CPFP Carve-Out for Fee-Prediction Issues in Contracting Applications (eg Lightning)

In an email conversation, David A. Harding pointed out that a parent transaction near the limit of 100,000 vbytes could have almost 10,000 outputs paying OP_TRUE. He further mentioned that if the children were limited to 10,000 vbytes each, this would allow relaying 100 mega-vbytes or nearly 400 MB data size, which is larger than the default maximum mempool size in Bitcoin Core. Johan noted that this opens up the potential for free relay of large amounts of data. They also discussed the possibility of special casing to only allow this for "LN commitment-like" transactions, but this would require application-specific changes. Meanwhile, they are moving forward with adding a 1 block delay to all outputs to utilize the current carve-out rule.