Towards assuring non-recurrence of faults leading to transaction outages: an experiment with stable business applications

Authors:
Anushri Agrawal;Ravindra Naik
Affiliations:
Tata Research Development and Design Centre, Hadapsar, Pune;Tata Research Development and Design Centre, Hadapsar, Pune
Venue:
Proceedings of the 4th India Software Engineering Conference
Year:
2011

Citing 5
Cited 0

Structured Programming with go to Statements

ACM Computing Surveys (CSUR)
Letters to the editor: go to statement considered harmful

Communications of the ACM
Software Reliability Engineered Testing

Software Reliability Engineered Testing
Cobol minefield detection

Software—Practice & Experience
Evaluating static analysis defect warnings on production software

PASTE '07 Proceedings of the 7th ACM SIGPLAN-SIGSOFT workshop on Program analysis for software tools and engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

Reducing the cost and efforts for maintaining a legacy business application is a major challenge faced by the industry. Capturing faults at the early stage of software life cycle is known to prevent certain kinds of defects in production -- we exploit this aspect to detect a class of software faults introduced while making changes to stable business applications. We discuss the results of an experiment with stable, back-office COBOL applications of a core banking system. With changes, the stable system exhibits frequent transaction outages (among other defects). We observed that most critical outages were due to few common causes, which were extremely difficult to identify by testing. We present our analysis and the results of the successful attempt to automatically detect the causes, using structural analysis and control flow analysis techniques.