Otto
your autonomous SRE · read-only · asks before it acts
live · in the cluster
Break down the last seven days of incidents and tell me where the noise comes from.
✓live agent · 42 incidents, topology and drift, 7 day hot window
✓your observability · 90 day baseline for the comparison
42 incidents in seven days. Two workloads produce 62% of them and they collapse into two fault classes.
Incidents by workload7d
checkout15
payments11
orders7
frontend5
postgres4
Topology drift7d vs baseline
ingressfrontendcheckoutpaymentsordersprometheusincident area · 26 of 42Δ +2 duplicate scrape targets
Three of the four monitoring incidents trace to scrape edges that drifted in, not to faults.
ingestion policyretention policycardinality policy
just now · Otto · 7-day window + long-term