Title: RUNTIME MONITORING TECHNIQUE TO HANDLE TAUTOLOGY BASED SQL INJECTION ATTACKS

Issue Number: Vol. 1, No. 3
Year of Publication: Nov - 2012
Page Numbers: 189-203
Authors: Ramya Dharam, Sajjan G. Shiva
Journal Name: International Journal of Cyber-Security and Digital Forensics (IJCSDF)
- Hong Kong

Abstract:


Software systems, like web applications, are often used to provide reliable online services such as banking, shopping, social networking, etc., to users. The increasing use of such systems has led to a high need for assuring confidentiality, integrity, and availability of user data. SQL Injection Attacks (SQLIAs) is one of the major security threats to web applications. It allows attackers to get unauthorized access to the back-end database consisting of confidential user information. In this paper we present and evaluate a Runtime Monitoring Technique to detect and prevent tautology based SQLIAs in web applications. Our technique monitors the behavior of the application during its post- deployment to identify all the tautology based SQLIAs. A framework called Runtime Monitoring Framework, that implements our technique, is used in the development of runtime monitors. The framework uses two pre-deployment testing techniques, such as basis-path and data-flow to identify a minimal set of all legal/valid execution paths of the application. Runtime monitors are then developed and integrated to perform runtime monitoring of the application, during its post-deployment for the identified valid/legal execution paths. For evaluation we targeted a subject application with a large number of both legitimate inputs and illegitimate tautology based inputs, and measured the performance of the proposed technique. The results of our study show that runtime monitor developed for the application was successfully able to detect all the tautology based attacks without generating any false positives.