[Lightning-dev] Penalty tx and RBF

[ZmnSCPxj]

Good morning Cezary and Rene,

The commitment transaction being RBF-able is not important, as old commitment transactions are in fact how theft attempts are made.

What is important to be RBF-able is the punishment transaction which spends the commitment transaction.
Fortunately:

1.  The punishment branch of the commitment SCRIPT has no CSV, so it can be spent immediately to get the old commitment confirmed via CPFP.
2.  The punishment branch requires only a single signature from one node, making RBF trivial.

I believe at least lnd already does RBF for breach claim transactions, I will have to review c-lightning code to see if onchaind does so.
I vaguely remember it being brought up some time ago for c-lightning.

Regards,
ZmnSCPxj
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxfoundation.org/pipermail/lightning-dev/attachments/20181123/e394e0a8/attachment.html>