Published on: 2014-11-06T23:50:53+00:00
A new section called "Confidence to include tx1" has been added to the GitHub repository for dgenr8's out-there project. The section discusses a potential attack on miners, in which a miner excludes any transaction that was first seen less than 30 seconds ago or double spent less than 30 seconds after it was first seen. The post mentions that if a miner were to perform this action, they should expect five of 10,000 nodes to delay their block. Hal Finney and Tom Harding commented on the idea, with Finney stating that it would need careful analysis and Harding suggesting that it may be possible to quantify and target the risk of including tx1.Gregory Maxwell has proposed an attack on Bitcoin in which a malicious miner can flood other miners with orthogonal double spends that he doesn't mine himself. This attack is based on transmitting three transactions to two miners at the same time, creating a situation where Miner A knows about tx1a before Miner B, and Miner B delays his block until he knows about tx1a. The attacker tries to create a conflict that only affects everyone else except Miner A. However, Miner A can neutralize his risk by excluding any tx1 known to be double-spent. The success of the attack depends on the timing between the malicious transactions. If they are sent 10 seconds apart, the attack has little chance of success. It will be possible to quantify and target the risk of including tx1a based on the time gap to tx2.In an online discussion on Bitcoin forums, user Tom Harding expressed caution about miners including a double-spend in their block. He mentioned several factors that could mitigate the risk, such as the time offset from the first spend received and most other nodes having seen the transaction. Another user named Matt raised concerns about the proposal, and two other users replied with additional insights.In an email conversation, Tom Harding acknowledged that without double-spend relay, a miner would not be able to detect conflicting transactions. He also stated that even with double-spend relay, the miner cannot be sure and may choose to include no transactions at all to be safe. This is because a malicious miner can flood the network with orthogonal double spends.In a conversation between Matt and Tom Harding, it was discussed that without double-spend relay, miners will not know about conflicting transactions. Tom plans to add first-double-spends relay according to a specific proposal. However, if a miner includes a double-spend in their block, they have to be very careful. They hope that only a small fraction of the network will delay their block based on the time offset from the first spend they received. Additionally, most other nodes should see their transaction. The best course of action for a miner is to exclude fast transactions and connect to everyone on the network to look for double-spends.The delay in relaying/accepting blocks based on local information is considered a bad idea as it can lead to the splitting of the network. Miners are incentivized to connect with everyone on the network and look for double-spends to avoid being delayed. Tom Harding proposed a solution to improve the ability of bitcoin users to rely on unconfirmed transactions without the need for hard or soft forks. The proposal has not been implemented yet, and feedback from Bitcoin Dev is appreciated.
Updated on: 2023-08-01T10:34:15.122397+00:00