How Uptime Checks Actually Work

When you run a check, we try to open the website from our servers and report what response we get. Here is what happens behind the scenes.

What happens during a check

  1. DNS lookup: We convert the domain to an IP address.
  2. Connection: We open a connection to the server (HTTP/HTTPS).
  3. Secure handshake: For HTTPS sites, we complete TLS setup.
  4. Request: We send an HTTP request.
  5. Response: We record the status code and timing.

How to read status codes

  • 2xx: Success (example: 200 OK)
  • 3xx: Redirect (example: 301, 302)
  • 4xx: Request/client issue (example: 403, 404, 429)
  • 5xx: Server issue (example: 500, 502, 503, 504)

What response time means

Response time is the delay between sending the request and receiving the first byte back. It includes network delay and server processing, but not the full page download.

Timeouts we use

  • Connection timeout: 10 seconds
  • Response timeout: 30 seconds

If these limits are exceeded, the check is marked as failed.

Why results can differ

  • We usually test one endpoint (often homepage).
  • Routes differ by network and region.
  • CDNs and load balancers can show mixed behavior.
  • Some sites block automated traffic.

Country-based checks

For regional testing, requests are routed through servers in selected countries. This helps detect geo-blocking, local outages, or region-specific filtering.