OK, I’m not really going to break it. I’m just going to beat it up a little bit.
This Thursday evening, I’ll be working my home turf at the North Texas SQL Server User Group monthly meeting, presenting “When ETL Goes Bad: Handling Errors and Anomalies in SSIS” at 6:15 at the Microsoft campus. In this hour-long presentation, I’ll go through some common error and data anomaly scenarios and demonstrate some ways to error-proof your packages.
Here’s the full synopsis of my presentation:
Even in the most reliable ETL system, it’s eventually going to happen: A package will fail, an output row count won’t match what is expected, a data anomaly will force a forensic search of the package infrastructure… In this session, we’ll talk about some of the things that can go wrong during package execution – and some of the defenses that can be built into SSIS to help head off those errors (or at least make the errors easier to troubleshoot). From error handlers to package restartability, and from logging methods to row count validation and remediation, we’ll cover various ways to keep your ETL flowing when problems arise.
After my presentation, my friend Dave Stein kicks off a new series he’s going on data warehousing. Hope to see you there!