come with new versions of the app, and the NAS shows an alert when a new release is out, checked weekly.

Management page

Scrutiny's page in the NAS dashboard (Apps → Scrutiny) shows whether it's running, its version, its address, the disks (each one's model, size and serial number, health, temperature and when it was last read) and when they were last read. It reads them now, restarts Scrutiny, shows the login (with the password), sets the ntfy topic for alerts and sends a test alert, and shows the disk reader's latest log lines. Only NAS admins can use the page (see Management pages).