Monitoring 34 NTP Synchronized Clocks from a Cron Job

Description and use

A Python cron job monitors 34 NTP‑synchronized LED clocks, polling their HTTP status and alerting when offsets exceed ~1 s.

Problem

34 wall‑clock displays across three buildings were not being monitored for time accuracy, risking incorrect time display.

Outcome

All 34 displays are now monitored; alerts fire when a display’s offset exceeds ~1 s or when it becomes unreachable, preventing unnoticed drift.

Workflow

• 1. Verify upstream NTP server sanity with ntplib.
• 2. Define human‑perceivable offset threshold (~1 s).
• 3. Load fleet list from JSON.
• 4. Concurrently poll each device’s /status endpoint.
• 5. Normalize JSON response to internal schema.
• 6. Evaluate offset, last_sync_age, and online status.
• 7. Trigger alerts if offset > threshold or device offline.
• 8. Schedule script via cron for regular monitoring.

Evidence

Each unit exposes an HTTP status endpoint.

Source

I initially monitored each display's offset from the NTP server and alerted at > 500 ms.

Source

The shape I normalize to internally looks like this: {"id":"BLD-A-F2-REG-01",...}

Source