Incident response process

A split section pairing a communication promise and three bullet commitments with a live incident feed of phase chips, timestamps, and update text.

A heading and lede over a bordered list of maintenance windows, each row carrying date, UTC time, affected components, impact, and a status chip.

Per region health cards with local latency and twelve month incident counts.

A centered heading and lede above a four column stat band, each cell pairing a large metric with its label and a method note.

Median and p99 latency as shadcn lines with the visible spike explained.

SLA targets beside trailing actuals and automatic remedies, published inline.

A one row live status banner with a pulsing dot linking the full status page.

A status list signup with a never marketing constitution and an email form.

The last 90 days as tick bars with the two degraded days named below, not in tooltips.

An aggregate status banner above per component rows with dots and uptime figures.

Dated incidents with severity chips, precise impact lines, and postmortem links.

Large quotes with avatars that alternate left and right alignment in a divided stack.