Off-Path Network Traffic Manipulation via Revitalized ICMP Redirect Attacks
Xuewei Feng, Qi Li, Kun Sun, Zhiyun Qian, Gang Zhao, Xiaohui Kuang, Chuanpu Fu, Ke Xu
摘要
ICMP redirect is a mechanism that allows an end host to dynamically update its routing decisions for particular destinations. Previous studies show that ICMP redirect may be exploited by attackers to manipulate the routing of victim traffic. However, it is widely believed that ICMP redirect attacks are not a real-world threat since they can only occur under specific network topologies (e.g., LAN). In this paper, we conduct a systematic study on the legitimacy check mechanism of ICMP and uncover a fundamental gap between the check mechanism and stateless protocols, resulting in a wide range of vulnerabilities. In particular, we find that off-path attackers can utilize a suite of stateless protocols (e.g., UDP, ICMP, GRE, IPIP and SIT) to easily craft evasive ICMP error messages, thus revitalizing ICMP redirect attacks to cause serious damage in the real world, particularly, on the wide-area network. First, we show that off-path attackers can conduct a stealthy DoS attack by tricking various public servers on the Internet into mis-redirecting their traffic into black holes with a single forged ICMP redirect message. For example, we reveal that more than 43K popular websites on the Internet are vulnerable to this DoS attack. In addition, we identify 54.47K open DNS resolvers and 186 Tor nodes on the Internet are vulnerable as well. Second, we show that, by leveraging ICMP redirect attacks against NATed networks, off-path attackers in the same NATed network can perform a man-in-the-middle (MITM) attack to intercept the victim traffic. Finally, we develop countermeasures to throttle these attacks.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper10
- Point Cloud Analysis for ML-Based Malicious Traffic Detection: Reducing Majorities of False Positive AlarmsChuanpu Fu, Qi Li, Ke Xu, Jianping WuCCS 2023 · 被引用 30 次
- Invisible Adversaries: A Systematic Study of Session Manipulation Attacks on VPNsYuxiang Yang, Ao Wang, Xuewei Feng, Qi Li 等INFOCOM 2026 · 被引用 1 次
- Detecting Unknown Encrypted Malicious Traffic in Real Time via Flow Interaction Graph AnalysisChuanpu Fu, Qi Li, Ke XuNDSS 2023
- FRAGJAM: DoS Attacks Using IP Reassembly CongestionYepeng Pan, Christian RossowUSENIX Security 2026
- Off-Path TCP Exploits: PMTUD Breaks TCP Connection Isolation in IP Address Sharing ScenariosXuewei Feng, Zhaoxi Li, Qi Li, Ziqiang Wang 等CCS 2025
它引用的顶会 Paper12
- Tranco: A Research-Oriented Top Sites Ranking Hardened Against ManipulationVictor Le Pochat, Tom van Goethem, Samaneh Tajalizadehkhoob, Maciej Korczynski 等NDSS 2019 · 被引用 826 次
- Network Hygiene, Incentives, and Regulation: Deployment of Source Address Validation in the InternetMatthew J. Luckie, Robert Beverly, Ryan Koga, Ken Keys 等CCS 2019 · 被引用 89 次
- Off-Path TCP Exploits: Global Rate Limit Considered DangerousYue Cao, Zhiyun Qian, Zhongjie Wang, Tuan Dao 等USENIX Security 2016 · 被引用 74 次
- The CrossPath Attack: Disrupting the SDN Control Channel via Shared LinksJiahao Cao, Qi Li, Renjie Xie, Kun Sun 等USENIX Security 2019 · 被引用 68 次
- Censored Planet: An Internet-wide, Longitudinal Censorship ObservatoryRam Sundara Raman, Prerana Shenoy, Katharina Kohls, Roya EnsafiCCS 2020 · 被引用 68 次
相关 Paper
- Man-in-the-Middle Attacks without Rogue AP: When WPAs Meet ICMP RedirectsXuewei Feng, Qi Li, Kun Sun, Yuxiang Yang 等S&P 2023
- ReDAN: An Empirical Study on Remote DoS Attacks against NAT NetworksXuewei Feng, Yuxiang Yang, Qi Li, Xingxiang Zhan 等NDSS 2025
- The Hijackers Guide To The Galaxy: Off-Path Taking Over Internet ResourcesTianxiang Dai, Philipp Jeitner, Haya Schulmann, Michael WaidnerUSENIX Security 2021 · 被引用 22 次
- Do (Not) Follow the White Rabbit: Challenging the Myth of Harmless Open RedirectionSoheil Khodayari, Kai Glauber, Giancarlo PellegrinoNDSS 2025
- Alias Equals Zone? Large-Scale and Stealthy Takeover of Domain Hosting Service via CNAME-Following Cross-Domain VerificationRuixuan Li, Xingyu Zhao, Yunyi Zhang, Baojun Liu 等USENIX Security 2026
