As a data engineer managing a crucial daily data processing pipeline that includes data ingestion, transformation, and loading into a data warehouse, you receive an alert about a job failure during the transformation task due to an unexpected cluster outage. What is the best course of action to minimize downtime and ensure data integrity?