Statistical
Error Growth Forecast
Updated Jul 28, 2026 · 1d ago
Spots an error signature whose arrival rate is climbing and projects when it reaches a volume that matters — as a time range, before the incident, not after.
Example alert
Exact wording varies — the detector generates titles from the anomaly it finds. This is representative of what an alert looks like when it fires.
How it works
Tracks how often each known error signature arrives, hour by hour, and fits a robust trend line that a single spike cannot fool. When the trend is statistically significant and the signature is already at a material volume, it projects the crossing and reports it as an earliest–likely–latest range rather than a single time. Resolution is one hour, so the forecast is honest about hours and never claims minutes; the hour still in progress is excluded. Low-confidence projections say so on the alert. Absolute floors keep a signature going from 1/min to 2/min silent despite the doubling. Learning period: 12 hours per signature.
Availability
Runs on these tiers:
Want to see this detector firing in the live demo?
Open alerts in the sandbox →Related detectors
- Volume Anomaly
Detects spikes, drops, and flatlines in log volume vs daily and weekly baselines per service.
- Silence Detection
Catches services that stop logging when they normally log every N seconds. The most dangerous failure mode: no errors, just absence.
- Outlier Detection
Multi-dimensional outliers in log feature space. Catches subtle anomalies that single-axis thresholds miss.
- Numeric Field Anomaly
Discovers the numeric fields your logs carry — payment amounts, cache hit rates, queue depths, inference scores — and flags when one drifts from its baseline.
- Post-Change Regression
Checks whether a deploy, config change or scale event actually made things worse, by comparing the window after the change against the matched window before it — per service.