Reconciliation Jobs
A reconciliation job (or “check”) compares data between two connections. It consists of two phases: extraction and comparison.
Extraction
Section titled “Extraction”DataRecs runs your configured query against each connection and pulls the result sets into a staging area. Extractors run as isolated Kubernetes Jobs, ensuring resource limits and tenant isolation.
Comparison
Section titled “Comparison”Once both extractions complete, a Comparator analyses the two result sets. It identifies:
- Matching rows — rows present in both sources with identical values.
- Missing rows — rows present in one source but not the other.
- Mismatched rows — rows present in both but with differing values in one or more columns.
Job groups
Section titled “Job groups”You can organise related jobs into job groups and run them together on a schedule or on-demand. Job groups track run history and aggregate status.
Scheduling
Section titled “Scheduling”Jobs can be triggered:
- Manually from the Console or CLI.
- On a schedule via the built-in scheduler (backed by Apache Airflow).
- Programmatically via the REST API.