Every service we monitor
179 services, each probed every 20 minutes from our own network. A service only appears here once someone has written a real description of it and we are genuinely checking it - There are no auto-generated placeholder pages on this site.
By category
Why services are grouped by category
Outages travel along dependencies. If several services in one category break together, look at the layer beneath them.
How a service gets into this directory
Two conditions, and neither is negotiable. We have to be able to monitor it meaningfully with an unauthenticated request to a public endpoint, and somebody has to have written a real description of it - What it is, how it fails, and what its symptoms look like from outside.
Why the directory stays deliberately short
The second condition is what keeps this list short. It would take an afternoon to generate a page for every domain anyone has ever searched "is X down" for, and sites in this category routinely have tens of thousands of exactly that: A hostname, a ping time, and nothing a person wrote. Ours is 179 services, and the limit on growing it is editorial capacity rather than engineering.
The trade is deliberate. A short directory of pages worth reading beats a long one of pages that answer nothing - And anything not listed is still one click away in the checker, which answers the question without leaving a thin page behind.
What the status beside each name means
Each name links through to the full 24-hour report chart, the recorded outage history and the response times behind it. How the checks work is the longer version.
What this directory is not
What the order does and does not mean
It is not a ranking. The order here is alphabetical, and on the home page it is by category; neither implies one service is more reliable than another. For that comparison, the uptime figure on each service's history page is the honest place to look - And the methodology page explains why even that is a floor rather than a verdict.
It is also not a list of services having problems. Most of the time nearly everything here is answering normally, which is what you would expect and is rather the point of keeping a record: The value is in what it shows over months, not in what it shows this afternoon.