Gapwise Status

Service health

Gapwise Status

Current production status for Gapwise applications, APIs, AI services, documentation, and operator-maintained systems.

Initializing monitoring
Current platform health

Loading current status…

Fetching the latest production snapshot.

Waiting for the latest checkAutomatic checks run every 15 minutes
Services tracked—Across production surfaces
Automatic monitors—Every 15 minutes
Healthy now—Last observed snapshot
Recorded incidents—Since history began

Production systems

Live service health

Current health, monitoring method, last confirmation, and endpoint-level diagnostics for every tracked Gapwise surface.

Live checks
Loading service health…
Freshness
—

The latest production snapshot is loading.

Automation coverage
—

Public production surfaces are automatically probed. Private workflows remain operator-confirmed until a safe synthetic check exists.

Probe cadence
15 min

Each automatic service receives three attempts per cycle, reducing false alarms from one-off network failures.

Observability

How to read this page

This page distinguishes measured evidence from operator confirmation instead of presenting invented uptime percentages.

◎
Automatic probe

Public endpoint checked repeatedly from GitHub Actions. HTTP outcome is recorded in the live snapshot.

◇
Operator-maintained

Used where a meaningful check requires authenticated state or a side effect that should not be synthesized continuously.

↗
Third-party boundaries

University of Toronto and other upstream systems can affect user experience while remaining outside Gapwise operational control.

Incident record

Recorded service history

Loading the recorded transition history…

Open incident history
Monitoring methodology and limitations

Automatic checks run every 15 minutes against public production surfaces. Each service is attempted three times before the cycle is classified.

Freshness protection prevents an old green result from being shown as current. A snapshot older than 60 minutes is explicitly marked delayed while still showing its last-known service states.

History records observed status transitions. It is not a continuous SLA ledger and does not backfill uptime percentages for periods without continuous measurement.

If a user-facing issue persists after a retry, contact [email protected]. Security reports belong at [email protected].