The Relationship Between Software Bug Type and Number of Factors Involved in Failures

Published

October 23, 2016

Author(s)

Raghu N. Kacker, David R. Kuhn, Zachary Ratliff

Abstract

Previous studies have defined different types of software bugs based on their complexity and reproducibility. Simple bugs, which involve only direct factors and are often easy to reproduce, have been called “Bohrbugs”, while complex bugs, with at least one indirect factor and generally much more difficult to reproduce, are called Mandelbugs. Locating Mandelbugs in software is often difficult and expensive during the development phase due to their complexity. In this paper we consider the relationship between different types of software bugs and the average t-way interaction involved in triggering them. Analysis suggests that Mandelbugs have a higher interaction strength when triggering their faults. This suggests that both direct and indirect factors play a role in triggering the fault. By using combinatorial testing methods, we can have high assurance of locating Mandelbugs as long as the indirect factors are present at the time of testing.

Proceedings Title

Proceedings of
Eighth IEEE International Conference on Software Testing, Verification and Validation
ICST 2016

Conference Dates

April 10-15, 2016

Conference Location

Chicago, IL

Conference Title

IEEE International Conference on Software Testing, Verification and Validation

Pub Type

Conferences

Download Paper

DOI Link

Keywords

Bohrbug, combinatorial testing, Mandelbug, software failure, t-way testing, test automation

Citation

Kacker, R. , Kuhn, D. and Ratliff, Z. (2016), The Relationship Between Software Bug Type and Number of Factors Involved in Failures, Proceedings of Eighth IEEE International Conference on Software Testing, Verification and Validation ICST 2016, Chicago, IL, [online], https://doi.org/10.1109/ISSREW.2016.26 (Accessed December 7, 2024)