Author: Bastien TEINTURIER 2020-06-22 07:35:20
Published on: 2020-06-22T07:35:20+00:00
In a discussion regarding the recent attacks on the Lightning Network (LN), Bastien Teinturier suggested using independent pay-to-preimage transactions as a technical solution. However, David Harding pointed out that this approach may lead to incentive issues and centralization. Miners who hide the preimage transaction in their mempool would have to be accomplices with the attacker, which could lead to a bidding war for the honest users, ultimately leading to the loss of money. Additionally, information about a node's mempool is partly trusted, and this could create a reputation-based centralizing force that pushes individual miners towards well-established pools. Harding suggested using independent pay-to-preimage transactions, as anyone who knows the preimage can mine the transaction, thereby eliminating reputational advantage or direct economies of scale. Moreover, if the defense is effective, the attack should rarely happen, and it should not have a significant effect on mining profitability. However, ZmnSCPxj noted that pay-to-preimage doesn't work with PTLCs, and Harding called for making pay-to-revealed-adaptor-signature possible using something like OP_CHECKSIGFROMSTACK. Regarding whether off-chain nodes should be heavily invested in on-chain operations, Harding said he didn't see how eltoo would help, as this problem involved an abuse of the final state. He added that from the perspective of a single LN node, it might make more sense to get the information and not share it, increasing security and allowing for lower routing fees compared to competitors. However, ignoring concerns about mining centralization, from the perspective of just the Lightning Network, investing in running nodes in mining pools to learn about attackers' txs and potentially share discovered preimages with the network off-chain doesn't sound unreasonable to him.
Updated on: 2023-06-03T00:49:31.387798+00:00