Skip to main content

Data Availability Monitoring

Audience: Operations, Management, IT Team

1. Overview​

This guide explains how data availability is measured, reported, and managed within the customer site deployment. Its purpose is to provide transparency and confidence, while ensuring issues are identified and resolved quickly without requiring constant manual oversight.

2. Data availability principle​

Each device is expected to deliver data continuously, evaluated in one-minute intervals.

  • 1 day = 1,440 expected minutes
  • Availability is measured end-to-end, up to final storage and reporting
  • This ensures reported data reflects real, usable information

3. Data Availability States Summary​

Availability StateWhat It MeansUser Impact
OK – On TimeData arrived within the expected 5-minutes time window and was processed successfully.Live and historical dashboards are complete, accurate, and final. No user action required.
OK – DelayData arrived successfully but later than expected due to temporary latency or buffering.Live dashboards may show delayed updates. Historical dashboards will update retroactively and become complete once processing finishes.
Delay – Device OfflineDevice temporarily offline due to network or connectivity interruption; data may arrive once reconnected.Temporary gaps or delays in live dashboards. Buffered data is typically recovered after reconnection, so historical dashboards usually become complete.
Missing – Pipeline Not IngestedData was generated but not processed by the ingestion pipeline.Live dashboards show gaps during the incident. Historical dashboards may be partially or fully backfilled later (footfall recoverable)
Missing – No PowerDevice not powered; no data generated.Live and historical dashboards show permanent data gaps for the affected period. Data cannot be recovered.
Missing – Counting Module Not RunningDevice powered on but counting module not running, so no usable data produced.Live and historical dashboards show permanent data gaps for the affected period. Data cannot be recovered.
Missing – UnknownData missing but root cause not yet classified; further investigation required.Live and historical dashboards may show gaps. Impact and recoverability are undetermined until Root Cause Analysis completes and the issue is reclassified.

4. Reports available to Customer​

4.1 Per-device availability (per day)​

Used to:

  • Validate installation stability
  • Identify isolated device issues
  • Support technical discussions when needed

4.2 System-wide availability (30 days)​

Used to:

  • Assess overall system health
  • Support management and SLA discussions
  • Track long-term reliability trends

5. Patching and corrections (what Customer sees)​

When corrections are applied:

  • The affected time window is logged

  • The action is recorded automatically

  • Monthly summaries reflect:

    • What was affected
    • What was done
    • Final data completeness

Customer does not need to manage patching workflows.

6. What Customer does NOT need to monitor​

The following are handled internally by FootfallCam:

  • Ingestion retries and reprocessing
  • Pipeline-level recovery
  • Internal escalation workflows
  • Root-cause diagnostics
  • Patch execution mechanisms

Only outcomes and transparency are surfaced.

7. Data Availability States, Responsibility & SLA​

Availability StateResponsible PartyAction RequiredSLA
OK – On TimeN/ANo actionN/A
OK – DelayFootfallCamRe-ingest data, validate completeness≤ 12 hours
Delay – Device OfflineCustomer ITCheck device power, network, LED status≤ 1 business day
Missing – Pipeline Not IngestedFootfallCamRe-ingest data, validate completeness≤ 12 hours
Missing – No PowerCustomer ITRestore power, verify device status≤ 1 business day
Missing – Counting Module Not RunningFootfallCamRemote diagnosis on the affected device(s)≤ 12 hours
Missing – UnknownFootfallCamInvestigate root cause≤ 1 business day