arrow
返回

Cefuzz: An Directed Fuzzing Framework for PHP RCE Vulnerability

delete2022-03-01
delete0
delete
OA
AI
DOI:10.3390/electronics11050758delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Current static detection technology for web application vulnerabilities relies highly on specific vulnerability patterns, while dynamic analysis technology has the problem of low vulnerability coverage. In order to improve the ability to detect unknown web application vulnerabilities, this paper proposes a PHP Remote Command/Code Execution (RCE) vulnerability directed fuzzing method. Our method is a combination of static and dynamic methods. First, we obtained the potential RCE vulnerability information of the web application through fine-grained static taint analysis. Then we performed instrumentation for the source code of the web application based on the potential RCE vulnerability information to provide feedback information for fuzzing. Finally, a loop feedback web application vulnerability automatic verification mechanism was established in which the vulnerability verification component provides feedback information, and the seed mutation component improves the vulnerability test seed based on the feedback information. On the basis of this method, the prototype system Cefuzz (Command/Code Execution Fuzzer) is implemented. Thorough experiments show that, compared with the existing web application vulnerability detection methods, Cefuzz significantly improves the verification effect of RCE vulnerabilities, discovering 13 unknown vulnerabilities in 10 popular web CMSes.

期刊

暂无期刊信息

机构

暂无机构信息
引用论文

引用论文

暂无论文信息