Component health, in real time. If something's degraded, you see it before your customers do.
Updated every 30 seconds from production probes — not a marketing page.
overall
All systems normal.
last incident · 17 days ago · resolved < 9 min
99.98%
uptime · 90 days
components
API · public
REST + auth · p95 latency 142ms
operational
60 days
99.99% 30d uptime
Send workers
SMTP fleet · 87% capacity headroom
operational
60 days
100% 30d uptime
IMAP idle pool
Reply ingest · 0 reconnects pending
operational
60 days
99.97% 30d uptime
iBrain (Claude)
AI classification · queue depth 0
operational
60 days
99.94% 30d uptime
Dashboard
Web app · cdn cache hit 94%
operational
60 days
100% 30d uptime
Webhooks delivery
Outbound · < 800ms p95
operational
60 days
100% 30d uptime
past incidents · 90 days
IMAP idle pool — elevated reconnect rate (resolved)
Anthropic regional capacity dip caused p95 latency to climb from 1.2s to 4.8s on classification requests. No errors, no missed classifications — just slower. Cleared automatically as upstream capacity returned.
Read replica failover during planned maintenance took longer than expected. Analytics endpoints returned 502 for ~7 minutes. Send, reply, and dashboard endpoints unaffected.