Fault

Concepts

• A fault is an incorrect data definition, step, or process in a software. <bib>vincenzi-etal:2007</bib>, incorrect instruction, incorrect statement

• A fault is a static defect in the software. <bib>ammann-offutt:2008</bib>, <bib>delamaro:slides:2009</bib>

• A fault is the manifestation of one or more errors. <bib>mathur:2008</bib>

Facts

• Faults are the result of the software complexity, new application domains, absence of methods, unpredictable situations and human mistake. <bib>delamaro:slides:2009</bib>

• Faults detected later have a higher correction cost. <bib>vincenzi-etal:2007</bib>

• A fault is commonly called error, bug or defect. <bib>vincenzi-etal:2007</bib>, <bib>mathur:2008</bib>

• A fault that causes an error and the executed path corresponds to the expected path is classified as a computational fault. <bib>vincenzi-etal:2007</bib>

• A fault that causes an error and the executed path does not correspond to the expected path is classified as a domain fault. <bib>vincenzi-etal:2007</bib>

• Data sensitive faults can be detected only by test cases with specific input values. <bib>vincenzi-etal:2007</bib>

• Data insensitive faults are detected by any test case, regardless of the actual input value. <bib>vincenzi-etal:2007</bib>

• A fault can mask the error of another fault. <bib>vincenzi-etal:2007</bib>

• The execution of a fault produces an error. <bib>vincenzi-etal:slides:2007</bib>

• A fault and its associated errors may cause one or more failures.

• For a given fault, not all inputs will trigger the fault into creating incorrect output (a failure). <bib>ammann-offutt:2008</bib>