Posted by danielabrozzoni
May 7, 2026/14:54 UTC
In the discussion around timestamp fuzzing techniques for network addresses, there is an agreement on the approach of aging timestamps to make them older rather than fresher. This strategy aims to mitigate the risk of addresses appearing too active, relying instead on self-announcements to refresh the timestamps when necessary. The concern here is that refreshing timestamps might allow nodes that have already left the network to remain listed, thus cluttering the address management system (addrman). More details on this issue can be found in a detailed commentary on a Bitcoin pull request, which outlines potential complications with refreshing node timestamps (33498 (comment)).
Another proposed solution involves adapting the timestamp based on whether the requester is on the same network as the responder. Specifically, if both parties are on the same network, the real timestamp would be used, while if they are on different networks, the timestamp would be aged by a random number between 1 and 10 days. This hybrid approach combines elements from two different strategies, ensuring that the timestamp only gets aged when interacting with requests from different networks, thereby reducing the risks associated with timestamp refreshing. This method not only addresses concerns about potential fingerprinting issues but also provides a nuanced way to handle node visibility across varied network conditions.
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