Skip to main content

💻 Software Issues

Factors affecting data reliability and accuracy that originate in the software pipeline and system architecture. See the 📘 Monitoring Tools Reference for tool details and the 🧩 Diagnostic Funnel for the general triage flow.

New deployment affects the real-time data pipeline​

How to verify:

  1. Check 📶 IoT Telemetry to confirm data is arriving from affected devices on schedule — if it is, but the dashboard/reports don't reflect it, the issue is downstream of the device, in the pipeline itself.
  2. Get the deployment timestamp from the engineering/release team and compare it against when the symptom started.
  3. Check whether the issue is isolated to one device (likely not a deployment issue) or spans many unrelated devices/zones at once (consistent with a pipeline-wide deployment issue).
  4. Check 🚨 Data Integrity Alerts across affected devices for a spike starting right after the deployment window.
Status: gap

No dedicated pre/post-deployment data validation check yet — currently relies on manual correlation via IoT Telemetry and Data Integrity Alerts after the fact.

Unexpected system downtime due to failure of any architecture component (e.g. Storm, batch processing, etc.)​

How to verify:

  1. Check 📶 IoT Telemetry and 🌐 Offline Dashboard across multiple, unrelated devices/zones at the same time — a simultaneous, widespread gap across otherwise-healthy devices indicates an infrastructure/pipeline component failure rather than device-level faults.
  2. Check with the engineering team for component-level health/restart logs (e.g. Storm topology restarts, batch job failures) for the affected time window.
  3. Check 🚨 Data Integrity Alerts for a system-wide spike in device_status alerts starting at the same time, which would corroborate a pipeline-side outage rather than a coincidence of individual device failures.
  4. Check 📝 User Logs (industry level) for any bulk configuration or infrastructure change applied around the same time.
Status: gap

No dedicated system component uptime dashboard yet — currently inferred from correlated device-level symptoms across zones.