Data Monitoring / Pipeline Freshness & Data Quality
Data pipelines: Row-count reconciliation
Source and target volumes disagree after a load. Compare counts using the same filters, batch boundary and deletion policy.
Environment and inputs: Data pipelines and analytical datasets; Apache Airflow is an orchestration example. Quality acceptance rules must match the dataset contract.
Failure scenario
Source and target volumes disagree after a load. Confirm the cause before choosing a fix.
Evidence to collect
Compare counts using the same filters, batch boundary and deletion policy. Record the time, affected scope and a healthy comparison interval.
Fix the confirmed cause
Investigate missing or duplicate keys before declaring a volume mismatch to be data loss. Write the expected result first. Change one relevant setting at a time.
Check the result and hand over
Repeat the check for row-count reconciliation using the same scope and a representative interval. Check that the symptom is resolved, expected records are present and downstream results are correct. If coverage, correctness or latency worsens, restore the recorded configuration through the agreed rollback process.
Scope and precautions
Prefer read-only access and limit queries to the relevant period or dataset. This example has not been tested on your systems. Test changes in an approved environment. Keep credentials and sensitive data out of shared evidence.