Investigating SWATCH-3478, we found that we do have tally_snapshots records that are duplicated.

What is a duplicated tally_snapshot? A record that have the same values for the following columns:

• org_id,
• product_id
• granularity
• snapshot_date
• sla
• usage
• billing_provider
• billing_account_id

Query to find duplicate records:

SELECT org_id, product_id, granularity, snapshot_date, sla, usage, billing_provider, billing_account_id, COUNT(*) AS duplicate_count 
FROM tally_snapshots ts 
GROUP BY org_id, product_id, granularity, snapshot_date, sla, usage, billing_provider, billing_account_id 
HAVING COUNT(*) > 1;

Document: https://docs.google.com/document/d/1PTxe7YJzWRqvDusOCgqnFdOtJJDTuLKQFS5GNgQ8ZuE/edit?tab=t.0

Acceptance Criteria

• Identify all the cases when a tally_snapshot is duplicated
• Investigate if we can add an unique constraint in tally_snapshots (to be done in SWATCH-3478). If not, change the index tally_snapshot_unique_constraint to not be unique.