Skip to main content
Custom monitors help you alert on a pipeline or a specific table before a data problem becomes a downstream incident. Open Monitors in the Artie Dashboard, or open a pipeline and use its Monitors section to add a monitor for that pipeline.

Create a monitor

1

Choose a pipeline and monitor type

Select the pipeline to monitor, then choose the condition you want Artie to evaluate.
2

Optionally limit the monitor to a table

Low data volume, high data volume, and high ingestion lag monitors can apply to the whole pipeline or one table. Replication-slot monitors apply to a PostgreSQL pipeline.
3

Preview the condition and configure alerts

Use the chart preview to check the selected condition against recent data. Set the detection method, threshold or anomaly settings, and recipients for alerts.
4

Save and review the monitor

Artie starts evaluating the saved monitor. Its detail page shows alert history and chart data for a selected time window.

Monitor types

For low and high data volume monitors, a threshold uses a configurable row-count rollup window. A replication-slot threshold can alert immediately or after the configured size has remained exceeded for a selected duration.

Detection and notifications

For low data volume, high data volume, and high ingestion lag, choose either:
  • Threshold to alert when the configured value is crossed.
  • Anomaly detection to alert when the metric departs from its expected range.
Choose whether alerts go to everyone, no one, or specific users. You can also include or exclude your company’s additional alert email recipients.
When Artie creates a pipeline, it creates a low data volume monitor if the pipeline does not already have one. The default alerts when the pipeline has no data for 12 hours. You can edit its threshold or delete it if that alert does not fit the pipeline.

Monitor history

Open a monitor to review its alert history alongside chart data. Use this view to validate whether a threshold or anomaly setting reflects normal pipeline behavior before changing the monitor.
The Analytics Portal and custom monitors are related but separate. Analytics lets you explore pipeline metrics with filters. A custom monitor evaluates the condition configured on that monitor and sends alerts to its configured recipients.