4 ms·
Interesting, thank you. And to avoid most risk, maybe the hackers could just mine the thieving transaction themselves? The hackers could operate a miner which
by FatalLogic 4y ago
Interesting, thank you. And to avoid most risk, maybe the hackers could just mine the thieving transaction themselves?
The hackers could operate a miner which privately adds that transaction every time it tries to solve a block, but broadcasts it only when they successfully got the block. It could be running for days before it succeeds, because there's probably no urgency. They can take all the fees, and it's difficult to frontrun because nobody knows the theft is happening until the block is already solved and it's almost too late.
- pcthrowaway 4y agoI mean it's not like they're risking a massive amount of ETH on the first try, but if they saw it get stolen the first time, sure, they might try sniping the token by running/bribing a miner if the payout was great enough (there are networks of miners that let you include transactions without getting front-run by paying them a fee)
- syzygyhack 4y agoYou are correct. Actually in practice it is far less complicated to achieve this, at least for Ethereum. You can simply use the Flashbots RPC to bundle your transaction and have it sent straight to a miner. No exposure to the public mempool = no frontrunning possible, and no need to run your own miner/validator.