CCS2025
Enhanced Web Application Security Through Proactive Dead Drop Resolver Remediation
Jonathan Fuller, Mingxuan Yao, Saumya Agarwal, Srimanta Barua, Taleb Hirani, Amit Kumar Sikder, Brendan Saltaformaggio
摘要
Dead Drop Resolver (DDR) malware evades traditional Command and Control (C&C) server takedowns by dynamically resolving C&C addresses hosted on popular web applications, such as Dropbox and Pastebin. These addresses are often manipulated (i.e., encoded or encrypted), rendering existing detection techniques largely ineffective. To tackle this challenge, we introduce VADER, a malware forensics system specifically designed for the proactive detection of dead drops. Analyzing a dataset of 100k malware samples collected in the wild, VADER identified 8,906 DDR malware samples from 110 families that leverage 273 dead drops across seven web applications. Additionally, it proactively uncovered 57.1% more dead drops spanning 11 web applications. Case studies revealed that over 40% of DDR malware samples employ sophisticated, layered de-manipulation algorithms, highlighting the prevalence and complexity of this evasion technique. Beyond detection, VADER enabled proactive remediation by discovering 13 previously unknown dead drops from a single DDR malware sample. This approach empowers web application providers to systematically scan their platforms, enabling the early detection and mitigation of dead drops. CCS Concepts • Security and privacy → Malware and its mitigation.
