Year: 2025
Type: dissertation
Abstract: HTTP (Hypertext Transfer Protocol) headers play a pivotal role in web application security. Recent studies suggest that HTTP Header Whitelisting(HWL) could be a robust defensive mechanism against various web vulnerabilities, including zero-day threats. This empirical study is designed to investigate the effectiveness of Header whitelist in mitigating some of the most prevalent HTTP header vulnerabilities in web applications. Specifically, it will identify which vulnerabilities can be effectively prevented by HWL, and which cannot, and explore the underlying reasons for these limitations. By conducting tests on HWL against ten commonly exploited HTTP header vulnerabilities, this research aims to provide a detailed analysis of its protective capabilities and limitations, thereby contributing to enhanced security practices in web applications. Through the study, we found that HWL is quite effective in mitigating some vulnerabilities that caused improper HTTP header parsing or the header values are limited, such as HTTP Request Smuggling (HRS) and Cache-Poisoned Denial-of-Service (CPDoS). However, it can’t mitigate some other vulnerabilities, such as JWT (JSON Web Token) authentication bypass attacks, as it needs dedicated function to parse the header. In addition, HWL requires additional configuration and maintenance, which makes the already complex system even more complicated.
Author: Fangxing, Xiong
Institutions: Tampere University, University of Tampere
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7132967154
