Uptime monitoring

Checks that tell you what broke, probed from six continents, and an alert rule you can state in one sentence: more than one region has to agree.

The measurement

A monitor runs one of 12 check types, from HTTP with assertions through DNS hygiene and domain expiry down to ping, traceroute, and the host agent. Every check runs dual-stack across all A and AAAA records with per-IP results, reports dns_ms separately from total latency, and attributes failures to a layer: your app, your DNS, the network, our resolver, or our own node. The probe network page covers the fleet itself.

• Intervals down to 10 seconds, floor depending on plan (300/60/30/15/10 s).
• Redirects followed up to 5 hops, every hop SSRF-validated.
• An http monitor carries TLS, security-header, and DNS-hygiene subchecks inline.

The alert rule

An incident opens when a quorum of regions confirms the failure, two by default, and within each region 75% of nodes must agree first. Below quorum nothing is swallowed: forming incidents are visible, and single-region drift lands on a watchlist instead of your phone. The full mechanics are in the docs.

The record

The incident’s start is the quorum confirmation, to the second, and the same boundary feeds your status page and your availability record. Maintenance windows suppress alerts and stay on the record. One measurement, every downstream story.

Start monitoring free, or read the monitor docs.