Research On Detection Of Buffer Overflow Vulnerability In The Linux

With the continuous development and extensive application of the Internet and computer technolo gy, people increasingly realize the importance of information and these dependence on computer systems highlight, accompanied by its own inherent vulnerability has become even more evident, the security of information and systems also need to face the unprecedented challenge. The reason for the problems of information and system security threats are varied, including buffer overflow vulnerability because of the widespread and easily use, has become a very common security threats and great danger with its widespread and easily use and causing high attention of academic and practical application areas. If the buffer overflow vulnerability is detected before exploiting, its information and systems security threats can be greatly reduced, so a study in this paper is that a buffer overflow vulnerability detection techniques for to propose and build a more convenient effective method for discovering vulnerabilities.Firstly, the buffer overflow vulnerability research background is introduced, followed by some sort of buffer overflow basic theory, including the principle of the buffer overflow vulnerability, formation reason, the different sub-categories with the principles and so on. On this basis, a detailed study of the existing buffer overflow detection has been finished. Existing detection methods include three categories: static buffer overflow vulnerability detection methods, dynamic buffer overflow vulnerability detection methods and combination of static and dynamic buffer overflow vulnerability detection methods. Static detection methods mainly rely on static detection tool to scan and analysis source code to get the final test results, and its advantage is to find vulnerabilities in advance before the threat generated and avoid losses; dynamic testing is the method used by the program is running and its advantage is higher correct detection rates, but can only be used to reduce losses after the attacks occurred. Therefore, this research focuses on innovative static detection methods.Given that the high prevalence of static detection of false positives and false negatives, and different static detection method can focus on detecting the different vulnerabilities, and therefore this paper constructs a buffer overflow vulnerability comprehensive testing approach. Papers regard reducing false positives and false negatives as the goal, integrated use multiple buffer overflow vulnerabilities of static detection tools, to make the test results of these tools can verify and complement each other with each other to get more accurate results. According to the comprehensive testing methods, the paper designed and implemented a comprehensive buffer overflow vulnerability inspection system. In order to obtain more accurate test results, the system will first detect a plurality of different types of tools run simultaneously and after the detection result of a comprehensive analysis of different tools. In this method, to implement the more important part of weight reliability calculation, credibility weights training module tool has been added to find the optimal weights. System test results show that compared with the single static detection tool, this method can effectively reduce the rate of false positives and false negative in buffer overflow vulnerability detection and improve the accuracy of test results.