rust-lightning
8 months agoMerge pull request #2549 from yanganto/socket-addr
Matt Corallo [Fri, 8 Sep 2023 20:08:08 +0000 (20:08 +0000)]
Merge pull request #2549 from yanganto/socket-addr

Rename SocketAddress from NetAddress

8 months agoStruct-ify blinded payment path intermediate node info
Valentine Wallace [Thu, 7 Sep 2023 21:11:30 +0000 (17:11 -0400)]
Struct-ify blinded payment path intermediate node info

8 months agoRemove unnecessary doc links
Valentine Wallace [Thu, 7 Sep 2023 21:10:21 +0000 (17:10 -0400)]
Remove unnecessary doc links

8 months agoFix blinded payment TLV ser to not length-prefix
Valentine Wallace [Thu, 24 Aug 2023 19:23:06 +0000 (15:23 -0400)]
Fix blinded payment TLV ser to not length-prefix

impl_writeable_tlv_based includes a length prefix to the TLV stream, which we
don't want.

8 months agoDerive Clone and Debug for blinded payment TLV structs
Valentine Wallace [Thu, 7 Sep 2023 17:32:51 +0000 (13:32 -0400)]
Derive Clone and Debug for blinded payment TLV structs

8 months agoMake blinded payment TLV fields public.
Valentine Wallace [Thu, 7 Sep 2023 17:32:30 +0000 (13:32 -0400)]
Make blinded payment TLV fields public.

These should've been made public when they were added for use in
BlindedPath::new_for_payment.

8 months agoSupport aggregating htlc_maximum_msat for BlindedPayInfo
Valentine Wallace [Sat, 19 Aug 2023 21:56:33 +0000 (17:56 -0400)]
Support aggregating htlc_maximum_msat for BlindedPayInfo

8 months agoSupport aggregating htlc_minimum_msat for BlindedPayInfo
Valentine Wallace [Sat, 19 Aug 2023 21:12:15 +0000 (17:12 -0400)]
Support aggregating htlc_minimum_msat for BlindedPayInfo

8 months agoCompute aggregated BlindedPayInfo in path construction
Valentine Wallace [Wed, 21 Jun 2023 00:27:57 +0000 (20:27 -0400)]
Compute aggregated BlindedPayInfo in path construction

8 months agoRename SocketAddress from NetAddress
Antonio Yang [Mon, 4 Sep 2023 02:34:21 +0000 (10:34 +0800)]
Rename SocketAddress from NetAddress

8 months agoIntroduce `ChannelPhase` enum
Duncan Dean [Wed, 26 Jul 2023 09:20:08 +0000 (11:20 +0200)]
Introduce `ChannelPhase` enum

We introduce the `ChannelPhase` enum which will contain the different
channel structs wrapped by each of its variants so that we can place
these within a single `channel_by_id` map in `peer_state` in the
following commits. This will reduce the number of map lookup operations
we need to do in `ChannelManager`'s various methods. It will also make
certain channel counting logic easier to reason about with less risk of
forgetting to modify logic when new channels structs are introduced for
V2 channel establishment.

8 months agoRemove outdated `Channel` TODO
Duncan Dean [Thu, 24 Aug 2023 13:32:05 +0000 (15:32 +0200)]
Remove outdated `Channel` TODO

8 months agoHave `path_to_windows_str` take reference to avoid `clone`s
Elias Rohrer [Fri, 8 Sep 2023 07:40:45 +0000 (09:40 +0200)]
Have `path_to_windows_str` take reference to avoid `clone`s

8 months agoMerge pull request #2371 from jkczyz/2023-06-offer-message-handling
Matt Corallo [Fri, 8 Sep 2023 01:52:38 +0000 (01:52 +0000)]
Merge pull request #2371 from jkczyz/2023-06-offer-message-handling

BOLT 12 Invoice payments

8 months agoMerge pull request #2472 from tnull/2023-08-add-kvstore
Matt Corallo [Thu, 7 Sep 2023 22:26:03 +0000 (22:26 +0000)]
Merge pull request #2472 from tnull/2023-08-add-kvstore

Replace `KVStorePersister` with `KVStore`

8 months agoConfigure BOLT 12 invoice payment retry strategy
Jeffrey Czyz [Tue, 5 Sep 2023 19:32:53 +0000 (14:32 -0500)]
Configure BOLT 12 invoice payment retry strategy

Replace a constant three retry attempts for BOLT 12 invoice payments
with a retry strategy specified when creating a pending outbound
payment. This is configured by users in a later commit when constructing
an InvoiceRequest or a Refund.

8 months agoUse u32 instead of usize in Retry::Attempts
Jeffrey Czyz [Tue, 5 Sep 2023 20:21:35 +0000 (15:21 -0500)]
Use u32 instead of usize in Retry::Attempts

An upcoming commit requires serializing Retry, so use a type with a
fixed byte length. Otherwise, using eight bytes to serialize a usize
would fail to read on 32-bit machines.

8 months agoAdd tests for send_payment_for_bolt12_invoice
Jeffrey Czyz [Thu, 31 Aug 2023 22:22:31 +0000 (17:22 -0500)]
Add tests for send_payment_for_bolt12_invoice

8 months agopub(crate) visibility for offers/test_utils.rs
Jeffrey Czyz [Thu, 31 Aug 2023 22:19:29 +0000 (17:19 -0500)]
pub(crate) visibility for offers/test_utils.rs

The test utilities for Offers are needed for testing message handling in
ChannelManager and OutboundPayments.

8 months agoSupport paying BOLT 12 invoices
Jeffrey Czyz [Fri, 1 Sep 2023 19:01:24 +0000 (14:01 -0500)]
Support paying BOLT 12 invoices

Add a send_payment_for_bolt12_invoice method to OutboundPayments for
initiating payment of a BOLT 12 invoice. This will be called from an
OffersMessageHandler, after which any retries are handled using the
Retryable logic.

8 months agoRename OutboundPayments::retry_payment_internal
Jeffrey Czyz [Wed, 6 Sep 2023 19:31:57 +0000 (14:31 -0500)]
Rename OutboundPayments::retry_payment_internal

It will be used for initial attempts at paying BOLT 12 invoices, so
rename it something that covers both that and retries.

8 months agoRefactor OutboundPayments::retry_payment_internal
Jeffrey Czyz [Fri, 1 Sep 2023 21:23:27 +0000 (16:23 -0500)]
Refactor OutboundPayments::retry_payment_internal

Consolidate the creation and insertion of onion_session_privs to the
PendingOutboundPayment::Retryable arm. In an upcoming commit, this
method will be reused for an initial BOLT 12 invoice payment. However,
onion_session_privs are created using a different helper.

8 months agoAdd PendingOutboundPayment::InvoiceReceived
Jeffrey Czyz [Wed, 30 Aug 2023 17:01:15 +0000 (12:01 -0500)]
Add PendingOutboundPayment::InvoiceReceived

When a BOLT 12 invoice has been received, a payment attempt is made and
any errors result in abandoning the PendingOutboundPayment. This results
in generating at PaymentFailed event, which has a PaymentHash. Thus,
when receiving an invoice, transition from AwaitingInvoice to a new
InvoiceReceived state, the latter of which contains a PaymentHash such
the abandon_payment helper can still be used.

8 months agoTest removing abandoned AwaitingInvoice payments
Jeffrey Czyz [Wed, 6 Sep 2023 20:17:01 +0000 (15:17 -0500)]
Test removing abandoned AwaitingInvoice payments

8 months agoTest for removing stale AwaitingInvoice payments
Jeffrey Czyz [Thu, 31 Aug 2023 01:22:18 +0000 (20:22 -0500)]
Test for removing stale AwaitingInvoice payments

8 months agoMove IDEMPOTENCY_TIMEOUT_TICKS to where it is used
Jeffrey Czyz [Wed, 6 Sep 2023 18:56:46 +0000 (13:56 -0500)]
Move IDEMPOTENCY_TIMEOUT_TICKS to where it is used

8 months agoAdd PendingOutboundPayment::AwaitingInvoice
Jeffrey Czyz [Mon, 17 Jul 2023 21:55:22 +0000 (16:55 -0500)]
Add PendingOutboundPayment::AwaitingInvoice

When a BOLT 12 invoice has been requested, in order to guarantee
invoice payment idempotency the future payment needs to be tracked. Add
an AwaitingInvoice variant to PendingOutboundPayment such that only
requested invoices are paid only once. Timeout after a few timer ticks
if a request has not been received.

8 months agoAdd `TestStore` implementation of `KVStore`
Elias Rohrer [Mon, 21 Aug 2023 14:17:35 +0000 (16:17 +0200)]
Add `TestStore` implementation of `KVStore`

8 months agoAdd benchmarking for `FilesystemStore`
Elias Rohrer [Fri, 4 Aug 2023 09:34:45 +0000 (11:34 +0200)]
Add benchmarking for `FilesystemStore`

We re-add benchmarking for `FilesystemStore` now that we switched over
to it.

8 months agoMigrate `FilesystemPersister` tests to `FilesystemStore`
Elias Rohrer [Fri, 4 Aug 2023 14:27:39 +0000 (16:27 +0200)]
Migrate `FilesystemPersister` tests to `FilesystemStore`

8 months agoMigrate to `KVStore`/`FilesystemStore`
Elias Rohrer [Tue, 1 Aug 2023 11:37:46 +0000 (13:37 +0200)]
Migrate to `KVStore`/`FilesystemStore`

Firstly, we switch our BP over to use `FilesystemStore`, which also gives us test
coverage and ensures the compatibility.

Then, we remove the superseded `KVStorePersister` trait and
the `FilesystemPersister` code.

8 months agoAdd `read_channel_monitors` utility
Elias Rohrer [Fri, 4 Aug 2023 14:20:50 +0000 (16:20 +0200)]
Add `read_channel_monitors` utility

This replaces the `FilesystemPersister::read_channelmonitors` method, as
we can now implement a single utility for all `KVStore`s.

8 months agoAdd `FilesystemStore`
Elias Rohrer [Tue, 1 Aug 2023 08:46:51 +0000 (10:46 +0200)]
Add `FilesystemStore`

We upstream the `FilesystemStore` implementation, which is backwards
compatible with `lightning-persister::FilesystemPersister`.

8 months agoAdd `test_utils`
Elias Rohrer [Fri, 4 Aug 2023 08:09:55 +0000 (10:09 +0200)]
Add `test_utils`

We add a utility function needed by upcoming `KVStore` implementation
tests.

8 months agoUpdate `lightning-persister` crate
Elias Rohrer [Mon, 21 Aug 2023 11:13:56 +0000 (13:13 +0200)]
Update `lightning-persister` crate

8 months agoAdd `KVStore` interface trait
Elias Rohrer [Tue, 1 Aug 2023 08:24:53 +0000 (10:24 +0200)]
Add `KVStore` interface trait

We upstream the `KVStore` interface trait from LDK Node, which will
replace `KVStorePersister` in the coming commits.

Besides persistence, `KVStore` implementations will also offer to `list`
keys present in a given `namespace` and `read` the stored values.

8 months agoAdd an InvoiceRequestFailed event
Jeffrey Czyz [Tue, 18 Jul 2023 19:41:50 +0000 (14:41 -0500)]
Add an InvoiceRequestFailed event

When an invoice is requested but either receives an error or never
receives a response, surface an event to indicate to the user that the
corresponding future payment has failed.

8 months agoMerge pull request #2134 from jbesraa/add_fromstr_to_netaddress
Elias Rohrer [Thu, 7 Sep 2023 12:16:25 +0000 (14:16 +0200)]
Merge pull request #2134 from jbesraa/add_fromstr_to_netaddress

implement fromstr trait to netaddress

8 months agoMerge pull request #2557 from waterson/stale-monitor-force-close-debugging
Matt Corallo [Thu, 7 Sep 2023 04:10:59 +0000 (04:10 +0000)]
Merge pull request #2557 from waterson/stale-monitor-force-close-debugging

Log each condition that was violated for a stale monitor

8 months agoFix misspelling, s/and/but/.
Chris Waterson [Wed, 6 Sep 2023 22:10:51 +0000 (15:10 -0700)]
Fix misspelling, s/and/but/.

8 months agoMerge pull request #2555 from tnull/2023-08-have-get-route-take-params
Matt Corallo [Wed, 6 Sep 2023 20:22:08 +0000 (20:22 +0000)]
Merge pull request #2555 from tnull/2023-08-have-get-route-take-params

Have `get_route` and `Route` take `RouteParameters`

8 months agoImprove `Route::get_total_amount` docs
Elias Rohrer [Wed, 6 Sep 2023 09:26:13 +0000 (11:26 +0200)]
Improve `Route::get_total_amount` docs

8 months agoInclude overpaid value in `Route::get_total_fees`
Elias Rohrer [Wed, 6 Sep 2023 09:17:51 +0000 (11:17 +0200)]
Include overpaid value in `Route::get_total_fees`

8 months agoHave `Route` hold `RouteParameters`
Elias Rohrer [Thu, 31 Aug 2023 13:10:09 +0000 (15:10 +0200)]
Have `Route` hold `RouteParameters`

8 months agoHave `get_route` take `RouteParameters`
Elias Rohrer [Thu, 31 Aug 2023 10:25:38 +0000 (12:25 +0200)]
Have `get_route` take `RouteParameters`

8 months agoImplement `from_str` trait for `NetAddress`
jbesraa [Fri, 25 Aug 2023 02:57:56 +0000 (05:57 +0300)]
Implement `from_str` trait for `NetAddress`

    - Add fuzz test for `NetAddress` `from_str` function

8 months agoLog each condition that was violated for a stale monitor
Chris Waterson [Wed, 6 Sep 2023 15:38:08 +0000 (08:38 -0700)]
Log each condition that was violated for a stale monitor

There are several conditions that can be violated which indicate a stale
monitor. This logs each that doesn't hold.

8 months agoFix indent in `message_signing.rs`
jbesraa [Mon, 4 Sep 2023 15:29:19 +0000 (18:29 +0300)]
Fix indent in `message_signing.rs`

8 months agoMove `zbase32` implementation to `base32` file
jbesraa [Fri, 25 Aug 2023 02:57:06 +0000 (05:57 +0300)]
Move `zbase32` implementation to `base32` file

8 months agoAdd RFC4648 base32 `encode` and `decode` functions
jbesraa [Fri, 25 Aug 2023 02:52:38 +0000 (05:52 +0300)]
Add RFC4648 base32 `encode` and `decode` functions

8 months agoMerge pull request #1924 from benthecarman/handle-coinbase-funding-channel
Matt Corallo [Tue, 5 Sep 2023 16:56:35 +0000 (16:56 +0000)]
Merge pull request #1924 from benthecarman/handle-coinbase-funding-channel

Handle if funding output is in a coinbase transaction

8 months agoMerge pull request #2548 from JosephGoulden/chain-monitor-logging
Elias Rohrer [Tue, 5 Sep 2023 07:19:27 +0000 (09:19 +0200)]
Merge pull request #2548 from JosephGoulden/chain-monitor-logging

Downgrade log message "Channel Monitor sync is still in progress" from info to debug

8 months agoAdd test for coinbase funding transactions
Duncan Dean [Fri, 1 Sep 2023 09:13:57 +0000 (11:13 +0200)]
Add test for coinbase funding transactions

8 months agoSupport creating coinbase funding transactions in tests
benthecarman [Wed, 19 Apr 2023 08:53:54 +0000 (03:53 -0500)]
Support creating coinbase funding transactions in tests

8 months agoHandle if funding output is in a coinbase transaction
benthecarman [Mon, 19 Dec 2022 08:02:22 +0000 (02:02 -0600)]
Handle if funding output is in a coinbase transaction

8 months agoDowngrade log message regarding Channel Monitor sync still being in progress from...
Joseph Goulden [Sun, 3 Sep 2023 11:48:56 +0000 (12:48 +0100)]
Downgrade log message regarding Channel Monitor sync still being in progress from info to debug

8 months agoMerge pull request #2522 from sr-gi/202308-set-feature
Matt Corallo [Fri, 1 Sep 2023 18:25:53 +0000 (18:25 +0000)]
Merge pull request #2522 from sr-gi/202308-set-feature

Adds a set of convenience methods to set non-custom features

8 months agoMerge pull request #2543 from optout21/txabort-typo
Matt Corallo [Fri, 1 Sep 2023 15:13:54 +0000 (15:13 +0000)]
Merge pull request #2543 from optout21/txabort-typo

[minor] Fix wrong msg type in SendTxAbort, typo

8 months agoAdds a set of convenience methods to set non-custom features
Sergi Delgado Segura [Fri, 25 Aug 2023 08:30:07 +0000 (10:30 +0200)]
Adds a set of convenience methods to set non-custom features

Currently only custom features can be set by specifying the feature bit. Add also the
ability to do so for regular features.

8 months agoAdd constructor to `RouteParameters`
Elias Rohrer [Thu, 31 Aug 2023 14:04:33 +0000 (16:04 +0200)]
Add constructor to `RouteParameters`

8 months agoFix wrong msg type in SendTxAbort, typo
optout [Thu, 31 Aug 2023 22:00:13 +0000 (00:00 +0200)]
Fix wrong msg type in SendTxAbort, typo

8 months agoMerge pull request #2468 from jkczyz/2023-08-offer-payment-id
Matt Corallo [Tue, 29 Aug 2023 19:29:21 +0000 (19:29 +0000)]
Merge pull request #2468 from jkczyz/2023-08-offer-payment-id

Offer outbound payments

8 months agoInclude PaymentId in payer metadata
Jeffrey Czyz [Fri, 21 Jul 2023 20:28:36 +0000 (15:28 -0500)]
Include PaymentId in payer metadata

When receiving a BOLT 12 invoice originating from either an invoice
request or a refund, the invoice should only be paid once. To accomplish
this, require that the invoice includes an encrypted payment id in the
payer metadata. This allows ChannelManager to track a payment when
requesting but prior to receiving the invoice. Thus, it can determine if
the invoice has already been paid.

8 months agoAdd a ChaCha20 utility for encrypting in place
Jeffrey Czyz [Thu, 24 Aug 2023 21:43:39 +0000 (16:43 -0500)]
Add a ChaCha20 utility for encrypting in place

Similar to ChaCha20::encrypt_single_block only encrypts in-place.

8 months agoAdd a ChaCha20 utility for encrypting a block
Jeffrey Czyz [Thu, 24 Aug 2023 21:31:16 +0000 (16:31 -0500)]
Add a ChaCha20 utility for encrypting a block

This hides an encryption implementation detail from callers.

8 months agoAdd an encryption key to ExpandedKey for Offers
Jeffrey Czyz [Thu, 24 Aug 2023 20:16:53 +0000 (15:16 -0500)]
Add an encryption key to ExpandedKey for Offers

Metadata such as the PaymentId should be encrypted when included in an
InvoiceRequest or a Refund, as it is user data and is exposed to the
payment recipient. Add an encryption key to ExpandedKey for this purpose
instead of reusing offers_base_key.

8 months agoRemove unnecessary #[allow(unused)]
Jeffrey Czyz [Thu, 20 Jul 2023 19:50:02 +0000 (14:50 -0500)]
Remove unnecessary #[allow(unused)]

8 months agoSplit InvoiceRequest::verify_and_respond_using_derived_keys
Jeffrey Czyz [Thu, 15 Jun 2023 22:13:55 +0000 (17:13 -0500)]
Split InvoiceRequest::verify_and_respond_using_derived_keys

InvoiceRequest::verify_and_respond_using_derived_keys takes a payment
hash. To avoid generating one for invoice requests that ultimately
cannot be verified, split the method into one for verifying and another
for responding.

8 months agoMerge pull request #2531 from optout21/channel-id-breaking
Elias Rohrer [Tue, 29 Aug 2023 11:30:46 +0000 (13:30 +0200)]
Merge pull request #2531 from optout21/channel-id-breaking

Add ChannelId pending change note; formatting

8 months agoMerge pull request #2537 from TheBlueMatt/2023-08-one-less-feature-dep
Matt Corallo [Tue, 29 Aug 2023 00:20:44 +0000 (00:20 +0000)]
Merge pull request #2537 from TheBlueMatt/2023-08-one-less-feature-dep

Drop dep `tokio`'s `io-util` feat as it broke MSRV and isn't useful

8 months agoPin the `memchr` test-only transitive dependency of `regex` 2023-08-one-less-feature-dep
Matt Corallo [Mon, 28 Aug 2023 19:45:37 +0000 (19:45 +0000)]
Pin the `memchr` test-only transitive dependency of `regex`

8 months agoDrop dep `tokio`'s `io-util` feat as it broke MSRV and isn't useful
Matt Corallo [Mon, 28 Aug 2023 18:39:04 +0000 (18:39 +0000)]
Drop dep `tokio`'s `io-util` feat as it broke MSRV and isn't useful

We use `tokio`'s `io-util` feature to provide the
`Async{Read,Write}Ext` traits, which allow us to simply launch a
read future or `poll_write` directly as well as `split` the
`TcpStream` into a read/write half. However, these traits aren't
actually doing much for us - they are really just wrapping the
`readable` future (which we can trivially use ourselves) and
`poll_write` isn't doing anything for us that `poll_write_ready`
can't.

Similarly, the split logic is actually just `Arc`ing the
`TcpStream` and busy-waiting when an operation is busy to prevent
concurrent reads/writes. However, there's no reason to prevent
concurrent access at the stream level - we aren't ever concurrently
writing or reading (though we may concurrently read and write,
which is fine).

Worse, the `io-util` feature broke MSRV (though they're likely to
fix this upstream) and carries two additional dependencies (only
one on the latest upstream tokio).

Thus, we simply drop the dependency here.

Fixes #2527.

8 months agoMerge pull request #2536 from waterson/test-channel-signer
Matt Corallo [Mon, 28 Aug 2023 20:43:08 +0000 (20:43 +0000)]
Merge pull request #2536 from waterson/test-channel-signer

Rename EnforcingSigner to TestChannelSigner

8 months agoFix BlindedPath::new_for_payment docs
Valentine Wallace [Mon, 28 Aug 2023 17:20:11 +0000 (13:20 -0400)]
Fix BlindedPath::new_for_payment docs

8 months agoMerge pull request #2528 from arik-so/arik/2023-08-2470-shorter-term-monitor-locks
Matt Corallo [Mon, 28 Aug 2023 17:07:03 +0000 (17:07 +0000)]
Merge pull request #2528 from arik-so/arik/2023-08-2470-shorter-term-monitor-locks

Release monitor write lock in between update iterations

8 months agoRename EnforcingSigner to TestChannelSigner
Chris Waterson [Mon, 28 Aug 2023 16:37:33 +0000 (09:37 -0700)]
Rename EnforcingSigner to TestChannelSigner

Since the advent of VLS, EnforcingSigner is only used now for testing.

8 months agoAdd ChannelId pending change note; formatting
optout [Mon, 28 Aug 2023 05:20:23 +0000 (07:20 +0200)]
Add ChannelId pending change note; formatting

8 months agoFix flaky aggregated HTLC revocation test.
Arik Sosman [Sat, 26 Aug 2023 00:34:10 +0000 (17:34 -0700)]
Fix flaky aggregated HTLC revocation test.

Releasing write locks in between monitor updates
requires storing a set of cloned keys to iterate
over. For efficiency purposes, that set of keys
is an actual set, as opposed to array, which means
that the iteration order may not be consistent.

The test was relying on an event array index to
access the revocation transaction. We change that
to accessing a hash map keyed by the txid, fixing
the test.

8 months agoRelease write lock between monitor update iterations.
Arik Sosman [Fri, 25 Aug 2023 19:31:33 +0000 (12:31 -0700)]
Release write lock between monitor update iterations.

Previously, updating block data on a chain monitor
would acquire a write lock on all of its associated
channel monitors and not release it until the loop
completed.

Now, we instead acquire it on each iteration,
fixing #2470.

8 months agoMerge pull request #2485 from optout21/channel-id-4struct1
Matt Corallo [Sun, 27 Aug 2023 05:18:00 +0000 (05:18 +0000)]
Merge pull request #2485 from optout21/channel-id-4struct1

Introduce new ChannelId struct

8 months agoUse new ChannelId type
optout [Fri, 25 Aug 2023 23:30:40 +0000 (01:30 +0200)]
Use new ChannelId type

8 months agoAdd new ChannelId type; unused
optout [Fri, 25 Aug 2023 23:15:46 +0000 (01:15 +0200)]
Add new ChannelId type; unused

8 months agoMerge pull request #2197 from jbesraa/feat/lockable_score_rw
Matt Corallo [Fri, 25 Aug 2023 18:48:49 +0000 (18:48 +0000)]
Merge pull request #2197 from jbesraa/feat/lockable_score_rw

add another lock to lockable_score

8 months agoMerge pull request #2248 from TheBlueMatt/2023-04-gossip-check
Elias Rohrer [Fri, 25 Aug 2023 12:10:39 +0000 (14:10 +0200)]
Merge pull request #2248 from TheBlueMatt/2023-04-gossip-check

Implement the UtxoSource interface for REST/RPC clients

8 months agoMerge pull request #2503 from valentinewallace/2023-08-fix-router-debug-panic
Elias Rohrer [Fri, 25 Aug 2023 10:46:37 +0000 (12:46 +0200)]
Merge pull request #2503 from valentinewallace/2023-08-fix-router-debug-panic

Fix debug panic in the case where a first hop has a channel with an introduction node

8 months agoMerge pull request #2466 from TheBlueMatt/2023-07-expose-success-prob
Elias Rohrer [Fri, 25 Aug 2023 10:40:36 +0000 (12:40 +0200)]
Merge pull request #2466 from TheBlueMatt/2023-07-expose-success-prob

Expose the historical success probability calculation itself

8 months agoSplit LockableScore responsibilities between read & write operations
jbesraa [Tue, 22 Aug 2023 15:57:06 +0000 (18:57 +0300)]
Split LockableScore responsibilities between read & write operations

    - Split Score from LockableScore to ScoreLookUp to handle read
      operations and ScoreUpdate to handle write operations
    - Change all struct that implemented Score to implement ScoreLookUp
      and/or ScoreUpdate
    - Change Mutex's to RwLocks to allow multiple data readers
    - Change LockableScore to Deref in ScorerAccountingForInFlightHtlcs
      as we only need to read
    - Add ScoreLookUp and ScoreUpdate docs
    - Remove reference(&'a) and Sized from Score in ScorerAccountingForInFlightHtlcs
      as Score implements Deref
    - Split MultiThreadedScoreLock into MultiThreadedScoreLockWrite and MultiThreadedScoreLockRead.
      After splitting LockableScore, we split MultiThreadedScoreLock following
      the same way, splitting a single score into two srtucts, one for read and
      other for write.
      MultiThreadedScoreLock is used in c_bindings.

8 months agoMerge pull request #2519 from Sharmalm/main
Matt Corallo [Thu, 24 Aug 2023 23:28:52 +0000 (23:28 +0000)]
Merge pull request #2519 from Sharmalm/main

Print contents of our own channel updates for broadcast in non-gossip logs

8 months agoRouter: account for blinded path fee, etc on first_hop<>intro hop add
Valentine Wallace [Tue, 15 Aug 2023 21:32:46 +0000 (17:32 -0400)]
Router: account for blinded path fee, etc on first_hop<>intro hop add

This previously led to a debug panic in the router because we wouldn't account
for the blinded path fee when calculating first_hop<>intro_node hop's available
liquidity and construct an invalid path that forwarded more over said hop than
was actually available.

This also led to us hitting unreachable code, see direct_to_matching_intro_nodes
test description.

8 months agoimproving message in log
Lalitmohansharma1 [Thu, 24 Aug 2023 14:51:08 +0000 (20:21 +0530)]
improving message in log

8 months agoFail UTXO lookups if the block doesn't have five confirmations 2023-04-gossip-check
Matt Corallo [Mon, 5 Jun 2023 17:22:36 +0000 (17:22 +0000)]
Fail UTXO lookups if the block doesn't have five confirmations

The BOLT spec mandates that channels not be announced until they
have at least six confirmations. This is important to enforce not
because we particularly care about any specific DoS concerns, but
because if we do not we may have to handle reorgs of channel
funding transactions which change their SCID or have conflicting
SCIDs.

8 months agoAdd a simple naive block cache in gossip sync lookups
Matt Corallo [Sun, 30 Apr 2023 02:06:19 +0000 (02:06 +0000)]
Add a simple naive block cache in gossip sync lookups

8 months agoMake the `P2PGossipSync` `UtxoLookup` exchangable without &mut self
Matt Corallo [Sun, 30 Apr 2023 00:48:57 +0000 (00:48 +0000)]
Make the `P2PGossipSync` `UtxoLookup` exchangable without &mut self

Because a `UtxoLookup` implementation is likely to need a reference
to the `PeerManager` which contains a reference to the
`P2PGossipSync`, it is likely to be impossible to get a mutable
reference to the `P2PGossipSync` by the time we want to add a
`UtxoLookup` without a ton of boilerplate and trait wrapping.

Instead, we simply place the `UtxoLookup` in a `RwLock`, allowing
us to modify it without a mutable self reference.

The lifetime bounds updates in tests required in this commit are
entirely unclear to me, but do allow tests to continue building, so
somehow make rustc happier.

8 months agoImplement the `UtxoSource` interface for REST/RPC clients
Matt Corallo [Sat, 29 Apr 2023 22:32:57 +0000 (22:32 +0000)]
Implement the `UtxoSource` interface for REST/RPC clients

In LDK, we expect users operating nodes on the public network to
implement the `UtxoSource` interface in order to validate the
gossip they receive from the network.

Sadly, because the DoS attack of flooding a node's gossip store
isn't a common issue, and because we do not provide an
implementation off-the-shelf to make doing so easily, many of our
downstream users do not have a `UtxoSource` implementation.

In order to change that, here we implement an async `UtxoSource`
in the `lightning-block-sync` crate, providing one for users who
sync the chain from Bitcoin Core's RPC or REST interfaces.

8 months agoMerge pull request #2515 from TheBlueMatt/2023-08-earlier-payment-hash-log
Matt Corallo [Wed, 23 Aug 2023 21:46:23 +0000 (21:46 +0000)]
Merge pull request #2515 from TheBlueMatt/2023-08-earlier-payment-hash-log

Include payment hash in more early payment logs

8 months agoStore a `HistoricalMinMaxBuckets` in `DirectedChannelLiquidity` 2023-07-expose-success-prob
Matt Corallo [Wed, 23 Aug 2023 00:46:18 +0000 (00:46 +0000)]
Store a `HistoricalMinMaxBuckets` in `DirectedChannelLiquidity`

This removes the need to reconstruct the struct in a number of
places by simply creating it up front.

8 months agoMove the bucketed history tracking logic into a scoring submodule
Matt Corallo [Sat, 20 May 2023 23:31:57 +0000 (23:31 +0000)]
Move the bucketed history tracking logic into a scoring submodule

8 months agoExpose the historical success probability calculation itself
Matt Corallo [Sat, 20 May 2023 23:28:18 +0000 (23:28 +0000)]
Expose the historical success probability calculation itself

In 3f32f60ae7e75a4be96d3d5adc8d18b53445e5e5 we exposed the
historical success probability buckets directly, with a long method
doc explaining how to use it. While this is great for logging
exactly what the internal model thinks, its also helpful to let
users know what the internal model thinks the success probability
is directly, allowing them to compare route success probabilities.

Here we do so but only for the historical tracking buckets.

8 months agoFind payment bucket in calculate_success_probability_times_billion
Matt Corallo [Sun, 9 Apr 2023 04:43:23 +0000 (04:43 +0000)]
Find payment bucket in calculate_success_probability_times_billion

This simply moves code which will simplify the next commit
somewhat.

8 months agoCorrectly apply penalty bounds on the per-amount penalties
Matt Corallo [Mon, 10 Apr 2023 22:54:48 +0000 (22:54 +0000)]
Correctly apply penalty bounds on the per-amount penalties

When we attempt to score a channel which has a success probability
very low, we may have a log well above our cut-off of two. For the
liquidity penalties this works great, we bound it by
`NEGATIVE_LOG10_UPPER_BOUND` and `min` the two scores. For the
amount liquidity penalty we didn't do any `min`ing at all.

This fix is to min the log itself first and then reuse the min'd
log in both calculations.