Alert Rules & IncidentsFrom alert rule to
From alert rule to
resolved incident.
Define the conditions that matter. Every time one fires, Trasys opens an incident, tracks MTTA and MTTR, and keeps the whole timeline — alerts, acks, escalations — in one place.
~/incidents
MTTA2m 14s
MTTR38m 02s
Open2
Postgres RLS transaction timeout
Investigating12m
ClickHouse p99 latency > 800ms
Monitoring1h 4m
Groq provider error rate spike
Resolved3h 20m
Why it matters
Alerting that turns into a workflow, not just a ping
Custom alert rules
Set thresholds on any metric, log pattern, or trace condition — cost, latency, error rate, uptime — and let Trasys watch it continuously.
Incident timeline & KPIs
Every incident tracks MTTA and MTTR automatically, rolled up org-wide so you can see whether response times are actually improving.
Escalation built in
Unacknowledged alerts climb the on-call chain automatically — see Escalation Policies for how shifts and rotations are configured.