Posted by cedarctic
Sep 4, 2025/14:08 UTC
The exploration of a stealthier variant of BGP hijacking reveals significant vulnerabilities within the Bitcoin network, demonstrating the feasibility of eclipse attacks through interception. By controlling a multi-peer Autonomous System (AS), attackers can manipulate BGP routes to intercept Bitcoin nodes' traffic, effectively becoming man-in-the-middle. This type of attack, distinguished by the use of BGP communities to limit announcement spread and evade detection, allows for the forwarding of traffic to legitimate destinations while intercepting or manipulating it. The attacker's ability to control all incoming and outgoing connections to a node enables potential abuses like wasting mining power, selfish mining, and double spending.
A proof-of-concept implementation highlighted the practicality of such attacks on the Bitcoin network. By targeting specific IP prefixes associated with Bitcoin nodes and using BGP communities to restrict the visibility of malicious announcements, an attacker could successfully hijack traffic intended for a large portion of reachable Bitcoin nodes. This was substantiated by analysis showing that over half of all reachable nodes are located within just 1000 IP prefixes, indicating a high concentration of nodes within a relatively small number of prefixes. Additionally, the rate at which prefixes were hijacked did not raise alarms due to the sheer volume of BGP announcements occurring daily, blending the malicious activity in with benign misconfigurations.
The research further delved into prefix vulnerability, finding a significant percentage of Bitcoin node prefixes susceptible to hijacking due to either permissive Route Origin Authorizations (ROAs) or lack thereof. In scenarios where ROAs were absent or overly permissive, the attacker's announcements often had a competitive edge over legitimate ones based on BGP's best path selection algorithm criteria like AS path length.
Mitigations proposed include leveraging networking data observable by nodes for defense, such as periodically rotating peers to disrupt the continuity of hijacked connections, using dynamic port negotiation to complicate passive fingerprinting efforts, and favoring connections to peers in RPKI-protected prefixes or those utilizing maximum-length prefixes. Additionally, incorporating traceroute capabilities into Bitcoin Core could enable route-aware peer selection, significantly hindering the effectiveness of such attacks by prioritizing path diversity and minimizing the attack surface.
This inquiry underscores the critical need for enhanced security measures within the Bitcoin network to protect against sophisticated interception attacks. By understanding the mechanics behind these attacks and implementing robust defenses, the resilience of Bitcoin's peer-to-peer network against external threats can be significantly improved.
TLDR
We’ll email you summaries of the latest discussions from high signal bitcoin sources, like bitcoin-dev, lightning-dev, and Delving Bitcoin.
We'd love to hear your feedback on this project.
Give Feedback