Skip to content

What do you need to do?

Docs

RealUptime + Cloudflare

Cloudflare sits in front of most of this setup rather than under it: DNS, the CDN, and often the origin itself (Pages or Workers). The check watches what a visitor actually gets back through Cloudflare's edge, and your status page can separately show Cloudflare's own status so an edge-wide incident does not read as your own outage.

What you are setting up

An HTTP check against your public domain (which resolves through Cloudflare, whether your origin is a VPS behind Cloudflare's proxy, Cloudflare Pages, or Cloudflare Workers), a status page, and, optionally, a vendor component that tracks Cloudflare's own incidents so the two are never confused.

Create the check

Point a check at the public domain, exactly as a visitor reaches it. This checks the whole path: DNS, Cloudflare's edge, and your origin, together, because that is what "is the site up" actually means to a visitor.

bash
export REALUPTIME_API_KEY=ru_live_...

realuptime checks create \
  --name "Site (via Cloudflare)" \
  --url https://www.example.com/ \
  --interval 60 \
  --regions iad,sjc,fra,nrt

If your origin is a server you control behind Cloudflare's orange-cloud proxy, see Monitor a Hetzner VPS or Monitor Docker containers to also watch the origin directly and the Monitor agent's own host metrics; a check against the public domain alone cannot tell an edge problem from an origin problem.

Add Cloudflare's own status as a vendor component

RealUptime's free outage tracker already watches major vendors, Cloudflare included, from four regions. Add it as a vendor component on your status page (Status pages → Components → Add vendor component → Cloudflare) so an incident on Cloudflare's side of the line shows up next to yours, attributed correctly instead of looking like your own outage.

bash
# No API key needed: reads the public /outages/cloudflare JSON endpoint.
realuptime outages cloudflare --region fra --json

See Status pages, "Vendor components", and Probe methodology for how the outage tracker tells a real vendor incident apart from a bot-defense challenge page.

Add a status page

Create a status page and add your own check plus the Cloudflare vendor component. See Status pages for custom domains and subscriber notifications.

What RealUptime does and does not see

The check sees the HTTP response returned through Cloudflare's edge, from each region you chose, exactly like a real visitor. It cannot tell you whether a failure is Cloudflare, your origin, or DNS without more checks (an origin-direct check bypassing the proxy is the usual way to split that apart, and is worth adding once you suspect the edge specifically).

There is no Monitor agent for Cloudflare Workers or Pages: Workers run in an isolate with no /proc or persistent process for an agent to run alongside, the same reason there is none for Vercel's edge runtime (see RealUptime + Vercel). RealUptime Errors is not supported on Workers today for the same reason: the SDK's transport needs a Node runtime.