bitcoin-dev

OP_Expire mempool behavior

OP_Expire mempool behavior

Original Postby Peter Todd

Posted on: March 13, 2024 03:32 UTC

The discussion on the delvingbitcoin forum highlights concerns regarding the potential for a "bandwidth-wasting vector" through the use of OP_Expire.

The concern stems from the possibility that an attacker could generate fee-paying transactions with sufficiently high fees to ensure mining in the next block or so, thus consuming significant bandwidth. This method, while costly due to the requirement of actual fee-paying transactions, is noted to be a way to misuse bandwidth if the attacker already intends to transact.

The argument further elaborates that this issue is not unique to OP_Expire. Similar to Replace-By-Fee (RBF) mechanisms, where any transaction can be replaced for a nominal additional cost (specifically 1 satoshi per virtual byte), OP_Expire operates under the same principle. Transactions employing OP_Expire would incur an additional cost of 1 sat/vB over the replaced transaction, which is considered a payment for the bandwidth consumed by such replacements. Essentially, the critique suggests that while OP_Expire introduces a new dimension to transaction handling and replacement, it does not fundamentally alter the economic incentives or costs associated with transaction replacement and bandwidth usage as established by existing practices like RBF.

For further insights into the discussion and technical perspectives shared, interested readers may explore the contributions of individuals including Peter Todd, whose insights can be found directly via petertodd.org or by reaching out at 'peter'[:-1]@petertodd.org.