Rust-Lightning is a Bitcoin Lightning library written in Rust

lightning, does not handle networking, persistence, or any other I/O

Rust-Lightning

. The main crate, lightning, does not handle networking, persistence, or any other I/O. Thus, it is runtime-agnostic, but users must implement basic networking logic, chain interactions, and disk storage. More information is available in the About section.

The lightning-net-tokio crate implements Lightning networking using the Tokio async runtime.

The lightning-persister crate implements persistence for channel data that is crucial to avoiding loss of channel funds. Sample modules for persistence of other Rust-Lightning data is coming soon.

Status

The project implements all of the BOLT specifications in the 1.0 spec. The implementation has pretty good test coverage that is expected to continue to improve. It is also anticipated that as developers begin using the API, the lessons from that will result in changes to the API, so any developer using this API at this stage should be prepared to embrace that. The current state is sufficient for a developer or project to experiment with it. Recent increased contribution rate to the project is expected to lead to a high quality, stable, production-worthy implementation in 2021.

Communications for Rust-Lightning and Lightning Development Kit happens through LDK slack.

About

LDK/Rust-Lightning is a generic library which allows you to build a lightning node without needing to worry about getting all of the lightning state machine, routing, and on-chain punishment code (and other chain interactions) exactly correct. Note that Rust-Lightning isn't, in itself, a node. There are various working/in progress demos which could be used as a node today, but if you "just" want a generic lightning node, you're almost certainly better off with c-lightning/lnd - if, on the other hand, you want to integrate lightning with custom features such as your own chain sync, your own key management, your own data storage/backup logic, etc., LDK is likely your only option. Some Rust-Lightning utilities such as those in chan_utils are also suitable for use in non-LN Bitcoin applications such as DLCs and bulletin boards.

We are currently working on a demo node which fetches blockchain data and on-chain funds via Bitcoin Core RPC/REST. The individual pieces of that demo are/will be composable, so you can pick the off-the-shelf parts you want and replace the rest.

In general, Rust-Lightning does not provide (but LDK has implementations of):

  • on-disk storage - you can store the channel state any way you want - whether Google Drive/iCloud, a local disk, any key-value store/database/a remote server, or any combination of them - we provide a clean API that provides objects which can be serialized into simple binary blobs, and stored in any way you wish.
  • blockchain data - we provide a simple block_connected/block_disconnected API which you provide block headers and transaction information to. We also provide an API for getting information about transactions we wish to be informed of, which is compatible with Electrum server requests/neutrino filtering/etc.
  • UTXO management - RL/LDK owns on-chain funds as long as they are claimable as a part of a lightning output which can be contested - once a channel is closed and all on-chain outputs are spendable only by the user, we provide users notifications that a UTXO is "theirs" again and it is up to them to spend it as they wish. Additionally, channel funding is accomplished with a generic API which notifies users of the output which needs to appear on-chain, which they can then create a transaction for. Once a transaction is created, we handle the rest. This is a large part of our API's goals - making it easier to integrate lightning into existing on-chain wallets which have their own on-chain logic - without needing to move funds in and out of a separate lightning wallet with on-chain transactions and a separate private key system.
  • networking - to enable a user to run a full lightning node on an embedded machine, we don't specify exactly how to connect to another node at all! We provide a default implementation which uses TCP sockets, but, e.g., if you wanted to run your full lightning node on a hardware wallet, you could, by piping the lightning network messages over USB/serial and then sending them in a TCP socket from another machine.
  • private keys - again we have "default implementations", but users can chose to provide private keys to RL/LDK in any way they wish following a simple API. We even support a generic API for signing transactions, allowing users to run RL/LDK without any private keys in memory/putting private keys only on hardware wallets.

LDK's customizability was presented about at Advancing Bitcoin in February 2020: https://vimeo.com/showcase/7131712/video/418412286

Design Goal

The goal is to provide a full-featured but also incredibly flexible lightning implementation, allowing the user to decide how they wish to use it. With that in mind, everything should be exposed via simple, composable APIs. More information about Rust-Lightning's flexibility is provided in the About section above.

For security reasons, do not add new dependencies. Really do not add new non-optional/non-test/non-library dependencies. Really really do not add dependencies with dependencies. Do convince Andrew to cut down dependency usage in rust-bitcoin.

Rust-Lightning vs. LDK (Lightning Development Kit)

Rust-Lightning refers to the core lightning crate within this repo, whereas LDK encompasses Rust-Lightning and all of its sample modules and crates (e.g. the lightning-persister crate), language bindings, sample node implementation(s), and other tools built around using Rust-Lightning for lightning integration or building a lightning node.

Tagline

"Rust-Lightning, not Rusty's Lightning!"

Contributing

Contributors are warmly welcome, see CONTRIBUTING.md.

Project Architecture

For a Rust-Lightning high-level API introduction, see ARCH.md.

License is either Apache-2.0 or MIT, at the option of the user (ie dual-license Apache-2.0 and MIT).

Issues

Collection of the latest Issues

TheBlueMatt

TheBlueMatt

good first issue
0

If an HTLC times out and we close a channel from the ChannelMonitor, we currently mark the ClosureReason as CommitmentTxConfirmed, even though the commitment tx has not yet confirmed, not to mention it was us who closed the channel! From a cursory glance, the fix is trivial - just interpret (and maybe rename) MonitorEvent::CommitmentTxConfirmed to mean "we force-closed" and then we can think about adding more detail, but I think its literally always due to HTLC timeouts.

TheBlueMatt

TheBlueMatt

good first issue
0

Probably should have thought of this before #1418, but oh well - ideally we'd be able to specify the payment retry count on a per-payment basis rather than globally in the InvoicePayer. This would likely imply refusing to retry things that weren't started using the InvoicePayer/after restart, but that may be okay? Would have to document that or make it configurable.

TheBlueMatt

TheBlueMatt

3

We currently use a BTreeMap for storing network graph data, as we need to be able to access it starting from a given key and continue iteration. This makes a sorted KV store like a BTree the obvious choice, but we don't actually need the keys to be sorted, we just need them to not change order as we iterate the graph.

Technically we also need to be able to check if a new key is below some previous iteration that is in-progress.

Supposedly the indexmap crate may offer something like what we need, but ultimately any hashtable that doesn't ever re-hash should suffice, and while creating our own probably won't be as efficient as hashbrown, it'd probably still be faster than a btree, and use way less memory compared to the substantial heap fragmentation that a btree creates.

TheBlueMatt

TheBlueMatt

4

Sensei ran into an issue where ldk-net-tokio (or some other async context) calls into ChannelManager, which handles some message and then goes to update the ChannelMonitor. Because Sensei is using the new KVStorePersister interface to simplify its persistence it ends up in a sync context in an tokio worker thread thread and then has to block on a future to talk to a (possibly remote) database. It then returns control to tokio by blocking on the future, but at the time we're still holding the channel and ChainMonitor locks, causing most other future execution to block.

While tokio should probably be willing to poll the IO driver in the block_on call from Sensei, even if it did it could still decide to run a different future which ends up blocking on the mutex already held and we're back to a deadlock.

Really if you're storing a monitor update via some async future you should use the TemporaryFailure return to do the monitor update asynchronously instead of blocking on IO with locks held, but we don't have good utilities for doing so today.

We need to (a) create a AsyncKVStorePersister (it has to either be runtime-specific because it needs the ability to spawn background tasks, as an alternative we could make the user provide a spawn method which we can call), and (b) adapt the background processor to using it for async manager/graph/score persistence.

TheBlueMatt

TheBlueMatt

good first issue
1

We have a few features defined in Cargo.toml with comments describing their use, but we don't expose them in the crate-level lib.rs documentation, making docs pretty much invisible. We should copy the text over, expand it, and clarify it so that you can see the features on docs.rs.

TheBlueMatt

TheBlueMatt

0

I thought we had an issue for this, but I guess not. We should really use the ChannelMonitorUpdates in the FilesystemPersister instead of always persisting the full ChannelMonitor.

Because we're moving towards the logic not being encapsulated in the FilesystemPersister in #1417 we'll have to do a bunch of work there - probably we'll want to move the read_channelMonitors thing into lightning using a get method added to the KVStorePersister trait, with the ability to roll forward the monitors using updates and then remove the updates after persisting the new monitor.

johncantrell97

johncantrell97

0

There's a TODO in the Persist implementation provided that says:

// TODO: We really need a way for the persister to inform the user that its time to crash/shut // down once these start returning failure. // A PermanentFailure implies we need to shut down since we're force-closing channels without // even broadcasting!

@TheBlueMatt thinks using Event::ChannelClosed here might make sense.

arik-so

arik-so

0

Currently, the channel manager assumes that as soon as a channel has six confirmations, its funding is locked. However, in the rare case that a channel gets funded from a coinbase transaction, those funds aren't spendable until the output matures 100 blocks later. We should add a flag to keep track of this edge case so we don't send the funding locked message prematurely.

TheBlueMatt

TheBlueMatt

good first issue
2

When we try to forward a payment but it fails, users may want to know about this, as it may imply something about rebalancing channels or at least may be something users want to collect stats about. As discussed in #1391 I don't think we want to expose payment hash here, but source + sink channels+nodes and whether we forwarded at all or failed without being able to forward would be useful info for users.

TheBlueMatt

TheBlueMatt

2

While you can do it already, passing information about a successful probe into ProbabilisticScorer is somewhat confusing today - you have to do it by calling the payment-failed method with a bogus scid to "blame". We should just wrap that call in a util function and expose it officially.

cc @jkczyz

Versions

Find the latest versions by id

v0.0.106 - Apr 03, 2022

API Updates

  • Minimum supported rust version (MSRV) is now 1.41.1 (#1310).
  • Lightning feature option_scid_alias is now supported and may be negotiated when opening a channel with a peer. It can be configured via ChannelHandshakeConfig::negotiate_scid_privacy and is off by default but will be on by default in the future (#1351).
  • OpenChannelRequest now has a channel_type field indicating the features the channel will operate with and should be used to filter channels with undesirable features (#1351). See the Serialization Compatibility section.
  • ChannelManager supports sending and receiving short channel id aliases in the funding_locked message. These are used when forwarding payments and constructing invoice route hints for improved privacy. ChannelDetails has a inbound_scid_alias field and a get_inbound_payment_scid method to support the latter (#1311).
  • DefaultRouter and find_route take an additional random seed to improve privacy by adding a random CLTV expiry offset to each path's final hop. This helps obscure the intended recipient from adversarial intermediate hops (#1286). The seed is also used to randomize candidate paths during route selection (#1359).
  • The lightning-block-sync crate's init::synchronize_listeners method interface has been relaxed to support multithreaded environments (#1349).
  • ChannelManager::create_inbound_payment_for_hash's documentation has been corrected to remove the one-year restriction on invoice_expiry_delta_secs, which is only applicable to the deprecated create_inbound_payment_legacy and create_inbound_payment_for_hash_legacy methods (#1341).
  • Features mutator methods now take self by reference instead of by value (#1331).
  • The CLTV of the last hop in a path is now included when comparing against RouteParameters::max_total_cltv_expiry_delta (#1358).
  • Invoice creation functions in lightning-invoice crate's utils module include versions that accept a description hash instead of only a description (#1361).
  • RoutingMessageHandler::sync_routing_table has been renamed peer_connected (#1368).
  • MessageSendEvent::SendGossipTimestampFilter has been added to indicate that a gossip_timestamp_filter should be sent (#1368).
  • PeerManager takes an optional NetAddress in new_outbound_connection and new_inbound_connection, which is used to report back the remote address to the connecting peer in the init message (#1326).
  • ChannelManager::accept_inbound_channel now takes a user_channel_id, which is used in a similar manner as in outbound channels. (#1381).
  • BackgroundProcessor now persists NetworkGraph on a timer and upon shutdown as part of a new Persister trait, which also includes ChannelManager persistence (#1376).
  • ProbabilisticScoringParameters now has a base_penalty_msat option, which default to 500 msats. It is applied at each hop to help avoid longer paths (#1375).
  • ProbabilisticScoringParameters::liquidity_penalty_multiplier_msat's default value is now 40,000 msats instead of 10,000 msats (#1375).
  • The lightning crate has a grind_signatures feature used to produce signatures with low r-values for more predictable transaction weight. This feature is on by default (#1388).
  • ProbabilisticScoringParameters now has a amount_penalty_multiplier_msat option, which is used to further penalize large amounts (#1399).
  • PhantomRouteHints, FixedPenaltyScorer, and ScoringParameters now implement Clone (#1346).

Bug Fixes

  • Fixed a compilation error in ProbabilisticScorer under --feature=no-std (#1347).
  • Invoice creation functions in lightning-invoice crate's utils module filter invoice hints in order to limit the invoice size (#1325).
  • Fixed a bug where a funding_locked message was delayed by a block if the funding transaction was confirmed while offline, depending on the ordering of Confirm::transactions_confirmed calls when brought back online (#1363).
  • Fixed a bug in NetGraphMsgHandler where it didn't continue to receive gossip messages from peers after initial connection (#1368, #1382).
  • ChannelManager::timer_tick_occurred will now timeout a received multi-path payment (MPP) after three ticks if not received in full instead of waiting until near the HTLC timeout block(#1353).
  • Fixed an issue with find_route causing it to be overly aggressive in using MPP over channels to the same first hop (#1370).
  • Reduced time spent processing channel_update messages by checking signatures after checking if no newer messages have already been processed (#1380).
  • Fixed a few issues in find_route which caused preferring paths with a higher cost (#1398).
  • Fixed an issue in ProbabilisticScorer where a channel with not enough liquidity could still be used when retrying a failed payment if it was on a path with an overall lower cost (#1399).

Serialization Compatibility

  • Channels open with option_scid_alias negotiated will be incompatible with prior releases (#1351). This may occur in the following cases:
    • Outbound channels when ChannelHandshakeConfig::negotiate_scid_privacy is enabled.
    • Inbound channels when automatically accepted from an OpenChannel message with a channel_type that has ChannelTypeFeatures::supports_scid_privacy return true. See UserConfig::accept_inbound_channels.
    • Inbound channels when manually accepted from an OpenChannelRequest with a channel_type that has ChannelTypeFeatures::supports_scid_privacy return true. See UserConfig::manually_accept_inbound_channels.

In total, this release features 43 files changed, 4052 insertions, 1274 deletions in 75 commits from 11 authors, in alphabetical order:

  • Devrandom
  • Duncan Dean
  • Elias Rohrer
  • Jeffrey Czyz
  • Jurvis Tan
  • Luiz Parreira
  • Matt Corallo
  • Omar Shamardy
  • Viktor Tigerström
  • dependabot[bot]
  • psycho-pirate

v0.0.105 - Mar 01, 2022

API Updates

  • Phantom node payments are now supported, allowing receipt of a payment on any one of multiple nodes without any coordination across the nodes being required. See the new PhantomKeysManager's docs for more, as well as requirements on KeysInterface::get_inbound_payment_key_material and lightning_invoice::utils::create_phantom_invoice (#1199).
  • In order to support phantom node payments, several KeysInterface methods now accept a Recipient parameter to select between the local node_id and a phantom-specific one.
  • ProbabilisticScorer, a Score based on learning the current balances of channels in the network, was added. It attempts to better capture payment success probability than the existing Scorer, though may underperform on nodes with low payment volume. We welcome feedback on performance (#1227).
  • Score::channel_penalty_msat now always takes the channel value, instead of an Option (#1227).
  • UserConfig::manually_accept_inbound_channels was added which, when set, generates a new Event::OpenChannelRequest, which allows manual acceptance or rejection of incoming channels on a per-channel basis (#1281).
  • Payee has been renamed to PaymentParameters (#1271).
  • PaymentParameters now has a max_total_cltv_expiry_delta field. This defaults to 1008 and limits the maximum amount of time an HTLC can be pending before it will either fail or be claimed (#1234).
  • The lightning-invoice crate now supports no-std environments. This required numerous API changes around timestamp handling and std+no-std versions of several methods that previously assumed knowledge of the time (#1223, #1230).
  • lightning-invoice now supports parsing invoices with expiry times of more than one year. This required changing the semantics of ExpiryTime (#1273).
  • The CounterpartyCommitmentSecrets is now public, allowing external uses of the BOLT 3 secret storage scheme (#1299).
  • Several Sign methods now receive HTLC preimages as proof of state transition, see new documentation for more (#1251).
  • KeysInterface::sign_invoice now provides the HRP and other invoice data separately to make it simpler for external signers to parse (#1272).
  • Sign::sign_channel_announcement now returns both the node's signature and the per-channel signature. InMemorySigner now requires the node's secret key in order to implement this (#1179).
  • ChannelManager deserialization will now fail if the KeysInterface used has a different node_id than the ChannelManager expects (#1250).
  • A new ErrorAction variant was added to send warning messages (#1013).
  • Several references to chain::Listen objects in lightning-block-sync no longer require a mutable reference (#1304).

Bug Fixes

  • Fixed a regression introduced in 0.0.104 where ChannelManager's internal locks could have an order violation leading to a deadlock (#1238).
  • Fixed cases where slow code (including user I/O) could cause us to disconnect peers with ping timeouts in BackgroundProcessor (#1269).
  • Now persist the ChannelManager prior to BackgroundProcessor stopping, preventing race conditions where channels are closed on startup even with a clean shutdown. This requires that users stop network processing and disconnect peers prior to BackgroundProcessor shutdown (#1253).
  • Fields in ChannelHandshakeLimits provided via the override_config to create_channel are now applied instead of the default config (#1292).
  • Fixed the generation of documentation on docs.rs to include API surfaces which are hidden behind feature flags (#1303).
  • Added the channel_type field to accept_channel messages we send, which may avoid some future compatibility issues with other nodes (#1314).
  • Fixed a bug where, if a previous LDK run using lightning-persister crashed while persisting updated data, we may have failed to initialize (#1332).
  • Fixed a rare bug where having both pending inbound and outbound HTLCs on a just-opened inbound channel could cause ChannelDetails::balance_msat to underflow and be reported as large, or cause panics in debug mode (#1268).
  • Moved more instances of verbose gossip logging from the Trace level to the Gossip level (#1220).
  • Delayed announcement_signatures until the channel has six confirmations, slightly improving propagation of channel announcements (#1179).
  • Several fixes in script and transaction weight calculations when anchor outputs are enabled (#1229).

Serialization Compatibility

  • Using ChannelManager data written by versions prior to 0.0.105 will result in preimages for HTLCs that were pending at startup to be missing in calls to KeysInterface methods (#1251).
  • Any phantom invoice payments received on a node that is not upgraded to 0.0.105 will fail with an "unknown channel" error. Further, downgrading to 0.0.104 or before and then upgrading again will invalidate existing phantom SCIDs which may be included in invoices (#1199).

Security

0.0.105 fixes two denial-of-service vulnerabilities which may be reachable from untrusted input in certain application designs.

  • Route calculation spuriously panics when a routing decision is made for a path where the second-to-last hop is a private channel, included due to a multi-hop route hint in an invoice.
  • ChannelMonitor::get_claimable_balances spuriously panics in some scenarios when the LDK application's local commitment transaction is confirmed while HTLCs are still pending resolution.

In total, this release features 109 files changed, 7270 insertions, 2131 deletions in 108 commits from 15 authors, in alphabetical order:

  • Conor Okus
  • Devrandom
  • Elias Rohrer
  • Jeffrey Czyz
  • Jurvis Tan
  • Ken Sedgwick
  • Matt Corallo
  • Naveen
  • Tibo-lg
  • Valentine Wallace
  • Viktor Tigerström
  • dependabot[bot]
  • hackerrdave
  • naveen
  • vss96

v0.0.104 - Dec 17, 2021

API Updates

  • A PaymentFailed event is now provided to indicate a payment has failed fully. This event is generated either after ChannelManager::abandon_payment is called for a given payment, or the payment times out, and there are no further pending HTLCs for the payment. This event should be used to detect payment failure instead of PaymentPathFailed::all_paths_failed, unless no payment retries occur via ChannelManager::retry_payment (#1202).
  • Payment secrets are now generated deterministically using material from the new KeysInterface::get_inbound_payment_key_material (#1177).
  • A PaymentPathSuccessful event has been added to ease passing success info to a scorer, along with a Score::payment_path_successful method to accept such info (#1178, #1197).
  • Score::channel_penalty_msat has additional arguments describing the channel's capacity and the HTLC amount being sent over the channel (#1166).
  • A new log level Gossip has been added, which is used for verbose information generated during network graph sync. Enabling the max_level_trace feature or ignoring Gossip log entries reduces log growth during initial start up from many GiB to several MiB (#1145).
  • The allow_wallclock_use feature has been removed in favor of only using the std and no-std features (#1212).
  • NetworkGraph can now remove channels that we haven't heard updates for in two weeks with NetworkGraph::remove_stale_channels{,with_time}. The first is called automatically if a NetGraphMsgHandler is passed to BackgroundProcessor::start (#1212).
  • InvoicePayer::pay_pubkey was added to enable sending "keysend" payments to supported recipients, using the InvoicePayer to handle retires (#1160).
  • user_payment_id has been removed from PaymentPurpose, and ChannelManager::create_inbound_payment{,_for_hash} (#1180).
  • Updated documentation for several ChannelManager functions to remove stale references to panics which no longer occur (#1201).
  • The Score and LockableScore objects have moved into the routing::scoring module instead of being in the routing module (#1166).
  • The Time parameter to ScorerWithTime is no longer longer exposed, instead being fixed based on the std/no-std feature (#1184).
  • ChannelDetails::balance_msat was added to fetch a channel's balance without subtracting the reserve values, lining up with on-chain claim amounts less on-chain fees (#1203).
  • An explicit UserConfig::accept_inbound_channels flag is now provided, removing the need to set min_funding_satoshis to > 21 million BTC (#1173).
  • Inbound channels that fail to see the funding transaction confirm within 2016 blocks are automatically force-closed with ClosureReason::FundingTimedOut (#1083).
  • We now accept a channel_reserve value of 0 from counterparties, as it is insecure for our counterparty but not us (#1163).
  • NetAddress::OnionV2 parsing was removed as version 2 onion services are no longer supported in modern Tor (#1204).
  • Generation and signing of anchor outputs is now supported in the KeysInterface, though no support for them exists in the channel itself (#1176)

Bug Fixes

  • Fixed a race condition in InvoicePayer where paths may be retried after the retry count has been exceeded. In this case the Event::PaymentPathFailed::all_paths_failed field is not a reliable payment failure indicator. There was no acceptable alternative indicator, Event::PaymentFailed as been added to provide one (#1202).
  • Reduced the blocks-before-timeout we expect of outgoing HTLCs before refusing to forward. This check was overly strict and resulted in refusing to forward som HTLCs to a next hop that had a lower security threshold than us (#1119).
  • LDK no longer attempt to update the channel fee for outbound channels when we cannot afford the new fee. This could have caused force-closure by our channel counterparty (#1054).
  • Fixed several bugs which may have prevented the reliable broadcast of our own channel announcements and updates (#1169).
  • Fixed a rare bug which may have resulted in spurious route finding failures when using last-hop hints and MPP with large value payments (#1168).
  • KeysManager::spend_spendable_outputs no longer adds a change output that is below the dust threshold for non-standard change scripts (#1131).
  • Fixed a minor memory leak when attempting to send a payment that fails due to an error when updating the ChannelMonitor (#1143).
  • Fixed a bug where a FeeEstimator that returns values rounded to the next sat/vbyte may result in force-closures (#1208).
  • Handle MPP timeout HTLC error codes, instead of considering the recipient to have sent an invalid error, removing them from the network graph (#1148)

Serialization Compatibility

  • All above new events/fields are ignored by prior clients. All above new events/fields are not present when reading objects serialized by prior versions of the library.
  • Payment secrets are now generated deterministically. This reduces the memory footprint for inbound payments, however, newly-generated inbound payments using ChannelManager::create_inbound_payment{,_for_hash} will not be receivable using versions prior to 0.0.104. ChannelManager::create_inbound_payment{,_for_hash}_legacy are provided for backwards compatibility (#1177).
  • PaymentPurpose::InvoicePayment::user_payment_id will be 0 when reading objects written with 0.0.104 when read by 0.0.103 and previous (#1180).

In total, this release features 51 files changed, 5356 insertions, 2238 deletions in 107 commits from 9 authors, in alphabetical order:

  • Antoine Riard
  • Conor Okus
  • Devrandom
  • Duncan Dean
  • Elias Rohrer
  • Jeffrey Czyz
  • Ken Sedgwick
  • Matt Corallo
  • Valentine Wallace

v0.0.103 - Nov 03, 2021

API Updates

  • This release is almost entirely focused on a new API in the lightning-invoice crate - the InvoicePayer. InvoicePayer is a struct which takes a reference to a ChannelManager and a Router and retries payments as paths fail. It limits retries to a configurable number, but is not serialized to disk and may retry additional times across a serialization/load. In order to learn about failed payments, it must receive Events directly from the ChannelManager, wrapping a user-provided EventHandler which it provides all unhandled events to (#1059).
  • get_route has been renamed find_route (#1059) and now takes a RouteParameters struct in replacement of a number of its long list of arguments (#1134). The Payee in the RouteParameters is stored in the Route object returned and provided in the RouteParameters contained in Event::PaymentPathFailed (#1059).
  • ChannelMonitors must now be persisted after calls that provide new block data, prior to MonitorEvents being passed back to ChannelManager for processing. If you are using a ChainMonitor this is handled for you. The Persist API has been updated to Optionally take the ChannelMonitorUpdate as persistence events that result from chain data no longer have a corresponding update (#1108).
  • routing::Score now has a payment_path_failed method which it can use to learn which channels often fail payments. It is automatically called by InvoicePayer for failed payment paths (#1144).
  • The default Scorer implementation is now a type alias to a type generic across different clocks and supports serialization to persist scoring data across restarts (#1146).
  • Event::PaymentSent now includes the full fee which was spent across all payment paths which were fulfilled or pending when the payment was fulfilled (#1142).
  • Event::PaymentSent and Event::PaymentPathFailed now include the PaymentId which matches the PaymentId returned from ChannelManager::send_payment or InvoicePayer::pay_invoice (#1059).
  • NetGraphMsgHandler now takes a Deref to the NetworkGraph, allowing for shared references to the graph data to make serialization and references to the graph data in the InvoicePayer's Router simpler (#1149).
  • routing::Score::channel_penalty_msat has been updated to provide the NodeId of both the source and destination nodes of a channel (#1133).

Bug Fixes

  • Previous versions would often disconnect peers during initial graph sync due to ping timeouts while processing large numbers of gossip messages. We now delay disconnecting peers if we receive messages from them even if it takes a while to receive a pong from them. Further, we avoid sending too many gossip messages between pings to ensure we should always receive pongs in a timely manner (#1137).
  • If a payment was sent, creating an outbound HTLC and sending it to our counterparty (implying the ChannelMonitor was persisted on disk), but the ChannelManager was not persisted prior to shutdown/crash, no Event::PaymentPathFailed event was generated if the HTLC was eventually failed on chain. Events are now consistent irrespective of ChannelManager persistence or non-persistence (#1104).

Serialization Compatibility

  • All above new Events/fields are ignored by prior clients. All above new Events/fields are not present when reading objects serialized by prior versions of the library.
  • Payments for which a Route was generated using a previous version or for which the payment was originally sent by a previous version of the library will not be retried by an InvoicePayer.

This release was singularly focused and some contributions by third parties were delayed. In total, this release features 38 files changed, 4414 insertions, and 969 deletions in 71 commits from 2 authors, in alphabetical order:

  • Jeffrey Czyz
  • Matt Corallo

v0.0.102 - Oct 18, 2021

API Updates

  • get_route now takes a Score as an argument. Score is queried during the route-finding process, returning the absolute amounts which you are willing to pay to avoid routing over a given channel. As a default, a Scorer is provided which returns a constant amount, with a suggested default of 500 msat. This translates to a willingness to pay up to 500 msat in additional fees per hop in order to avoid additional hops (#1124).
  • Event::PaymentPathFailed now contains a short_channel_id field which may be filled in with a channel that can be "blamed" for the payment failure. Payment retries should likely avoid the given channel for some time (#1077).
  • PublicKeys in NetworkGraph have been replaced with a NodeId struct which contains only a simple [u8; 33], substantially improving NetworkGraph deserialization performance (#1107).
  • ChainMonitor's HashMap of ChannelMonitors is now private, exposed via Chainmonitor::get_monitor and ChainMonitor::list_monitors instead (#1112).
  • When an outbound channel is closed prior to the broadcasting of its funding transaction, but after you call ChannelManager::funding_transaction_generated, a new event type, Event::DiscardFunding, is generated, informing you the transaction was not broadcasted and that you can spend the same inputs again elsewhere (#1098).
  • ChannelManager::create_channel now returns the temporary channel ID which may later appear in Event::ChannelClosed or ChannelDetails prior to the channel being funded (#1121).
  • Event::PaymentSent now contains the payment hash as well as the payment preimage (#1062).
  • ReadOnlyNetworkGraph::get_addresses now returns owned NetAddress rather than references. As a side-effect this method is now exposed in foreign language bindings (#1115).
  • The Persist and ChannelMonitorUpdateErr types have moved to the lightning::chain::chainmonitor and lightning::chain modules, respectively (#1112).
  • ChannelManager::send_payment now returns a PaymentId which identifies a payment (whether MPP or not) and can be used to retry the full payment or MPP parts through retry_payment (#1096). Note that doing so is currently not crash safe, and you may find yourself sending twice. It is recommended that you not use the retry_payment API until the next release.

Bug Fixes

  • Due to an earlier fix for the Lightning dust inflation vulnerability tracked in CVE-2021-41591/CVE-2021-41592/CVE-2021-41593 in 0.0.100, we required counterparties to accept a dust limit slightly lower than the dust limit now required by other implementations. This appeared as, at least, latest lnd always refusing to accept channels opened by LDK clients (#1065).
  • If there are multiple channels available to the same counterparty, get_route would only consider the channel listed last as available for sending (#1100).
  • Persist implementations returning ChannelMonitorUpdateErr::TemporaryFailure from watch_channel previously resulted in the ChannelMonitor not being stored at all, resulting in a panic after monitor updating is complete (#1112).
  • If payments are pending awaiting forwarding at startup, an Event::PendingHTLCsForwardable event will always be provided. This ensures user code calls ChannelManager::process_pending_htlc_fowards even if it shut down while awaiting the batching timer during the previous run (#1076).
  • If a call to ChannelManager::send_payment failed due to lack of availability of funds locally, LDK would store the payment as pending forever, with no ability to retry or fail it, leaking memory (#1109).

Serialization Compatibility

  • All above new Events/fields are ignored by prior clients. All above new Events/fields, except for Event::PaymentSent::payment_hash are not present when reading objects serialized by prior versions of the library.

In total, this release features 32 files changed, 2248 insertions, and 1483 deletions in 51 commits from 7 authors, in alphabetical order:

  • 1nF0rmed
  • Duncan Dean
  • Elias Rohrer
  • Galder Zamarreño
  • Jeffrey Czyz
  • Matt Corallo
  • Valentine Wallace

v0.0.101 - Sep 23, 2021

API Updates

  • Custom message types are now supported directly in the PeerManager, allowing you to send and receive messages of any type that is not natively understood by LDK. This requires a new type bound on PeerManager, a CustomMessageHandler. IgnoringMessageHandler provides a simple default for this new bound for ignoring unknown messages (#1031, #1074).
  • Route graph updates as a result of failed payments are no longer provided as MessageSendEvent::PaymentFailureNetworkUpdate but instead included in a new field in the Event::PaymentFailed events. Generally, this means route graph updates are no longer handled as a part of the PeerManager but instead through the new EventHandler implementation for NetGraphMsgHandler. To make this easy, a new parameter to lightning-background-processor::BackgroundProcessor::start is added, which contains an Optional NetGraphmsgHandler. If provided as Some, relevant events will be processed by the NetGraphMsgHandler prior to normal event handling (#1043).
  • NetworkGraph is now, itself, thread-safe. Accordingly, most functions now take &self instead of &mut self and the graph data can be accessed through NetworkGraph.read_only (#1043).
  • The balances available on-chain to claim after a channel has been closed are now exposed via ChannelMonitor::get_claimable_balances and ChainMonitor::get_claimable_balances. The second can be used to get information about all closed channels which still have on-chain balances associated with them. See enum variants of ln::channelmonitor::Balance and method documentation for the above methods for more information on the types of balances exposed (#1034).
  • When one HTLC of a multi-path payment fails, the new field all_paths_failed in Event::PaymentFailed is set to false. This implies that the payment has not failed, but only one part. Payment resolution is only indicated by an Event::PaymentSent event or an Event::PaymentFailed with all_paths_failed set to true, which is also set for the last remaining part of a multi-path payment (#1053).
  • To better capture the context described above, Event::PaymentFailed has been renamed to Event::PaymentPathFailed (#1084).
  • A new event, ChannelClosed, is provided by ChannelManager when a channel is closed, including a reason and error message (if relevant, #997).
  • lightning-invoice now considers invoices with sub-millisatoshi precision to be invalid, and requires millisatoshi values during construction (thus you must call amount_milli_satoshis instead of amount_pico_btc, #1057).
  • The BaseSign interface now includes two new hooks which provide additional information about commitment transaction signatures and revocation secrets provided by our counterparty, allowing additional verification (#1039).
  • The BaseSign interface now includes additional information for cooperative close transactions, making it easier for a signer to verify requests (#1064).
  • Route has two additional helper methods to get fees and amounts (#1063).
  • Txid and Transaction objects can now be deserialized from responses when using the HTTP client in the lightning-block-sync crate (#1037, #1061).

Bug Fixes

  • Fix a panic when reading a lightning invoice with a non-recoverable signature. Further, restrict lightning invoice parsing to require payment secrets and better handle a few edge cases as required by BOLT 11 (#1057).
  • Fix a panic when receiving multiple messages (such as HTLC fulfill messages) after a call to chain::Watch::update_channel returned Err(ChannelMonitorUpdateErr::TemporaryFailure) with no ChannelManager::channel_monitor_updated call in between (#1066).
  • For multi-path payments, Event::PaymentSent is no longer generated multiple times, once for each independent part (#1053).
  • Multi-hop route hints in invoices are now considered in the default router provided via get_route (#1040).
  • The time peers have to respond to pings has been increased when building with debug assertions enabled. This avoids peer disconnections on slow hosts when running in debug mode (#1051).
  • The timeout for the first byte of a response for requests from the lightning-block-sync crate has been increased to 300 seconds to better handle the long hangs in Bitcoin Core when it syncs to disk (#1090).

Serialization Compatibility

  • Due to a bug in 0.0.100, Events written by 0.0.101 which are of a type not understood by 0.0.100 may lead to Err(DecodeError::InvalidValue) or corrupt deserialized objects in 0.100. Such Events will lead to an Err(DecodeError::InvalidValue) in versions prior to 0.0.100. The only such new event written by 0.0.101 is Event::ChannelClosed (#1087).
  • Payments that were initiated in versions prior to 0.0.101 may still generate duplicate PaymentSent Events or may have spurious values for Event::PaymentPathFailed::all_paths_failed (#1053).
  • The return values of ChannelMonitor::get_claimable_balances (and, thus, ChainMonitor::get_claimable_balances) may be spurious for channels where the spend of the funding transaction appeared on chain while running a version prior to 0.0.101. Balance information should only be relied upon for channels that were closed while running 0.0.101+ (#1034).
  • Payments failed while running versions prior to 0.0.101 will never have a Some for the network_update field (#1043).

In total, this release features 67 files changed, 4980 insertions, 1888 deletions in 89 commits from 12 authors, in alphabetical order:

  • Antoine Riard
  • Devrandom
  • Galder Zamarreño
  • Giles Cope
  • Jeffrey Czyz
  • Joseph Goulden
  • Matt Corallo
  • Sergi Delgado Segura
  • Tibo-lg
  • Valentine Wallace
  • abhik-99
  • vss96

v0.0.100 - Aug 17, 2021

API Updates

  • The lightning crate can now be built in no_std mode, making it easy to target embedded hardware for rust users. Note that mutexes are replaced with no-ops for such builds (#1008, #1028).
  • LDK now supports sending and receiving "keysend" payments. This includes modifications to lightning::util::events::Event::PaymentReceived to indicate the type of payment (#967).
  • A new variant, lightning::util::events::Event::PaymentForwarded has been added which indicates a forwarded payment has been successfully claimed and we've received a forwarding fee (#1004).
  • lightning::chain::keysinterface::KeysInterface::get_shutdown_pubkey has been renamed to get_shutdown_scriptpubkey, returns a script, and is now called on channel open only if lightning::util::config::ChannelConfig::commit_upfront_shutdown_pubkey is set (#1019).
  • Closing-signed negotiation is now more configurable, with an explicit lightning::util::config::ChannelConfig::force_close_avoidance_max_fee_satoshis field allowing you to select the maximum amount you are willing to pay to avoid a force-closure. Further, we are now less restrictive on the fee placed on the closing transaction when we are not the party paying it. To control the feerate paid on a channel at close-time, use ChannelManager::close_channel_with_target_feerate instead of close_channel (#1011).
  • lightning_background_processor::BackgroundProcessor now stops the background thread when dropped (#1007). It is marked #[must_use] so that Rust users will receive a compile-time warning when it is immediately dropped after construction (#1029).
  • Total potential funds burn on force-close due to dust outputs is now limited to lightning::util::config::ChannelConfig::max_dust_htlc_exposure_msat per channel (#1009).
  • The interval on which lightning::ln::peer_handler::PeerManager::timer_tick_occurred should be called has been reduced to once every five seconds (#1035) and lightning::ln::channelmanager::ChannelManager::timer_tick_occurred should now be called on startup in addition to once per minute (#985).
  • The rust-bitcoin and bech32 dependencies have been updated to their respective latest versions (0.27 and 0.8, #1012).

Bug Fixes

  • Fix panic when reading invoices generated by some versions of c-lightning (#1002 and #1003).
  • Fix panic when attempting to validate a signed message of incorrect length (#1010).
  • Do not ignore the route hints in invoices when the invoice is over 250k sats (#986).
  • Fees are automatically updated on outbound channels to ensure commitment transactions are always broadcastable (#985).
  • Fixes a rare case where a lightning::util::events::Event::SpendableOutputs event is not generated after a counterparty commitment transaction is confirmed in a reorg when a conflicting local commitment transaction is removed in the same reorg (#1022).
  • Fixes a remotely-triggerable force-closure of an origin channel after an HTLC was forwarded over a next-hop channel and the next-hop channel was force-closed by our counterparty (#1025).
  • Fixes a rare force-closure case when sending a payment as a channel fundee when overdrawing our remaining balance. Instead the send will fail (#998).
  • Fixes a rare force-closure case when a payment was claimed prior to a peer disconnection or restart, and later failed (#977).

Serialization Compatibility

  • Pending inbound keysend payments which have neither been failed nor claimed when serialized will result in a ChannelManager which is not readable on pre-0.0.100 clients (#967).
  • Because lightning::chain::keysinterface::KeysInterface::get_shutdown_scriptpubkey has been updated to return a script instead of only a PublicKey, ChannelManagers constructed with custom KeysInterface implementations on 0.0.100 and later versions will not be readable on previous versions. ChannelManagers created with 0.0.99 and prior versions will remain readable even after the a serialization roundtrip on 0.0.100, as long as no new channels are opened. Further, users using a lightning::chain::keysinterface::KeysManager as their KeysInterface will have ChannelManagers which are readable on prior versions as well (#1019).
  • ChannelMonitorUpdates created by 0.0.100 and later for channels when lightning::util::config::ChannelConfig::commit_upfront_shutdown_pubkey is not set may not be readable by versions prior to 0.0.100 (#1019).
  • HTLCs which were in the process of being claimed on-chain when a pre-0.0.100 ChannelMonitor was serialized may generate PaymentForwarded events with spurious fee_earned_msat values. This only applies to payments which were unresolved at the time of the upgrade (#1004).
  • 0.0.100 clients with pending Event::PaymentForwarded events at serialization-time will generate serialized ChannelManager objects which 0.0.99 and earlier clients cannot read. The likelihood of this can be reduced by ensuring you process all pending events immediately before serialization (as is done by the lightning-background-processor crate, #1004).

In total, this release features 59 files changed, 5861 insertions, and 2082 deletions in 95 commits from 6 authors.

v0.0.99 - Jul 09, 2021

API Updates

  • lightning_block_sync::poll::Validate is now public, allowing you to implement the lightning_block_sync::poll::Poll trait without lightning_block_sync::poll::ChainPoller (#956).
  • lightning::ln::peer_handler::PeerManager no longer requires that no calls are made to referencing the same SocketDescriptor after disconnect_socket returns. This makes the API significantly less deadlock-prone and simplifies SocketDescriptor implementations significantly. The relevant changes have been made to lightning_net_tokio and PeerManager documentation has been substantially rewritten (#957).
  • lightning::util::message_signing's sign and verify methods now take secret and public keys by reference instead of value (#974).
  • Substantially more information is now exposed about channels in ChannelDetails. See documentation for more info (#984 and #988).
  • The latest best block seen is now exposed in ChannelManager::current_best_block and ChannelMonitor::current_best_block (#984).
  • Feerates charged when forwarding payments over channels is now set in ChannelConfig::fee_base_msat when the channel is opened. For existing channels, the value is set to the value provided in ChannelManagerReadArgs::default_config::channel_options the first time the ChannelManager is loaded in 0.0.99 (#975).
  • We now reject HTLCs which are received to be forwarded over private channels unless UserConfig::accept_forwards_to_priv_channels is set. Note that UserConfig is never serialized and must be provided via ChannelManagerReadArgs::default_config at each start (#975).

Bug Fixes

  • We now forward gossip messages to peers instead of only relaying locally-generated gossip or sending gossip messages during initial sync (#948).
  • Correctly send channel_update messages to direct peers on private channels (#949). Without this, a private node connected to an LDK node over a private channel cannot receive funds as it does not know which fees the LDK node will charge.
  • lightning::ln::channelmanager::ChannelManager no longer expects to be persisted spuriously after we receive a channel_update message about any channel in the routing gossip (#972).
  • Asynchronous ChannelMonitor updates (using the ChannelMonitorUpdateErr::TemporaryFailure return variant) no longer cause spurious HTLC forwarding failures (#954).
  • Transaction provided via ChannelMonitor::transactions_confirmed after ChannelMonitor::best_block_updated was called for a much later block now trigger all relevant actions as of the later block. Previously some transaction broadcasts or other responses required an additional block be provided via ChannelMonitor::best_block_updated (#970).
  • We no longer panic in rare cases when an invoice contained last-hop route hints which were unusable (#958).

Node Compatibility

  • We now accept spurious funding_locked messages sent prior to channel_reestablish messages after reconnect. This is a known, long-standing bug in lnd (#966).
  • We now set the first_blocknum and number_of_blocks fields in reply_channel_range messages to values which c-lightning versions prior to 0.10 accepted. This avoids spurious force-closes from such nodes (#961).

Serialization Compatibility

  • Due to a bug discovered in 0.0.98, if a ChannelManager is serialized on version 0.0.98 while an Event::PaymentSent is pending processing, the ChannelManager will fail to deserialize both on version 0.0.98 and later versions. If you have such a ChannelManager available, a simple patch will allow it to deserialize. Please file an issue if you need assistance (#973).

v0.0.98 - Jun 20, 2021

0.0.98 should be considered a release candidate to the first alpha release of Rust-Lightning and the broader LDK. It represents several years of work designing and fine-tuning a flexible API for integrating lightning into any application. LDK should make it easy to build a lightning node or client which meets specific requirements that other lightning node software cannot. As lightning continues to evolve, and new use-cases for lightning develop, the API of LDK will continue to change and expand. However, starting with version 0.1, objects serialized with prior versions will be readable with the latest LDK. While Rust-Lightning is approaching the 0.1 milestone, language bindings components of LDK available at https://github.com/lightningdevkit are still of varying quality. Some are also approaching an 0.1 release, while others are still much more experimental. Please note that, at 0.0.98, using Rust-Lightning on mainnet is strongly discouraged.

v0.0.14 - May 04, 2021

v0.0.13 - Mar 22, 2021

v0.0.12 - Nov 24, 2020

  • Large API overhauls to make several major API surfaces much easier to work.
  • C/C++ Bindings are a part of this release, however are still considered beta status. They are expected to mature as more language-specific binding work is built on top.

Information - Updated May 19, 2022

Stars: 727
Forks: 186
Issues: 203

PumpkinDB a Rust based key value store

A immutable key-value db engine written in Rust and easily extended

PumpkinDB a Rust based key value store

krill-kmip-ttlv - A library for (de)serializing KMIP TTLV

The OASIS Key Management Interoperability Protocol specifications which define message formats for the manipulation of cryptographic material on a key management server

krill-kmip-ttlv - A library for (de)serializing KMIP TTLV

gpg-tui is a Terminal User Interface for GnuPG

Terminal User Interface for gpg provides but it tries to bring a more interactive approach to key management

gpg-tui is a Terminal User Interface for GnuPG

kmip-ttlv - A library for (de)serializing KMIP TTLV

The OASIS Key Management Interoperability Protocol specifications which define message formats for the manipulation of cryptographic material on a key management server

kmip-ttlv - A library for (de)serializing KMIP TTLV

Dorea is a key-value data storage system

Very simple and easy to use key-value system for use in your next application

Dorea is a key-value data storage system

A toy key-value storage for DB study

Use snappy compression algorithm

A toy key-value storage for DB study

Remove duplicate key member lines from sorted NDJSON

Remove duplicate key member lines from sorted jlsort)

Remove duplicate key member lines from sorted NDJSON

Key-Value Server

s somekeyname Some string as a value I guess

Key-Value Server

Key Management System (KMS) for curve Secp256k1

=====================================

Key Management System (KMS) for curve Secp256k1

RSA-key-encryption

A simple RSA cryptography library

RSA-key-encryption

Keyhouse is a skeleton of general-purpose Key Management System

Keyhouse is not an off-the-shelf system, and it's not ready for production

Keyhouse is a skeleton of general-purpose Key Management System

Shappy key-value database created by Rust

Copyright (©) 2021 Sh-Zh-7

Shappy key-value database created by Rust
Facebook Instagram Twitter GitHub Dribbble
Privacy