FUZZVPN: Finding Vulnerabilities in OpenVPN
summary
FUZZVPN is a man‑in‑the‑middle fuzzer that learns OpenVPN’s control‑channel MSC (passive + active learning) and runs 1,000 scenarios (60 s each, 5–6 hours) against OpenVPN 2.6.4 in UDP and TCP modes. It finds two DoS replay/flood attacks, 17 configuration options with input‑validation bugs (e.g., port 60000 parsed to 4464 via getaddrinfo truncation), an early‑data case where server sends data the client never accepts, and a client‑auth downgrade that degrades a victim’s performance.
AI-written summary · about
chapters (10)
- 0:00FUZZVPN and OpenVPN intro
- 4:47OpenVPN security study
- 7:38FUZZVPN research
- 8:40Message sequence learning
- 12:15FUZZVPN design and testing
- 14:24Setup and vulnerability detection
- 15:08Config input validation bugs
- 16:33OpenVPN DoS attacks
- 17:41Early data transmission flaws
- 19:08Conclusion and future work
related talks
45mAutomatic Discovery of Evasion Vulnerabilities Using Targeted Protocol Fuzzing
39mFuzzing Session: Finding Bugs and Vulnerabilities Automatically
49mAnalysis of DTLS Implementations ... State Fuzzing
39mDiving into Windows Remote Access Service for Pre-Auth Bugs
37mFinding Vulnerabilities in IoT Embedded Devices Using Linux OS and Open Source Tools
27mWIFI-Important Remote Attack Surface: Threat is Expanding
01h 39mFuzzing in the open: Integrate your project in OSS-Fuzz for continuous fuzzing
28mFuzz Testing of Envoy