System Status

What we monitor

A machine-readable probe is available at /api/health. It runs the same two checks as this page — a database round trip and a synthetic resolution of a canary short link through the live lookup path — and returns HTTP 503 when a check fails. A check with nothing behind it reports not_configured instead of passing.

These are point-in-time checks from the single instance that rendered this page. They are not a multi-region measurement. Redirect observability is structured logs only: every resolved scan writes one PII-free line with its outcome, cache hit and timings.

We do not publish an uptime percentage yet. We will only do so once it is backed by aggregated redirect metrics, not marketing.

What each state means

Operational means the check actually ran and passed. It is a real probe rather than a hand-set badge, but it is not necessarily a measurement of this second: results are reused for up to a minute, and the timestamp below says when they were taken.

Degraded means the check ran and failed. It does not necessarily mean the whole service is down: the redirect edge and the dashboard fail independently, and a degraded dashboard does not stop a printed code from resolving.

Not monitored means there is no probe behind that component right now, so we are not claiming anything about it. We would rather show a grey light than a green one nobody measured.

If a code is not resolving

Check the link in your dashboard first. A code stops resolving for reasons that are not outages: the link was paused, it reached an expiry date, it hit a scan limit, or its destination failed a safety check. Each of those shows on the link itself, and each is reversible without reprinting anything.

If this page reports everything operational and your code still fails, tell us with the short code and roughly when you scanned it. For anything security related, use the contact in our security.txt.