Author: Rusty Russell 2015-08-12 01:06:57
Published on: 2015-08-12T01:06:57+00:00
The email exchange between Joseph Poon and Anthony Towns discusses how double-spend attacks are known to occur with Bitcoin, and with the Lightning Network, confirmation times matter less as it is off-chain. However, when setting up a channel with an untrusted counterparty, one will wait for N confirmations of their anchor transactions. If N is too small, and one can afford to do a double-spend despite M (M > N) confirmations as long as it gains them $X, they can open anonymous channels capable of receiving at least $X and start the double-spend fork. They then construct multiple lightning channels simultaneously, fund them at $d each, and wait for N confirmations so the new channels are active. They quickly route multiple payments from their new channels to their anonymous channels until they can't send anymore, publish the double-spending fork, so that their $d*n never got spent and close their original anonymous channels gaining $X. The only people worse off are the ones who opened the $d channels after N confirmations, and any intermediary hubs are fine. With onion routing, none of the ripped-off hubs need know where the money ended up. There are constraints, including how many channels one can open in M-N blocks, having >$X funds available in the first place to commit to the double spend, and how much capacity the lightning network actually has in routable bitcoin.
Updated on: 2023-05-23T19:21:50.207916+00:00