A Constraint-Based System Fault Injection Method

Abstract

The invention discloses a constraint-based system fault injecting method. The method comprises the steps of: classifying constraints into atomic condition and conjunction normal form clause levels according to atomic condition inversion and the conjunction normal form clause inversion strategies and solving a group of negative forms of the specification constraints at the levels to serve as basesof generating fault data; replacing constraint variables in the inverted constraints with normal input combinations, gradually carrying out slack replacement on the constraint variables and finding minimum-variation fault data which have the least number of slack variables and are closest to a real situation through a constraint solving operation; carrying out faulty injection on a to-be-tested system by taking the minimum-variation fault data as test inputs to obtain fault injecting results. The method disclosed by the invention has the beneficial effects that by adopting a new fault injecting technology, the more real and more sufficient fault data are generated; the error detecting capability of the fault injection is strengthened; and faults close to the normal state are also beneficial for a tester to repair and eliminate the defects at the same time.

Description

technical field [0001] The invention belongs to the technical field of computer system development, and in particular relates to a constraint-based system fault injection method. Background technique [0002] Fault injection is a system reliability (Dependability) verification technology. It detects the reliability of the system by deliberately introducing faults into the system to be verified in the experiment, and detecting and observing the system behavior after the fault is introduced. The existing fault injection methods generally include injecting abnormal data and injecting abnormal codes. The former, such as E.Jeong et al., generate abnormal data according to the fault model, and inject the abnormal data into the system through message interception, etc.; the latter, such as R.Natella et al., inject abnormal codes in the source layer or binary layer to indirectly generate faults State, to achieve the purpose of fault injection. [0003] In terms of fault data injec...

AI Technical Summary

Problems solved by technology

This method can generate fault data that is logically unreasonable relative to the constraints, but in the black-box test scenario, it is difficult to know which parts of the constraints correspond to the decision predicates on the program path, so the constraints in SAGE are reversed method is not suitable for black box testing scenarios

On the other hand, the SAGE method obtains a new combination of input data through constraint solving for testing. The generated fault data cannot be guaranteed to be close to the real situation, and the defect detection ability of fault injection cannot be guaranteed.

Images