Service status¶
Live health of every KATE General service, checked in your browser against the platform's public status API. A background job on the server samples every service every 5 minutes, so the uptime percentages below are real history, not just a snapshot.
Auto-refreshes every 30s · uptime windows: last 24h / 7d ·
What's monitored¶
| Service | What "operational" means |
|---|---|
| API | The Cloudflare Worker's API layer is serving requests. |
| Database (D1) | A test query against the edge database succeeds. |
| Media storage (R2) | The audio/upload bucket is reachable. |
| Cache (KV) | The read-through cache is reachable. |
| AI | The Workers AI binding is available. |
| OCR service | The kate-ocr service responds. |
| Web application | The student/teacher/manager web app loads. |
| Docs site | This documentation site is up. |
The mobile app is not listed separately — it consumes the same API and Media services shown above, so its backend health is reflected there.
How it works¶
- The Worker exposes a public, CORS-enabled
GET /api/statusthat checks each dependency live, andGET /api/status/historythat returns recorded uptime percentages. - A scheduled cron runs every 5 minutes, samples every service, and appends the result to a rolling ~7-day history in KV — so uptime is measured continuously, even when no one is watching this page.
- This page reads both endpoints from your browser and refreshes every 30 seconds. The web-app and docs-site rows are checked directly from your browser.
Want alerting or a public status page with incidents?
The self-hosted open-source Uptime Kuma can point at https://kategeneral.long-cao.dev/api/status (or /health) for notifications (email, Slack, Telegram) and a shareable incident history. The built-in monitor above needs no extra infrastructure.