React Status Page

Copy Page

A vendor without a status page is asking customers to learn about outages from their own monitoring. This is the per-service uptime history and the incident thread that go with it.

Installation

Props

Overall status is the worst service

The banner takes the minimum health across every service — it never rounds up to green. A page that says "All systems operational" while one component is degraded is worse than having no status page, because it converts a transient outage into a credibility problem. The ranking is explicit: outage beats degraded beats maintenance beats operational.

maintenance is a distinct state rather than a flavour of outage, so planned work does not look like a failure — and does not quietly hide one either.

Incident threads, not incident headlines

Each incident carries an ordered list of updates with investigating, identified, monitoring, and resolved statuses. The thread is the useful artefact: a customer wants to see that you found it, understood it, and confirmed the fix — with times attached. A single "resolved" line reveals nothing about whether you knew what happened.

Uptime bars derive their colour from the daily percentage when no explicit state is given (≥99.9% operational, ≥95% degraded, below that an outage).

The printed average is clamped so it never rounds up to a perfect 100.00%. A 90-day window containing a single 99.85% day averages to roughly 99.996%, which naive formatting renders as 100.00% — sitting directly beside a strip with a visible amber bar, so the page contradicts itself. Anything short of a genuinely flawless window displays as at most 99.99%. It is the same rule as the overall banner: never round up to green.

Feed real measurements; a status page with invented history is the single fastest way to destroy the trust it exists to build.