Matt Morehouse

Fundamentally, how much do we save by doing batched sweeps? Other implementations don't do this at all. I imagine we could make RBFs much simpler and less error-prone without the...

> I think the general idea here is we don't need to care about mempool rules anymore if `TestMempoolAccept` can be used. We just retry creating txns until `TestMempoolAccept` allows...

> Yeah that's the next step, to map all the errors. `TestMempoolAccept` gives a detailed `RejectReason` which can be used to decide the next step. Besides, how could rule 5...

> Jut to be sure, I'm referring rule numbers [here](https://github.com/bitcoin/bitcoin/blob/master/doc/policy/mempool-replacements.md) and I think you mean rule 2 right? Ah, sorry, it's not an RBF rule at all! What I was...

> Detecting bad inputs was sort of a side effect of the old back off process. With the new approach, we'll use `testmempoolaccept`, so we'll never even broadcast this bad...

A rebase would also be nice -- it seems there's some already merged changes in the diff.

Replying to https://github.com/lightningnetwork/lnd/pull/8434#issuecomment-1971644848 on the original PR. > @morehouse I think this PR needs to be viewed in context with the `protofsm` PR and the rbf-close PR above. For new...

> What do you mean by saying next-block deadline? That we only have 1 block left till it times out? What is this min feerate? Here's an example graph of...

> Thank you for catching this! I forgot to mention I wanna more opinions here re how we decide the deadline in this case. I don't think we can use...

So the basic idea is to synchronize the entire contractcourt and sweeper on each block? Do arrows in the above diagrams represent passing around the latest block info? It seems...