HTTP(s)

Requests a URL and judges the response code.

Use it when

You want to know a website or endpoint is reachable and not erroring.

Don't use it when you care that the content is right — a broken site that returns 200 passes this check. Use Keyword for that.

What to enter

Exactly when it's DOWN

• With an expected status set: any other status code.
• Without one, following redirects: anything outside 200–399 (we follow up to 10 redirects; more than that fails).
• Without one, not following redirects: anything outside 200–299. A 3xx is reported as a redirect that was not followed.
• A connection, DNS or TLS failure.
• No answer within the timeout.

What it doesn't do

• No custom request headers, and no request body. The dashboard has no fields for them. That makes POST and PUT of limited use here — you can send the method, but not what normally goes with it.
• No authentication. Anything behind a login will see the login page, which usually returns 200 and passes.
• No JavaScript. We fetch the HTML the server sent, not the page a browser would build from it. A site that renders entirely client-side returns an almost-empty document, which is still a 200.
• No content check at all. That is Keyword.

We send User-Agent: UptimeCraft-Agent/0.1, which is also how you can exclude our checks from your analytics.

Example

A marketing site: URL https://example.com, method GET, expected status blank, follow redirects on, interval 1 minute, threshold 2.

The same thing over the API: POST /api/v1/monitors with {"name": "Marketing site", "monitorType": "HTTP", "url": "https://example.com"}. Full reference in the API documentation.

Alerts

Down, recovered, and latency if you set a threshold.

Plan notes

Interval floor and available regions come from your plan; see settings.