In the context of designing a data pipeline on Azure Databricks that utilizes constraints to ensure data integrity, consider a scenario where a batch of data fails to meet the defined constraints. The pipeline is configured to process financial transactions for a banking application, where data accuracy and integrity are paramount due to regulatory compliance requirements. Given the need to maintain high data quality and adhere to strict compliance standards, what is the most appropriate default behavior of the pipeline when a constraint violation is encountered? Choose the best option.