STATUS/LIVE ●
All systems operational
Availability here is graded by an external monitor probing from outside our network every sixty seconds. We cannot quietly edit it, which is the point.
PROBES: HTTP, SMTP, DNS, ICMP · 60s INTERVAL · ALERTS PAGE A HUMAN
Something looks down from where you are? Tell us, we page on it.
STATUS/METHOD
How the numbers are made
An external monitoring service — infrastructure we rent but do not control — probes the cluster and key services every sixty seconds from multiple continents. Failures must confirm from more than one location before they count, which filters out the internet’s ordinary hiccups, and every confirmed incident is timestamped into the public history automatically. We chose an external judge specifically because a status page a host can edit is a diary, not a record.
STATUS/SLA-MATH
The SLA promises 99.9% monthly availability — at most ~43 minutes of unplanned downtime in a month. Miss it and credits apply per the SLA page without you having to argue for them. Planned maintenance is announced in advance, scheduled in low-traffic windows, and listed separately, because mixing maintenance into uptime math is how other hosts reach 100%.