This page is our own dogfood — built on the same status-pages product our customers use. Each component above is backed by a DevFlow SLO; the “current” column is the trailing 90-day attainment.
| Component | State | SLO target | Current (90d) |
|---|---|---|---|
| Probe network (writes) | operational | 99.95% | 99.97% |
| Public API | operational | 99.9% | 99.94% |
| Dashboard | operational | 99.5% | 99.71% |
| Webhook delivery | operational | 99.5% | 99.62% |
A network-bench rotation upstream of our eu-west-1 webhook fan-out caused 12 minutes of 5xx delivery to ~6% of customers with EU webhook subscribers. Auto-resolved at 11:42 UTC. Postmortem on /blog.
A scheduled DB-index rebuild in the dashboard read replica overlapped with EU peak traffic. Dashboard latency p95 spiked to 4.5s for 8 minutes. No data loss; check execution unaffected.
An upstream provider outage in ap-northeast-1 forced a region failover. Multi-region monitors with quorum across other regions saw no failed checks. Single-region ap-northeast-1 monitors showed null results during the 47 minutes.