{"type":"rich","version":"1.0","author_name":"npub19helcfnqgk2jrwzjex2aflq6jwfc8zd9uzzkwlgwhve7lykv23mq5zkvn4","author_url":"https://nostr.ae/npub19helcfnqgk2jrwzjex2aflq6jwfc8zd9uzzkwlgwhve7lykv23mq5zkvn4","provider_name":"njump","provider_url":"https://nostr.ae","html":"📅 Original date posted:2020-04-22\n📝 Original message:\n\u003e Indeed, that is what I’m suggesting\n\nGotcha, if this is indeed what you're suggesting (all HTLC spends are now\n2-of-2 multi-sig), then I think the modifications to the state machine I\nsketched out in an earlier email are required. An exact construction which\nachieves the requirements of \"you can't broadcast until you have a secret\nwhich I can obtain from the htlc sig for your commitment transaction, and my\nsecret is revealed with another swap\", appears to be an open problem, atm.\n\nEven if they're restricted in this fashion (must be a 1-in-1 out,\nsighashall, fees are pre agreed upon), they can still spend that with a CPFP\n(while still unconfirmed in the mempool) and create another heavy tree,\nwhich puts us right back at the same bidding war scenario?\n\n\u003e There are a bunch of ways of doing pinning - just opting into RBF isn’t\n\u003e even close to enough.\n\nMhmm, there're other ways of doing pinning. But with anchors as is defined\nin that spec PR, they're forced to spend with an RBF-replaceable\ntransaction, which means the party wishing to time things out can enter into\na bidding war. If the party trying to impeded things participates in this\nprogressive absolute fee increase, it's likely that the war terminates\nwith _one_ of them getting into the block, which seems to resolve\neverything?\n\n-- Laolu\n\n\nOn Wed, Apr 22, 2020 at 4:20 PM Matt Corallo \u003clf-lists at mattcorallo.com\u003e\nwrote:\n\n\u003e\n\u003e\n\u003e On Apr 22, 2020, at 16:13, Olaoluwa Osuntokun \u003claolu32 at gmail.com\u003e wrote:\n\u003e\n\u003e\n\u003e \u003e Hmm, maybe the proposal wasn't clear. The idea isn't to add signatures to\n\u003e \u003e braodcasted transactions, but instead to CPFP a maybe-broadcasted\n\u003e \u003e transaction by sending a transaction which spends it and seeing if it is\n\u003e \u003e accepted\n\u003e\n\u003e Sorry I still don't follow. By \"we clearly need to go the other direction -\n\u003e all HTLC output spends need to be pre-signed.\", you don't mean that the\n\u003e HTLC\n\u003e spends of the non-broadcaster also need to be an off-chain 2-of-2 multi-sig\n\u003e covenant? If the other party isn't restricted w.r.t _how_ they can spend\n\u003e the\n\u003e output (non-rbf'd, ect), then I don't see how that addresses anything.\n\u003e\n\u003e\n\u003e Indeed, that is what I’m suggesting. Anchor output and all. One thing we\n\u003e could think about is only turning it on over a certain threshold, and\n\u003e having a separate “only-kinda-enforceable-on-chain-HTLC-in-flight” limit.\n\u003e\n\u003e Also see my mail elsewhere in the thread that the other party is actually\n\u003e forced to spend their HTLC output using an RBF-replaceable transaction.\n\u003e With\n\u003e that, I think we're all good here? In the end both sides have the ability\n\u003e to\n\u003e raise the fee rate of their spending transactions with the highest winning.\n\u003e As long as one of them confirms within the CLTV-delta, then everyone is\n\u003e made whole.\n\u003e\n\u003e\n\u003e It does seem like my cached recollection of RBF opt-in was incorrect but\n\u003e please re-read the intro email. There are a bunch of ways of doing pinning\n\u003e - just opting into RBF isn’t even close to enough.\n\u003e\n\u003e [1]: https://github.com/bitcoin/bitcoin/pull/18191\n\u003e\n\u003e\n\u003e On Wed, Apr 22, 2020 at 9:50 AM Matt Corallo \u003clf-lists at mattcorallo.com\u003e\n\u003e wrote:\n\u003e\n\u003e\u003e A few replies inline.\n\u003e\u003e\n\u003e\u003e On 4/22/20 12:13 AM, Olaoluwa Osuntokun wrote:\n\u003e\u003e \u003e Hi Matt,\n\u003e\u003e \u003e\n\u003e\u003e \u003e\n\u003e\u003e \u003e\u003e While this is somewhat unintuitive, there are any number of good\n\u003e\u003e anti-DoS\n\u003e\u003e \u003e\u003e reasons for this, eg:\n\u003e\u003e \u003e\n\u003e\u003e \u003e None of these really strikes me as \"good\" reasons for this limitation,\n\u003e\u003e which\n\u003e\u003e \u003e is at the root of this issue, and will also plague any more complex\n\u003e\u003e Bitcoin\n\u003e\u003e \u003e contracts which rely on nested trees of transaction to confirm (CTV,\n\u003e\u003e Duplex,\n\u003e\u003e \u003e channel factories, etc). Regarding the various (seemingly arbitrary)\n\u003e\u003e package\n\u003e\u003e \u003e limits it's likely the case that any issues w.r.t computational\n\u003e\u003e complexity\n\u003e\u003e \u003e that may arise when trying to calculate evictions can be ameliorated\n\u003e\u003e with\n\u003e\u003e \u003e better choice of internal data structures.\n\u003e\u003e \u003e\n\u003e\u003e \u003e In the end, the simplest heuristic (accept the higher fee rate package)\n\u003e\u003e side\n\u003e\u003e \u003e steps all these issues and is also the most economically rationale from\n\u003e\u003e a\n\u003e\u003e \u003e miner's perspective. Why would one prefer a higher absolute fee package\n\u003e\u003e \u003e (which could be very large) over another package with a higher total\n\u003e\u003e _fee\n\u003e\u003e \u003e rate_?\n\u003e\u003e\n\u003e\u003e This seems like a somewhat unnecessary drive-by insult of a project you\n\u003e\u003e don't contribute to, but feel free to start with\n\u003e\u003e a concrete suggestion here :).\n\u003e\u003e\n\u003e\u003e \u003e\u003e You'll note that B would be just fine if they had a way to safely\n\u003e\u003e monitor the\n\u003e\u003e \u003e\u003e global mempool, and while this seems like a prudent mitigation for\n\u003e\u003e \u003e\u003e lightning implementations to deploy today, it is itself a quagmire of\n\u003e\u003e \u003e\u003e complexity\n\u003e\u003e \u003e\n\u003e\u003e \u003e Is it really all that complex? Assuming we're talking about just\n\u003e\u003e watching\n\u003e\u003e \u003e for a certain script template (the HTLC scipt) in the mempool to be\n\u003e\u003e able to\n\u003e\u003e \u003e pull a pre-image as soon as possible. Early versions of lnd used the\n\u003e\u003e mempool\n\u003e\u003e \u003e for commitment broadcast detection (which turned out to be a bad idea\n\u003e\u003e so we\n\u003e\u003e \u003e removed it), but at a glance I don't see why watching the mempool is so\n\u003e\u003e \u003e complex.\n\u003e\u003e\n\u003e\u003e Because watching your own mempool is not guaranteed to work, and during\n\u003e\u003e upgrade cycles that include changes to the\n\u003e\u003e policy rules an attacker could exploit your upgraded/non-upgraded status\n\u003e\u003e to perform the same attack.\n\u003e\u003e\n\u003e\u003e \u003e\u003e Further, this is a really obnoxious assumption to hoist onto lightning\n\u003e\u003e \u003e\u003e nodes - having an active full node with an in-sync mempool is a lot\n\u003e\u003e more\n\u003e\u003e \u003e\u003e CPU, bandwidth, and complexity than most lightning users were\n\u003e\u003e expecting to\n\u003e\u003e \u003e\u003e face.\n\u003e\u003e \u003e\n\u003e\u003e \u003e This would only be a requirement for Lightning nodes that seek to be a\n\u003e\u003e part\n\u003e\u003e \u003e of the public routing network with a desire to _forward_ HTLCs. This\n\u003e\u003e isn't\n\u003e\u003e \u003e doesn't affect laptops or mobile phones which likely mostly have private\n\u003e\u003e \u003e channels and don't participate in HTLC forwarding. I think it's pretty\n\u003e\u003e \u003e reasonable to expect a \"proper\" routing node on the network to be\n\u003e\u003e backed by\n\u003e\u003e \u003e a full-node. The bandwidth concern is valid, but we'd need concrete\n\u003e\u003e numbers\n\u003e\u003e \u003e that compare the bandwidth over head of mempool awareness (assuming the\n\u003e\u003e \u003e latest and greatest mempool syncing) compared with the overhead of the\n\u003e\u003e \u003e channel update gossip and gossip queries over head which LN nodes face\n\u003e\u003e today\n\u003e\u003e \u003e as is to see how much worse off they really would be.\n\u003e\u003e\n\u003e\u003e If mempool-watching were practical, maybe, though there are a number of\n\u003e\u003e folks who are talking about designing\n\u003e\u003e partially-offline local lightning hubs which would be rendered\n\u003e\u003e impractical.\n\u003e\u003e\n\u003e\u003e \u003e As detailed a bit below, if nodes watch the mempool, then this class of\n\u003e\u003e \u003e attack assuming the anchor output format as described in the open\n\u003e\u003e \u003e lightning-rfc PR is mitigated. At a glance, watching the mempool seems\n\u003e\u003e like\n\u003e\u003e \u003e a far less involved process compared to modifying the state machine as\n\u003e\u003e its\n\u003e\u003e \u003e defined today. By watching the mempool and implementing the changes in\n\u003e\u003e \u003e #lightning-rfc/688, then this issue can be mitigated _today_. lnd 0.10\n\u003e\u003e \u003e doesn't yet watch the mempool (but does include anchors [1]), but\n\u003e\u003e unless I'm\n\u003e\u003e \u003e missing something it should be pretty straight forward to add which mor\n\u003e\u003e or less\n\u003e\u003e \u003e resolves this issue all together.\n\u003e\u003e \u003e\n\u003e\u003e \u003e\u003e not fixing this issue seems to render the whole exercise somewhat\n\u003e\u003e useless\n\u003e\u003e \u003e\n\u003e\u003e \u003e Depends on if one considers watching the mempool a fix. But even with\n\u003e\u003e that a\n\u003e\u003e \u003e base version of anchors still resolves a number of issues including:\n\u003e\u003e \u003e eliminating the commitment fee guessing game, allowing users to pay\n\u003e\u003e less on\n\u003e\u003e \u003e force close, being able to coalesce 2nd level HTLC transactions with the\n\u003e\u003e \u003e same CLTV expiry, and actually being able to reliably enforce multi-hop\n\u003e\u003e HTLC\n\u003e\u003e \u003e resolution.\n\u003e\u003e \u003e\n\u003e\u003e \u003e\u003e Instead of making the HTLC output spending more free-form with\n\u003e\u003e \u003e\u003e SIGHASH_ANYONECAN_PAY|SIGHASH_SINGLE, we clearly need to go the other\n\u003e\u003e \u003e\u003e direction - all HTLC output spends need to be pre-signed.\n\u003e\u003e \u003e\n\u003e\u003e \u003e I'm not sure this is actually immediately workable (need to think about\n\u003e\u003e it\n\u003e\u003e \u003e more). To see why, remember that the commit_sig message includes HTLC\n\u003e\u003e \u003e signatures for the _remote_ party's commitment transaction, so they can\n\u003e\u003e \u003e spend the HTLCs if they broadcast their version of the commitment (force\n\u003e\u003e \u003e close). If we don't somehow also _gain_ signatures (our new HTLC\n\u003e\u003e signatures)\n\u003e\u003e \u003e allowing us to spend HTLCs on _their_ version of the commitment, then if\n\u003e\u003e \u003e they broadcast that commitment (without revoking), then we're unable to\n\u003e\u003e \u003e redeem any of those HTLCs at all, possibly losing money.\n\u003e\u003e\n\u003e\u003e Hmm, maybe the proposal wasn't clear. The idea isn't to add signatures to\n\u003e\u003e braodcasted transactions, but instead to CPFP\n\u003e\u003e a maybe-broadcasted transaction by sending a transaction which spends it\n\u003e\u003e and seeing if it is accepted. You only need to\n\u003e\u003e know the transaction's exact format (ie txid, which we do, since we sent\n\u003e\u003e a signature for it long ago) to do this, you\n\u003e\u003e don't have to actually *have* the fully-signed transaction (and you\n\u003e\u003e don't).\n\u003e\u003e\n\u003e\u003e \u003e In an attempt to counteract this, we might say ok, the revoke message\n\u003e\u003e also\n\u003e\u003e \u003e now includes HTLC signatures for their new commitment allowing us to\n\u003e\u003e spend\n\u003e\u003e \u003e our HTLCs. This resolves things in a weaker security model, but doesn't\n\u003e\u003e \u003e address the issue generally, as after they receive the commit_sig, they\n\u003e\u003e can\n\u003e\u003e \u003e broadcast immediately, again leaving us without a way to redeem our\n\u003e\u003e HTLCs.\n\u003e\u003e \u003e\n\u003e\u003e \u003e I'd need to think about it more, but it seems that following this path\n\u003e\u003e would\n\u003e\u003e \u003e require an overhaul in the channel state machine to make presenting a\n\u003e\u003e new\n\u003e\u003e \u003e commitment actually take at least _two phases_ (at least a full round\n\u003e\u003e trip).\n\u003e\u003e \u003e The first phase would tender the commitment, but render them unable to\n\u003e\u003e \u003e broadcast it. The second phase would then \u003cinsert something something\n\u003e\u003e \u003e scriptless scripts here\u003e enter a new sub-protocol which upon conclusion,\n\u003e\u003e \u003e gives the commitment proposer valid HTLC signatures, and gives the\n\u003e\u003e responder\n\u003e\u003e \u003e what they need to be able to broadcast their commitment and claim their\n\u003e\u003e \u003e HTCLs in an atomic manner.\n\u003e\u003e \u003e\n\u003e\u003e \u003e -- Laolu\n\u003e\u003e \u003e\n\u003e\u003e \u003e [1]: https://github.com/lightningnetwork/lnd/pull/3821\n\u003e\u003e\n\u003e\n-------------- next part --------------\nAn HTML attachment was scrubbed...\nURL: \u003chttp://lists.linuxfoundation.org/pipermail/lightning-dev/attachments/20200422/ec41b541/attachment-0001.html\u003e"}