Author: Carla Kirk-Cohen 2023-03-14 19:09:02
Published on: 2023-03-14T19:09:02+00:00
On the latest jamming mitigations call, participants discussed different aspects of local reputation tracking to counteract spamming. The meeting was structured around a communally sourced agenda and there was no strong preference to introduce a rotating chair for meetings. As for circuit breaker update, the UI is being redesigned for a more user-friendly experience. The main discussion centered on approaches to local reputation tracking, and there were many proposals suggested on the mailing list. There was a suggestion that a simpler approach to reputation should be taken before proposing more ambitious algorithms for tracking whether a peer has "good" reputation. This pragmatic approach would make solutions more easily observable to end-users. Several reputation schemes have been discussed so far, which depend on the addition of an "endorsement" field to update_add_htlcs to help peers communicate whether they believe a HTLC is unlikely to be part of a jamming attack. The possibility of starting without an "endorsement" field was floated; however, many participants believed that it was an important part of informing forwarding decisions. Instead, the possibility of simplifying the local reputation metric that is used to decide whether a peer's behavior is considered "good" was raised as a simplifying alternative. The question of introducing upfront fees if we have an effective local reputation mechanism that can identify bad behavior was posed, and opinions varied. It was noted that the original motivation for local reputation paired with upfront fees was to address attackers who target their attacks to sit just beneath a "good" threshold.The discussion also touched on the shifting landscape for the types of attacks they are trying to mitigate. Instead of proposing a mitigation and then thinking about custom attacks for that exact solution, it was suggested that they collect a set of different attack strategies they wish to defend against, then compare different solutions' effectiveness. They agreed to use an issue on the repo that is used to administer these meetings to track various forms of attacks. Finally, they spent some time discussing the availability of data and possibility of using simulations to compare various approaches. A few approaches to empirically examining the problem were discussed, including "shadow" deployment of a reputation metric, just logging outcomes, to see how it would work on nodes today, data gathering on a collection of nodes in the network to feed into a more realistic simulation, and simulation of various payment flows using the real network graph.
Updated on: 2023-06-01T19:17:34.023496+00:00