dbt
Sifflet brings your dbt model runs and test results next to the data they produce. Failed or skipped model runs show up on the affected tables and in the lineage graph, and your dbt tests become Sifflet monitors, with their run history, alerting, and incidents. Lineage built from your dbt project shows which downstream tables and dashboards a failed model or test affects. See dbt Collected Data for everything Sifflet imports.
Choose Your Deployment
Sifflet integrates with both dbt Core and dbt Cloud. Both integrations import the same metadata, but they get the dbt artifacts in different ways:
| Deployment | How Sifflet gets the artifacts | Identity Sifflet uses | Network access | Guide |
|---|---|---|---|---|
| dbt Core | Your pipeline uploads the artifacts to Sifflet after each dbt command | A Sifflet access token, used by your pipeline | Outbound from your pipeline to the Sifflet API; Sifflet needs no access to your network | dbt Core |
| dbt Cloud | Sifflet fetches the artifacts of your latest job runs from the dbt Cloud API, on a schedule | A read-only dbt Cloud service token | From Sifflet to the dbt Cloud API over the internet | dbt Cloud |
If you use dbt Cloud but can't build your jobs the way the dbt Cloud integration requires, you can use the dbt Core integration instead: upload the artifacts of your dbt Cloud runs to Sifflet from your own pipeline.
Related Pages
- Pipeline Alerting: get notified when a dbt model fails.
- dbt impact analysis in your pull requests: GitHub Action and GitLab CI/CD component.
Updated 11 days ago