System status
Live platform health, current incidents, historical uptime, and where postmortems live. This page is a signpost; the live board is at status.nebbos.ai .
Live board
The live status board is at status.nebbos.ai . It surfaces the current health of every user-visible surface:
- REST API
- MCP endpoint
- WebAuthn ceremony
- Cradle attestation
- Export pipeline
- Audit-chain verify
- Each region’s data plane
Colour codes: green (nominal), yellow (degraded — investigating), red (incident — declared).
Incident lifecycle
| Phase | What happens |
|---|---|
| Declared | Substrate telemetry or a synthetic probe fires an alert; incident declared and posted to status board within 5 min. Names affected surface, observed impact, estimated blast radius. |
| Updated | Every 30 min for Sev-1, every hour for Sev-2, until resolution. Each update names what has changed since the last update — no filler. |
| Mitigated | Impact stops. Incident stays on the board with the mitigation named; underlying cause may still be under investigation. |
| Resolved | Root cause identified, fix landed. Incident closes with a link to the postmortem placeholder. |
| Postmortem | Every Sev-1 and Sev-2 incident publishes a postmortem within 7 days of resolution. Blameless, five-whys shape, corrective actions with owners. |
Historical uptime
- 90 days of rolling uptime per surface, updated hourly, on the live board
- 12 months of incident summaries listed under the board with per-incident postmortems linked
Every measurement is computed against the synthetic-probe methodology named in SLA.
Subscribe to status
RSS, JSON, and email digests all available from the live board. For high-severity events, subscribers get pushed within two minutes of declaration.