horizOn Status
Updated at –
Availability*
Last 90 days
Response times*
Incident history
Last 90 days
* A service counts as unavailable only while all of its regions are unreachable at the same time. Response times are measured inside each region.
How we measure
Availability: a separate monitoring node checks the backend and the frontend of every region every two minutes. A service counts as unavailable only while all of its regions are unreachable at the same time, because requests move to a working region as long as one is up.
Response times: each region calls its own backend and frontend once a minute from inside the cluster. They show how fast the server answers, not the network path from a player to the region.
If the monitoring node itself is offline, its results for that time are ignored, so those minutes do not count as downtime. The history covers the last 90 days.