Bitcoin Optech #152: Lightning Node Payments And More

This week’s publication describes a proposal to permit LN nodes to receive payments with out conserving their non-public keys online the entire time. Additionally integrated are our popular sections with the summary of a Bitcoin Core PR Overview Membership meeting, announcements of fresh system releases and initiating candidates, and descriptions of distinguished changes to popular Bitcoin infrastructure system.

News

• Receiving LN payments with a largely offline non-public key: in 2019, developer ZmnSCPxj proposed an different formulation to encapsulate pending LN payments (HTLCs) that might perchance perchance maybe lower the quantity of community bandwidth and latency required to just gain a price. More no longer too prolonged within the past, Lloyd Fournier suggested the root might perchance perchance maybe additionally be old to permit a node to just gain a pair of incoming LN payments with out conserving its non-public keys online. The postulate does hang some downsides:
  • The node would still need its non-public keys to ship a penalty transaction if valuable.
  • The more payments the node bought with out the usage of its non-public key, the more onchain prices would should be paid if the channel used to be unilaterally closed.
  • The receiving node would lose privacy—its instantaneous check might perchance perchance maybe be ready to search out out that it used to be the last receiver of a price and no longer correct a routing hop. Nevertheless, for some finish-user nodes that don’t route payments, this might occasionally also already be evident.
• Within these limitations, the root appears workable and adaptations of it bought discussion on the mailing record this week, with ZmnSCPxj making ready a transparent presentation. Fournier later suggested enhancements to the root.

  Enforcing the root would require several significant LN protocol changes, so it appears unlikely to be one thing users will hang access to in both the short or medium term. Nevertheless, any individual looking out to minimize the need for LN receiving nodes to withhold their keys online is impressed to analyze the root.

Bitcoin Core PR Overview Membership

On this month-to-month half, we summarize a latest Bitcoin Core PR Overview Membership meeting, highlighting among the vital questions and answers. Click on a check below to gaze a summary of the retort from the meeting.

Prune g_chainman usage in auxiliary modules is a refactoring PR (#21767) by Carl Dong that is section of a project to de-globalize g_chainman as a significant step in direction of modularizing the consensus engine. This would decouple scheme and enable more focused checking out. A longer-term aim is to utterly separate the consensus engine from non-consensus code.

The review club discussion began with the following overall questions earlier than diving deeper into the code changes:

• This PR is a refactoring and might perchance perchance maybe no longer swap any real looking behaviour. What are many ways we can verify that?

  Reviewing the code in moderation, running the tests, at the side of test coverage, inserting asserts or custom logging, building with –enable-debug, running bitcoind with the changes, and stepping by the code with debuggers fancy GDB or LLDB.

• This PR is section of a higher project to modularize and separate the Bitcoin Core consensus engine. What are some advantages of doing that?

  This would perchance construct it more straightforward to reason about, withhold, configure and test the code. It can perchance maybe snarl a minimal API for security and maintainability, with configuration alternate choices to pass non-global info. Shall we create scheme with variable parameters, offering more possess watch over over checking out these objects with diversified configurations.

• What’s the ChainstateManager to blame for?

  The ChainstateManager class provides an interface for creating and interacting with one or two chainstates: initial block download (IBD) and an no longer significant snapshot.

• What does CChainState attain?

  The CChainState class stores the most modern ideally suited chain and provides an API to update our native info of its utter.

• What’s the CChain class?

  The CChain class is an in-memory listed chain of blocks. It contains a vector of block index pointers.

• What’s the BlockManager to blame for?

  The BlockManager class maintains a tree of blocks kept in m_block_index that is consulted to stumble on the most-work chain tip.

• What’s cs_main?

  cs_main is a mutex that protects validation-specific info (moreover, for now, many diversified issues). The title formulation extreme half significant, as it safe info in significant.cpp, and the code that is now in validation.cpp and net_processing.cpp old to be in a single file called significant.cpp).

• Conceptually, when we discuss over with the “validation” section of the codebase, what does that embrace?

  Validation stores and maintains our ideally suited scrutinize of the block chain and associated UTXO build. It additionally entails an interface to put up unconfirmed transactions to the mempool.

Releases and initiating candidates

Fresh releases and initiating candidates for popular Bitcoin infrastructure projects. Please establish in mind upgrading to fresh releases or helping to test initiating candidates.

• LND 0.13.0-beta.rc5 is a initiating candidate that adds improve for the usage of a pruned Bitcoin plump node, permits receiving and sending payments the usage of Atomic MultiPath (AMP), and increases its PSBT capabilities, amongst diversified enhancements and malicious program fixes.

Necessary code and documentation changes

Necessary changes this week in Bitcoin Core, C-Lightning, Eclair, LND, Rust-Lightning, libsecp256k1, Hardware Pockets Interface (HWI), Rust Bitcoin, BTCPay Server, Bitcoin Enchancment Proposals (BIPs), and Lightning BOLTs.

• Bitcoin Core #22051 adds improve for importing descriptors for taproot outputs into the Bitcoin Core wallet. This PR permits wallet users to receive funds to taproot outputs and is the prerequisite for an open PR that implements plump improve for users to receive to and exercise from taproot outputs.
• Bitcoin Core #22050 drops improve for version 2 Tor onion products and companies (hidden products and companies). Version 2 products and companies are already deprecated and the Tor project has introduced that they’ll change into inaccessible in September. Bitcoin Core already supports version 3 onion products and companies (gaze E-newsletter #132).
• Bitcoin Core #22095 adds a test to envision how Bitcoin Core derives BIP32 non-public keys. Although Bitcoin Core has continuously derived these keys precisely, it used to be no longer too prolonged within the past came staunch by that some diversified wallets had been incorrectly deriving moderately more than 1 out of 128 keys by failing to pad extended non-public keys (xprivs) that had been lower than 32 bytes prolonged. This doesn’t straight finish in a lack of funds or a discount in security, nevertheless it does construct complications for users who construct an HD wallet seed in a single wallet and import it into one other wallet or who construct multisignature wallets. The test vector applied in this PR is additionally being added to BIP32 to support future wallet authors steer sure of this self-discipline.
• C-Lightning #4532 adds experimental improve for upgrading a channel—rebuilding the most modern commitment transaction so that it might maybe perchance incorporate fresh features or structural changes, e.g. converting to the usage of taproot. The protocol begins with a demand for quiescence, an settlement that neither occasion will ship any fresh utter updates unless the quiescence duration is done. For the length of this period, the nodes negotiate the changes they hang to construct and put into effect them. At last, the channel is restored to plump operation. C-Lightning for the time being implements this all the arrangement by connection reestablishment when the channel has already been in a duration of forced utter of being inactive. Varied proposals for channel upgrades had been mentioned in E-newsletter #108 and the creator of this PR wants the feature in section to work on the “simplified HTLC negotiation” described in E-newsletter #109. This specific PR permits upgrading old channels to enhance option_static_remotekey, which C-Lightning first added improve for in 2019, gaze E-newsletter #64.
• LND #5336 adds the flexibility for users to reuse AMP invoices non-interactively by specifying a fresh price secret. The default invoice expiry for AMP invoices created by LND is additionally bumped to 30 days in whisper to facilitate the aforementioned reuse mechanism.
• BTCPay Server #2474 adds the flexibility to test webhooks by sending spurious events that hang all their long-established fields nevertheless dummy info. This mirrors checking out features readily accessible on centrally hosted Bitcoin price processors corresponding to Stripe and Coinbase Commerce.