{"type":"rich","version":"1.0","author_name":"npub1jdl3plz00rvxwc6g2ckemzrgg0amx5wen4kfvs3laxtssxvk9cvsf3gh0m","author_url":"https://nostr.ae/npub1jdl3plz00rvxwc6g2ckemzrgg0amx5wen4kfvs3laxtssxvk9cvsf3gh0m","provider_name":"njump","provider_url":"https://nostr.ae","html":"📅 Original date posted:2022-09-29\n📝 Original message:\u003e Right, good catch, this does require new logic to handle this case.\nAs Gloria points out, this should be doable, and is definitely worth\nadding (those CSV 1 on every other output are really hacky, glad to\nfind a way to get rid of them).\n\nFor the record, it turns out ephemeral anchors + v3 solves this already, as\nthe anchor must be spent, and the parent tx may only have one child.\nSomehow I missed this implication for a few months. It's great news if we\ncan directly source fees from any output claimable, including HTLCs!\n\nOn Thu, Sep 29, 2022 at 5:15 AM Bastien TEINTURIER \u003cbastien at acinq.fr\u003e wrote:\n\n\u003e Hi Gloria, Greg,\n\u003e\n\u003e \u003e I interpret most of the discussion around limitations as ideas for\n\u003e \u003e future improvements rather than criticisms of the proposal\n\u003e\n\u003e As far as I'm concerned, definitely!\n\u003e\n\u003e My current understanding is that the main change/improvement that would\n\u003e make sense here is restricting the whole v3 package's size (instead of\n\u003e just the child) via committing to a specific value in the taproot annex\n\u003e (also note that it's probably not just the v3 package's size, it should\n\u003e be the whole unconfirmed package including potential v2 unconfirmed\n\u003e ancestors).\n\u003e\n\u003e While I think this would be very valuable and would like to see this\n\u003e happen, I believe that can be done in a second, separate step since this\n\u003e would make relay policy stricter (some v3 transactions that previously\n\u003e propagated wouldn't propagate under this new rule). As long as you are\n\u003e able to find a path to miners through upgraded peers that use this annex\n\u003e approach, you should be able to resolve ACP pinning issues?\n\u003e\n\u003e I'm curious to know how other people feel about that: is it ok to do\n\u003e later or should we try to implement this for the first release of v3\n\u003e transactions?\n\u003e\n\u003e The other change mentioned (making OP_TRUE standard and allowing outputs\n\u003e that are below dust) can be added later, as those won't be standard until\n\u003e we start allowing them, so there shouldn't be any backwards-compatibility\n\u003e issue with postponing this change. But maybe it's still worth having from\n\u003e the get-go, even though it may take a bit more time? Again, I'm curious to\n\u003e have other people's opinion here, I'd be happy to get all of those directly\n\u003e in the first release of v3 transactions, but I don't know how much\n\u003e implementation will have to go into that.\n\u003e\n\u003e \u003e For clarification, package RBF is ParentTx*s*(plural), and\n\u003e ChildTx(singular),\n\u003e \u003e so it might be a bit more complicated than we're thinking\n\u003e\n\u003e Right, good catch, this does require new logic to handle this case.\n\u003e As Gloria points out, this should be doable, and is definitely worth\n\u003e adding (those CSV 1 on every other output are really hacky, glad to\n\u003e find a way to get rid of them).\n\u003e\n\u003e Thanks,\n\u003e Bastien\n\u003e\n\u003e Le lun. 26 sept. 2022 à 18:48, Gloria Zhao \u003cgloriajzhao at gmail.com\u003e a\n\u003e écrit :\n\u003e\n\u003e\u003e Hi Greg, Antoine, Bastien,\n\u003e\u003e\n\u003e\u003e Thanks very much for the feedback! I interpret most of the discussion\n\u003e\u003e around limitations as ideas for future improvements rather than criticisms\n\u003e\u003e of the proposal (please correct me if I'm wrong). I'll try to respond to as\n\u003e\u003e much as possible.\n\u003e\u003e\n\u003e\u003e Also I realize that I didn't contextualize this proposal clearly enough;\n\u003e\u003e it is very tailored for LN Penalty and definitely doesn't close all pinning\n\u003e\u003e attacks possible (sorry for confusing anyone). I also agree that some bits\n\u003e\u003e can be a little ugly or tack-on; I would definitely prefer a comprehensive\n\u003e\u003e RBF revamp to fix all our problems and enable other fee-bumping strategies\n\u003e\u003e such as\n\u003e\u003e sign-ANYONECANPAY-then-bring-your-own-fees-by-adding-inputs-at-broadcast. I\n\u003e\u003e was hoping to get some ideas with the \"RBF Improvements\" post in January,\n\u003e\u003e but it doesn't seem like we're much closer to a workable proposal. I think\n\u003e\u003e this is a minimally-invasive step that works for Lightning today, a small\n\u003e\u003e fix similar to CPFP carve out.\n\u003e\u003e\n\u003e\u003e \u003e As you likely know from previous discussions the biggest scenario this\n\u003e\u003e does not fix in my estimation is ANYONECANPAY situations. If the parent\n\u003e\u003e transaction can be \"inflated\" by tacking on additional inputs, this means\n\u003e\u003e the total weight of the parent tx lowers the effective feerate of the\n\u003e\u003e package.\n\u003e\u003e\n\u003e\u003e (For more context to other readers I wrote an explanation for this in\n\u003e\u003e \"SIGHASH_ANYONECANPAY Pinning\" section of RBF ML post).  Yes, this\n\u003e\u003e unfortunately doesn't fix any of the existing pinning attacks for single\n\u003e\u003e transaction RBF but also doesn't make them worse. This boils down to adding\n\u003e\u003e an incentive compatibility rule that ensures you can't replace a\n\u003e\u003e transaction with something that will confirm slower. Package RBF has an\n\u003e\u003e ancestor feerate-based rule for this (note it is quite conservative and not\n\u003e\u003e perfect).\n\u003e\u003e\n\u003e\u003e So in the scenario above with the \"inflated\" parent that was signed ACP,\n\u003e\u003e the replacement would be rejected because the package ancestor feerate is\n\u003e\u003e lower than the feerate of what is being replaced. But it is imperfect\n\u003e\u003e (explained below) and thus I wouldn't recommend it for single transaction\n\u003e\u003e replacement. So that attack still exists for single transactions, yes.\n\u003e\u003e\n\u003e\u003e The strategy of using ACP to bring-your-own-fees has its own challenges\n\u003e\u003e but hopefully has no current use cases as you say. AFAIK LN Penalty is not\n\u003e\u003e affected by this since it doesn't use ACP, though obviously I agree we\n\u003e\u003e should fix it for the future.\n\u003e\u003e\n\u003e\u003e So when I said \"this is intended for fee-bumping presigned txns in\n\u003e\u003e contracting protocols,\" I should have said \"this is intended for\n\u003e\u003e fee-bumping presigned txns specifically using CPFP and anchor outputs.\"\n\u003e\u003e Apologies for forgetting to contextualize, I've been sitting on this for\n\u003e\u003e too long.\n\u003e\u003e\n\u003e\u003e \u003e The other scenario it doesn't really fix is where HTLC/commitment-like\n\u003e\u003e transactions are being resolved in a batch, but due to relative time\n\u003e\u003e constraints, you may want to accelerate some and not others. Now you must\n\u003e\u003e pay higher rates to replace all of the transaction bumps. This is a\n\u003e\u003e \"self-pin\" and \"get good at utxos noob\" type problem, but it's something\n\u003e\u003e that axing rule#3 in favor of a Replace-by-ancestor-feerate system would\n\u003e\u003e get us.\n\u003e\u003e\n\u003e\u003e I understand you to mean \"if you don't have enough UTXOs and you're\n\u003e\u003e forced to batch-bump, you over-pay because you need to bring them all to\n\u003e\u003e the highest target feerate.\" Isn't this kind of separate, wallet-related\n\u003e\u003e problem? Contracting or not, surely every wallet needs to have enough UTXOs\n\u003e\u003e to not batch transactions that shouldn't be batched... I don't see how a\n\u003e\u003e replace-by-ancestor-feerate policy would make any difference for this?\n\u003e\u003e\n\u003e\u003e Also in general I'd like to reiterate that ancestor feerate is not a\n\u003e\u003e panacea to all our RBF incentive compatibility concerns. Like individual\n\u003e\u003e feerate, unless we run the mining algorithm, it cannot tell us exactly how\n\u003e\u003e quickly this transaction would be mined.\n\u003e\u003e\n\u003e\u003e We're estimating the incentive compatibility of the original\n\u003e\u003e transaction(s) and replacement transaction(s), with the goal of not letting\n\u003e\u003e a transaction replace something that would have been more incentive\n\u003e\u003e compatible to mine. As such, we don't want to overestimate how good the\n\u003e\u003e replacement is, and we don't want to underestimate how good the original\n\u003e\u003e transactions are. This rule \"The minimum between package feerate and\n\u003e\u003e ancestor feerate of the child is not lower than the individual feerates of\n\u003e\u003e all directly conflicting transactions and the ancestor feerates of all\n\u003e\u003e original transactions\" is a conservative estimate.\n\u003e\u003e\n\u003e\u003e \u003e Would kind of be nice if package RBF would detect a \"sibling output\n\u003e\u003e spend\" conflict, and knock it out of the mempool via the other replacement\n\u003e\u003e rules? Getting rid of the requirement to 1 block csv lock every output\n\u003e\u003e would be quite nice from a smart contracting composability point of view.\n\u003e\u003e\n\u003e\u003e Interesting, so when a transaction hits a mempool tx's descendant limit,\n\u003e\u003e we consider evicting one of its descendants in favor of this transaction,\n\u003e\u003e based on the RBF rules.\n\u003e\u003e Cool idea! After chewing on this for a bit, I think this *also* just\n\u003e\u003e boils down to the fact that RBF should require replacements to be better\n\u003e\u003e mining candidates. As in, if we added this policy and it can make us evict\n\u003e\u003e the sibling and accept a transaction with a bunch of low-feerate ancestor\n\u003e\u003e junk, it would be a new pinning vector.\n\u003e\u003e\n\u003e\u003e \u003e If you're a miner and you receive a non-V3, second descendant of an\n\u003e\u003e unconfirmed V3 transaction, if the offered fee is in the top mempool\n\u003e\u003e backlog, I think you would have an interest to accept such a transaction.\n\u003e\u003e\n\u003e\u003e \u003e So I'm not sure if those two rules are compatible with miners\n\u003e\u003e incentives...\n\u003e\u003e\n\u003e\u003e The same argument can be made for the 26th descendant of a mempool\n\u003e\u003e transaction; it's also not entirely incentive-compatible to reject it, but\n\u003e\u003e that is not the *only* design goal in mempool policy. Of course, the\n\u003e\u003e difference here is that the 25-descendant limit rule is a sensible DoS\n\u003e\u003e protection, while this 1-descendant limit rule is more of a \"help the\n\u003e\u003e Bitcoin ecosystem\" policy, just like CPFP carve-out, dust limit, etc. I can\n\u003e\u003e of course understand why not everyone would be in favor of this, but I do\n\u003e\u003e think it's worth it.\n\u003e\u003e\n\u003e\u003e \u003e \u003e 4. A V3 transaction that has an unconfirmed V3 ancestor cannot be\n\u003e\u003e \u003e \u003e    larger than 1000 virtual bytes.\n\u003e\u003e\n\u003e\u003e \u003e If I understand correctly the 1000 vb upper bound rational, it would be\n\u003e\u003e to constraint the pinning counterparty to attach a high fee to a child due\n\u003e\u003e to the limited size, if they would like this transaction to be stuck in the\n\u003e\u003e network mempools. By doing so  this child has high odds to confirm.\n\u003e\u003e\n\u003e\u003e Yeah exactly, the \"Rule 3 pin\" is done by adding a child that's high-fee\n\u003e\u003e (so you have to pay that much to evict it). Because they *don't* want this\n\u003e\u003e tx to confirm, normally, this child would be really large. If they only\n\u003e\u003e have 1000vB for the child, they can't increase the replacement cost without\n\u003e\u003e also fee-bumping the transaction to make it confirm faster.\n\u003e\u003e\n\u003e\u003e \u003e As of today, I think yes you can already fingerprint LN transactions on\n\u003e\u003e the  spec-defined amount value of the anchor outputs, 330 sats. There is\n\u003e\u003e always one of them on post-anchor commitment transactions. And sadly I\n\u003e\u003e would say we'll always have tricky fingerprints leaking from unilateral LN\n\u003e\u003e closures such as HTLC/PTLC timelocks...\n\u003e\u003e\n\u003e\u003e \u003e I agree with you, this isn't worse than today, unilateral closes will\n\u003e\u003e probably always be identifiable on-chain.\n\u003e\u003e\n\u003e\u003e Great to hear that there is no privacy worsening!\n\u003e\u003e\n\u003e\u003e Best,\n\u003e\u003e Gloria\n\u003e\u003e\n\u003e\u003e On Mon, Sep 26, 2022 at 5:02 PM Greg Sanders \u003cgsanders87 at gmail.com\u003e\n\u003e\u003e wrote:\n\u003e\u003e\n\u003e\u003e\u003e Bastien,\n\u003e\u003e\u003e\n\u003e\u003e\u003e \u003e This may be already covered by the current package RBF logic, in that\n\u003e\u003e\u003e scenario we are simply replacing [ParentTx, ChildTx1] with\n\u003e\u003e\u003e [ParentTx, ChildTx2] that pays more fees, right?\n\u003e\u003e\u003e\n\u003e\u003e\u003e For clarification, package RBF is ParentTx*s*(plural), and\n\u003e\u003e\u003e ChildTx(singular), so it might be a bit more complicated than we're\n\u003e\u003e\u003e thinking, and currently the V3 proposal would first de-duplicate the\n\u003e\u003e\u003e ParentTx based on what is in the mempool, then look at the \"rest\" of the\n\u003e\u003e\u003e transactions as a package, then individually. Not the same, not sure how\n\u003e\u003e\u003e different. I'll defer to experts.\n\u003e\u003e\u003e\n\u003e\u003e\u003e Best,\n\u003e\u003e\u003e Greg\n\u003e\u003e\u003e\n\u003e\u003e\u003e On Mon, Sep 26, 2022 at 11:48 AM Bastien TEINTURIER via bitcoin-dev \u003c\n\u003e\u003e\u003e bitcoin-dev at lists.linuxfoundation.org\u003e wrote:\n\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e Thanks Gloria for this great post.\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e This is very valuable work for L2 contracts, and will greatly improve\n\u003e\u003e\u003e\u003e their security model.\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e \u003e \"Only 1 anchor output? What if I need to bump counterparty's\n\u003e\u003e\u003e\u003e commitment tx in mempool?\"\n\u003e\u003e\u003e\u003e \u003e You won't need to fee-bump a counterparty's commitment tx using CPFP.\n\u003e\u003e\u003e\u003e \u003e You would just package RBF it by attaching a high-feerate child to\n\u003e\u003e\u003e\u003e \u003e your commitment tx.\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e Note that we can also very easily make that single anchor spendable by\n\u003e\u003e\u003e\u003e both participants (or even anyone), so if you see your counterparty's\n\u003e\u003e\u003e\u003e commitment in your mempool, you can bump it without publishing your\n\u003e\u003e\u003e\u003e own commitment, which is quite desirable (your own commitment tx has\n\u003e\u003e\u003e\u003e CSV delays on your outputs, whereas your counterparty's commitment tx\n\u003e\u003e\u003e\u003e doesn't).\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e \u003e \"Is this a privacy issue, i.e. doesn't it allow fingerprinting LN\n\u003e\u003e\u003e\u003e transactions based on nVersion?\"\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e I agree with you, this isn't worse than today, unilateral closes will\n\u003e\u003e\u003e\u003e probably always be identifiable on-chain.\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e \u003e Would kind of be nice if package RBF would detect a \"sibling output\n\u003e\u003e\u003e\u003e spend\"\n\u003e\u003e\u003e\u003e \u003e conflict, and knock it out of the mempool via the other replacement\n\u003e\u003e\u003e\u003e rules?\n\u003e\u003e\u003e\u003e \u003e Getting rid of the requirement to 1 block csv lock every output would\n\u003e\u003e\u003e\u003e be\n\u003e\u003e\u003e\u003e \u003e quite nice from a smart contracting composability point of view.\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e +1, that would be very neat!\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e This may be already covered by the current package RBF logic, in that\n\u003e\u003e\u003e\u003e scenario we are simply replacing [ParentTx, ChildTx1] with\n\u003e\u003e\u003e\u003e [ParentTx, ChildTx2] that pays more fees, right?\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e \u003e 1) I do think that we should seriously consider allowing OP_TRUE to\n\u003e\u003e\u003e\u003e become\n\u003e\u003e\u003e\u003e \u003e a standard script type as part of this policy update. If pinning is\n\u003e\u003e\u003e\u003e solved,\n\u003e\u003e\u003e\u003e \u003e then there's no reason to require all those extra bytes for \"binding\"\n\u003e\u003e\u003e\u003e an\n\u003e\u003e\u003e\u003e \u003e anchor to a specific wallet/user. We can save quite a few bytes by\n\u003e\u003e\u003e\u003e having\n\u003e\u003e\u003e\u003e \u003e the input be empty of witness data.\n\u003e\u003e\u003e\u003e \u003e 2) If we allow for a single dust-value(0 on up) output which is\n\u003e\u003e\u003e\u003e immediately\n\u003e\u003e\u003e\u003e \u003e spent by the package, anchors become even easier to to design. No\n\u003e\u003e\u003e\u003e value has\n\u003e\u003e\u003e\u003e \u003e to be \"sapped\" from contract participants to make an anchor output.\n\u003e\u003e\u003e\u003e There's\n\u003e\u003e\u003e\u003e \u003e more complications for this, such as making sure the parent\n\u003e\u003e\u003e\u003e transaction is\n\u003e\u003e\u003e\u003e \u003e dropped if the child spend is dropped, but maybe it's worth the\n\u003e\u003e\u003e\u003e squeeze.\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e I also think both of these could be quite useful. This would probably\n\u003e\u003e\u003e\u003e always\n\u003e\u003e\u003e\u003e be used in combination with a parent transaction that pays 0 fees, so\n\u003e\u003e\u003e\u003e the\n\u003e\u003e\u003e\u003e 0-value output would always be spent in the same block.\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e But this means we could end up with 0-value outputs in the utxo set, if\n\u003e\u003e\u003e\u003e for\n\u003e\u003e\u003e\u003e some reason the parent tx is CPFP-ed via another output than the\n\u003e\u003e\u003e\u003e 0-value one,\n\u003e\u003e\u003e\u003e which would be a utxo set bloat issue. But I'd argue that we're probably\n\u003e\u003e\u003e\u003e already creating utxo set bloat with the 330 sat anchor outputs\n\u003e\u003e\u003e\u003e (especially\n\u003e\u003e\u003e\u003e since we use two of them, but only one is usually spent), so it would\n\u003e\u003e\u003e\u003e probably be *better* than what we're doing today.\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e Thanks,\n\u003e\u003e\u003e\u003e Bastien\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e Le lun. 26 sept. 2022 à 03:22, Antoine Riard via bitcoin-dev \u003c\n\u003e\u003e\u003e\u003e bitcoin-dev at lists.linuxfoundation.org\u003e a écrit :\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e Hi Gloria,\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e Thanks for the progress on package RBF, few early questions.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e \u003e 2. Any descendant of an unconfirmed V3 transaction must also be V3.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e \u003e 3. An unconfirmed V3 transaction cannot have more than 1 descendant.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e If you're a miner and you receive a non-V3, second descendant of an\n\u003e\u003e\u003e\u003e\u003e unconfirmed V3 transaction, if the offered fee is in the top mempool\n\u003e\u003e\u003e\u003e\u003e backlog, I think you would have an interest to accept such a transaction.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e So I'm not sure if those two rules are compatible with miners\n\u003e\u003e\u003e\u003e\u003e incentives...\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e \u003e 4. A V3 transaction that has an unconfirmed V3 ancestor cannot be\n\u003e\u003e\u003e\u003e\u003e \u003e    larger than 1000 virtual bytes.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e If I understand correctly the 1000 vb upper bound rational, it would\n\u003e\u003e\u003e\u003e\u003e be to constraint the pinning counterparty to attach a high fee to a child\n\u003e\u003e\u003e\u003e\u003e due to the limited size, if they would like this transaction to be stuck in\n\u003e\u003e\u003e\u003e\u003e the network mempools. By doing so  this child has high odds to confirm.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e I still wonder if this compatible with miner incentives in period of\n\u003e\u003e\u003e\u003e\u003e empty mempools, in the sense that if you've already a V3 transaction of\n\u003e\u003e\u003e\u003e\u003e size 100Kvb offering 2 sat/vb, it's more interesting than a V3 replacement\n\u003e\u003e\u003e\u003e\u003e candidate of size 1000 vb offering 10 sat/vb. It could be argued the former\n\u003e\u003e\u003e\u003e\u003e should be conserved.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e (That said, the hard thing with any replacement strategy we might\n\u003e\u003e\u003e\u003e\u003e evict a parent transaction *now* to which is attached a high-feerate child\n\u003e\u003e\u003e\u003e\u003e *latter* making for a utxo considered the best ancestor set. Maybe in the\n\u003e\u003e\u003e\u003e\u003e long-term miners should keep every transaction ever accepted...)\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e \u003e (Lower bound) the smaller this limit, the fewer UTXOs a child may use\n\u003e\u003e\u003e\u003e\u003e \u003e to fund this fee-bump. For example, only allowing the V3 child to\n\u003e\u003e\u003e\u003e\u003e have\n\u003e\u003e\u003e\u003e\u003e \u003e 2 inputs would require L2 protocols to manage a wallet with\n\u003e\u003e\u003e\u003e\u003e high-value\n\u003e\u003e\u003e\u003e\u003e \u003e UTXOs and make batched fee-bumping impossible. However, as the\n\u003e\u003e\u003e\u003e\u003e \u003e fee-bumping child only needs to fund fees (as opposed to payments),\n\u003e\u003e\u003e\u003e\u003e \u003e just a few UTXOs should suffice.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e Reminder for L2 devs, batched fee-bumping of time-sensitive\n\u003e\u003e\u003e\u003e\u003e confirmations of commitment transactions is unsafe, as the counterparty\n\u003e\u003e\u003e\u003e\u003e could enter in a \"cat-and-mouse\" game to replace one of the batch element\n\u003e\u003e\u003e\u003e\u003e at each block to delay confirmation of the remaining elements in the batch,\n\u003e\u003e\u003e\u003e\u003e I think.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e On the other hand, I wonder if we wouldn't want a higher bound. LN\n\u003e\u003e\u003e\u003e\u003e wallets are likely to have one big UTXO in their fee-bumping reserve pool,\n\u003e\u003e\u003e\u003e\u003e as the cost of acquiring UTXO is non-null and in the optimistic case, you\n\u003e\u003e\u003e\u003e\u003e don't need to do unilateral closure. Let's say you close dozens of channels\n\u003e\u003e\u003e\u003e\u003e at the same time, a UTXO pool management strategy might be to fan-out the\n\u003e\u003e\u003e\u003e\u003e first spends UTXOs in N fan-out outputs ready to feed the remaining\n\u003e\u003e\u003e\u003e\u003e in-flight channels.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e \u003e 1. The rule around unconfirmed inputs was\n\u003e\u003e\u003e\u003e\u003e \u003e originally \"A package may include new unconfirmed inputs, but the\n\u003e\u003e\u003e\u003e\u003e \u003e ancestor feerate of the child must be at least as high as the\n\u003e\u003e\u003e\u003e\u003e ancestor\n\u003e\u003e\u003e\u003e\u003e \u003e feerates of every transaction being replaced.\"\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e Note, I think we would like this new RBF rule to also apply to single\n\u003e\u003e\u003e\u003e\u003e transaction package, e.g second-stage HTLC transactions, where a\n\u003e\u003e\u003e\u003e\u003e counterparty pins a HTLC-preimage by abusing rule 3. In that case, the\n\u003e\u003e\u003e\u003e\u003e honest LN node should be able to broadcast a \"at least as high ancestor\n\u003e\u003e\u003e\u003e\u003e feerate\" HTLC-timeout transaction. With `option_anchor_outputs\" there is no\n\u003e\u003e\u003e\u003e\u003e unconfirmed ancestor to replace, as the commitment transaction, whatever\n\u003e\u003e\u003e\u003e\u003e the party it is originating from, should already be confirmed.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e \u003e \"Is this a privacy issue, i.e. doesn't it allow fingerprinting LN\n\u003e\u003e\u003e\u003e\u003e transactions based on nVersion?\"\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e As of today, I think yes you can already fingerprint LN transactions\n\u003e\u003e\u003e\u003e\u003e on the  spec-defined amount value of the anchor outputs, 330 sats. There is\n\u003e\u003e\u003e\u003e\u003e always one of them on post-anchor commitment transactions. And sadly I\n\u003e\u003e\u003e\u003e\u003e would say we'll always have tricky fingerprints leaking from unilateral LN\n\u003e\u003e\u003e\u003e\u003e closures such as HTLC/PTLC timelocks...\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e \u003e \"Can a V2 transaction replace a V3 transaction and vice versa?\"\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e IIUC, a V3 package could replace a V2 package, with the benefit of the\n\u003e\u003e\u003e\u003e\u003e new package RBF rules applied. I think this would be a significant\n\u003e\u003e\u003e\u003e\u003e advantage for LN, as for the current ~85k of opened channels, the old V2\n\u003e\u003e\u003e\u003e\u003e states shouldn't be pinning vectors. Currently, commitment transactions\n\u003e\u003e\u003e\u003e\u003e signal replaceability.\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e Le ven. 23 sept. 2022 à 11:26, Gloria Zhao via bitcoin-dev \u003c\n\u003e\u003e\u003e\u003e\u003e bitcoin-dev at lists.linuxfoundation.org\u003e a écrit :\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e Hi everyone,\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e I'm writing to propose a very simple set of mempool/transaction relay\n\u003e\u003e\u003e\u003e\u003e\u003e policies intended to aid L2/contract protocols. I realized that\n\u003e\u003e\u003e\u003e\u003e\u003e the previously proposed Package Mempool Accept package RBF [1]\n\u003e\u003e\u003e\u003e\u003e\u003e had a few remaining problems after digging into the RBF logic more\n\u003e\u003e\u003e\u003e\u003e\u003e [2].\n\u003e\u003e\u003e\u003e\u003e\u003e This additional set of policies solves them without requiring a huge\n\u003e\u003e\u003e\u003e\u003e\u003e RBF overhaul.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e I've written an implementation (and docs) for Bitcoin Core:\n\u003e\u003e\u003e\u003e\u003e\u003e https://github.com/bitcoin/bitcoin/pull/25038\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e (You may notice that this proposal incorporates feedback on the PR -\n\u003e\u003e\u003e\u003e\u003e\u003e thanks Suhas Daftuar, Gregory Sanders, Bastien Teinturier, Anthony Towns,\n\u003e\u003e\u003e\u003e\u003e\u003e and others.)\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e If you are interested in using package RBF/relay to bump presigned\n\u003e\u003e\u003e\u003e\u003e\u003e transactions, I think you may be interested in reviewing this\n\u003e\u003e\u003e\u003e\u003e\u003e proposal.\n\u003e\u003e\u003e\u003e\u003e\u003e This should solve Rule 3 pinning and perhaps allow us\n\u003e\u003e\u003e\u003e\u003e\u003e to get rid of CPFP carve-out (yay!). I'm keen to hear if people find\n\u003e\u003e\u003e\u003e\u003e\u003e the 1-anchor-output, 1000vB child limit too restrictive. Also, if you\n\u003e\u003e\u003e\u003e\u003e\u003e find a\n\u003e\u003e\u003e\u003e\u003e\u003e pinning attack or something that makes it unusable for you, I would\n\u003e\u003e\u003e\u003e\u003e\u003e really really like to know.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e Note that transactions with nVersion=3 (\"V3 transactions\") are\n\u003e\u003e\u003e\u003e\u003e\u003e currently non-standard in Bitcoin Core. That means **anything that was\n\u003e\u003e\u003e\u003e\u003e\u003e standard before this policy change would still be standard\n\u003e\u003e\u003e\u003e\u003e\u003e afterwards.** If you don't want your transactions to be subject to\n\u003e\u003e\u003e\u003e\u003e\u003e these rules, just continue whatever you're doing and don't use\n\u003e\u003e\u003e\u003e\u003e\u003e nVersion=3. AFAICT this shouldn't break anything, but let me know if\n\u003e\u003e\u003e\u003e\u003e\u003e this would be disruptive for you?\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e **New Policies:**\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e This includes:\n\u003e\u003e\u003e\u003e\u003e\u003e - a set of additional policy rules applying to V3 transactions\n\u003e\u003e\u003e\u003e\u003e\u003e - modifications to package RBF rules\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e **V3 transactions:**\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e Existing standardness rules apply to V3 (e.g. min/max tx weight,\n\u003e\u003e\u003e\u003e\u003e\u003e standard output types, cleanstack, etc.). The following additional\n\u003e\u003e\u003e\u003e\u003e\u003e rules apply to V3:\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e 1. A V3 transaction can be replaced, even if it does not signal BIP125\n\u003e\u003e\u003e\u003e\u003e\u003e    replaceability. (It must also meet the other RBF rules around fees,\n\u003e\u003e\u003e\u003e\u003e\u003e etc. for replacement to happen).\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e 2. Any descendant of an unconfirmed V3 transaction must also be V3.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e *Rationale*: Combined with Rule 1, this gives us the property of\n\u003e\u003e\u003e\u003e\u003e\u003e \"inherited\" replaceability signaling when descendants of unconfirmed\n\u003e\u003e\u003e\u003e\u003e\u003e transactions are created. Additionally, checking whether a transaction\n\u003e\u003e\u003e\u003e\u003e\u003e signals replaceability this way does not require mempool traversal,\n\u003e\u003e\u003e\u003e\u003e\u003e and does not change based on what transactions are mined. It also\n\u003e\u003e\u003e\u003e\u003e\u003e makes subsequent rules about descendant limits much easier to check.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e *Note*: The descendant of a *confirmed* V3 transaction does not need\n\u003e\u003e\u003e\u003e\u003e\u003e to be V3.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e 3. An unconfirmed V3 transaction cannot have more than 1 descendant.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e *Rationale*: (Upper bound) the larger the descendant limit, the more\n\u003e\u003e\u003e\u003e\u003e\u003e transactions may need to be replaced. This is a problematic pinning\n\u003e\u003e\u003e\u003e\u003e\u003e attack, i.e., a malicious counterparty prevents the transaction from\n\u003e\u003e\u003e\u003e\u003e\u003e being replaced by adding many descendant transactions that aren't\n\u003e\u003e\u003e\u003e\u003e\u003e fee-bumping.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e (Lower bound) at least 1 descendant is required to allow CPFP of the\n\u003e\u003e\u003e\u003e\u003e\u003e presigned transaction. The contract protocol can create presigned\n\u003e\u003e\u003e\u003e\u003e\u003e transactions paying 0 fees and 1 output for attaching a CPFP at\n\u003e\u003e\u003e\u003e\u003e\u003e broadcast time (\"anchor output\"). Without package RBF, multiple anchor\n\u003e\u003e\u003e\u003e\u003e\u003e outputs would be required to allow each counterparty to fee-bump any\n\u003e\u003e\u003e\u003e\u003e\u003e presigned transaction. With package RBF, since the presigned\n\u003e\u003e\u003e\u003e\u003e\u003e transactions can replace each other, 1 anchor output is sufficient.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e 4. A V3 transaction that has an unconfirmed V3 ancestor cannot be\n\u003e\u003e\u003e\u003e\u003e\u003e    larger than 1000 virtual bytes.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e *Rationale*: (Upper bound) the larger the descendant size limit, the\n\u003e\u003e\u003e\u003e\u003e\u003e more vbytes may need to be replaced. With default limits, if the child\n\u003e\u003e\u003e\u003e\u003e\u003e is e.g. 100,000vB, that might be an additional 100,000sats (at\n\u003e\u003e\u003e\u003e\u003e\u003e 1sat/vbyte) or more, depending on the feerate.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e (Lower bound) the smaller this limit, the fewer UTXOs a child may use\n\u003e\u003e\u003e\u003e\u003e\u003e to fund this fee-bump. For example, only allowing the V3 child to have\n\u003e\u003e\u003e\u003e\u003e\u003e 2 inputs would require L2 protocols to manage a wallet with high-value\n\u003e\u003e\u003e\u003e\u003e\u003e UTXOs and make batched fee-bumping impossible. However, as the\n\u003e\u003e\u003e\u003e\u003e\u003e fee-bumping child only needs to fund fees (as opposed to payments),\n\u003e\u003e\u003e\u003e\u003e\u003e just a few UTXOs should suffice.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e With a limit of 1000 virtual bytes, depending on the output types, the\n\u003e\u003e\u003e\u003e\u003e\u003e child can have 6-15 UTXOs, which should be enough to fund a fee-bump\n\u003e\u003e\u003e\u003e\u003e\u003e without requiring a carefully-managed UTXO pool. With 1000 virtual\n\u003e\u003e\u003e\u003e\u003e\u003e bytes as the descendant limit, the cost to replace a V3 transaction\n\u003e\u003e\u003e\u003e\u003e\u003e has much lower variance.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e *Rationale*: This makes the rule very easily \"tacked on\" to existing\n\u003e\u003e\u003e\u003e\u003e\u003e logic for policy and wallets. A transaction may be up to 100KvB on its\n\u003e\u003e\u003e\u003e\u003e\u003e own (`MAX_STANDARD_TX_WEIGHT`) and 101KvB with descendants\n\u003e\u003e\u003e\u003e\u003e\u003e (`DEFAULT_DESCENDANT_SIZE_LIMIT_KVB`). If an existing V3 transaction\n\u003e\u003e\u003e\u003e\u003e\u003e in the mempool is 100KvB, its descendant can only be 1000vB, even if\n\u003e\u003e\u003e\u003e\u003e\u003e the policy is 10KvB.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e **Package RBF modifications:**\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e 1. The rule around unconfirmed inputs was\n\u003e\u003e\u003e\u003e\u003e\u003e originally \"A package may include new unconfirmed inputs, but the\n\u003e\u003e\u003e\u003e\u003e\u003e ancestor feerate of the child must be at least as high as the ancestor\n\u003e\u003e\u003e\u003e\u003e\u003e feerates of every transaction being replaced.\"\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e The package may still include new unconfirmed inputs. However,\n\u003e\u003e\u003e\u003e\u003e\u003e the new rule is modified to be \"The minimum between package feerate\n\u003e\u003e\u003e\u003e\u003e\u003e and ancestor feerate of the child is not lower than the individual\n\u003e\u003e\u003e\u003e\u003e\u003e feerates of all directly conflicting transactions and the ancestor\n\u003e\u003e\u003e\u003e\u003e\u003e feerates of all original transactions.\"\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e *Rationale*: We are attempting to ensure that the replacement\n\u003e\u003e\u003e\u003e\u003e\u003e transactions are not less incentive-compatible to mine. However, a\n\u003e\u003e\u003e\u003e\u003e\u003e package/transaction's ancestor feerate is not perfectly representative\n\u003e\u003e\u003e\u003e\u003e\u003e of its incentive compatibility; it may overestimate (some subset of\n\u003e\u003e\u003e\u003e\u003e\u003e the ancestors could be included by itself if it has other high-feerate\n\u003e\u003e\u003e\u003e\u003e\u003e descendants or are themselves higher feerate than this\n\u003e\u003e\u003e\u003e\u003e\u003e package/transaction). Instead, we use the minimum between the package\n\u003e\u003e\u003e\u003e\u003e\u003e feerate and ancestor feerate of the child as a more conservative value\n\u003e\u003e\u003e\u003e\u003e\u003e than what was proposed originally.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e 2. A new rule is added, requiring that all package transactions with\n\u003e\u003e\u003e\u003e\u003e\u003e mempool conflicts to be V3. This also means the \"sponsoring\"\n\u003e\u003e\u003e\u003e\u003e\u003e child transaction must be V3.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e *Note*: Combined with the V3 rules, this means the package must be\n\u003e\u003e\u003e\u003e\u003e\u003e a child-with-parents package. Since package validation is only\n\u003e\u003e\u003e\u003e\u003e\u003e attempted if the transactions do not pay sufficient fees to be\n\u003e\u003e\u003e\u003e\u003e\u003e accepted on their own, this effectively means that only V3\n\u003e\u003e\u003e\u003e\u003e\u003e transactions can pay to replace their ancestors' conflicts, and only\n\u003e\u003e\u003e\u003e\u003e\u003e V3 transactions' replacements may be paid for by a descendant.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e *Rationale*: The fee-related rules are economically rational for\n\u003e\u003e\u003e\u003e\u003e\u003e ancestor packages, but not necessarily other types of packages.\n\u003e\u003e\u003e\u003e\u003e\u003e A child-with-parents package is a type of ancestor package. It\n\u003e\u003e\u003e\u003e\u003e\u003e may be fine to allow any ancestor package, but it's more difficult\n\u003e\u003e\u003e\u003e\u003e\u003e to account for all of the possibilities. For example, it gets much\n\u003e\u003e\u003e\u003e\u003e\u003e harder to see that we're applying the descendant limits correctly if\n\u003e\u003e\u003e\u003e\u003e\u003e the package has a gnarly, many-generation, non-tree shape. I'm also\n\u003e\u003e\u003e\u003e\u003e\u003e not sure if this policy is 100% incentive-compatible if the sponsor\n\u003e\u003e\u003e\u003e\u003e\u003e is not a direct descendant of the sponsee.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e Please see doc/policy/version3_transactions.md and\n\u003e\u003e\u003e\u003e\u003e\u003e doc/policy/packages.md in the PR for the full set of rules.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e **Intended usage for LN:**\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e Commitment transactions should be V3 and have 1 anchor output. They\n\u003e\u003e\u003e\u003e\u003e\u003e can be signed with 0 fees (or 1sat/vbyte) once package relay is\n\u003e\u003e\u003e\u003e\u003e\u003e deployed\n\u003e\u003e\u003e\u003e\u003e\u003e on a significant portion of the network. If the commitment tx must\n\u003e\u003e\u003e\u003e\u003e\u003e be broadcast, determine the desired feerate at broadcast time and\n\u003e\u003e\u003e\u003e\u003e\u003e spend the anchor output in a high feerate transaction. I'm going to\n\u003e\u003e\u003e\u003e\u003e\u003e call the broadcasted commitment tx \"the parent\" and the attached\n\u003e\u003e\u003e\u003e\u003e\u003e fee-bumping tx \"the child.\"\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e - This child must be V3.\n\u003e\u003e\u003e\u003e\u003e\u003e - This child must be at most 1000vB. Note this restricts the\n\u003e\u003e\u003e\u003e\u003e\u003e   number of inputs you can use to fund the fee bump. Depending\n\u003e\u003e\u003e\u003e\u003e\u003e on the output types, this is around 6-15.\n\u003e\u003e\u003e\u003e\u003e\u003e - One child may fund fees for multiple commitment tx (\"batched\n\u003e\u003e\u003e\u003e\u003e\u003e   fee-bumping\").\n\u003e\u003e\u003e\u003e\u003e\u003e - To do a second fee-bump to add more fees, replace the\n\u003e\u003e\u003e\u003e\u003e\u003e   *child* with a higher-feerate tx. Do not try to attach a grandchild.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e Otherwise, never try to spend from an unconfirmed V3 transaction. The\n\u003e\u003e\u003e\u003e\u003e\u003e descendant limits for V3 transactions are very restrictive.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e **Expected Questions:**\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e \"Does this fix Rule 3 Pinning?\"\n\u003e\u003e\u003e\u003e\u003e\u003e Yes. The V3 descendant limit restricts both you and your counterparty.\n\u003e\u003e\u003e\u003e\u003e\u003e Assuming nodes adopted this policy, you may reasonably assume that you\n\u003e\u003e\u003e\u003e\u003e\u003e only need to replace the commitment transaction + up to 1000vB.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e \"Only 1 anchor output? What if I need to bump counterparty's\n\u003e\u003e\u003e\u003e\u003e\u003e commitment tx in mempool?\"\n\u003e\u003e\u003e\u003e\u003e\u003e You won't need to fee-bump a counterparty's commitment tx using CPFP.\n\u003e\u003e\u003e\u003e\u003e\u003e You would just package RBF it by attaching a high-feerate child to\n\u003e\u003e\u003e\u003e\u003e\u003e your commitment tx.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e \"Is this a privacy issue, i.e. doesn't it allow fingerprinting LN\n\u003e\u003e\u003e\u003e\u003e\u003e transactions based on nVersion?\"\n\u003e\u003e\u003e\u003e\u003e\u003e Indeed it may be unrealistic to assume V3 transactions will be in\n\u003e\u003e\u003e\u003e\u003e\u003e widespread use outside of L2. IIUC, unilateral closes are already\n\u003e\u003e\u003e\u003e\u003e\u003e obvious LN transactions because of the HTLC inputs. For e.g.\n\u003e\u003e\u003e\u003e\u003e\u003e cooperative closes and opens, I think it makes sense to continue using\n\u003e\u003e\u003e\u003e\u003e\u003e V2. So, unless I'm missing something, this shouldn't make it worse.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e \"So a V3 transaction that doesn't signal BIP125 replaceability is\n\u003e\u003e\u003e\u003e\u003e\u003e replaceable? Is that a backward compatibility issue?\"\n\u003e\u003e\u003e\u003e\u003e\u003e Yes it's replaceable. It's not an issue AFAICT because,\n\u003e\u003e\u003e\u003e\u003e\u003e under previous policy, the V3 transaction wouldn't have been\n\u003e\u003e\u003e\u003e\u003e\u003e in the mempool in the first place.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e \"Can a V2 transaction replace a V3 transaction and vice versa?\"\n\u003e\u003e\u003e\u003e\u003e\u003e Yes, otherwise someone can use V3 transactions to censor V2\n\u003e\u003e\u003e\u003e\u003e\u003e transactions spending shared inputs. Note if the\n\u003e\u003e\u003e\u003e\u003e\u003e original V3 transaction has an unconfirmed V3 parent, this would\n\u003e\u003e\u003e\u003e\u003e\u003e violate the \"inherited V3\" rule and would be rejected.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e Thanks for reading! Feedback and review would be much appreciated.\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e [1]:\n\u003e\u003e\u003e\u003e\u003e\u003e https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2021-September/019464.html\n\u003e\u003e\u003e\u003e\u003e\u003e [2]:\n\u003e\u003e\u003e\u003e\u003e\u003e https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2022-January/019817.html\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e\u003e Best,\n\u003e\u003e\u003e\u003e\u003e\u003e Gloria\n\u003e\u003e\u003e\u003e\u003e\u003e _______________________________________________\n\u003e\u003e\u003e\u003e\u003e\u003e bitcoin-dev mailing list\n\u003e\u003e\u003e\u003e\u003e\u003e bitcoin-dev at lists.linuxfoundation.org\n\u003e\u003e\u003e\u003e\u003e\u003e https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev\n\u003e\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e\u003e _______________________________________________\n\u003e\u003e\u003e\u003e\u003e bitcoin-dev mailing list\n\u003e\u003e\u003e\u003e\u003e bitcoin-dev at lists.linuxfoundation.org\n\u003e\u003e\u003e\u003e\u003e https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev\n\u003e\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\u003e _______________________________________________\n\u003e\u003e\u003e\u003e bitcoin-dev mailing list\n\u003e\u003e\u003e\u003e bitcoin-dev at lists.linuxfoundation.org\n\u003e\u003e\u003e\u003e https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev\n\u003e\u003e\u003e\u003e\n\u003e\u003e\u003e\n-------------- next part --------------\nAn HTML attachment was scrubbed...\nURL: \u003chttp://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20220929/fa260464/attachment-0001.html\u003e"}