Hey Product Hunt 👋 I built SteadyStack after one too many 3am pages for outages that never happened. The cause is almost always the same: your monitor checks from a single location, that location has a bad second a dropped packet, a slow TLS handshake, a route flap somewhere between the prober and your server and it pages you. Your service was fine the whole time. Do that enough and your team learns to ignore alerts, which is genuinely worse than having no monitoring at all. How it works. Every monitor is checked from multiple regions each cycle. When a region sees a failure it immediately re-checks before that failure counts toward anything that kills single dropped packets silently. Then a quorum of regions has to independently confirm before an incident opens: 2-of-3 on the free tier, 4-of-7 on paid. A region that's just slow gets excluded from the count rather than counted as a failure, so one congested route can neither trigger an alert nor suppress one. The part I actually use most is the classification. Some regions failing while a majority succeed isn't an outage it's usually a CDN edge or a geo-routing rule, and it gets labelled regional degradation instead of paging anyone. Different problem, different response. Free tier is 50 monitors with multi-region quorum included, 1-minute checks on your first 10, and commercial use permitted in writing. Launch-cohort terms are grandfathered permanently. I've watched enough free tiers change underneath people that I wanted that in writing rather than implied. Known limits, since someone will ask: the free tier runs 3 regions rather than the full 7, and we don't have probe coverage in Africa or the Middle East at all our locations page says so rather than padding the map. A genuine global outage still pages you; a problem affecting only those users may not surface. And if you're happily self-hosting Uptime Kuma genuinely, keep doing that. It's excellent software. This is for when one box's opinion isn't enough. I'll be here all day. Ask me anything, especially about the consensus logic or the edge cases where it gets weird.

SteadyStack

Uptime monitoring that never pages you for nothing

Uptime monitoring that never pages you for nothing