How Telescope Works
A look under the hood at how Telescope monitors your sites.
Telescope continuously checks your monitors from multiple regions and alerts you when something goes wrong. Here's how it all fits together.
The check cycle
- You create a monitor and set a check frequency (e.g., every 30 seconds).
- Telescope's workers pick up your monitor when it's due and run the appropriate check.
- The result — status, response time, any errors — is recorded as an event.
- If failures cross your alert threshold, you get notified through your configured channels.
- When the monitor recovers, the alert is automatically resolved.
Reliability
- No single point of failure — multiple worker instances run in parallel across regions. If one goes down, the others continue checking.
- No double-checking — Telescope guarantees each monitor is checked by exactly one worker per cycle, even with many instances running.
- Graceful degradation — if a check result can't be written immediately, it's buffered and retried. No data is silently dropped.
What's next
- HTTP Monitoring — how HTTP checks work, including WAF detection.
- Alerts — how alerts fire, deduplicate, and resolve.
- Database & Data Retention — how check data is stored and aged out.