From dave at dtrt.org Sun Nov 25 18:47:17 2018 From: dave at dtrt.org (David A. Harding) Date: Sun, 25 Nov 2018 13:47:17 -0500 Subject: [Lightning-dev] [PATCH] First draft of option_simplfied_commitment In-Reply-To: <87bm6jp42a.fsf@rustcorp.com.au> References: <87bm6jp42a.fsf@rustcorp.com.au> Message-ID: <20181125184717.mrj5xz2mdpkjusim@email> On Wed, Nov 21, 2018 at 12:47:17PM +1030, Rusty Russell wrote: > I'm also starting to implement this, to see what I missed! > > - Feerate is fixed at 253 [satoshis per 1,000 weight] IIUC, this is just over Bitcoin Core's default minimum relay fee of 0.00001000 BTC/vByte. That works right now, as mempools are nowhere near full, but if they fill up again and the BIP133 feefilters are increased by any amount, nodes will no longer relay transactions with minimum feerates. In that case, how does the commitment transaction get relayed in order for its `to_*_pushme` outputs to be spent for CPFP fee bumping?[1] There's currently some text in the PR about using sighash_single|sighash_anyonecanpay for use with RBF, but I don't think that can apply to spending the commitment transaction, as that would allow not just adding new inputs and outputs, but also removing all but the 'singled' output (allowing theft of its value). -Dave [1] I don't think Bitcoin Core currently relays transaction packages consisting of parents below the relay fee limit and children sufficiently above the limit to pay for their parents. This has certainly been discussed, so maybe I'm wrong and it is available or maybe it'll be available in the future. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: