Stale
YAML type: stale | Category: Data quality
Flags series whose last observed point is older than maxAgeDays.
Description
Stale monitors freshness of each series relative to the expected latest date (respecting ignoreLastNDays in scheduled processing). Optional maxReportAgeDays stops reporting very old ended series.
How it works
Stale measures how old the latest observed point is compared with the expected latest date (scheduled processing respects ignoreLastNDays). If that age exceeds maxAgeDays, the series is reported as stale.
Optional maxReportAgeDays stops alerting once a series has been stale for too long -- useful for retired products or regions that will never update again.
Setup
Set maxAgeDays to your freshness SLA. Use maxReportAgeDays for products or regions that retire so they stop alerting forever.
Configure detectors on the data view Analysis tab or in YAML. Validate with Data Preview before enabling production schedules.
Settings
| Parameter | Default | Description |
|---|---|---|
maxAgeDays |
3 |
Maximum allowed age of the latest point |
maxReportAgeDays |
null | Stop reporting after this age (null = indefinitely) |
Shared series filters
Most detectors also accept optional series filters in params:
| Parameter | Purpose |
|---|---|
maxGapAmount |
Skip series when gap amount is too high (null = no limit). Catalog create-default is often 5. |
minValue |
Skip series whose average value is below this (null = no limit). Catalog create-default is often 1000. |
maxValue |
Skip series whose average value is above this (optional; omit/null = no limit) |
useSpecialDates |
When true, suppress anomalies that overlap special dates |
When a filter skips a series, Data Preview shows the reason.
Configuration example
analysis:
detectors:
- type: stale
params:
maxAgeDays: 3
maxReportAgeDays: 90
Using the detector
Stale anomalies indicate missing recent data. Check Monitoring load status and the source connector first.
Use cases
- Warehouse load SLA monitoring
- Detecting discontinued series that stopped updating
- Ops dashboards for feed health
Specifics
Not a value anomaly -- combine with Gap for mid-history holes.